1620_Users_Group_Western_Region_196406 1620 Users Group Western Region 196406
1620_Users_Group_Western_Region_196406 1620_Users_Group_Western_Region_196406
User Manual: 1620_Users_Group_Western_Region_196406
Open the PDF directly: View PDF .
Page Count: 555
Download | |
Open PDF In Browser | View PDF |
1620 USERS GROUP WESTERN REGION MINUTES OF THE MEETING JUNE 17-19, 1964 DENVER, COLORADO ROBERT R. WHITE WESTERN REGION SECRETARY :_*_ _ ·+tt&.... t*nst_ttW*_*1IiIIIIIIIiIIM.......~~HIllliMtrt!tt_"=___......'""",'·b_MIIIiiiiiIooIIoI_ _ _"_bt"iIoIiIiH...,_ _ _ _ _ _ _ _ _ _ _ _ _ _ _..... ,,~,.Millld'Hlilillit _'MiIIIIIiIq"'Hsd"'b~t"_h*,*_tt_MlMilitt***_" 1' o CONTENTS 1. Roster of Attendees 2. Minutes of the Eighth Meeting of the Western Region 1620 Users Group 3. Sound-off Session 4. Agenda 5. A Least Squares Solution tor a Range Measuring Instrumentation System; Oliver Lee Kingsley and Burton L. Williams 6. Boundary Value Problems in Ordinary Differential Equations with Constant Coefficients; Riohard Rosanoft and Gordon Nab 7. a) b) 8. 9. o Reader; R. C. Steinbaoh A 519 Simulator; R. C. Steinbach Simultaneous Linear Equations with Complex Coefficients; H. Kuffel Applications of Numerical Filters in the Power Spectral Analysis of Stationary Time Series; Alexander A. J. Hoffman 10. IBM 1620 Assists Student Counselors at Junior College; 11. 1620 Computer Utilization in a Wind Tunnel Data Acquisition System; Stanley E. Wisniewski 12. 1620 1PL-V; Wendell T. Beyer 13. Petroleum Exploration and Production Application for the IBM 1620 and Plotter; Jack L. Morrison 14. A Control System Approach to Automatic Jet Engine Testing; Aubrey D. Wood 15. Generalized Filter Hetwork A/C Steady State Analysis Program; D. H. O'Herren 16. FORTRAN II - Debugging Techniques and Aids; Leon P. Goldberg 17 • FORTRAN II and the 1443; Lanny L. Hotfman 18. A Survey ot the Beginning Programming Paul S. Chan Cou~se; Clarence B. Germain 19. FORTRAN "Teach" Problems; Wendell L. Pope 1 .. ------.--,~~~~~~----.------~--~---- ._--- 20. A Load-and-Go SPS with Monitor Control; Kenneth M. Lochner and Glenn R. Ingram 21. Examples of 1620 Use in College Administration; Noel T. Smith 22. Automatic Processing of Autospot and Automap Programs with the 1620-1311 Disk System; Jack T. Dunn and Ernie G. Moore 23. General Ray Trace Program; D. H. O'Herren (presented at Tempe, Arizona, December 1963) . ...• ,- ".? - - " ....-.-.. ---~--- ....--...... ------ ... · . · ____.z.·n...S.eer____ HNJ. . .WI~ •••- -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ ~IH_NtWtMM"1~f"e_'t."*!_'i_illfI'_!tt:u_',,,_naM.aU--M· ~ 1,620· USERS GROUP WESTERN' REGION SUMMER MEET J NG .JUL. Y 1:7. 18. 1.9. 1·964 DlENYER.CfLOR,AOO o ROSTER OF ATT9C)EES 1032 MRS. BETTY CJLSICl( 1084 GEORGIA STATE COLL. ATLANTA ,GA. 1118 NANCY PAQUIN PATRICK 1118 PUeL l'e tEAL TH SERV. LANNY L. HOFFMAN GUGGENHEIM LABS 1177 1258 1.290 o 3016 RENe: SEUIGNY ..JR. HAYES INTERNATIONAL. HUNTSVILLE-ALA. 1238 FRANCES K. DURKAN 1273 S. HAMER ..JUDITH KOERNER ARGONNE NAT. L.AB. NEW YORK.N.Y. IDAHO FALLS • l'OAtiO ROBERT D. WEEMS N.C,. STATE CQL.LEGE RALE I,GH. N.C. 1352 THOMAS P. SODANO 3041 EUGENE C. EWING MEMPHIS. TENN. ARTHUR P. WOODS 4R. ARMCO STEEL CORP MIDOL..ETOWN. orn 0 3055 BARNEY T. WATSON VA HO:SPITAl.. OMAHA ,NEB. Me PHERSON.KAN. ..J. RICHARD BURROWS H.O.+R •• ENG! NEERS OMAHA. NE:a. WALKER R. HURD OP I ,BOARO OF EOUC. NAT cooP REFINERY AS 3082 PAUL A,. BlCIiCFORO . OU MED RES COMP CNTR OtlLLO. TEX. .LAS YEGAS.NEV. LOWELL A. RASMUSSEN ~AMAUSER CO. TACCMA ••ASH. 5096 BURTON L. WILL'lAMS 5;104 :w. CLAIII< ,CHASE JR. 5'117 C~GE WEteELL ·,·L. 'POPE UT AH·~T.ATE tIN IV. 5131 o A08I!A'HOfI'lI'MAN 5'130 A.A• .J ..NOP'FMAN TEXAS CMRISTlAN 'UNIV FORT WOR:fit.,TEX. 51.3:1 E.d,• .SVANS co. MAR·TIN: CO. MART!:" DEMVEA'::.COLO. D£NVER.·COLO. .....·.8·ISENIWS MARTIN 51:33 ca. .000000t CQL.:O. 5133 ASTRtlC 1E'1RIIIEHDJlAN BOULDER" COLO" L.08AN=.UTAH $131 DR. ~i.R • •U:t ... TEXAS. GOL1.... OF A•• KI:NGSVILL.EtTEX • NAT eTA .ATMOS RES "AMCGUYER • WASH>. !S,1a6 BOYD·c. ,NORA I S BUREAU OF· RE,C. SACRAMENTO, CAL. WHiJT£ SANDS.N.M. ELT. . DAVID ·IMER u.s.P'....s·• WHJ-TE SANDS ....R. • ,108 MARRY CASTLE -'R. PIEIN!EA'NATUAAL GAS AMAAlLLO.TEX. !S095 TED J. NCKE...A DENVERtCOLO. UNIV. Opr Te;XAS .086 GL;DMR •..INGRAM SUNl)STRAND AVIATION SUNDSTAAND AV·IATION DENWEReCOLO. 5060 JOHN A.;F£ALING MONTANA STATECGLL. BOZEIIfA... 1IlWJNT. z. 5058 PARK.ARIZ CLARE...-rT 'JIIEN' $ COLL. CLAREMUNT:t CAL. LIT'OPIELD PARK.ARIZ 5041 D . . . . OiNEAREN GOODYEAR AEROSPACE LEWIS:· • • •OMSLE MAS. . '+ . . tJAt«iiER .AIIARtLLO.·:FEtC. . 51';43 RALPH.D·.····,.,.J'NE MAsaI+_NGER AfllARl'WO • rEX • GERALD W. LeCKE TEXAS ·.T·EGM:. i::CGLLEGe: 5 ·LtJIIBOCk .·TEx • &&wuuuaaw=",aa:4a:::a:::uiS .!£I£AXi! I lill!;g !i.t. (.£44 MM. 41 $. i. BaaL. ; 1" .... ..iiDSEPN:·:A.· STRML ·U.S•. ·. . .T_R·BUREAU 5,I'E .aCflAMMrO.CAL. ••)~7.J"R8AO:SA"PF NDFtTH,··AM&RI,CttNAWI:,A. 5150 '_SO 1!UGDE"8AKEA MALL·t._TON CG·. 5;1:50 GEORGE V.COPLAND MALLr,tattRTGN CO'. DUNCaN,.'GKt.,A. IJUNCAN,.~OI UGHL t N FT RAYS t leasuring Squipment or s i 7TIp 1 y D>~~. An ins trune nta t ion Ry ste.:n cap.1;) 1e <)f gi vi ng Sue 1 i dean three s?ace position 2sti::1ates is termed as a :xr?tJ~'·!E system. ca 1 ;):'m/D>;'~ A typi- system usually consi sts of three non-coli near instrument or eq'.lip:11e'l1t si tes used to neasure range. The typical solution equations are the classical deterministic set that rejects th,c minor image solution. A four or more T)?-1E system presents a problem because a slisht error in any range ~ill not produce a set of classical approach. ho~ogeneous ~ea~urenent space position estimates by the There is a need for a good method to combine o 15 Leu iii.tila.iSaaE::: alii: hUM $Ii the set of overdetermined measurements into a single space paint estimate. The least squares method will provide the required space point estimate if the set of measurements are unhiased. The least squares equations developed ~inimize the sumS of squares of the error set of range measurements. The observational equation from "C:.]hich the error equati.on is derived is written: (1) R • ml = R. + E . 1 ml o where R • denotes the measured range from the i -th 'J:··18 to the ml tracked vehicle. denotes the true range frdm ~. 1 the i -.th DHZ to the tracked vehicle. B Tn i denotes the measurement error associ ated vli th th·e observa- tion taken from the i-th D>fE. The error equation is easily obtained from the ob~ervational equation, thus: (2) Eml. = :l . ml Ri The true value needs to be replaced by so~e suitable spproximation. Later, the true value td 11 be estimated by t1:'? fin:!l soilltion or 16 . - - - - - - - -....• --~ . . . . .- ...~.. ---.-~.--.- ...... - - -....... ~. . . o reduction equ.qtions. The assu:nption is made that the true value can be represented by the linear terms of a Taylor series about some nearby poi nt R , 'o1here: o R0 =0 ~ (X, Y , Z ) 000 (3) From the i-th DHE, the range to the point (X0 , 0 Y ,0 Z ) is written: (4a) R . /1/(X a 01 0 _ Xi)2 + (y _ 1.)2 + (z 0 1 Z )2 _ i 0 For any space position (X, Y, Z), the equation becomes: = (4b) Ri 1\, 'veX - Xi):! + .- (Y: Yi)~ -:r + (Z ... Zi) The linear Taylor series representation is written: =.~. +~1i I 01 2> X :to (5) 1 (x- X) 0 o +~RIH -d Y I (Z-~) o 0 0 where ~ TU~ x (.'j " = (X o (Y The 1 01 0 Ri/ Z 0 (zo - z0 ) /Po. 01• = '~ighted - Y. )/R i 010 = ;~ c;:: - X.) /R . error sums of squares for k measurements from k DME sites is written: (6a) k ~ i-1 ... 2 J!,mi k = 2i-1 2 10 17 I I ==_i,SMMweU;';:" ,mIMi. ,1.,".*=140£4",4; IUki. 4¥ to . ,14 $, . ' .. ; .. A4GUP o where wi is the weight given to each measurement. The errors sums of squares for equally weighted measurements is written: (6b) k k ~ i·1 E2. mt = ~i -~ :2i=l 2 Generally, the components of the instrumentation system are near enough alike in performance and behavior that equation (6b) is applicable. For a system that consists of heterogeneous distance measuring equipment (system components), each weight that would be inversely proportional to the equipments range variance would be appropriate. The error sums of squares are minimized ~vi th respect to the range measurement parameter which is a function of the three orthogonal components X, Y and Z. o The three resulting equations thus formed are A '" ~ called the normal equations from which the estimates X, Y and Z are obtained: (7a) (7b) ~x ~l d ~l ';;>Y (7c) ~ ~z ~l (R . ml -Ria (R . - mt (Rml. - =0 Ri~ = R.0 =0 1 0 o 18 II '",HlltP '9IWt"i"' \ t 7$ e »t htHtt :!!indtt t'dtritt*HWt:dYitf tl .. - '"¥tnar! j o The constant terms not involving (X-So), (Y-Y ) and (2-2 ) are placed o 0 on the rig~t (8a)(f,2 ~~:?. Ex hand side of the equation: -K. )(X 0 1 0 -x. )~X + (Y -1 )(X -X.) 61 + (.Z 1 1 0 0 1 .0 -z. 1 )(X -X. 0 1 )j)-;;~ J 01 ~ (')282 (Sb -. ~. K -- T,2 :\oi C> f'\. • if} + (y -Yo )(Y -y.) 61 + (Z -,~.)A X t.rz;.'''0 -X.)(.2 1 0 1 + (Y -Y.)( 7, -3.).6 Y + (2 -ZI )(Z -Zi 0 1 0 0 1 1 0 0 1 0 0 )A~1j<'. :J =c 2 =c 3 01 'ti.lere [ 1< C -Z~)(Y..-l.)6-;) 1 0 1 :J r(X 0 -:\i)(Y0 -Yo1 ).6X C" 1 =~ i=l (., . - -.. i) (X - 1\. " ') m1 • 0 k C2 =ii=l f .... 3 r~ro1 =.t i =1 /"\, L~ ~{ = ,"' .... l\ = /'0 ?; ) / d I.. 01. D J J .}(.7.0 - 2.1 ) /'l 01 01 -X 0 L\ Y = '" Y .. .., 1 .-R 01.}(y0 -Y.)/?. 1 01 ml .-;: k ,.. U' 0 - Y a 7 '''0 o 19 lOW _ saawZlaa:i1::aiii:aS::i4iQAi Itiil mal%1I.:.';. ;:;;AQP4ii. i.UXA :. - - ---- --.........-...-- ---- "-~-----------~--.--~~~.............. The equations can be written in coapact fora by matrix notation: - (9a) or (9b) SolYing tor A A - If.. equation c (9b): (10) The final solution tor X, Y, and Z can be written: o (11) The necessar,r start point (X.'Yo,Zo') can be obtained fram a deterministic solution for three range measureaents*. The region for convergent solutions has not been tully explored at this tiae. * Armijo, Larry, "Determination of Trajectories Using Range Data from Three Non-Co1inear Radar Stations", Technical Memorandua ,.66, USASMSA, Sept. 1960, WSMR,N.M. 0' 20 o III A NEr~OD FOR ESrrIAT.l!N-; IN~ENTA.TION SYSTErvl PRECISION The term precision estimate refers to the standard deviation estimates for the coordinate data X, Y and Z from the instrumentation system. If there exi sts a cormnon range measur.ement vari ance (II"~), then by use of the relative variance-covariance matrix, A -1 , from equation (10) esti- mates of the component variances can be obtained. The diagonal elements A-1 are used to estimate the component variance: from the 0 0 A22 0 (12) = /\2 .1 = 0 1 If no such set of C i exist the functions () w X X s- a> -0 s- O - ..c c a> > a> V') I a> C 0 a:: X a> a.. E 0 U CI a> ..c C en 0 0 0:::: ~ 0 Q. 0 ~ X X M . ~ ::> .2' u.. o - 12 - 32 1$ tit NniM"tt'" I IV. 0 BASIS FUNCTIONS AND BOUNDARY CONDITIONS 1 1 1 The solution of a homogenous fourth order ordinary differential equation' with constant coefficients whose characteristic polynomial has the roots ± a ± i (3 can be represented as a linear combination of the functions. te (a + i(3)x' . " e (a - i~)x (- Q' + i~)x (- a - iB)x\ , e" e (4. 1.) Of course, this is not the only set which could have been selected. Such a set is called a system or basis of functions if the member functions are linearly independent. A test for the linear independence is provided by the Wronskian determinant. This is a determinant whose first row is the system itself and whose jth row is made up of J-Ist derivative of the function in the corresponding column. We shall have a great deal more to say about linear independence in the next section of the paper. For the present, we wish to show how the programming and del?ugging of the problem are simplified by writing the derivative boundary conditions in terms of the Wronskian. Recall the form of the derivative boundary conditions. n - 1 i j = L= 1 2, n . J 1 h.. d -. T 1, J dx J at x = ~x) 0 = F. (4. 2) 1 and x c =L We see that . 1 d J - T (x) j - 1 (4.2a) dx is a column vector. We also see that recognition of the boundary conditions in the form (4. 2) permits us to deal separately 'with the specification of boundary conditions (bij) and the determination of the vector (4.2a). For our 11 th order problem the matrix (b i , j) was computed from a coded input. This provided a flexibility which waf? most useful when numerical difficulties were seen in the boundary conditions themselves. Such a scheme suggests the pos sibility of writing a generalized program. o - 13 - 33 Notice the vector (4.2a) is a function of X and exists in an N space. On the other hand, the matrix (b i , j) can be written as N linearly independent conditions at either X = 0 or, X = L. Thus~ for the two-point boundary value, problem N conditions must be selected from 2N derivative conditions and possibly some integral conditions. This draws attention to the fact the boundary conditions selected must be such as to specify a unique solution. We shall not give adequate coverage to this problem in this paper. Let us now examine the vector (4. 2a). It has a physical meaning without regard to the basis or coordinate system in which the solution is written. That is to say it could be written at various values of X as a table of numbers which would be independent of the manner in which it was obtained. To analyze it symbolically, however, we must assume a basis of functions, say {¢iJ. The first element of (4. 2a) (j = 1) is the zeroth derivative, or the function (4. 3) where the a are the constants of integration to be determined by the K equations 4. 2. But: . 1 d J - T (x) C" dx j - I n = 2: k = 1 d J. 1 - CP'k (x) j - I dx a W{x)a k k = (4.4) where W{x) is the matrix of functions from the Wronskian determinant. The recognition of these matrix products is the key to relieving the program of unmanageable detail. As will be seen later, a factor in the choice of the basis is the ease with which the Wronskian may be developed by a simple set of do loops. The (bi, j) matrix may be checked separately. Any linear independence may be displayed in easier-to-recognize form. The program has pattern. As we have indicated, the specification of boundary conditions which are sufficient and compatible involves more than we can discuss here. The reader is referred to Ref. 19· o - 14 - II "ill "'I v. SELECTION OF THE SYSTEM OR BASIS FUNC TIONS FOR THE SOL UT10N Given an ordinary differential equation whose characteristic polynomial contains the non-repeated roots ± a ± iB there are many choice s of functions for the solution. Consider, for example, the se four base s or systems: e {a + iB}x e {Q' - iB)x e {- a + iB}x e{- a - iB}x I (5. t) Cosh(ax} Cos(Bx}, Sinh{ax} Cos(6x}, Cosh(ax}Sin{Bx}, Sinh{ax} Sin{ Bx} I {5. ~} {5. 3} and: Z =L I - X fe -ax cos{Bx), e -ax, -az -az J sin(Bx}, e cos{Bz}, e sin{Bz} (5.4) The question arises, is there a choice? If these functions are mathematically equivalent, which they are, can one set be superior to another for digital programming? The answer is yes. Let us first dispose of (5. 1) on the arbitrary basis that we prefer not to.perform complex arithmetic if we can avoid it. The usual textbook treatment is to point out that the functions must be linearly independent. For the mathematician, this con,dition is me t for all of the bases under discussion. For the digital programmer, however, this situation may be quite different. If the domain L of the solution is large enough, all of the sets of functions expressed in the computer number set become linearly dependent: Lim aL . - co , I sinh{aL} = 1 cosh{aL} Lim -aL -aL. (({L) e cos(6L} = a, Lim e SIn tJ aL -co L _ co = Lim -aL ± iBL e aL - co = 0 (5. 5) o - 15 - 35 Ott> t o In the machine , this breakdown in linear independence becomes exact. That is to say, we may have coshaL = sinhaL to the last digit. Of course one may expect computational difficultie s long before the los s of the last tragic digit. Is there any way out of this dilemma? Again the answer is yes. The moment we realize that the finite number of digits in the calculation limits our ability to produce the "exact" solution we begin to consider analogies with approximate methods. We seek functions to represent our solution which "look like" the solution. Clearly basis (5.4) is greatly superior in this light. Only (5.4) of the bases considered, directly represents a function which arises at disturbances at the boundaries and is damped as it proceeds to the interior of the region. Comparing basis (5.4) with basis (5.2) it is seen that the functions of (5. 2) approach each other in exactly the range in which they become large. The functions of (5.4), on the other hand, approach each other in exac tly the range in which they drop out of the solution. It is not surprising that the basis (5.4) is superior when we take note of the fact that it contains more physical information. Only basis (5.4) is cognizant of the location of the disturbance caused by mismatching strains at X = L. o Let us inquire into the physical significance of positive real parts of characteristic polynomial roots. In the initial value problem, with time as the independent variable, positive real parts have been used as criteria of stability. When space is made the independent variable, and the problem is formulated as an initial value problem, a numerical instability is quickly associated with positive real parts. Consider for example, the differential equation (5. 6) which is the homogenous equation for the deflection of a railroad track. Ref. 15, the solution is shown in several form s including the me thod of initial conditions. In o - 16 - 36 t ',~', ", y(;x) = yo c osh(A. x) co's (A. ;x) e + 2~' fcoS h (>-";x) sin{>-..;x) + sinh{>-";x) cos{>-.. ~Q 2o s inh(>-";x) s in{>-..;x) 30 2>-" EI 4>-.. EI {c o X)} osh(>-";x) sin(>-,,;xl - sinh(>-";x) cos {>-.. ;xl I (5. 7) where Yo = displacement at e0 = ~ Q 0 0 slope at origin = bending = origin moment at origin shear ,at origin In fac t, the solution is damped as one in ove s away from a local load or disturbance. The method of initial conditions, however, requires that this damped function be composed of a linear combination of rapidly e;xpanding func tions. c In the finite arithmetic of the digital computer, this means that if I a fly settles on the railroad track in Denver and the initial conditions are determined to the full capacity of the computer number set some place this side of Los Angeles, the railroad tracks will be ripped and torn in the most terrible carnage since World War II. With time as the independent variable, the model is realis tic. Positive e;xponentials mean positive feedback. The rate of growth of the function is proportional to the function with a positive coefficient of proportionality. lilt takes money to make money. II "Population growth is e;xplosive in a favorable environment"-"chemical reactions become e;xplosive if the rate of the reaction increases with the reaction products" (including the final ene-rgy as a product). But time moves on. Functions do not have causes today and develop into effects yesterday. Space coordinate s, however, are quite arbitrary. One does not need to establish a coordinate system to pull a glued joint. Thus, the physical meaning of the positive e;xponential in our glue line problem is seen to be a stereotyped pattern of coordinate system specification. o - 17 - 37 .. - - - - - - - - - -..--- .~~~--~~- '$ o It is well known that the choice of a coordinate system can make the solution of a problem easier or more difficult (Ref. 21). In at least some problems it has great numerical consequences (Ref. 5). Interestingly enough, since the boundary conditions supply the information for the evaluation of the constants of integration, we see that they contain vital information about the origion of the coordinate system. One of the problem s in specifying boundary conditions is to make sure they produce a unique solution in a defined coordinate system. Stres s analysts have traditionally handled positive exponentials by testing such a parameter as a over the domain. If aL is large (say aL > 6), he uses basis (5.3) as a special b~sis. He sets the arbitrary constants associated with positive exponentials equal to zero (when aL> 6) and calls his solution the semi-infinite case. We think the choice of basis (5.4) is superior for several reasons. For one thing, the characteristic polynomial may contain more than one set of complex roots with positive real parts. One set may correspond to a semiinfinite case and another to a short case. This situation actually occurs in our 11 th order problem. Additionally, we should like to pick one basis and avoid programming more than one basis of functions. o If our mission in this paper were only to show that the basis (5.4) is superior for our problem, we could certainly bring this discussion to a close now. Our interest, however, is broader. We wish to explore the relative merits of bases in hopes that we may obtain criteria which help us in the solution of some other problem. We have identified part of our numerical difficulties as arlslng from a breakdown in linear independence due to the finite nature of the computer number set. Linear independence is established if there are non-zero values of the Wronskian and Gramian. For basis (5.3) or (5.4), we rediscovered an ancient device for writing the Nth derivative which will render our discussion easier. Consider one of the functions, say e -axcos «(3x) ~ dn-. ~= -ae = e -ax cos«(3x) -ax cos«(3x) - (3e o - 18 - -ax sin(f3x) (5.8) But ide ntify a + iB= R(cos(e) + c i sin(e) ) -ax R cos (Bx - e) (5. 9) It can be shown that: J -ax d e · cos(Bx} -ax j . ------ = e (-R) cos(Bx - je} j dx (5.10) And since Z = J -QZ d e cos(Bz) ---~-j dx j d e -QZ dx dz L - X sin(Bz) j d;Z = -1 = = e (5.11) -az j R sin(je + c Bz) In the seventh order problem we used the basis e -a z -a z -aZz 1 cos((\z), e 1 sin(B1z), e I 1 , d¢> = e -axR (-c os (e) cos (Bx) - sin( e) sin( B:x» df{ = - e l (5.1Z) o - 19 - 39 · t*t t The jth row of the Wronskian determinant for this basis is: o (5. 13) Consider this .Wronskian if L is large. When X = L, the first three columns of the determinant become very small. If X = 0, the last three become very small. It is shown in Ref. 22 that for an Nth order differential equation which does not contain the (N - 1st) derivative, the Wronskian is constant through the domain. To see that this is the fact for this Wronskian, it is helpful to factor the Wronskian into the three factors. = [R] [ W (X) ] o [T (X) ] [E (X ) ] (5. 14) Where [R 1 is the diagonal determinant whose elements are Rl j, [E(X)l is the diagonal determinant whose elements are the exponential terms -0:' e 1 -0:' Z e 1 Z and T (x) is the remaining determinant whose jth row may be written (5.15) then [R 1 '" R (l 1 + + 0:'1 -(0:'1 2 + + 3 +4 + 5 + z) 0:'2) (x [E(X) 1 = e + 6) '" R 21 1 (5.16) o - 20 - 4u t' tt * ••t " Iii I 'I I With considerable manipulation of rows, the IT(X)] determinant can be seen to contain factors of the form 2 cos (/3x) + c sin 2«(3x) The Wronskian of basis (5. 2) is less manageable, as may be seen from the single term: w ,7 1 324. = (a 5 - lOa (3 + 5a (3 ) Slnh(ax) cos«(3x) 4 2 3 5 . - (5a /3 - lOa /3 + /3 )cosh(crx) sln(/3x) (5.17) Numerical difficulties arise because columns 1 and 2 and columns 5 and 6 of the complete basis may be obtained from each other 'by replacing sinhcrx w~th coshax or vice versa. Thus, the linear independence depends upon the ability to distinguish the hyperbolic functions at large values of X with a computer number set. Through considerable algebra the Wronskian may be shown to contain the factors (cosh2(ax) - sinh 2(ax) and (cos 2 (/3x) + sin2(/3x) ). The difficulties with basis (5.3) interestingly eno,ugh do not arise in the same manner as with (5.2). Except for a minor phase difference in angles /3x and 6z the most significant difference between (5.3) and (5.4) lies in their respective[E(X)ldeterminants. For basis (5.3), this becomes (-a e l - a 1 - a 2 + cr 1 + a 1 + ( 2 )x 0 = e = 1 so that on this consideration the basis is seen to be at least as good as basis (5.4). A similar pattern is observed if we consider the Gramian determinant. For a basis of functions (>i)' the Gramian determinate is given by L G.. = 1,J J o >. >. dx 1 J (5~18) -21 - 41 C 'I" N"f ' ,) \!:Iiriei'M. 'ribW'Wrii'iliit""Nfl'!W't"!![t"WM1f'l"ifHt"H\"ij"I"'HfW,'*\8*WillNNWW'''j"''' U1 fW'+':i1I:Wlli''mW'¥"J ""'111,""11" W"'lftl'ij"eHWUtfltt""tlbiWN!wH" 'U"I!!'W t 'IIIetlllfUUlfU,,",.,'Mltt", 1 ttt h $ tttt rMtI ....... tt±ris·t*rt:tW#rt. we prefer. to normalize the Gramian as to. o L J o G~:C = i,j 1>. 1>. dx 1 J (5.. 19) L L f f 1>.2 dx 1 o o 1>. J 2 dx In this form we may think of the functions 1>i and 1>j as being coordinate vectors in a function space. The terms Gi . are then cosines of angles between the coordinate vectors. One may ~~so think of them as simple correlation coefficients between the base functions. Now consider the functions e-axcos(f3x) and e- az cos(f3z). Then L f G~:C o 1, 3 e -ax cos (f3x) e -az cos ( Bz) dx o = L L f e -2ax 2 cos (f3x)dx J e -2az 2 cos (f3z )dx o o L e -ax .f cos(f3x) cos(6z) dx 0 = L J e -2ax 2 cos «(3x) dx (5. 20) 0 If L is great the definite integral in the denominator will be approximated by the lower limit so that the order of magnitude of G 1 3 is e -aL. Thus, in this basis the functions separate into nearly orthogo~al sets as sociated with the respective boundaries. o - 22 - 42 A similar situation holds for basis (5.3). The exponentials f(ill out of the numerator before integration and the denominator contains e aL . For basis (5.2), however: o L J G* 1,2 cosh(ax) cos(0x} sinh(ax} <;:os(0x)dx o = L L f 2 2 cosh (ax) cos (Bx)dx .22 sInh (ax) cos (Bx}dx o L J (e 2ax - e - 2ax 2 ) cos (0x)dx o = L J (e 2ax + 2+ e -2ax 2 ) cos (0x}dx (e 2ax -2+e -2ax 2 ) cos (0x}dx o ---_t (5. 21) 1 How may we distinguish between basis (5. 3) and (5. 4)? recall that our solution is in the form T (x) = ~ (5.22) a. . (x) 1 Let us 1 o This can be seen to be the vector T (x) in a function space, written in terms of its components ai along the coordinate axes i(x}. We have seen that basis (5. 2) is unattractive because the axes i become parallel. For the basis (5.3), the problem is the scaling of the axes. The length of the coordinate vectors is exactly the integral in the denominator of the Gramian. Thus, for a basis with distinct axes, such as (5.3), the choice of a second coordinate origin essentially normalizes the coordinate vectors. Let us now consider the full matrix for the derivative boundary conditions. We may partition th is matrix into four partitions as , -- - -----,-------- , S 21 I S 22 (5.23) o - 23 - 43 ,t t "\!!ii"'iillif'H'd'lli:ill:l!l:tlYf!l¥t:H±l"W!!'!tW$!!'tttt eln"Wnr!l!!!!WM'Ilf¥l!fIll'"w"W',,!'!t,,'rt!lI!!!!I11WnN'MH'ltI" o II",,'' ' !! "f'IW"",'''$7'' 1m •• hila tt t • tt ' •. "$ *> Where above the horizontal line we write the conditions to be met at X = 0 and below the conditions to be met at X = L. To the left of the vertical line we write solution components associated with exponentials with negative real parts and to the right exponentials with .positive real parts (or for basis 5. 4 solution components in Z). We have two matrix e,quations (at X = 0 and X represented as: [ B1 [R 1 [T (X n [E (X ) 1 = = L). They may be (5. 24) [51 The matrice s [B 1 are rectangular. They and the matrice s [R 1 are unchanged from basis (5.3) to (5.4). The changes in [T(X)l are only differences in phase angle of the trigonometric functions. The matrix [E (X)l , however, is very different. A t X = 0, the E matrix for basis (5.3) is the identity matrix. The E matrix for basis (5.4) is the diagonal matrix L -Q' I 1,1,1, e -0:' 1 e 1 L This same matrix multiplies the E matrix of (5. 3) at X = L. To give the E matrix of basis (5.4) o -0:' e 1 L -Q' e 2 L If we denote the complete matrix for the boundary conditions for basis (5. 3) as S and the diagonal transformation matrix L -0:' 1,1,1, e I 1 -0:' e 1 L as C, we may write the boundary conditions in basis (5.4) as [51 [Cl [Al = [Fl except for the difference s in phase angle noted above. If we chose to pre-multiply [A 1 , rather than post-multiply [Cl J it is seen that C represents a scaling of the coefficients [A 1. This problem is identified by Lanczos (Ref. 5) as artificial ill-conditioning. His recommendation is just the sort of rescaling accomplished by the choice of basis (5.4). o - 24 - !,:I "1 I I o VI. REMARKS AND CONCLUSIONS Using basis (5.4), we finally see the boundary condition equations in the partitioned form of the previous section in a physical light. The matrix Sll represents the semi-infinite problem at X = 0_ The matrix S22 represents the semi-infinite problem at X = L. The matrices S12 and S21 represent cross-coupling between the semi-infinite solutions. The final four figures show the solution and components for the seventh order problem solved with basis (5.4). Figure 4 is a plot of the shear stress for a typical joint. Figure 5 shows the components of the shear stress for this particular problem. The components associated with complex roots were too small to show on the same scale. Notice that the physical problem - two transients moving in from the boundaries, the solution, the solution components and the partitioned form of the matrix - all reflect the same pattern. Figure 6 shows the peel stress solution. ~Figure 7 shows the components of the peel stress. Rl is the absolute value of the complex root 0'1 + i f3 1 - R2 is the value of 0'2. Recall with basis (5. 4) the derivatives contain successively higher powers of the moduli RiO Now Rl is 6. 78, whereas, R2 is 1. 77. One is not too surprised, then, when a-which is given by _£ a- (X) 2 CJ dT(X)) dx contains larger components or the functions associated with the roots ±O'l ± i f31Notice, also, that the magnitude of the real parts serve to determine how local the effects will be. Notice the components associated with ±O'1 ±i-f31 are much more rapidly damped than the components associated with 0' 2This same problem was solved using basis (5.2). An interesting comparison of the numerical difficulties is p,rovided by comparison of the conditioning number of the matrices which had to be inverted. The conditioning number is the ratio of the largest Eigenvalue to'the lowest. o - 25 - 45 #dbtt'J'J.tLtW' rittH±>tbW' HIt"'.*' '11 lb .t o The logarithm of the conditioning number provides an estimate of the number of digits which will be lost in the inversion. For the basis (5.2) solution, a 6 x 6 matrix was inverted. The conditioning number was 2.6. 10 8 • For the basis (5.4), two 3 x 3 matrices are inverted (solution by partioning Ref. 10). Their conditioning numbers were 1.10.10 2 • Our conclusions follow HaITlming' s beautiful statement (Ref. 1): "The Purpose of computing is insight, not numbers." Recognition of matrix products added tremendously to our insight and provided an unusual opportunity to see the nature of our numerical difficulties. We confess to a strong interest in writing more solvable equations. The work which has been done on best approaches to the problem of solving poor equations, while very useful, has already run its course. Nothing but more digits will improve on the best methods available. The problem of writing better equations is certainly not simple. Nor do we feel that we now know how. We do believe that the close imitation of the physical problem is a good clue. Further, for this problem we identified two mechanisms which could affect the equations. The basis (5. 2) led to badly skew axes, the basis (5. 3) to badly scaled axes. The double coordinate system improved the scaling. It is interesting to note that the normal equations of a least square approximation problem become highly skewed"if the coordinate origin is very distant from the center of gravity of the function being approximated. The pos sibili ty of writing a gene ralized program for the clas s of problems treated here looks good. If we were to write it, the first thing we should like to do is be very sure of our polynomial root routines. Another difficulty would be choosing the boundary conditions before the nature of the characteristic polynomial was established. o - 26 - 46 o E Cl) ::0 0 d: ~ ~ Cl) -0 ~ 0 - ...c: c Cl) > Cl) Vl M Cl) ...c: ' +- 0 C 0 ::J C 0 Vl ~ N 0 Cl) ...c: Vl . ""¢ Cl) ~ ::J .~ u.. ~ 0 0 0 0 an 0 0 0 ~ 0 0 0 (W) 0 0 0 N 0 0 0 ~ 0 - 27 - 47 o o /ase o 1.77X a 4000 7 9 -1.77 Z /i 3000 ~ 2000 1000 ---I o ~ (7 o " ~ 1 ff 2 3 4 5 • 'Ti(X) Figure 5. Components of Shear Solution (Note: Complex Components Were Too Small for Scale) ~~ 6000 5000 4000 3000 N .~ 2000 1000 0 -1000 1 \ o .. / 1 Figure 6. "'c.o"' o 2 3 II 5 • U(X) 4 Peel Stress Sol ution Curve for Seventh Order Problem c ~-~ ~ ------- o o o 1 cos (til :x) e 5000 o -C1/ X R1=6.78 R2=1.77 011 =4.796 til =4.794 012=1.77 4000 3000 Ui (:x:) 3 Ti (x:) - lE.a (d.ci:x: :3 1i(:X:») _t ci c;,l ci:x: 2000 IN o 1000 o -1000 c.n e -OIl X sin (til :x: ) e- OiaX Il'C o E< » 1 2 3 4 ./11 5 Figure 7. Components of Peel Stress Sol ution, Seventh Order Problem * ·Curves are labeled with corresponding functions from the basis of the shear solution. As can be seen from peel stress expression, the true functions of this diagram are linear combinations of high order derivatives of the respective shear basis functions. o REFERENCES 1. Ham.m.ing, R. W. Num.erica.! Methods for Scientists and Engineer s. New York: McGraw-Hill Book Co .. Inc. (1962). 2. Goland, M., and Eric Reisner. "The Stre sses in Cem.ented Joints, " Journal of Applied Mechanics (March 1944), pp. A17-A27. 3. National Physical Laboratory. Modern Com.puting Methods. Edition, London: Her Majesty's Stationery Office (1961). 4. Burnside, W. S., and A. W. Panton. The Theory of Equations, Vol. I. New York: Dover Publications, Inc. (1928). 5. Lanczos, C. Applied Analysis. Inc. (1956). 6. Wilkinson, J.H. "The Evaluation of the Zeros of IlI-CondltJ.;,)ned Polynom.ials," Num.erische Mathem.atik, Vol. I, pp. 150-180. 7• Lance, G. N. Num.erical Methods for High-Speed Com.puter s. Iliffe & Sons, Ltd. (1960). 8. Golom.b, M., and M. Shanks. Elem.ents of Ordinary Differential Equations. New York: McGraw-Hill Book Co., Inc. (1950). 9. Kaplan, W. Advanced Calculus. Publishing Co., Inc. (1953). Second Englewood Cliffs, N. J.: Prentice Hall, Cam.bridge: London: o Addison Wesley 10. Fadeeva, V.N. Com.putational Methods of Linear Algebra. Dover Publications, Inc. (1958). New York: 11. Kaplan, W. Ordinary Differential Equations. Publishing Co., Inc. (1958). 12. Ralston, A., and Henry Wil£' Mathem.atical Methods for Digital Com.puters. New York: John Wiley & Sons, Inc. (1960). 13. Hildebrand, F. B. Introduction to Num.erical Analysis. McGraw-Hill Book Co" Inc. (1956). Palo Alto: Addison Wesley New York: o - 31 - 51 rr_ I I,,,,,,, IMIM o 0, :'1 . ·1 1''ILi'''NIfI''Iopl'i''!!1 2' I , " 't.t,' '1IfPf'?1I tt ., t I 1Ft , t h II tttttttwitrrrbtt 14. Householder, A. S. Principles of Numerical Analysis. McGraw-Hill Book Co., Inc. (1953). New York: 15. Hetenyi, M. Beams on Elastic Foundation. University of Michigan Press (1946). 16. Courant, R. Differential and Integral Calculus, Vol. 1. Inter scienc p Publisher s, Inc. (1934). 17. Pa ige, L. J., and J. D. Swift. Ginn and Company (1961). 18. Churchill, R. V. Fourier Series and Boundary Value Problems. New York: McGraw-Hill Book Co., Inc. (1941). 19. Lanczos, C. Linear Differential Operators. D. Van Nostrand Co., Ltd. (1961). 20. Edwards, J. An Elementary Treatise on the Differential Calculus. London: MacMillan and Co. (1892). 21. Morse, P. M., and H. Feshbach. Methods of Theoretical Physics. New York: McGraw-Hill Book Co., Inc. (1953). 22. Ince, E. L. Ordinary Differential Equations. Publications, Inc. (1926). 23. Ford, L. R. Differential Equations. Co., In c • (1 9 55) • Ann Arbor: The New York: Elements of Linear Algebra. o - 32 - New York: New York: New York: Dover New York: McGraw-Hill Book READER A program to read and execute elementary machine language laboratory exercises R. C. Steinbach (5145) Introduction Grossmont College is one of California's many public two-year colleges. These colleges provide three educational programs: (1) General education courses for the community, (2) Technical-vocational courses, (3) Transfer courses for students going on to four year institutions. Within the techni- cal-vocational area Grossmont College has a data processing program containing a one year (four units per semester) computer programming course which begins with machine language. Students are capable of writing miniature machine language programs after approximately two lecture hours. The pro- gram described here monitors the student programs, allowing the student to C see his program executed and relieving the instructor of the job of reading machine language programs. Student Program Format During the first six weeks of the programming course the students are assigned specific problems to code. end of this paper. as follows: Examples of these problems can be found at the For each problem, each student hands in a deck of cards (See Figure 1) each student's program. The first card or Header Card is used to identify This card contains the student's name beginning in column one and ending with a record mark. It also contains a five digit identification number beginning in column 75 and a record mark in column 80. Reader uses this latter record mark to recognize the header card. ~a~ Cards follow the header card. The Pro- The student machine language program is punched 72 digits (6 instructions) per card into as many cards as is necessary to a maximum of ten. A record mark in column 73 of a program card indicates 52 o .et to" \ 't t» tt rt •• t t. t .j I eM trW $\ d" t • 1 j t • H..... * t" tt. rt ee UH¢&ttss' trbtMt* htritnaz Page 2. o that column one of the next card follows column 72 of the card just read. Thus the last card (it may be the first and hence the only card of the program) has no record mark in column 73. All programs return control to READER with a branch to 00000. This allows a manual restart (INSERT, RELEASE, START) if the student program hangs up and has not destroyed the READER program. OPTIONS During the time that the student has no knowledge of input/output instructions READER outputs the work area so that the student (and the instructor) may check the program results. This output may be suppressed using console switch 3 after the student is familiar with output instructions. The output device, either card punch or typewriter, for READER may be selected using console switch 4. This latter option allows remarks from READER to be output on the same device required of the student in a given problem. A TYPICAL RUN For each problem, the programs written by the students form a single deck which follows the READER object deck and four special data cards. (See Figure 2) The first speciaL data card contains program identification, console switch settings and tabulator information for the operator. The next three cards contain data for the student work area, e.g. numbers to add or subtract, negative numbers to count. It is advisable to add an instructor written solution to this deck of 4 special data cards so that the students can see the right answers and see o one way of writing the program. first student program. As far as READER is concerned, this is the Note that the 4 special data cards and the instructor 53 Iii I, Page 3. written program form a package which separates the reader object deck from o the deck of student programs and which is easy to include for any given assignment. READER types the program identification and operator message and halts. It then reads the three data cards. initializes the student work area and reads and executes the student programs as follows: 1. Search for Header Card. (Go to 3 when found; go to 2 on last card indicator.) 2. Type "All programs read" and halt. Press start to read next 4 special data cards and new batch of student programs. 3. Type student identification number. 4. Input student program. output student name and number of cards required for program. 5. Branch to student program. Return to 6 is automatic by student o or manual by operator. 6. Output work area if switch 3 is on. 7. Initialize student work area. 8. Go to 1. REMARKS One should list the student program deck before doing anything else so that there is a permanent record of who turned in what. This is at least a par- tial defense against a charge of deck shuffling at execution time. A clumsy student can wipe out core with a TF or TR. The only thing to do is reload the READER. but at least you have his identification number on the typewriter. A loopt checkstoPt or bad operation code can be noted by hand on the typewriter output and the READER restarted manually. o it &sttdt t t •• tttt tnt t * st t j t Ht tt # .'" , t*, •••• $ ••• mtr6{trttrri' it. eWh" W#tt I it "j Page 4. o It is possible for a student to read the next student's program as data. As soon as this is obvious, a STOP, INSERT, R/S, will restart the READER. A comparison of the initial listing and the run listing will determine who was left out and his (their) program(s) can be placed at the end of the student program deck. Conclusion I would appreciate comments and criticism from any interested person. I do not plan to submit this to the Users Group Library until at least one more class has tried the system; they may think up new ways of giving the READER trouble. o 55 :1 ! o *~ 0 0 0 0 0 0 .... CD 0 0:: Z Z « ~ () () 0) ~ w l- 0 (f) « I I ...J z z z en z 0 ~ 0 ~ a:: ~ en z - ~ e 0 « a:: a:: « <.9 (.) 0 a:: w 0:: ::) (9 l.J.... a:: 0.. W (J) 0 « * LLJ I w :E c( z I 0 56 o o o / STUDENT PROGRAMS ,d I / Ii INSTRUCTOR'S PROGRAM DATA :1 ,tI ill (3 CARDS) i PROGRAM I D, OPERATOR MESSAGE ------ ( READER OBJECT DECK ~ I III I ----------------------Y L -__ ~ -'1 FIGURE 2 i ji 1IIIjli ' I ~I I LAB EXERCISE I Numbers, described below, are in storage with the most significant digit flagged. Address of least Number Number of Digits significant digit A 1.1 B 6 2 C D 3 4 7006 7010 7016 7021 Assume no overflow, numbers are integers. Replace A by A+B Replace C by C-B Replace D by D-658 1.2 Assume no overflow, numbers are integers. Replace A by the integer A-2B+C-D 1.3 Assume no overflow. o Assume decimal locations as follows: xxx. xxx .xx C = x.xxx D = xx.x A = B = Replace A by A - D Replace D by C + D Replace C by C + 2.93 o 58 trh t o ttt _.ew LAB EXERCISE 3 Note: Memory addresses above 11000 are available for your use. digit of your program is in 07300. The first 3.1 Return the carriage on the typewriter. Type out the numerical contents of 7001 - 7009, space the typewriter, type out the alphameric contents of 7030 - 7047. Return the carriage and type the numeric contents of 7030 - 7047. There are no record marks in place. 3.2 Return the carriage, type your name (25 oharacter maximum), tabulate and type your code number. 3.3 As input to your program have one card with your name beginning in col. 1, and the words "1620 I/O PROGRAM" in col. 32-47, and a second card with 5 zeros, 5 ones, 5 twos, etc., and 5 nines in col. 1-50. Duplicate the two cards. 3.4 I will supply you with 3 cards which you will use as input to your program. Each card will have the following format: o A five digit number A in col. 6 - 10. A nine digit number B in col. 17 - 25. You are to punch out three cards with the following format: A and B as above A+B with low order digit in col. 40 A.B with low order digit in col. 60 There are no flags on the input cards, and there should be no flags on the output cards. o 59 et. 'I I o LAB EXERCISE 5 5.1 Type a message to turn on console switch 2 and then halt. If the switch is not set properly repeat the message and halt. Continue this process until the switch is on. 5.2 Two flagged 4 digit integers have their units position in 7005, and 7010 respectively. If the n-th integer is less than 2222} equal to 2222 greater than 2222 put a {I } 2 3 in 7011 + n 5.3 35 flagged 4 digit integers have their units position in 7004, 7008, ••• , 7000 + 4n, ••• , 7140. Tabulate the typewriter and type the number of negative numbers in the list. 5.4 Three 5 digit integers are located in 7005, 7010, and 7015 respectively; arrange them in ascending order in locations 7020, 7025, 7030. c o 60 ,*.' rirttfri .trtrttMnttrtrtt ttirt . . t t w j tttM .c. wrtftttt t' A 519 Simulator R. C. Steinbach (5145) o Introduction Card reproduction on the 1620 is not new; approach is to insert 371111100500 391111100400 the most straight forward 4900000 R/S. The problem becomes slightly more complex if information is to be deleted, the columns permuted, sequence numbers added, and/or information gang punched into the cards. This paper describes one method of handling these other possibi1i- ties. Method During the first phase, 'the simulator sets up a table of source addresses. The first entry in the table is the address of the two digit field to be placed in column one of the output deck; the second entry addresses the source field for column two; etc. During the second phase, a card is read into an input buffer, 80 two-digit fields are transmitted from the appropriate source (the source table is addressed indirectly) sequentially into an output buffer. A card is punched and the next card read, and so on. Format Cards The deck to be reproduced is preceded by three format cards called INPUT, OUTPUT, and EMIT. All three cards must be there, however, the INPUT and EMIT cards may be blank. The input format card identifies the source of characters from the deck to be reproduced; the output format card identifies the destination of all characters to be punched in the new deck; the emit format card contains characters to be gang punched into all cards of the o new deck. The simulator produces the source table by scanning the output format 61 =-= .. ",,' "' .... '''''.'''"-'-'--,,~."'''''."=, ... "=." Page #2. card. All c,olumns that are blank in the output format card will be blank in the new, or output, deck. A field of l's in the OUTPUT card indicates that the source is the same field on the old, or input, deck. A fIeld of 2's (up to 5) indicates a sequence number field on the output deck. Note that this requires the OUTPUT card to be scanned from right to left. ' A field of 3's indicates that characters are to be emitted from the corresponding columns of the EMIT card. If a field of any other character, e.g. AAA or »»), is encountered on the OUTPUT card, then the INPUT card is searched for a corresponding field. The location of the field on the INPUT card determines the columns to be picked up in the old deck; the location of the field on the OUTPUT card determines the destination in the new deck. If the OUTPUT card contains a character other than the four special characters (blank, 1, 2, 3), that same character must appear on the input format card; furthermore, the field length defined must be the same. "Format card mismatch" is typed cards. an~ If either of these conditions fail, c the program will then accept new format Figure 1 shows an eKample of the three format cards. Anomalies Although it is not immediately obvious, the method chosen to set up the source table allows one field of the input deck to be placed in more than one field of the output deck. To accomplish this, a field indication on the INPUT card appears in several (non-adjacent) fields of the OUTPUT card. Two non- adjacent fields on the input card designated by the same non-special character will not be correctly interpreted. Sequence numbers (even of different length) may also be punched in several non-adjacent fields. o Modifications Often, one wishes to change the emit characters whenever a master card is 62 wwtrWllt.,"!!y" p " ","'I'"t'lll"" w'QUtl""""flfW'" . , '''*,!',?,m"".,rtM'$",.'P ! , Mel" 'I't WJ'I!Wfft"fflMWN"""MfI' '" tiP .,n. !'rOY • It tt t t ",'_ • 'tb »or tittt bt uw tntt Page #3. o detected. The variety of ways in which a master card may be indicated, and the number of possible reactions to a master card suggests one of the following manual solutions to the problem rather than a fully automated system. If there are just a few decks headed by master cards, the same INPUT and OUTPUT cards may be used with a different EMIT card. The Master card may be used for an EMIT card if the master card is not to be duplicated and the characters to be emitted are in the correct columns. If there are many master cards in a particular run, they may be detected using a compare or compare immediate after each card is read. A special routine is then added to the source deck to transmit characters from the Master card to the EMIT card image. The bulk of the routine can be instruc- t ions of the form TF ENIT-2+2 1:ecn, IN-2+21:mcn where ecn stands for emit column number and mcn stands for master column number. o With the d'lZ~tect routine and transmit routine added, the source deck is reassembled. Conclusions Any suggestions on ways to improve this program will be greatly appreciated. It will be submitted to the Users Group Library after these improve- ments are incorporated. o 63 c t- => a.. z 0 I w a:: :::::> (!) lJ.... J 1 !"tnt. tI' ttt . tit. tr IT * . t',' h It rent'. ,I t' rt Htt ht et t t t H.rHhr. o GOOOIiEAR GOODYEAR AEROSPACE CORPORATION AR I ZONA DIVISION LITCHFIELD PARK. ARIZONA o SIMULTANEOUS LINEAR ~UATIONS WITH CCIU'LEX COEFFICIENTS N. Kuffel AAP-18906 May 1, 1961& o 65 SIMULTANEOUS LINEAR EQUATIONS WITH CClWLEX COEFFICIENTS N. Kuffel INTRODUCTIOIf This program 801T•• a1ll1ltaneoue linear equations with complex coefficient. resulting in complex roote. It vas original~ developed to aolve large 8.18tema and baa applications in mechanical and electrical engineering problema. Ot the I1UJIlerOU8 programs available for ma~ix inversion and simul taneoua o 'equationa, very fev take into account the under-and-overfiow problems , . encountered on large matrix qatema. There are no }rograma published at the present time tor the 1620 tor solutions of complex simultaneous tiona, and 'fiery fev available even tor other machines. are available on the 162~ tor real systems. e~a Several. progr81U '1'h1a program w1ll 80lTe up to 20 aillul taneous linear equftioDS with oomplex coetticients. Two tOJ'IU ot output results, A+jB and Ke j , are available . tor either a 8pecified limited DUJlber of unknowns, or for all unknowna up to 20. The progr_ 18 written in Fortran with Format and requires 40 K IlSOr.r. Qi"D a I178tea ot H 81Dltaneoua linear' equationa, in N unknowns, v1th coaplex (or real) coeftioients, the progra solves forth. desired nuaber of unknowns in terms ot complex numbers. In certainaituatioDs, onlY a tew ot w.rous unkn01lD8 are needed. Tho8e desired can be rearranged to appear tirst in the equations. By ap8city1ng the muaber desired, only' o that DWIber v11l. be solved tor, saving considerable coapt tar time in the cue ot large systems. 66 ----_*_____H_tt_.h_.______ ~ -*-r--_._____ t.____. *___ -.. M_t_~_· .'M.'_*_____________tt_t_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ * __ "".,., o .tr II %'1• • SIMULTANEOUS LINEAR PltUATICJlS WITH C()fPLEX COEFFICIENTS Those equations to be solved are set in determinants of the fonal IZI • la' +j where a and b are the coefficients. Ibi The' application of Cramer.' Rule gives ~I IZI • fel +j 1-1 +j • ( 1& I ICI Idl Ib J + USI • ~I 1&,2 ,a, -j Ibl lal -j Ibl leil ) (,a, + ~ + lei, ~I Ie I) 11'12 where Z 1. the determinant ot the coefficients and is the same determinant with the coefficients or the desired unknown replaced by the constant terms. (&) All determinants are evaluated by the triangular method, in which all elements to one side or the leading diagonal are computed to be zero. The determinant i8 equal to the product of the eleMents in the leading diagonal of the triangular determinant. This method of evaluation is preferable to that of expansion in terms of ainors or the pivotal method because of the storage and time problem imolved in the large complex systems. Previous programs have made it necessary to do a manual rearrangement ot data when a sero element is encountered on the diagonal, resulting e1 ther trom the original coetficients or trom subsequent compltat1ons. This program will check elements in the same column of the remaming rows of the determinant for a non-sero element. It such a value 18 found, a row interchange is performed, changing also the sign ot the determinant. It no non-sero element is found we 'haye the case of a sero determinant. It this occurs for the coetficient detena1nant, a me8sage 1s typed out ani a different _thod of solution JIlUst be found for this cue of a nonsingular solution. A .ero nu.rator determinant evaluates an unknown equal to zero, which is the correct result. o o.-er and underflow proble.. are quite common in matrix problems when dOing accumulative operations J such .a coml'Qting t.he product ot the diagonal e1811l8nt& oftha deterJlinant. A scaling procedure baa eliminated such ditt1culties in this program. Before mul t1plying, each diagonal element 1.IJ scaled to the range between .1 and 1.0, storing an accumulative characteristic -2- 67 SDlTLTANEOUS LINEAR EQUATIONS WITH COMPLEICOEFFICIENTS o (or power· of 'ten) tor the determinant, which is ()utput with the product and then applied in the final division ot deteminants 80 that the end 'results have the correct magnitude. EepeciaJ.q :Ill the case ot large 87Ste1l8, this program has been found to be as 8tt1ci~nt even for real systems as most existing programs, particu- larly because of the row interchange and scaling procedures. As many as 20 eQ.uations in 20 unlmOWllS may be handledby' this program on a hOI machine, which is minimum core for the program. The largest system . run up to this time baa been 18 equations, but no difficulties can ,be foreeeen on 8.l\Y larger problems because of the 8Caling procedure. The reeul.ta are indicated in two forms. The actual outputs are the real and imaginary parts of the 8olution, as well as the magnitude and phase angle. These will give results in the tormal A + jB and Ke jt G where A • real part B • imaginar,y part I • magnitude t • phase 8l'Igl, in degrees t • tan-1 B I SUMMARY This program has been used numerous tim·es for several months nov, on systelft8 trom 3 equations to 18, both partial. and complete solutions. on the 1620 MOD II have runa Execution tiIles 3rd and bth order - 1 lIlin. lSth order - 20 Ilin. 17th order - 29 Ilin. o It ehould be noted that theae times are dependent on the original set up ot the coefficients and how m&ny' row interchanges are nece8slll7. 68 -3- *: ! ,.1,..,r""Miur ,t t ..... " t r r r tt: r t ztt In t. t,) *' tt hi _ $ ht de HMstrttsttitstt· "m tiM tt)L SIMULTANEOUS LINEAR I!XtUATIONS WITH COMPLEX COEFFICIENTS The program. is written in Fortran with Format and uses an ABSOLUTE VALUE aubrout1ne. Thi8can be easily changed in the 80urce program it the subroutine 18 not relld1:q available. Although the program Jresently begins at 6600, there 18 _ple storage to recompile with a starting position of 8)00 for other _chine configurations. It would be a silllple matter to change input and output DIOdes to tit other needs and equ1pnent. No serise switches are used. SUIple input and output data follow in Appendix A and a JrOgram listing is in Appendix B. o o 69 -1&- APPENDIX A Sample inplt and output data listing tollow. Input data tollows the aame o sequence tor all programs although Case 1 will be the only one described. . .~ Case 1 - 3 Input . 1 order complex system, canplete solution st Card - NSOL • 3 (number of solutions desired) - 13 format Note statements 500 and 101 in program listing (Appendix B) 2nd Card - N· • :3 (order or· system) - I3 format ot the coefficients) - both values are on the same card in Elh.8 N X N (9) Cards - AR and AI (real and imagin&J7 parts format and are entered row-wise. Note statement 100 in Appendix B. N(:3) Cards - FR and FI (real and imaginary parts or the constanta) both values on the same card as were the coefficients. output Real and imaginary parts ot the input coefricients Real and imaginary parts otthe input constanta Real and imaginary diagonal produots, value of the coefficient determinant, scale factors for the produots and the determinant, phase angle and magnitude. Real am imaginary diagonal products, value of the determinant and scale factors for NSOL(:3) solutions which include real and imaginary (A an~ B), phase angle (t) and magnitude (X). parJ CatSe 2 - 4th order real system, complete solution ease :3 - 4th order real system, partial solution Case 4- 'fd order real system, sero determinant o 7U LN'i#!Htt"i:i!.1·dt6·H"JWi~WMlJjJ""tMiMWi "•.•,# fW'bwrw'i#tPtiJliH¥t' ""fit W'litiiW*Mtw "'w'\\tiriW ""'iib' tit" '.' b' 1fut'W'J/,'i"i:iINibtt' 'It 'f 'il'YrrWwlttW''b' HfW 'f'IWWifY-B'm"tfiJ'''twt2'',eiilI''!'tmnUWW'H IPl/W II pmllr:"r' , Wi '"rnltl"""Wm', ! t t rt h t wrtrris Sample Case 1 Page 2 o Input 3 3 +.20110300E+04+.13140000~+03 -.20~50000£+04-.22700000E+Ol +.OOOOOOOOE-99+.0000000 0E - 99 -.20550000E+04-.22700000E+Ol· +.161U2980E+U5+.21747GOOE+03 -.14170000E+05-.1b~OOOOOE+03 +.00000000E-99+.00000000E-99 -.14170000E+05-.18500000E+03 +.22498000E+OS+.13500000E+03 +.OOOOOOOO~-99+.000000UOE-99 +.OUOOOOOOE-99+.00000000E-99 +.G3300000E+04+.00000000~-Y9 o 7I t• ...._ .. Page 3 Sample Output C~~e, " 1 o SOLUTIO~ OF SIMULTANEOUS LINEAR EQUATIONS WITH COMPLEX CUEFFICIENTS PROG. 223-63 ORDER 3 REAL I r'lAGI NARY .20110300E+04 -.20550000E+04 .000000001:-99 -.20550000E+04 • 16 10298 0 E +0 5 -.14170000E+05 .OOOOOOOOE-99 -.1'-:-170000E+05 .22498000E+05 .131'tOOOOE+03 -.22700000E+Ol .OOOOOOOOE-99 -.22700000E+Ol .21747000E+03 -.18500000E+03 .OOOOOOOOc-99 -.1i3500000E+03 .185UOOOOE+03 c O~,j ST td'J T S .OOOOOOOOE-99 .OOOOOOOOE-99 • [j 3300 000 E +0'+ .OOOOOOOOE-99 .OOOOOOOOE-99 .OOOOuOOOt-99 REAL PROD. .22940795E-OI IMAG PROO. DETERMINANT .25515200[-02 .53279032E-03 0 f'~UlTIPLY REt\L ANO H,'IAGIf,u\R.Y PRODUCTS BY I.OF (·':UL T I PLY OETE;':j'·jJ NAi'JT BY 1.0E 26 = • (, 3 4 6':,. 6 L t: +OlD f ~ R t: E S = • 2 308 22 5 1 E- () 1 ~~ 1. a~ PH A S E td ~ G L E i"i A GNIT U 0 r~ REA L P RU D • t l" I Iii AG PRO D • 1. 3 L) ET E Ri',( I NA1'1 T .2425606P.E-Ol .34347995E-03 .Sd847480E-03 1 MULTIPLY REAL AND IMAGI~ARY P~OOUCTS BY I.OE MULTIPLY UETERMINANT BY I.eE 26 REAL P~OD. .23717163E-Ol IMAG PROD. 13 DETERMI~ANT .lH60~040E-n2 .56596678E-03 2 r-'ULTIPLY REAL AND Ii:1l\GIi'JAkY Pi.{ODUCTS dY l.OE tl: Ul TIP LY 0 ET E:;( ;'<1 I /.! Ar\J T BY l. 0 E 26 REAL PROD. INAG PROD. DETE~MIN~NT .23433940E-ul .21191024E-02 .55364013E-03 3 j'' l UL TIP LY REA LAN [) I r;;~ GIL ARY PRO UUCTS d Y 1. 0 ~ ~llJ L TIP LY 0 ET ER~!I :"J Ar·JT BY 1. () E 26 13 1J SOLUTIONS OF THE SIMULTANEOUS LINEAR EQUATIUNS ORDER 3 REAL I r;At;I NARY PHASE AiJGLE ~;'iA(~f"! o I TUDE 72 Page o b met tri±it tittbstt· r o_..sri t r th WI teeti tnrM,. . . . . trtr #& ttt tt 4 .10460585[:+01 -.10137221f:-OO .10301~13E+Jl - • 3 3 it- :; 4 it-:~ 1 E- 0 1 • 10 1 <) 1 (, 2 Gt: +0 1 -.209t~0622E-Ol -.553:'1769E+Ol -.18600954E+Ol -.11793317E+Ol .J.05095BHE+Ol .103066~·3F.+Ol • J_ 0 1 C) 3 7 i5t) E + 0 1 o o 73 Page 5 Sample Case 2 Input 4 4 +.30000000E+Ol+.OOOOOOOO~-99 +.20000000E+Ol+.OOOOOOOOE-99 -.10000000E+Ol+.OOOOOOOOE-99 +.lOOOOOOOE+Ol+.OOOOOOOOt-99 +.10000000E+Ol+.00000000E-99 -.10000000E+Ol+.UOOOOOOOE-99 -.20000000E+Ol+.OOOOOOOOE-99 +.40000000E+Ol+.OOOOOOOOE-99 +.20000000E+Ol+.OOOOOOOOE-99 +.30000000E+Ol+.OOOOOOOOE-99 +.lOOOOOOOE+Ol+.OOOOOOOOE-99 -.20000000E+Ol+.OOOOOOOGE-99 +.50000000E+Ol+.OOOOOOOOc-99 -.20000000E+Ol+.OOOOOOOOE-S9 +.30000000E+Ol+.00000000E-99 +.20000000E+Ul+.000000UO~-99 +.lOOUOOOOE+Oi+.OOOOOOOOE-99 +.30000000E+Ol+.OOOOOOOOE-99 -.20000000E+Ol+.OOOOOOOOE-99 +.OOOOOOOOE-99+.00000000E-99 o o - .rrttt ... rMn II Page rt ttt ·W'U'EW' t h r d • t ttt st trtt dtt *: trte H *• ttr_.. Sample Case 2 Output 6 o SOLUTION OF SIMULTANEOUS LINEAR EQUATIONS WITH COMPLEX COEFFICIENTS PROG. 223-63 ORDER I t-\A GI NAR Y REAL .30000000t:+0 1 .20000000E+()1 -.lOOOOOOOE+Ol .100U0000f+Ol .10000000E+Ol -.10000000E+Ol -.20000000E+C 1 .40000000E+Ol .20000000E+Ol .3()OOOOOOc+{J1 .100000001::+01 - • 20 0000 00 [-: + 0 1 .500000001::+01 -.20()()OOOOE+Ol .300000001::+01 .20000000[:+01 .OOOOOOOOl:-99 .OOOOOOOOt-99 .OOOOOOOOE-99 .000000001.:-99 .OOOOOOOOt-99 .OOOOOOOOE-99 .OOOOOOOOE-99 • OOU 00000 c- C1 9 .OOOOOOOOE-99 .OOOOOOOOE-99 .OUOOOOOOl-99 .()()OOOOOOF.-99 .OOOOOOOOt-99 .OOOOOOUQE-99 .OOOOOUOGE-09 .OOOOOOOOE-SY C l) N S T A~! T S .10000000E+Ul .30000000E+Ol -. ~~ 00 00000 E +0 1 .OOOOOOCOF:-99 REAL PROD. • OOOOOOOl) \:-99 .OOOOOOOOi::-99 .00000000 r:-9<; .00000000l:-99 IMAG PROO. DfTFRMINANT .49999993E-02 .nOOOOOOOf-99 .24999993E-04 0 t.1 UL TIP l Y Ret, L A1'1 D I jVl AG 11\1 A~), Y P FJJ Due T S BY 1. 0 E 1"1 Ul TIP LY D~.: T ER 1'"1 I j\J I~ (~ T BY 1. 0 E 8 PHASE Af'JGLE MAGNITUDE = = .(;OOOOOOOE:-99DEGREES .49~99992E-02 1.DE * 4 REAL PROD. I~1AG FRUD. DET~::r-(;'lIi'JANT .19000000E-OO .00000000E-99 .3610000UE-Ol 1 i.~, UL TIP LY n. cAL t\ !'J D I fll' AG I I\! ARY P t;( [J l") UCT S E Y 1. a f: t': U LTIP L Y D~: T ER t'-'1 I Nt, NT B Y 1. 0 E REA L PRO D• -.28999992E-02 I 1'1: i~ G PRUD• .J0000000t-99 ...., i. 4 DET L: ;~ f.'; I NAN T .84099953E-05 2 MULTIPLY REAL AND IMAGINARY PPUOUCTS BY I.OE MULTIPLY DETERMINANT BY l.OE 8 o 4 REA.L PROD. rr"1AG PP,OD. DETERMI!~t\NT -.50999986E-02 .OOOOOOOOE-99 .26009985E-04 3 MULTIPLY PlAL AND IMAGINARY PRODUCTS BY 1.DE ~ULTIPLY DETERMINANT BY 1.OE 3 4 4 75 ....& 'r ""1 .;.;.~~ 'r"~p • '~·-·t '-'----t'j·.. '-·'·!··,e!·-t··''''·!·''·--.-·····t·-.·--"-.. Page ~·.~.-, 7 o RcAL PPCO. I j"lAG PROD. DETER~·: I:\~ANT .48999998E-02 .OOOOOOOOE-99 .24009998E-04 4 MULTIPLY REAL !.\NO H"1AGINAr~Y PRODljCTS l3Y 1.OF. MU L TIP L Y 0 E T Ek f--i I rJ tHJ T b Y 1. 0 E - c) SOL UTI OI'J S 0 F THE S I ;~ UL TAI'd: f) US L I f\! EAR E QUA T I UN5 ORDER 4 REAL .38000004E-OO -.:57999987E-OO -.10199998E+Ol .98000006E-07 I!\iAGINARY -.DOOOOOOOE-99 -.OOOOOOOOE-99 -.00000000 F:-99 -.00000000E-99 PHt\SE ANGLE !·itiGi\' I TUDE .OOOOOOOOE-99 .OOOOOOOOE-99 .OOOOOOOOE-99 .OOOOOOOOE-99 .380n0003E-OO .579999[~6E-OO .10199997t:+Ol .9?()nC00:5f:-07 o 76 = re.sr1 "ts WsttH j" _ t t t rSrz t t. = 1m rtn', • Sample Case 3 Input ' Page 8 o 4 4 +.30000000E+Ol+.OOOOOOOOE-99 +.20000000E+Ol+.OOOOOOOOE-99 -.10000000E+Ol+.OOOOOOOOE-99 +.lOOOOOOOE+Ol+.OOOOQOOOE-99 +.lOOOOOOOE+Ol+.OooodoOOE-99 -.lOOOOOOOE+Ol+.OOOOOOOOE-99 -.20000000E+Ol+.OOOOOOOOE-99 +.40000000~+Ol+.OOOOOOOOE-99 c +.20000000E+Ol+.OOOOOOOOc-99 +.30000000E+Ol+.OOOOOOOOE-99 +.lOOOOOOOE+Ol+.OQOOOOOOE-99 -.?OOOOOOOE+Ol+.OOOOOOOOE-99 +.50000000E+Ol+.OOOOOOOOE-9 Q -.20000000E+Ol+.OOOOOOOOE-99 +.30000000E+Ol+.OOOOOOOOE-99 +.20000000E+Ol+.OOOOOOOOE-99 +.lOOOOOOOE+Ol+.OOOOOOOOc-99 +.30000000E+Ol+.OOOOOOOOE-99 -.20000DOOE+Ol+.OOOOOOOOE-9~ +.OOOOOOOOE-99+.00000000E-99 o 77 " . r~ ."_ Page 9 Sample Case 3 OUtput 0 '· ' Ii\iEAR EUU/~T IOi'lS COEFFICIENTS SOLUTlfJN UF S I~~UL T/ i'EOLiS L WITH COMPL~X PROG. or~DER I 2~~3-63 4 REAL I t-I/\ GIN AR Y .00000000[:-99 .OOOOOOOOt-99 .OOOOOOOOE-S9 .OOOOOOOOE-99 .ClOOOOOOOE-99 .OOOOOOOOE-99 .00000000[-99 .OOOOOOOOE-99 .OOOOOOOOE-99 .00000000c-99 .OOOOOOOO!:-99 .OOOOOOOOE-99 .OOOOOOQOE-99 .OOOOOOOOE-99 .OOOOOOOOE-99 • 30000000E -C) 2 .20000000[-()2 -. lOOOOOOOE -0 2 .lOOOOOOOE-01 .lOOOOOOOE-02 -.lOOOOOOOE-02 -.20000000E-02 .lrOOOOOOOE~02 .200()OOOOE-02 .3000(lOOOE-U2 .lOOOOOOOt:-02 -.20000000E-02 .5 0000000 E -0 2 -.?OOOOOOOE-LJ2 .30000000E-02 .20000000E-02 • noo OCO(jO E-9 c() !\~ S TAN T S • 1 0 0 0 0 000 t; -u 2 • 30 0 0 () 0 0 0 E - C;: • OO()CuuOO f-:;9 -. 20000000~: -u 2 .00000000£::-99 ~-':J9 • 0 0 0 0 0 (j 0 () .O(J()O:JC'OQ ~~ - q ( Ii .--"" ,,) 9 .O(lOOOCO()f-99 I i'ii I.:, c-,. p~: U0 • DE T E k [il I j\J A~~ T • 00000 000 F - 99 • 2 4999993 E - 0 1+ 0 t-t U L TIP LY I~ f: A L Ar\j D Iii 1\ G I ;'J A~ Y PRO () UC T S B Y 1. 0 :.: t~ULTIPLY DETf:Ri\III\IAf\jT BY 1.0E -16 REA L PRO D• .4999999 3 E -0 ~~ - f; = • 0 COO (! 0 UU :: - q c) D t Gk E t-: s I PH AS E M ..JG L E MAG NIT U 0 E == • L~ 9 '-) 9 9 99 2 C- 0 2 :;, REAL PROD. I~AG ].. 0::: - 8 PROD. OETERMINA~T .19000000[-02 .OOOOUOOOE-99 .36100000E-05 1 f.'i UL TIP LY \:;J. t.: :\ L ~ iJ l) H·~ f\ (; I i'l A ~~. Y P k f.-F~ UC T S B Y 1. 0 F:. ~~ULTIPLY REA L PRO D• DclER~lIf\NT !:iY 1.GE -16 [) ErE p, >; TNAf~ T I [vi AG P PJ J D• -.28999992E-02 .OOOOQOOOE-99 .84099953E-052 t., U L T I ;:) LY I-' EA L Af\j D I ~1;\ G I !. L-\ !~ Y P R() [) UC T S UY 1. 0 E ~1 UL TIP L YJ rJ E T ER H I NI~ f\J T ~) Y 1. 0 E - 16 SOLUTIONS OF THE ORDER SIMULTA~EUUS .' -c· LI~fAR -8 o EQUATIUNS 4 78 ------ .-.--.--~--------.-~-.--.--~-~-.-.--.-~--- -- -~- ~ ~ -~~- m '!!!!!¥I'UPWMWlIP'IIIH"W ,I,. • tt· • t H. t rt tr the trW = Page 10 IiJ,GINARY .3800000 l tE-OO -.57999987i:-OO -.OOOOO()OOf--99 -.OOOUOOOOE-99 PHASE !\I\JGL E .OOOOOOOOE-99 .OOOOOf)OOE-99 ~1 AGiJ I T UDE .3G000003E-OO .~7999Yd6F-OO o o 79 ' I' ... Page II Sample Case Input . • ........ , ~, ,to - . . . ,..- 4 o 3 3 +.lOOOOOOQE+Ol+.OOOOOOOOE-99 +.20000000E+Ol+.OOQOOOOOE-99 +.20000000E+Ul+.oodpoOOOE-99 +.30000000E+Ol+.00000000E-99 +.lOOOOOOOE+Ol+.OO~OOOOOE-99 +.10000000E+Ol+.00000000E-99 +.20000000E+Ol+.OOOOOOOOE-99 +.20000000E+Ol+.OOOOOOPOE-99 +.20000000E+Ol+.OOOOOOOOE-99 +.lOOOOOOOE+02+.00000000E-99 +.50000000E+Ol+.OOOOOOOOE-99 +.15000000E+02+.00000000E-99 C''-' ' ~ '\ o 8U we ttttrsttttttt • • •1 thr« 1 t mi't _* ",=,U,,'!!W'!J[!W5Ll5" rt ' . " t Page 12 ew·''':"''',, tiS" l' .M"',,·I"2 rr"'rrf)5fWfYtYWWltW"! Sample Case OUtput 4 o SOLUTION OF SIMULTANEOUS LINEAR EQUATIONS WITH COMPLEX COEFFICIENTS PROG. 223-63 D,RDER 3 Rt:AL I 1-1,.,\ GIN ARY .OOOOOOOOE-99 .00000000 [:::-99 .OOOOOOOOE-99 .10000000E+Ol .200UOOOOE+Ol .20000000E+Ol .3()OOOOOOE+Ol .OOOOOUOOt-99 .OOOO()OOOE:-99 .10()OOOOOE+Ol .10000000E+Ol .00000000[-99 • 2CJOOOOOO[~ +() 1 .2uOOOOOOE+Ol .2000000Uf-:+Ol .O'JOOOOOOE-99 .OOOOOOOOE-99 • OOOOGOOO E-(~'':1 CUNS TAI\TS .lOOUOOUOE+02 .50UIJOOOUt:+Ol .150000COE+02 o J Z E RU DE T E R MI j\J A t ~ T - .COClOOOOOE-99 • 0 0 (J lJ 0 0 0 0 t: - 99 .OOOOOOOOE-99 US t UI F FER EN T ~1f: THO D 0 F S () L UTI 0 ;\1 o 81 APPENDIX B PAGE 01 SOURCE PROGRArll] 07000 C SOL UTI 0 ;\1 07000 C S I HUL TAN E 0 US LIN EAR E CUt\ T I Ur,; S trJ I THe II tv; P LEX (J F COEFFICIENTS USING CRAMERS RULE 07000 C PROG NO 223-63 07000 C o E T E R'''11 NAN T S 07000 C E L E,.,1 E NT SEN T ERE D R0 \.J - \..' I S E E V !~ L U ATE 0 B Y THE T R I A N GU L.4 R :'1 ET t-l (J [) 07000 C 07000 C AR-REAL PART OF DETERMINANT 07000 C A I - I ~1 AGIN A~~ Y PAR T 07000 C REA LAN D Hi; AGIN ARY PAR T S 0 F A C (j F. F FIe lEN T E ~~ T ERE D 07000 C FR-REAL PART OF CONSTANT 07000 C FI-If·1AGINARY PART CF CONSTAt-.!T 07000 C N-ORDER OF 07000 C N SOL . - 07000 o It·') ENS ION 07000 DIMENSION XR(20), XI(20) 07000 CONV 07048 07072 07096 07118 07142 • 07296 07320 (J F Ut: T t=. R ~/I I NAi'.J T ~:lf'\ S AI·H:: C .~ R j) TERM TER~~ THE SY STEt'i N U MB ER 0 F S (J L UTI 0 f\J S fH:': S I ~ E D (E qUA L T 0 [j R AR ( 2 0 , 2 0 ) , h I ( 2 () , 2 a ) , F R ( ;: 0) , F I ( ? 0 ) , ~.! R ( L F: :-; S T H A f'~ f'J) 2 0 , 2 () ) , \,.r I ( 2 0 , 2 0 ) = 180./3.14159265 500 READ 101, NSOL READ 101, N 10 1 FOR t~ AT (I 3 ) PUNCH 104 104 FOR tv, A T (I / 1 5 X 41 H S[) L UTI 0 N 0 F S I MU L T A j\j E LJ U S LIN EAR t: 0 U :4 T I (J (\J S) PUNCH 105 105 FORMAT (22X, 25HwltH COMPLEX CUEFFICIENTS/29X12HPROG. 223-63/) o I 07552 07576 07712 C PUNCH 119, N , 119 FORMAT{5HORDER,I31/5X,4HREAL,11X,9HIMAGINARY) INPUT AND PUNCH MATRIX 82 p" o t "J'P!!SP*W'fttMW!!'NP' "t ., nmwW',w.m.Inni.uRn W' "'Ynrrm'Hl''tW·h'' '2fif"#rWij[ddS'HiflW'PW.... ··MEWWFiiiE'S·W-Y·!Wwi fijiMHi·WI .t'''ttffl'fm" ]HZ' 'Nrnil!!'!'" .·Mt!'W'BW'''\'''!l"··fHf 'MfflWt''t'Urt'tN M! "8 nffYf'l'lm." .wlm WilE j7' J5!F"liWw 'Pt'i ['i? t [! PAGE 02 07712 DO 1 I 07724 DO 1 J 07736 RE/~D 07892 = = 1, N 1, N 100, AR(I,J), AI(I,J) 100 FORMAT (E14.8, EI4.B) AR ( I , J ), AI ( I , J ) 0"7 92 0 P U i\l CHI 16, 08076 C SET UP WORKING MATRIX 08076 08232 WI ( I , J) 08388 1 CO I'll T I ~.J UE PUI\!CH 08460 c· 103 084B4 08532 AI ( I , J ) ~ 103 FOR ;·i AT (/9 H CON S T t\ N T S ) I N PUT AND P Uf\; CH CO r·J S T i~ NT S C 08532 DO 2 I = 1, N 08544 READ 08628 2 100, FP,{I), PUf'~Cr! 116, FR(I), FItI) Fl(I) 4 08748 08784 08832 =0 LIM = N-l SIGN = 1.0 MN 50 08868 C DIAGONALIZATION UF Ut:TU<.:·'iI~jANT 08868 DO 25 I = 1,Llr'~ 08880 OB916 08964 o 09300 L = 1+1 1 8 0 E N = WR ( NUH , I ) ~:~ vJ R ( 1': ur"l, I ) + \. .J I ( NUM, I ) ::~ ~,I I (N U""\ , I ) IF(OEN) 14, 15, 14 09356 14 If(NUM-I) 914,914, 24 09424 15 NUr-'1 = NUi';+ 1 83 " , . mm,...i,,, PAGE 03 o 09472 IF (NUtvi-f'J) 09540 53 I F (f·1 N· ) 4 , 09596 4 PUNCH 110 09620 PRINT 110 09644 09788 18, 4, 18, 53 5 110 FORMAT(//42HZERO DETERMINANT - USE DIFFERENT METHOD OF9H SOLUTION) STOP 09836 24 DO 16 J ;:: 1, N 09848 WRT ;:: WR(I,J) 09944 vJ I T ;:: WI(I,J) 10040 WR(I,J) ::: WR ( NUt:i , J ) 10196 ~~ ::: ~~ 10352 WR ( N Ut·l t J ) ;:: 10448 16 WIH.JUM,J) I (I , J ) I ( NLJ t:j , J ) ~'!R T C'·' , 10580 C ;:: WIT . I CHANGE SIGN OF DETERMINANT IF ROWS ARE INTERCHANGED 10Seo SIGN;:: -SIGN 10628 914 DO 23 J 10640 WRT ::: WR(J,I) 10736 WIT == ~~I(J,I} 10832 DO 23 K == 10844 Xl ;:: WR T t-J R ( I , K ) - WI ( I ,K ) ::q'J I T * 11072 X2 ;:: WR(I,K>*WIT+WI(I,K)*WRT == L, N I , N 11288 11660 12020 WI ( J , K) 23 CONTINUE 1.2128 C 12128 ;:: ~J I ( ,-I· t K ) - ( WR ( I , I ):~ X2 - \~ I ( I , I ) ~:, Xl) IDE N ADJUST ,"4AGNITUOE TO AVUID OVER OR U~~DERFLO~J 5 IS;:: 0 o ts O>AGE tt 't , In PII""YPMP=Y'R"fll!flNtMPWUW W "f!!1'"!W'E'l'iI'f1f'P'flHiff'YR!P"MJ!'!"M!UWI'E'P ''''wi'f',,'1'1 INS n 'PWw.;iI&WbkWr r ,WiWWM""#tt""'1!prWWtiitMi1tl"ffiltn "'15ft'mCW JI 'fl'i'MTI'IMWMlWifW'Pi:'1*iI'!!f·'''' I'. 04 12176 = DO 200 I 12164 220 AB\4R = 1, N ABSF(WR(I,I) I F ( AS ~~ R ) 200, ·200, 213 12272 211, 200, 210 12328 213 IF(ABWR-1.) 12396 211 IF(ABWR-.l) 212, 200, 200 12464 212 TTEN = = 10. 12500 IS 12548 GU TO 214 12556 210 TTEN = • 1 12592 1 S = I S+ 1, 12640 012808 12976 12984 214 15-1 \~R(I,I) = WR(I,I)*TTEN WI ( I , I ) = WI(I,I)*TTEN GO TO 220 200 CONTINUE 13020 C EVALUTION OF DETERMINANT TAKING PRODUCT OF DIAGONAL ELEMENTS 13020 DO 7 I 13032 J 13080 PROD R 13428 PRO D I 13764 \tJ R ( I , I) 13860 0 r • -riP t d ' = 2, N ;: 1-1 =(WK ( J , J ) *\-J R ( I , I ) - WI ( J, ,J ) ::q~ I ( I , I ) ) =(~J R ( I , I ) ~('fJ I ( J , J ) + ~'I I ( I , I ) ~:~ WR ( J , J ) ) ;: PRO 0 R 7 WI ( I , I) = P ROD I = PRODR*SIGN 13992 PRODR 14040 PRODI ;: PRODI*SIGN 14088 DET 14184 IF (DET), Ill, 121, III 14240 = PRODR*PRODR+PRODI*PROOI 121 IF(MN) 4, 4, 111 ·e 8. t.t c PAGE 05 14296 111 PUNCH 115 14320 115 FORMAT (/2X, lOHREAL PROD., 7X, lOHIMAG PROD. 7X, 11HDETERMINANT) 14478 PUNCH 116, PRODR, PRODI, DET, MN 14538 116 FORMAT (E14.8, 3X, E14.8, 3X, EI4.8, = 14602 ISO 14650 PUNCH 117, IS 14674 IS+IS 117 FORMAT (lOX,44HMULTIPLY REAL AND IMAGINARY PRODUCTS BY I.OE, 15) 14818 14842 IS) PU~·jCH 118, I SD 11 8 FOR j\1 /..\ T (lOX, 28 Ht·1 ULTIP L Y 0 ET E Rt·1 I Nt~ i" T 3 Y 1. () E, I5 ) 14954 IFU'lN) 8, 9,8 15010 C DETERMINANT OF THE COEFFICIENTS IS SAVED FOR LATER COMPUTATIONS 15010 9 BOT :: = DET 15046 ISZ 15082 PRDIZ :: PROD! 15118 PRDRZ 15154 PHID =ATAN~(PRODI/PRODR)*CO~V 15226 AMAG = SQRTF(DET) 15262 PUNCH 109, PHID 15286 109 15368 15404 15486 C 15486 C IS = PR()OR FUR~AT(/13HPHASE ANGLE =, E14.8, 7HDEGREES) PUNCH 125, AMAG, IS 125 FORMAT(llH~AGNITUOE =, E14.8, 7H SET UP DETERMINANTS WITH * 1.0E,I5/) COE~FICIENTS OF UNKNO~NS REPLACED BY KN OvJN T ER ~1S 15486 DO 10 MN 15498 DO 11 I 15510 DO 11 J = 1, NSOL = 1, N = 1, N 86 tttttnttttrttttrs rts 0 • s • tt _ tsn" , WI "'j'ffii"B'WP'5M IL'S"),IY'.H"'rr!P'w :l!WJYIII8!·"''ij'''!m' N' , r::" MNW/.'MDMI'H'fW'tMrW''!'Wf!!! Itt n "M.5BM.' TIl! 'c"'r:"lJIl',!,U"MnfB'" !IY"FWF2!"'f'".'N"WW'W'WW+rt' I' ''l"!I!!'UmWMMII'f-rru'tUH'''fP'm:r 'Yn!'fltwmP(": PAGE 06 ~~ R ( 15522 = AR ( I , J ) = AI(I,J) = 1, N I,J) 15678 11 WI ( I , J ) 15906 DO 12 J 15918 ~, 16038 12 WI ( J R ( J , t~1 N ) ,(if: I'~ ) = FR(J) ;: F I ( ,J) 16194 GO TO 50 16202 C SOLUTION OF THE UNKNOWNS 16202 C POWER OF 10 READJUSTS 16202 8 VAL = = = XR(hN) 16430 XI(MN) 16574 10 CONTINUE PUI'JCH ~AGNITUDE (lO.**(IS-ISZ)/SOT 16298 016610 TocnR~ECT (PRODR*PRDRl+PROUI*PRUIZ)*VAL (PRORZ*PRODI-PROOR*PRDIZ)*VAL 106 1 0 6 F 0 Kt·1 AT (I /1+ 6 H S () L UTI Uj'J S DF THE S I t'-i Ul T!\ ~\! E 0 US lIN FAR. E (~1 UJ'. T ION S / ) PUNCH 3, N 1676h 16790 3FURMAT(5HORDERI4//~X4HREAL1IX9~II~AGINARY7X11HPHASE = 17026 DO 13 I 17038 IF(XR(I» 17118 123 PHID 1715't GO TO 17162 1 2 2 PHI 0 17282 1 24 17486 17594 01767() 17706 17730 = ANGLE7X9HMAGNITUDL/} 1, NSUL 122, 123, 122 90. 124 = ATAN F ( X I ( I ) / XR ( I ) ) ~:: CON V Ar", AG :;: S(.) RTF ( XR ( I ) ;:~ XR ( I ) + XI ( I ) ;:' XI ( I ) ) PUN CH 1 20, XR ( I ), XI . ( I ), PHI 0, A1\1 AG 120 FORMAT(E14.8, 3X, E14.b, 3X, F14.8, 3X, E14.8) 13 CONTINUE PRINT 900 gOO FORMAT(31HPAUSf, PUSH START FOR NEXT CASE) 8" o PAGE 07 17816 PAUSE 17828 GO TO 500 17836 END SYMBOL TABLE 39999 SIN 399b9 SINF 39979 COS 39969 CDSF 39959 ATAN 39949 ATANF 39939 EXP 39929 EXPF 39919 39909 39899 39889 39879 LOG LOGF SQRT SQRTF 39869 39859 35859 31859 31659 ABSF ABSFF AR AI FR FI 31459 WR 27459 ~~I 23459 XR XI 23059 CONV 23259 23049 23039 23029 23019 23009 22999 35869 31869 31669 31469 27 L.. 69 c 23469 23269 23069 18000000+03 31 Lt15926+01 000 ),'<0500 *0101 *0101 221)89 NSOL 22979 N 22969 *0104 22959 *0104 22949 ::::0105 22939 ~::o 105 22929 l!~O 119 22919 ~~o 119 22909 :::0001 22899 I 22889 J 22879 *0100 22.869 ~:~O 100 22859 ~:(o 116 22849 ~(O 116 o >tt ttt OPAGE 08 22839 22829 22819 22809 22199 22789 22779 22769 22759 22749 22139 22729 22719 22709 103 103 '::0002 ~::O ;:~o MN 0000 L I ~1 0001 )::0050 SIGN 10000000+01 *0023 NUJlt1 L ;::0018 22699 DEN 22689 22679 22669 22659 22649 001 ':'0014 ;:::0015 *0914 22639 ;::00 S 3 ::::0004 *0005 *0110 t,:0 110 :~OO24 22629· 22619 022609 ··22599 22589 ~:OOI6 22579 ~'JR T 22569 ~-J IT 22559 K 22549 Xl 22539 X2 22529 22519 22509 22499 22489 22479 22469 22459 22449 22439 22429 22419 22409 22399 22389 22379 22369 22359 022349 . 22339 22329 UOz IS 200 *0220 ;~o A8 ~'JR *0213 ;~o 211 *0210 10000000+00 *0212 TTEN 10UOOOOO+02 ):~o 214 ~:~OOO 7 PRUDR PRODl DET :::0111 *0121 *0115 )~o 115 22319 ISO 89 0 PAGE 09 22309 ,~o 22299 ::~ 11 7 011 "7 22289 :;'0118 22279 :~o 118 22269 *0008 22259 ::~OOO9 2224:9 22239 22229 22219 BOT PRDIZ 22209 PHID 22199 22189 22179 22169 22159 22149 22139 22129 22119 22109 22099 22089 22079 22069 22059 22049 22039 22029 22019 ISZ PRDRZ A~1AG ~;0109 109 ~:~O 125 ~:~o >;'0125 10 ::'0011 *0012 ::~OO VAL :;:0106 01 () 6 *0003 ::~ ~'OOO3 Ci :',::0013 ::(0122 :'''0123 90000000+02 ~:~O 124 ;:~ 01 ? 0 22009 ;;'0120 21999 !:~09 GO 21989 ~~O900 c 90 o APPLICATIONS OF NUMERICAL FILTERS IN THE POWER SPECTRAL ANALYSIS OF STATIONARY TIME SERIES BY ALEXANDER A. J. HOFFMAN TEXAS CHRISTIAN UNIVERSITY FORT WORTH, TEXAS o Presented At Western Region 1620 Users Group Meeting Denver, Colorado June 17, 1964 o 91 o We will focus our attention on the spectral analysis of finite length recordings of a physical process which is assumed to be random in nature. For deterministic functions such as periodic and aperiodic functions a harmonic analysis is usually carried out by Fourier series analysis and by Fourier integral analysis, respectively. The discrete line spectrum for a periodic function and the continuous spectrum for the aperiodic C~·i I function may be determined analytically because these deterministic functions are "known for all values of time". Random series are a class of functions which are not deterministic and do not lend themselves to the same harmonic analysis techniques used for deterministic functions: that is, statistical methods must be used. The Tukey techniquei which is used here, is applicable o to random time series which very closely approximate a ~2 1'.. li "l!!",'WW'l!1itrrS!l'F1'ffi"MVt'tf Wffil§wl',!"W'!,5!1 UP T'"Mn'IBeftHfti&'Ui:'tWP=tHfbW,jijiwli+"tMi ' '')' ' 'I MM,"l"¥ttif:@'j"jHjjt*ij:r#iijni'ti"!ijt"ri;WhLf"i'jh"r"r II II -2- o stationary random ergodic process. This computational procedure yields the variance spectrum of a time series. Other names for the resultant computation are power density spectrum, second-degree spectrum, or quadratic ~ll spect~umi of which refer to the distribution of variance as a function of frequency. One begins with a recording of a physical process which is assumed to represent a sample of a random process. o The record must be free of "pure tone" or periodic components and transients. After sampling the record at equi-spaced intervals the linear trends and average should be removed. Briefly, the Tukey method consists of computation of statistical estimates of the spectrum of a finite discrete time series by a numerical approximation of the o Wiener-Khinchine equations. The procedure involves two 93 -3steps. First, one computes a set of mean lagged products of the time series. Another hame for the set of mean lagged products is the autocorrelation function. The raw power spectral estimates are computed by application of a discrete finite Fourier cosine transform to the autocorrelation function. This transformation gives the desired frequency domain representation of the time series. Systematic statistical errors resulting from use of a finite amount of data appear in the raw power spectral estimates. c The Tukey technique to obtain improved spectral estimates involves a smoothing or refining operation performed on the raw estimates. Slide 1 shows the Tukey equations. Slide 2 shows an example of a time series to which one might apply the Tukey analysis. Slide 3 shows the power densi ty spec-truro of the time o r tttt st. 1"z -4- o series. Eighty percent confidence intervals are shown. Your attention is directed to the fact that the power density qraph has an upper bound at a point marked fN and that no power estimates of higher frequency are plotted. This upper band set is known as the Nyguist frequency and is a function of the length of the sampling interval. A full discussion of sampling theory is beyond the scope of this presentation. However, a few brief remarks are in order. When a continuous function is sampled at equi-spaced intervals, the question should be asked: "How well will the discrete set of sampled values represent the original function?" A continuous function of time is completely determined by its values at equally spaced intervals provided that the continuous function contains no o frequenctes higher than, say, W cycles per second, and · 9 t.) ~ -5- the ordinates are given at points spaced 1/2 W seconds apart, the series extending for all time. G This is a statement of the popularly referred to Shannon theorem. Under consideration here is an analysis which is to be based on sampled values obtained from continuous records which are not infinite in extent and are not band limited. Analysis based on finite amounts of data is common to statistical work. c Of immediate concern is the selection of the sampling interval and the problem of aliasing. Consider two sine waves of equal amplitude, but different frequencies. (See Slide 4) Attention here is directed to a particular set of sine waves, differ¥ing in frequency, but having a common set of equally spaced sample values. Thus, given only the sampled values, a sine wave of a given frequency may o .. m• t t -6- o be confused with a sine wave of higher frequency. Specifically, if a harmonic time function X(t) is sampled at equally spaced time intervals At, then a frequency =_1_ 2~t called the Nyguist or folding frequency, exists such that the functions with frequencies f + nfN ' for n = 0,2,4, .•. , o are not distinguishable. Obviously, then, power contributed to a power spectrum at a given frequency f cannot be distinguished from powers contributed by frequencies f + n f of frequencies is known as aliasing. N . This translation If the data actually contain power at frequencies greater than f N , this power will be "folded back" into the principal band which extends from 0 to f N . o Power that is folded back results in a distortion of the true power spectrum in 97 · _._ ...•..... _ ........ _.... '''r ... _""",,,,,,,,',.,,,,,,,,.''-_',,",",_ , ~""""""" _ _ _ _ _ _ _ __ -7- o the principal band. To make the effect of aliasing negligible it is necessary to select a sampling interval IIsmall enough ll to place the Nyguist frequency beyond all significant power contributi ons. Associated with each spectral estimate there is a confidence interval which depends on the number of degrees of freedom in the computation. If one assumes o the distribution of the data to be Gaussian and that the distribution of the variability in the spectral estimates follows the so called "chi-square" distribution, then the number of degrees of freedom may be computed by the convenient formula: k = m~ (N - ~ ) where k = number of degr~es N = number of sampled values m = number of the maximum log o • tsstrtr err -8- o The confidence intervals are then computed using the number of degrees of freedom. As the number of degrees of freedom is increased the confidence intervals decrease in size and the computed estimates are more reliable. The number of degrees of freedom is, generally speaking, directly proportional to the number of data points and inversely proportional to the maximum o number of lags. Acquisition of more data may be impossible or economically unfeasible and reducing the number of lags reduces the number of spectral points in the frequency range from zero to the Nyguist frequency. This brings us to the point of this paper. In many physical processes the power density decreases very rapidly with increasing frequency. Often at the higher frequencies the power density of the process under o investigation is of the same order of magnitude as the -9- noise background. o One must sample the processes often enough to avoid aliasing which would cause the noise to "fold back" into the frequency range of interest. Then one must take many lags and compute many power density estimates in order to have a good look at the lower frequencies. The consequences of this are large confidence intervals and much computation. In order to get around this problem one can operate o on the original sampled data with a linear operator which is often called a numerical filter because of its mathematical resemblance to an electrical filter. Through use of filters one can change the frequency spectrum in a known and desireable way. In particular, a low-pass filter may be used to suppress the power near the Nyguist frequency and not significantly disturb the low frequency spectrum of a time o series~ 1HO __* .m 3.T'""" HttM -10- o Slide 5 shows a' power density spectrum computed before and after low-pass filtering. Slide 6 shows a comparison between the mathematical model of an electrical filter which operates on a continuous electrical signal and a linear operator (a numerical filter) which operates on a set of equi- spaced sample values of a time series. Note that the time domain representation of the electrical filter is characterized by W, the impulse response or memory of the filter. The time domain representation of the linear operator is simply an array of numbers. In the frequency domain both the electrical and numerical filters have representations called the frequency response. It can be shown that the numerical filter is simply a numerical approximation to the mathematical model of o the electrical filter. i 0 j -11- o Slide 7 shows a plot of the coefficients of a low-pass filter. Slide 8 shows the frequency response of both a high-pass and a low-pass filter. After the time series has been operated on by say, a low-pass filter, the new time series may be resampled usina a larger sampling interval. That is, the set of sampled values may be decimated by taking every other value, every third value, etc. c A new lower Nyguist frequency is associated with the power spectrum of the new time series since the new sampling interval is larger than the original one. The low-pass filter has suppressed the power at the higher frequencies and thus all but eliminated possible distortion caused by aliasing. Now the low frequency range may be investigateQ using fewer lags and thus keep the size of the confidence intervals small. 1 t) 2 o -12- o After the power spectrum has been computed the effect of the filter is removed using the frequency domain representation of the filter. In various applications high-pass, band-pass as well as low-pass filters have been used. Such computations are used in geophysical applications such as analysis of temporal variations in the earth1s magnetic field and o in biomedical applications such as analysis of EEG recordings. Slide 9 shows a macro-flow chart of a computer program, written in 1620 Fortran lIz to accomplish the computations discussed in this presentation. Listings of the program are available from the author. (User 5130). o 1 0 ;) z r ., "{ --4~ >- t- -I~ VJ + H Z uJ o d.~ uJ g ~ o uJ Z " H u.. ul -;> + ? + f > ~ -\ti " 0-0 o£ o Mr_ tt' . " r t E"',mnl"fI,.,,g:\rrW'P·''Z'T!!',,'JijM'MWMMH'tS''= l ' o .::::: > o o - - a:: o I I ! ! • ~ ...=' ~ U L&I A. \I) >..... ..... '"z uJ o G ~ UJ ~ o Q... -I -o o o FIGURE 9 SINE WAvES OF rlF"FERENT ~ FREOUENCIES WITH THE SAME SE T OF EOUALL Y SPACED SAMPLE \ALUES '\Q! 0 SLIl)£ + / r o 1{'~ SLII)£S o o ~ 0 - & ..aCtTKIC-"L 'i H (ot)· PSLTall . ~H : \1"(to) e . ~lI'~t H (*) ... f.~ NUM.a,.·-&. . ~~_"ij~ ~". ~; Ca (to -"t) 'Nh.\ dt G.(-t.)~) • - e ,, -2;-. u O 'f I' zl&J : ~ I ::::> <-? g I t Ii l I _4L_--~-",---, . -.. . . . . . . ._ . __.-..1..------.....-.--.,tw. . (j FRE(JUENCY o IN f,./2 ~r-, CYCL.ES PER KILO.;x-.CONO FtGlJR E 14 THE F,REOUENCY F~E. SPONSEOF LOW~ PASS' .. I\~Nn H'(;H . . ~'AC;S FII.'TERS, TABLE' I. t 1- J c o tt H tr " » tn.t• • h!' j t ttt T' '3 .r, ltfrnu ) t' .'")" . t I' , Ii ! ! man IBM 1620 ASSISTS STUDENT COUNSELORS AT JUNIOR COLLEGE o Paul S. Chan IBM CORPORATION 3610 - 14th Street Riverside, California May 18, 1964 o 11 3 . g'&,"' TABLE OF CONTENTS • I I 1. Abstract 2. Introduction 3. Purpose of the Study 4. Data 5. Method of Analysis 6. Results 7. Sununary 8. Appendices (a) (b) (c) (d) Correlations between Test Scores and Final Grades Scattergrarn of SCAT T vs. Chemis~y IA Scattergram of Mathematics Placement vs. Chemistry lA Smnmary of equations. C: c - r Mm.t s t ,tt n.' Int. ! . t't.s _ $ I. It __ *ttttrrt_r. htttrtsm o ABSTRACT IBM 1620 ASSISTS STUDENT COUNSELORS AT JUNIOR COLLEGE The present study reports findings, based on the com.puted results from the IBM 1620, concerning the extent to which test scores on the college freshman testing program - such as the ACE, SCAT, Co-operative English Tests - are able to predict academic success or failure in specific junior college courses. Scattergrams have been created for those correlations of highest significance to assist counselors in estimating the incoming student I s aptitude for college level study and in making a more accurate appraisal of the student1s competence in a particular subject area. o Paul S. Chan May 15, 1964 o 11 5 * INTRODUCTION Unlike private colleges, the state colleges, or the state university, California I s public junior colleges are required by law to admit any resident of their districts who is a high school graduate or who is over 18 and able to profit from instruction. Junior college adrninistrators have interpreted this as m.eaning that they cannot deny admission to any applicant who has reached his 18th birthday, although virtually all now have retention policies which deny re ... enrollm.ent to students who fail to m.aintain a "satisfactory" grade point average ~ At one t:ii:me, many administrator s interpreted the legislative m.andate to m.ean that they could not set any qualification for registration in any class. An apparent change in legislative sentiment has com.bined with the realities of P?st-war enrollment pressures to cause most junior colleges to search for som.e equitable means of screening from. classes (particularly from. transfer classes) those students who have little opportunity to succeed. The freshm.an testing program. has been an established practice at Riverside City College, a public junior college, for the past years. Although the counselors and admissions officers have been m.aking extensive use of these tests to assist in laying out the academ.ic path of m.any students, there have been no attem.pts until recently to m.ake regular evaluations of the measuring instrum.ents in use. Recently an IBM 1620 was installed at the college. One of the first projects to use the system was ab attem.pt to determ.ine the relationship between the test scores and the final grades in specific courses. It is anticipated that the results will im.prove placem.ent of students in appropriate sections or courses, and selection of students for particular areas of concentration or preprofessional training. -2- c sH o ctt: mitt ,.. 'Nt·., II"$"n2'",", ,,,,".'MY,.,!!,,,:·!!" 1, PURPOSE OF THE STUDY The battery of tests - ACE, SCAT and others - were administered to the in-corning new students for the dual purpose of counseling and placement. Since this investigation was the initial application, the present study was to demonstrate the validity of the battery for these purposes. Another aim of the study was to modify the battery to include only those tests best suited for the screening program. Excessive overlap of abilities measured by one test and those measured by another results in a waste of the student 1 s time. Also, too great an array of scores for academic counselors might prove more confusing than helpful. It was anticipated, too,' that critical cut-off scores could be developed for 'eachtest, making it both practical and possible to advise the individual student, upon the basis of his score, just what his chances for success of failure in a specific course would be. -3- o 1j 7 DATA This study involved over BOO students who were enrolled in Psychology 49, a freshInan orientation course, and who had completed one or more of 2S.courses which the college wished to examine. There were fifteen predictor s. The se included: (1) three scores from the ACE (Quantitative, Linguistic, and Total) (2) the R. C. C. Arithmetic Competency Test of 40 items (3) three scores from the School and College Ability Tests (SCAT, Quantitative, Verbal, and Total) (4) six scores from the Cooperative English Tests, Form lA-1960 EDITION (Vocabulary, Level of Comprehension, Speed of Comprehension, Total Reading, English Expression, and Total English) (S) overall high school grade point averages (to obtain this figure academic subjects and othe rs such as typing, speech, journalism, and music courses were used. Physical education, military science and driver education were not used. Shop cour ses were used where it was the student's high school major.) (6) academic grade point averages (to obtain this figure only solids such as English, foreign languages, math at the algebra and higher level, history and sciences, but not including general science, were used.) C ACE and ArithInetic scores were easily obtairied because they are a part of the placement battery of tests required of all new students. The SCAT and Cooperative English test scores were obtained by testing in the Psycho- I logy 49 clas ses and the two high school grade point averages were rather tediously obtained by employing an individual to compute the figures by hand. The courses included chiefly transfer courses with a few not-transit:-. ~ type courses and represented a cross-section of the major divisions within the college. DEPARTMENT Anthropology Art Biology Business Business Business Business Chemistry Chemistry Descriptive· Title Couroe No. 2 lA 1 lA lBA (hour) SOA (SlA) BlA (SOA) lA 2 Cultural Anthropology History and Appreciation of Art General Biology Principals of Accounting Business Law Elementary Accounting Business Mathematics Chemistry Introductory General Chemistry o -4- 1 1 ti =__ ·_t...t•••••____. . . . ._. . . ._ _ ._._·. . . . . 1II. . ._ DEPARTMENT 0 o Electronics English Geography History History History Math Music Nursing Philosophy Physical Science Physics Political Science Psychology Sociology Spanish t .r.. * ....l.!.&__ .W. . __ m.. .. .In.•.P'.. . . . ., .t . . .St. . . . . . . . .• ._. .• .•______ __J' _. . . .. . . .t . t .tt. ._ . 'I( _ _ _ _•_•~ _ _• _ Descriptive Title Course No. 51 1A 1 3 4A 6A 3A 20 1A 6A 1 2A 3 lA 1 1 Electrical Fundamentals of Electr onic s English Composition Introductory Physical Geography American History History of European Civilization Political and Social History of the US Analytic Geometry and Calculus History and Appreciation of Music Introduction to Nursing Introductory Philosophy Introduction to Physical Science General Physics American Political Institutions General Psychology Introduction to Sociology Elementary Spanish This battery of tests was originally selected to provide a basis for predicting over-all scholastic success and success in specific subject-matter areas: The ACE for general scholarship, with its Q and L sub-scores for areas of primarily quantitative and verbal content respectively; the Cooperative English Test for English and other areas which require considerable reading; the Mathematics Tests for placement in mathematics and allied physical science courses. Final grades of the students in each of the chosen freshman courses were compared with their scores on each of the tests, The courses were chosen from four areas: Language, Humanities, Social Science s, and Natural Sciences. o -4B- 11 9 ,I ...=................- .... -..:;;." ~=-===~.= =~~= c METHOD OF ANALYSIS Results of the test battery were separated into ten test variables: (1) Three scores, the Q, L, and T, were derived from the ACE; (2) Six scores froITl the Co-operative English Test; (3) Three scores, the V, Q, and T, froITl the SCAT; and (4) One score froITl MatheITlatics PlaceITlent Test. All the test scores and course grades were recorded in punched cards. An analysis prograITl was written in Fortran. Coefficients of correlation were co:mputed by the 1620 between scores on each of these tests and final grades in each course. To substantiate the validity of the results, besides the correlation coefficient, regression line coefficients, standard error of estiITlate, and standard error of regression coefficient b, the significance of r and of b were analyzed. A SUITlma ry of equations for these calculations can be found in Appendix D. c -5- 12u -..... N CORRELATIONS BETWEEN TEST SCORES AND FINAL GRADES MATH cases Q ACE L (EXp~ 297 268 326 196 233 397 491 tt 363 196 362 290 362 363 084 275 089 656 Tt 286 122 287 596~\- 407 457 11 441 018 100 404 051 144 169 488 ft 438 525 TT 612 ft 436 1: 185 217 400 121 209 342 218 326 ft 165 406 tf 348 245 376 306 241 4S3"~ 433 1: 160 416 tt 439 360 334 TT 033 316 ft 421 357 457 tT 582 TT 290 295 Humanities Anthropology Art lA Music 20 Philosophy 6A average 22 18 22 24 131 488 1: 463 fT 415 399 287 009 233 117 159 Social Science Geography l8 201 Natural Science Biology Chemistry lA Chemistry 2 Electronics Mathematics 3A Nursing lA Physical Science average 40 55 30 13 23 25 21 125 578~'" 127 534 ft 278 ~': tT o SCAT 5 ~Tr~ 044 101 CO-OP ENGLISH 3 4 ~Sp~ 064 599~': 2 ~Le~ 33 442~': 1 ~Vo) Language Spanish .............. T 189 386 TT 359 144 144 450 TT 218 596~\- l33 326 458~~ 6 (Tot Eng) V 179 232 040 546 ft 389 027 251 228 7941: 404 154 495 688~': 675 Tf 449~': 379 ft 106 048 421fT 405 324 310 449 260 473~': 425 368 503 ft 377 295 Q T 281 161 362 1T 182 478~': 737~': 091 367 382 396 065 330 049 388 330 045 203 159 519 TT 018 501 fT 299 336 548 11 616 1: 093 242 547~': 536 fT 116 598 1: 492 324 452 fT 538 ft 342 391 fT 208 411ft 392 413 579"\065 275 375 TT 266 Tf J.24 484 126 384 11 328 232 146 551 tf 501 305 294 528 ft 320 456~~ 375 fT 329 ft 405 fT 588 n 225 558"\491 385 Indicates .01 level of significance .05 level of significance ~ndicates o o o RESULTS The results of this study are reported in Appendix A, a table presenting the correlations between test results and course grades. Within each curricular area, the average correlation with each test is also given. All the correlations coefficients in Appendix A at the .0 I level of significance are marked with an asterisk and at the .05 level with double :crimes. Grades in some courses appear to correlate relatively well with scores on all the tests, while those in orther courses showed low correlations with most of the test scores. For example, biology has 12 out of 13 subscore s with correlation at either· .05 or .01 level of significance and chemistry has ten out of 13, whereas Spanish and electronics h?~Te only one out of 13 at .05 level of significance. Some explanations may be offered for this phenomenon. One is that the differential magnitude of the correlations depends partly on the magnitude of the reliabilities of the grades in those courses. Sectionings of a course will certainly be a factor to affect the magnitude. Another factor is that grades in some courses .are based on objectivetype examinations, while in others on a more subjective basis. The relatively high predicitive power of the mathematics placement test in the Natural Science Division is more or less expected. However, an almost equivalent result was found in the Q part of the SCAT Test. This is an indication that it may be possible to obtain the same predictive information from either of the tests, so duplication of student's effort can be avoided. It is quite unexpected that Spanish correlates with only the total score of the SCAT Test in the entire battery. Also, electronics correlates only at .05 level of significance, with Q part of the SCAT Test. It is possible that this phenomenon is due to the fact that SCAT Tests involves not only the psychological functions commonly measured by tests of verbal ability, but als~ a particular type of reasoning ability important in academic success which is not assessed by any othe r tests employed in the present battery. G The two parts, speed of comprehension and total reading, of the Cooperative English Test show high correlation with geography. This can be explained because of the fact that the Social Studies courses normally require more speed in reaidng and in comprehension. The significant correlation at .01 level between philosophy and the vocabulary part of the Cooperative English Test certainly implies the requirements to succeed in the course. -6- c 122 o In general, the six JR rts of the Co-operative English Test correlate relatively better than any of the three parts in the ACE Tests, with all the selected courses. This is illustrated by the ,r-values of .794 with Art, .458 with Music, etc. The tendency was noted also for correlations to be relatively high or low with reference to separate courses rather than to the different tests •. It was hypothesized that this phenomenon might be the result of difference among the courses in inter-section standardization reliability of grading, or use of objective examinations. o -7- o !, 1 r.. tJ SUMMARY o In this study of the value of a battery of aptitude and achieve:ment tests for the prediction of junior college fresh:man grades, test scores were correlated with final grades in a variety of fresh:man courses. The individual correlations appeared s:mall, but the relative predictive power was de:monstrated clearly. The following :major conclusions concerning the predictive significance of the present battery appear to be warranted: (a) Overlapping of tests in the battery used is evidenced, suggesting that such an extensive array of exa:minations is so:mewhat superflous and repetitive. Both over-all and individual course predic·tions could be :made with even greater accuracy with a :more abbreviated battery. (b) Fro:m the scattergra:m, it was found that it is feasible to deter:mine the cut-off score in screening and to obtain :more insight in the statistical probability of achieve:ment of a student in a particular course. (c) o Because of the s:mall nu:mber of cases in this particular study, a caution against placing too :much weight on individual test scores in guidance, selection or place:ment is in order. -8- o 1 • Wtrt m t."• • 2 =. *. tt tr t ms: • rtFn SCATTERGRAM OF SCAT T-SCORE F D 99-95 2 3 1 2 2 2 84-80 2 3 5 1 79-75 2 3 1 74-70 1 3 4 69-65 4 1 64-60 3 1 1 1 59-55 1 49-44 TOTAL A 1 89-85 54-50 o B 3 94-90 C C 1 1 14 14 19 4 3 VB. r II' r, $ CHEMISTRY IA _* tt h j1' c SCATTERGRAM OF MATHEMATICS PLACEMENT VS. • CHEMISTRY lA F D C 44-40 B A 1 39-35 3 3 11 3 34-30 9 7 7 1 29-25 4 4 1 24-20 1 2 1 C:! TOTAL 17 14 19 5 3 o - -----------~~-------- ----------- ------------- --------------------------------- ----------------------------------- - - - - - - - - - - • _r t me. r_'. * t 1m't'm!m. o St. _ • t rt t t = _ rtt t APPENDIX (D) SUMMAR Y OF EQUA TIONS (1) '- (6~)2.. \1\;3"('2- _ Variance s V\ CV\- \) = V\~~"1. C2..~)"1- _ Y\(\t\-\) (2)· Regression Line (3 ) V\ ~X1 - L~ 2.~ V\ L:-~ ~--=(_L>)( ") i b -- ~ ~~(h~-bLrk) Correlation Coefficient f1- ~ j[( h ~i').-h~ , Ln L y.~ -{2.-,c y. Jl (\ ~ 1~ - ( 'i J)~ J (4) Standard Error of Estimate (5) Standard Error of Regression Coefficient b / :")~. '- ! ~. S I)(r-~ ~ ,---' -,,-,-<-- (6) Signficance of r COITlpare l.n..[ with the critical value in statistical table for 2 variables and n-2 degrees of freedom. (7) Signfic anc e of b Compare with the critical value in statistical table for n-2 degrees of freedom. t -:..1* o -12- 127 mti' .. \ G\ ,I 1620 COMPUTER UTILIZA TION IN A WIND TUNNEL DATA ACQUISITION SYSTEM by Stanley E. Wisniewski Programming Operations Section NORTHROP DATA PROCESSING Presented to the 1620 Users Grou p Brown Palace Hotel Denver I Colorado 18 June 1964 NORTHROP CORPORATION o 120 • t trtt _m _ 1_.b. _M.b. .r· t ••• #*1 _ _ or •• NtH o TABLE OF CONTENTS PAGE . . . . . . . . . . . .. . . . . . ... . . ..... . .. .. . ..... .. .. . . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WIND TUNNEL TESTING FACILITIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA ACQUISITION SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMPUTER UTILIZATION •• . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COMPUTER HARDWARE MODIFICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT ...... _. 0, I' 1 2 3 4 5 6 SOFTWARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 . . . . . . . . . . . .. . . . . .... . . . .. . .. . . . . . . . . . . 8 RECORD FORMAT ....................................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. CONCLUDING REMARKS. 10 REFERENCES 10 o 1 2 !J i ",.-1 _______________________________________________ NORTHROPCORPORATION ABSTRACT This paper describes how the IBM 1620 computer was teamed with a high-speed digital data acquisition system and two tape units to perform on-line processing of wind tunnel test data. The total installation is located in the Northrop Norair wind tunnel complex comprised· of three tunnels: subsonic, transonic-sup~rsonic, and hypersonic. The processing installation provides a central data acquisition and reduction function for all three tunnels, even simultaneously when necessary. The high-speed data acquisition section scans, measures, and digitizes test data, in- . troduces identification information, and records the data on magnetic tape for instantaneous reading by the 1620, in a read-after-write manner. The 1620 then reduces the data into tabulations meaningful to the aerodynamics research engineers, enabling them to make early evaluation of test run results and to proceed with model changes if called for. During off-line operations, the computer is available for other applications, and has full control of the tape units. o 130 1 tr '"M'!?'ID!" .m. tt • • .• nrt reth••_ --______________________________________________ NORTHROPCORPORATION o INTRODUCTION Today, more than ever, competition in the aerospace industry is very keen and time is one of the most important elements to be utilized. For this reason, a company that makes use of wind tunnels nlust also have a satisfactory test data acquisition system and a means of automatically reducing the collected data as soon as it becomes available. In the following paragraphs you will learn how we at Northrop have improved our techniques in this area. Our wind tunnels will be described as well as our data acquisition system to which a 1620 computer is coupled. Also of interest will be the changes we designed into the 1620 computer to make it suitable for our applications and the programs we have written to fulfill our objectives. To the general public, wind tunnels are environmental chambers used to test model planes,but to the aerodynamicist, wind tunnels are probably the most superior devices used in aeronautical and aerospace research and development. Because of modern wind tunnels, today's test pilots are no longer the nerveless stunt men of the past, but professional engineers. Wind tunnels offer both fast and accurate data as well as the ability to simulate the different types of atmospheric conditions of any time of day or year. However, they are by no means new tools. Years before the Wright Brothers famed flight at Kitty Hawk, wind tunnels, crude as they were, gave valuable aerodynamics data which proved the feasibility of powered flight. The original wind tunnel employed by Orville and Wilbur Wright is on exhibit at the Air Force Museum, Wright- Patterson Air Force Base in Dayton, Ohio. o 131 2 :.::...;,:;..;.,'.,.......- j ....... -.-.~.'""" ..; . " 1 ~ I ;1 ~~~~~~~~~~~~~~~~~~~~~~~~_NORTHROPCORPOR~ION j C'I~ • • WIND TUNNEL TESTING FACILITIES Dominated by its 100,000-cubic-foot vacuum sphere is the supersonic-hypersonic wind tunnel facility at Hawthorne, California. This space age test facility provides test velocities from Mach 0.5 to Mach 14 with temperatures to 3000 degrees and simulated altitudes to 200,000 feet. To my knowledge, no privately-owned wind tunnel in the United States can produce the combined heat, pressure, velocity and run time that are obtainable with the one at Hawthorne. This relatively new, dual-circuit facility provides a greatly expanded capability for aerodynamics testing on advanced aircraft, missiles and space systems. It consists of two separate wind tunnel circuits: transonic-supersonic (Mach 0.5 through Mach 5) and hypersonic (Mach 6 through Mach 14). Design models can be tested for periods of at least 30 seconds in the supersonic circuit and up to one minute in the hypersonic circuit. The hypersonic tunnel can accommodate up to six 30-second runs each eight-hour shift. More test runs of proportionately shorter duration are possible. Test sections, in which the models are mounted for aerodynamic study, measure two feet square in the supersonic circuit and 30 inches in diameter in the hypersonic circuit. A special "free jet" section in the hypersonic circuit allows removal of a model from the air flow while air flow is being established, thus protecting the model from excessive heat loads. The pressing of a button promptly injects the model into the flow stream. In a transonic or supersonic run, air passes from storage through a settling chamber (to smooth the airflow and remove any turbulence), is expanded through a nozzle (to establish Mach number), flows through the test section and then is forced through a "second throat" to reduce its velocity and to recompress it to atmospheric pressure before it exhausts through a muffler. C> In a hypersonic run, air must be expanded so much (to achieve the higher velocities) that its temperature could actually be reduced to a point where the air would turn to liquid. To prevent liquefaction, an electrically fired heater containing a 16-ton bed of 3/8-inch alumina pebbles heats the air to temperatures as high as 3000 degrees Fahrenheit before it reaches the hypersonic nozzle. When the air is cooled by expansion, its temperature is therefore still high enough to keep it from liquefying. From the test section, the hypersonic air passes through a "second throat" as in the supersonic circuit, to reduce velocity and then through a cooler to remove heat. It is then discharged into a large 100,000 cubic-foot vacuum sphere. The vacuum sphere is essential to hypersonic operations in order to achieve the high velocities desired in the test section. With storage pressure fixed at 3,200 pounds per square inch, the required pressure ratio obviously cannot be met by discharging the "used" air to atmospheric pressure (14.7 pounds per square inch). A low-pressure atmosphere is necessary and this is the function of the vacuum sphere. About 100 feet from the supersonic-hypersonic facility and in another building is the' 7' x 10' subsonic wind tunnel which went into operation in the year 1956 and was used in the very successful development of the Northrop T-38 Talon supersonic trainer, F- 5 fighter, and Laminar Flow Control (LFC) airplane. During those tests, the output of test data was punched onto cards, carried to a remotely-located IBM 704 computer installation, processed and returned in a relatively-long turn-around time (normally about three days; on emergency basis about four hours). 3 o 132 C" o trinbtt trtaHtttrtntt **_ oem tMttt . . t _ trsttr« Ht --------------------.----------------------------NDRTHROPCDRPORATION DATA ACQUISITION SYSTEM Today in the same building that houses the subsonic tunnel, is the data acquisition system, which we are very proud to possess. It was designed and built to our specifications by the Astrodata Corp. It serves all three of our wind tunnels. The data from any two of the three remotely-located tunnels can be transmitted to this center simultaneously. o The analog data, supplied by transducers at each of the tunnels is digitized by an analog-to-digital converter (ADC) in the central data system. The digital data from the ADC is then sent to the formatting generator where it is joined by other digital data from the model-position encoders, the time-of-day clock and also the switch settings from both the transmitting site and the central data system. The switch settings provide fixed information such as the barometric pressure, the test number, the run number and the date. The formatting generator then assembles and prepares the data for recording on magnetic tape. The records produced by the formatting generator are of variable length and automatically padded to contain an integral multiple of six characters, so that the resulting magnetic tape recordings can be used with both the 7090 and 1620 mM computers. The ability to read the system-generated tapes by the 7090 computer proved very valuable during system checkout, because the 1620 computer was not adapted to handle magnetic tapes until later. There are two types of records produced by the data acquisition system. The first of these is the title run record which identifies the test run by a test number, a run number, four parameters, the barometric pressure, the day and the time of day, and the model position by roll, yaw and pitch. The activation of the title push button switch will initiate output of a title run record consisting mainly of the above information provided through manuallyset, thunlbwheel switches. The second of the two types of records produced by this system is the data record. A data record is generated when the data circuit is closed (manually or automatically). The data record consists of an identification header, the time of day, the model position, and data from all site input channels programmed for the specific test. o 133 4 - - - - - - - - - - - - - - - - - - - -_ _ _ _ _ NORTHROP CORPORATION o COMPUTER UTILIZATION The 1620 computer employed is a Mod I with 40,000 core storage positions. It is equipped with most of the special, built-in features (indirect addressing, hardware divide, and floating point arithmetic). This computer is attached to the data acquisition system by an umbilical cord; it has been programmed to read and reduce the data as it is being recorded on anyone of the two magnetic tape units. The reading is accomplished in a readafter-write manner, termed "eavesdropping." The information is introduced into the computer by the read gap, which is pOSitioned a distance of .300-inch behind the write gap of a two-gap read-write head, almost immediately after the information is written onto the tape by the system. The normal function of the read gap, which is to provide parity checking during the recording process, was extended to make this possible. The two magnetic tape units used are Datamec D2020. These units are IBM compatible, using either 200 bpi or 556 bpi tape formats at 30 ips tape speed. The Central Data System (CDS) records at the 556 bpi density. Eavesdropping allows the computer to sample the data as it is being recorded without interfering with the recording process itself. During the eavesdropping or on-line mode, as it is sometimes called, all the tape units are under the control of the CDS. Upon receipt of a signal from the 1620, the CDS causes the first character and associated parity bit to be transmitted to the 1620. Each character and associated parity bit continues to be transmitted until the longitudinal redundancy check character (LRCC) is encountered. The corAputer cannot initiate tape movement by attempting to read a tape while in this mode; therefore, a read tape instruction hangs up the computer until the CDS moves the tape to record new information. Besides the eavesdropping mode, the computer is also able to operate in an off-line mode. During the off-line mode, a selected tape unit (any of the two) may be read or written by the 1620 as if it were its own. These two modes of operation are manually selected. C' Reduced punched card data is generally generated and plotted off-line during tests. An IBIVI 407 printer is also available in this center and is used to print much of the punched card output. c 5 ,rl .. H -- _. 91!!!M'V"'Il"mYW I 'U! 'p,.,,'!'?'! .,'.'UM'$I' 'nnw 'mI!liU mm"" "'!"11 ¥'ruWIIWIliWIfWt'W'"iHt/iit'+H ----------------------------------------------__ NORTHROPCORPORATION o COMPUTER HARDWARE MODIFICATIONS The 1620 computer performing the data reduction is unique. Three new instructions had to be designed and the computer modified to permit their use for this special application. The design and implementation of these instructions into the computer required . several months. In addition, other instructions were adapted to permit the reading, writing and other handling of magnetic tapes. The three new instructions pertain specifically to the use of magnetic tape. BST, backspace magnetic tape (36XXXXX01300), REW, rewind magnetic tape (36XXXXX02300) and WEF, write end of file (36XXXXX01200) Two instructions that refer to paper tape normally, RNPT, read numerically paper tape and WNPT, write numerically paper tape, were modified to read magnetic tape (RMT) and write magnetic tape (WMT), in the numerical mode. 0 '1 ,I RMT, read magnetic tape (36YYYYY00300) and WMT, write magnetic tape (38YYYYY00200) In order to allow for tape redundancy and end of file testing, the functions of the following sense switch testing codes were extended. BC 1, branch console switch 1 on (46YYYYY00100) and BC2, branch console switch 2 on (46YYYYY00200) When a BC1 instruction is executed, a branch takes place if either sense switch 1 is on or if a tape redundancy occurs. Likewise, the BC2 instruction also serves two purposes: a branch will occur if either sense switch 2 is on or an end of file mark is sep..sed. These two sense switches must be in their off position during magnetic tape operations. The redundancy and end of file indicators are not reset by any of these two instructions; they are reset only when the selected tape is pet into motion again. o 135 6 1 '.·" • - - - - - - - - - - . - - - - - - - - - -_ _ _ _ NORTHROP CORPORATION SOFTWARE Although the rnagnetic tapes normally may be read and written with FORTRAN coded programs by utilization of the paper tape statements, the tapes produced by this system can only be read by SPS or machine type programs. This is due to the various field widths contained within the records written by the system. The problem of reading tapes was quickly resolved by the writing of an SPS subprogram that could be called by and loaded with FORTRAN coded programs. The SPS subprogram was designed to operate in two modes. The first of these modes, as directed by the arguments of the FORTRAN program, causes a compacted system record to be read from the tape. Each of the fields of the record is then extracted and expanded to a six-character field width. Flags are placed over the leftmost positions of each of the fields and the fields are then transmitted to their prescribed COMMON locations as integers. The second of the two modes requires the subprogram to search through the tape (not used during the on-line operation) for a particular title run record that agrees with the run and test numbers indicated in the arguments of the calling program. When the appropriate record is found, the information fronl the record is processed in the same manner as it was in the first mode. If the record is not found an indicator is placed in a communication field, reserved in COMMON for this purpose. The two tape reading modes of the subprogram~ described above, are very useful and make the 1620 an even more important asset to the overall system. The first of these two modes provides the user with integer data that. is FORTRAN-compatible. The second mode, in addition to performing the same task as the first mode~ assists in retrieving previously recorded data. o 7 Itt .H ** ttttt ssttts tt M • tt. _. t -------------------------------------_____________ NORTHROPCORPORATION 0' ....1/ RECORD FORMAT The formatting generator produces t\·vo types of records and these are the title run record and the data record. The purpose of t he title run record is to provide identification for the data records that follow it. The title run record consists of 54 characters of the following information: Characters 1 - 5 Characters 6 - 12 Characters 13 - 24 Characters 25 - 32 Characters 33 - 35 Characters 36 - 40 Characters 41 - 48 Characters 49 - 54 c Information Retrieval Aid (IRA) Time of Day (TOD) Model Position (Pitch, Yaw and Roll) Parameters from the CDS to be used for computations or further identification Day (00 1 thru 366) Barometric Pressure Paran1eters from the transmitting site to be used for computations or identification Test and Run numbers The IRA indicates the record type and identifies the test site. also gives the number of channels (data words) that were recorded. In the data record, it The data records produced by the formatting generator are of a variable length. The length varies with the number of data words that are recorded. Characters 1 - 24 of the data record are of the same format as those of the title run record. Characters 25 and above represent data words. Each data word consists of four characters. As many as 100 data words can be recorded in one record. The record is automatically padded to contain an integral multiple of six characters. o 8 'f' .. I! r t! ~ IRG RUN C 2 1 1 1 0 1 1 o 1 X 0 X II I ) \ MODEL POSIT:ON TEST PARA. 2 PARA 1 y----- BAR. PRESS. ~. DAY PARA 2 50 49 48 X X X o 0 o 0 1 X X X X X X X X X X X 36 35 34 33 X X X X o 0 o 0 100 1 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X LX X X X_~ X X X X X X X X X X X X X~ ~~ - XX l"X PARA 1 ROLL ------"YAW PITCH TOO IRA ~-T~--~--~--~--~--~--~--~--~--~~--v-----~------~----~--~ 58 57 56 55 54 53 52 51 X X X X X -X o 0 o 0 o 0 1 0 XX X X X X f- 1 X 0 CENTRAL (FIXED DATA SWITCHES! ---,-- L R r-----"-----... C X SITE (FIXED DATA SWITCHES) 47 46 45 X X X 0 o 0 o 0 1 X X X X X X 44 43 42 X X X o 0 o X o 0 X X X 41 X 0 1 X 40 39 38 37 X X X X o 0 o 0 0 000 X X X X X X X X X X X X 32 31 X X o 0 0 o X X X X X X 30 29 28 X X X o 0 o 0 1 0 X X X X X X X X X 27 26 25 X X X 0 o 0 o 0 1 X X X X X X X X X 24 23 22 X X X o 0 o X o 0 X X X X X X X X 21 20 19 18 X X X X 0 o 0 o 1 X o 0 X X X X X X X X X X X X X~ ~~ X~ ~>< L~~ ~ X X X X » 17 X 0 1 X X X X 16 15 14 X X X o 0 o X o 0 13 12 11 10 9 8 X X X X X X X 0 o 0 o 0 o 0 1 0 o 0 000 X X X X X X X X· X X X X X X X X X X X X X X X X X X X X X X X X X -~~ 2< X;~ 2<_ X 6 5 4 3 2 X X X X o 0 o 0 o 1 X 000 X X X X 0 X X X X X X X X X X X X X X X X 1 1 0 1 1 0 1 0 CHARAC,Eh: ~ 2 } TRACt<: . I TAPE DIRECTION TITLE RUN RECORD Ii A .. { IRG L R - It MODEL POSITION DATA ~ --------., l~ I r ---, PAD DATA WORD 1 DATA WORD 0 if I: ROLL YAW PITCH TOO C ---~C 2 1 X 1 1 0 1 1 o 1 X 0 X X 1 0 ( I IRA It ~ 40393837363534333231302928272625242322212019181716151413121110987654321 X 1 1 1 1 X X X X X X X X X X X X X X X X X X X )( X X X X X X X X X X X 1 Xo 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 o 0 r-X o 0 o 0 X 0 0 1 X o 0 1 X o 0 1 X o 0 1 X o 0 1 0 o 0 000 1 X 0 o 0 1 ")( 1 1 1 1 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0 1 ")( o 0 o 0 X X X X X X X X X X X X X X X X X X X X X X X X X X o X X X X 0 X1 1 1 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 l Xo 0 o 0 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 0 ;HARACTER ; } TRAe, Z TAPE DIRECTION Q !I z DATA RECORD ::a Q "'D X. IN THE ABOVE ILLUSTRATIONS \:':..\~,'3 ~·'-'_l\T EITHt-.R A n OR A 0 MAY BE GENERATED. Q ::a "'D Q ::a t S z ....... c,.,", C1 (') ~ o • . t tr 'ern t • '!!'t zt_t • •F,z.fII. • _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ NORTHROP CORPORATION o CONCLUDING REMARKS This paper has described how an automatic data acquisition and processing system was developed to perform a vital function in the modern wind tunnel complex at Northrop Norair. The major benefits of this computerized system can be stated as follows: 1. It reduces wind tunnel data immediately when it is most needed. 2. It permits quicker and more effective adjustments to be made to the model within the test chamber. 3. It shortens the time spent in carrying out a series of tests. REFERENCES "The Programming Gap in Real Time Systems", R. V. Head, Datamation, February 1963. "Pitfalls & Safeguards in Real Time Digital Systems", W. A. Hosier, Datamation, May 1962. "Air in a Hurry", H. M. Karaszewski, Compressed Air, February 1963. "Flight Simulation Without Forward Velocity", Henry M. Karaszewski, Compressed Air, December 1959. "Northrop Inaugurates Mach 14 Wind Tunnel", Northrop Technical Digest, November 1962. o 10 r. F. _ _ _ _ _-.-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ NORTHROP CORPORATION °1 1043 OVERALL VIEW OF WIND TUNNEL COMPLEX, SHOWING SUPERSONIC AT LEFT AND SUBSONIC AT EXTREME RIGHT HYPERSONIC FACILITY C''\, " ) 12113 MODEL OF SUPERSONIC - 11 HYPERSONIC CIRCUIT c $ • ""WI"U""W'YM. 7. *. 'P J tr sm." rs _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ NORTHROP CORPORATION CENTRAL DATA ACQUISITION SYSTEM o SIGNAL CONDITIONING CABINET AT SUPERSONIC SITE 12 _r_t: o 1 1620 IPL-V A NON- NUMERIC PROBLEM SOLVING TOOL by Wendell Terry Beyer c An essay submitted to the Department of Mathematics of the University of Oregon in partial fulfillment of the requirements for the degree of Master of Arts Apr i 1 1964 1 li -') • o .__• *_ r r __ =_'tt mil t • •,' r m - Acknowledgments The development of the 1620 IPL-V system was done in part while the author was an IBM/WDPC Research Assistant at the University of Oregon. This assistantship was provided by the western Data Processing Center at the University of California. Many long hours on an IBM 1620 computer were freely provided by the University of Oregon Statistical Laboratory and Computing Center. o o 1 43 o Preface This paper is composed of three section. Section I introduces the need for computer languages similar to IPL-V, section II outlines the IPL-V language, and section III describes the IPL-V implementation for the IBM 1620 computer. A detailed description of the IPL-V instructions and a sample problem are contained in the appendix. A list of selected references is given at the end. c o I I tt ttmtssz b - rtt tt t n o .' .'W' S t "11,.",,==' 9'WYM'"YIti'P"":rrw'W'W",,,"'lnr :.' IN'f"'!J 'FZMiHW1iWi!IiiNjW,QIjWW'Brwft",w,["\!1'!Y'"''Jlmn III I Stored program digital computers were initially developed as devices for performing complex arithmetic calculations at high speeds. At first, the task of programming these machines was burdensome because all programming was done in machine language. However, programming languages were soon developed as an aid to the programmer, beginning with low level assembly languages for specific machines and eventually evolving into high level, machine independent languages such as FORTRAN, ALGOL, and COBOL. Due to the arithmetic origins of the computer, these languages were designed to assist o the programmer in the coding of arithmetic or numeric prob 1ems. For a long time, however, it had been known that the digital computer, with its ability to analyze data and take differential action, was not inherently limited in scope to numeric problems. Indeed the problem of translating source statements from a high level language like FORTRAN into machine code is itself a problem basically non-numeric in nature. Other problems for which computer solutions were sought include chess, bridge, analytic differentiation and integration, language translation, pattern recognition, study of learning and self-organizing systems, information retrival, o 1 45 2 1 Q theorem proving, and most recently theory developing. 1 As interest in these and similar problems grew, certain questions arose. Is the present form of digital computer, designed with numeric computations in mind, It necessarily the best for non-numeric problems? what better designs might there be? If not, Is it in fact possible to believe that one design will be capable of handling the majority of non-numeric problems? Is it possible to develop a high level language which will do for non-numeric computation what FORTRAN does for numeric computation? Today these questions remain largely unanswered. No one has succeeded in developing a high level language designed for non-numeric computing although work is being carried on in this area. Some computer designs have been developed which seem to yield a better method of attack on non-numeric problems than that afforded by numeric computers. To more fully appreciate the problems confronting the designer of a non-numeric computer, it is necessary to examine some of the common characteristics of the nonnumeric problems listed above. These problems cover a wide variety of topics and one might suspect that there is little in common among them; however, four character-. istics do appear in most of the problems. First, each problem is non-numeric in part or .in 146 rt r±. tr t t tt_ i . . ft. i mrlmIZEI'rma,Y'""""mm t t .'''!' = !!I'm", , In It • tr m: 3 o whole. The great computational power of modern numeric computers is not needed. Second, in most of the problems there is a need for a unit of data more complex than a simple number or array of numbers. For example in analytic differentiation, some method of representing algebraic formulae is needed. In language translation or theorem proving some method of representing syntax or theoretical relationships must be provided. Third, in many of the problems the assrgnment of specific areas of memory to contain certain types of information is difficult or impossible since the form, o structure, and amount of information is not known at the time a program is set in action. For example, in many cases it is not known what form a self-organizing system will take, or what concepts, and hence information. a theory developing program will yield. Fourth, it is often desirable to have certain portions of a program call on themselves as subroutines. This is called recursion and is useful in differentiation or game playing where a routine may call on itself to look ahead a move. A successful non-numeric computer, if it is to have general applicability, must be designed to meet these four needs. o numeric work Similarly, any language aimed at non- must fill these needs. 1 Ii 7 . - _. _.__ .""-'.-."""- ,=....;..:;=-..:..;;..;="-'-""'""---_ ..... ~~='"~~-~--""=~--.-- 4 IPL-V is an abbreviation for Information Processing Language V, a highly successful and widely used language designed for non-numeric computing. The IPL languages were developed at the RAND Corporation by Newell, Shaw and Simon, beginning in 1954 with IPL-I, a language for playing chess. Of the IPL languages, IPL-V is the only one which has seen widespread use. The language is well-documented and a manual for programmers is available. 1 In the next section IPL-V is outlined and the manner in which it meets the four problems posed above is discussed. o lSee reference [3J. o 148 m ttn • trtm_. o $ sr. tst nt rt _ * m j t _tr'm mttrt II The IPL-V language may be regarded as an assembly language for a non-numeric computer, the IPL computer, or as a medium level language which is machine independent and is executed on numeric computers by an interpreter program. It is interesting to note that an IPL computer has never been built, and all work done in IPL-V is accomplished by means of interpreters. Nevertheless it is useful to describe the IPL-V language in terms of the IPL computer. It is the function of the IPL computer to manipu- o 1ate symbol s, that is, to accept as data, members of a certain set of symbols, to store these symbols in memory, move the symbols from one location to another, compare the symbols', make decisions bas~d on these comparisons. organize the symbols in memory in a meaningful manner and produce as output a sequence of symbols. For this reason IPL-V is often referred to as a symbol-manipulation language. The memory of the IPL computer is divided into cells. and it is the addresses of these cells which form the symbol manipulated by the computer. That is, an IPL symbol is the address of a cell in the IPL memor y. o trary. The meaning assigned to these symbols is arbiThus regardless of the contents of cell 14613, -5- 149 6 the address 14613 may represent New york City in a military problem, Act II of Hamlet in a literature analysis. or the principle of mathematical induction in a theorem proving problem. Since it is inconvenient for a programmer to deal directly with memory addresses, the IPL-V language allows a more convenient external representation of The thirty-six characters ABC ••• Z $ symbols. - "I" / ) ( and, are called regional characters. =. + At the beginning of his program a programmer may assign to each regional character a continuous block of cells in memory. The block of cells assigned to say A is called the A region and the individual cells in this region are referred to by the symbols AD (or simply A) for the first cell, A1 for the second cell, etc. Any symbol naming a cell in one of the thirty-six regions is called a regional symbol. The assembler translates regional symbols into the corresponding addresses. In addition the IPL computer has the ability to transform the address of any regional cell into the correct regional symbol during output operations. The address of any cell not assigned to a region is a non-regional symbol, and may be represented by the programmer in a variety of ways. Each cell in the IPL memory contains two digits called the P and Q digits and two addresses called the SYMB for symbol and the LINK. A typical cell in memory 150 Itt o de • __ *_* _. 1 ttsttts t't m $, ,n It t =* H t _ tm _ ••t "h.d h to. =wrzre 7 is represented by the following diagram: , , P Q SYMB LINK The individual portions of a cell are not addressable. Cells may be used for one of three purposes: to contain an instruction for the IPL computer, to contain data, or to contain information necessary to the functioning of the IPL computer. There are a fixed number of cells of the third type and three regions are automatically set aside to contain them. The H, W, and J regions always contain the same cells in memory. o The cells of the H region function as registers and indicators in the IPL computer. The W region contains some cells usable by the programmer as temporary storage and other cells used in exercising a certain degree of control over the operation of the computer. Each cell in the J region represents ~nd contains the first instruction of a built in subroutine, of which there are 188 in a complete system. With the exception of the H, W, and J cells, any cell in memory may be used to contain data or an instruction, and during the course of a program. may contain both. A cell containing data may be of two types. o A data term is a cell containing special alphanumeric or 151 8 numeric information. while the P and Q digits indicate the type of information~ o A standard data cell is a cell used to store an IPL symbol. The symbol is stored in the SYMB and the P and Q digits indicate the type of symbol. 3 The LINK of a standard data cell also contains a symbol, the use of which will be described below. data cell containing the symbol I0 I 0 I 1461 3 "+" A might look as follows: 00000 P Q SYMB LINK where 14613 is the address of the first cell in the "+" region. Unless attention is to be called to the P and Q di gi ts, thi s wi 11 be represented by + ,0 1 • o 2All data terms have a Q digit of 1 which serves to distinguish them from standard data cells which have a Q digit of 0, 2, or 4. The P digit of a data term indicates the type of information stored in the data term as fo 11 ows : P=O Decimal integer P=l Floating point number P=2 Alphanumeric Octal number P=3 3Standard data cells usually have a P digit of 0 although they may be specially marked by a P digit of 1. The Q digit indicates the type of symbol contained in SYMB as follows: Q=O Q=2 Q=4 SYMB is regional SYMB is local SYMB is internal o r o _ • •• tt » _ t r 1Mff 9 The data terms playa rather minor role in the computer, usually serving as storage locations for numeric information; while the role of the standard data cell is central to the operation of the computer. In dealing with symbols of arbitrary meaning, the IPL computer answers the first need of a non-numeric computer, that of dealing with non-numeric information. These symbols do double duty, serving sometimes as the addresses of cells in memory and at other times representing the concept assigned by th~ programmer. However, the IPL symbol, being an address, is basically no more complex than a number. o The need for a complex unit of data is fulfilled by the list, a basic unit of data in the IPL computer. A list is a sequence of data cells whicharejoined together by having the link of each cell contain the address or name of the following cell. A list of the symbols Al, 87, c4, and Al in that order would be represented by the following diagram; M4 ->/ A1 ,-+-->/ 87 ,--+-/-->/ c4 -1---->/ Al ,0/. where the arrows indicate the cell referenced by the LINK of a cell. Note the use of the symbol 0 in the link of the 1as t ce 11 • Th iss ymbo 1 i s ca 11 ed the ter mi na t i on symbol and indicates that the list terminates at that o point. The name of the first cell in the above list is 153 10 o M4 and the list is also referred to by that symbol. Given cell M4, any symbol on the list may be reached by passing from link to link. Far more complex structures may be created by using the SYMB of some cells on a list to contain the names of other lists. The Q digit of a cellon a list may be used to indicate whether the SYMB contains an abstract symbol or the name of a sublist which is to be considered part of the structure. 4 The number of structures possible is limited only by the programmer's imagination, but for simplicity only lists will be considered below. Because of the list, IPL-V is called a listprocessing language, as are other languages which use the same concept. The language contains subroutines for c list manipulations such as copying, printing, searching, or erasing lists. An example is a subroutine which will 4 For example., the algebraic expression A-1'B+C/(D+A) may be represented by the structure El below which expresses the structure of the expression in a manner not possible in a linear list representation. El->/ ->1 ,0/ >1 + >1 1_.>1 C >/ L >I~ 1_>1 0 >1 >1 A >1 "it: + >1 B A ,01 1°1 c *11: _ rt t tt t !,:: ' $ '=uwm ,"'M"" W _ t. r tr ''M''''.'''.' t •• 11 o test whether a given symbol occurs on a list or not. consider again the list: M4 -> I A1 ,-+---> I 87 ,-11-----:> I C4 ,-11--> I A 1 ,0 I The location of the first cell of the list is important. Since the name of the list is M4, the firSt cell of the list must be cell M4, but the location of the remaining three cells is unimportant to the structure of the list. This fact has important consequences. When inserting a new symbol on a list, it is not necessary to disturb the original cells of the list. For example, the symbol D5 may be inserted between 87 and o C4 on list M4 above by finding any unused cell anywhere in memory, placing the symbol D5 in that cell, and rearranging the links as follows: M4->1 Al '----'----I >I~ r>/ ->/05 1 C4 1-+-1->1 Al ,01 +=1 In this way a solution is achieved for the third problem of non-numeric computers, memory assignment. A block of memory need not be reserved for expansion of a data structure, since in expanding, a data structure may make use of any unused cells in memory, whether they lie in a continuous block or not. Even the names data structures may be kept on lists. of new Onl) the total o 155 · - .. .. - -'- ... ... ~ -~. .. . . . -- .•. .__ ....•. - -.. -..... , 12 number of cells in memory is of concern to the IPL-V '-~.- ... ' ...... ... ---..• - .... " ~ -- ....... ... o programmer. It might. seem that locating an unused cell in memory would be difficult, but this problem is handled in an elegant and efficient manner. After assembly, all unused cells are linked together to form a list named H2 and called the available space~. During processing when a cell is needed, one is removed from H2 for use; and when a cell is no longer needed by the programmer, it is returned to H2. The list organization also allows cells to be used as though they were capable of storing more than one symbol. Suppose for the moment we have a symbol stored in cell WO, say A7, and we need to temporarily store a second symbol, say 83, also in WOe We execute an IPL instruction causing the computer to push down cell WOe That is, an unused cell is removed from H2, inserted behind cell WO, and a copy of the symbol in WO is placed in the new cell, creating the following list: WO->I A7 , -+--~>IA7 ,01 o 15 G itt r· • .t _ t $'.0. ..! m_ , . It., _ * Wnr$M at . • d" h 13 o Now that a copy of A7 has been made, B3 may be placed in WOe WO - > /._B..:;..3--&-=--> 1 A7 ,0 I We may go even further and store C8 in WO before removing B3, by pushing down WO again, then storing C8. WO ->1 C8 I -+--->1 B3 -+----->1 A7 ,01 The list created in this manner is called a push down list but is no different from any other list. When the symbol c4 is no longer needed in cell WO, a pop up instruction is executed. c; This operation copies the second symbol on the list into the first cell and removes the second cell from the list. returning it to H2. wo->I B3 1-+1-.....;>/ A7 ,0/ One more pop up, and WO is returned to its original state. wo->I A7 ,0/ The preceeding sequence of events may be summarized by writing the push down list vertically. push store A7 down> A7 B3 A7 o :> push store 8 pop B3 down> B3 C8 > C up > A7 B3 B3 A7 A7 - - E' 'I' 14 The push down and pop up instructions enable a subroutine and main routine to use the same storage cefls. A set of working cells, WO through W9, are provided for temporary storage. When a subroutine needs temporary storage, some of these cell s are pushed down, then used as storage. Any information stored by the main routine in these cells is preserved by the push down operation. Before terminating, the subroutine pops up these cells, returning them to their original state. The ability of the IPL computer to allow recursion, the fourth need of a non-numeric computer, is also based on the push down operation. The cell Hl, called the current instruction address cell t contains at any given c time the address of the instruction currently being executed by the IPL computer. When an instruction is completed, the address of the next instruction is obtained and placed in Hl. Like any other cell in the memory, H1 may be pushed down. When one routine calls on another as a subroutine, Hl is pushed down by the computer;. saving the address of the instruction in the main routine where processing is suspended. The address of the first instruction in the subroutine is placed in Hl and that instruction is executed. Processing now continues along the subroutine and the computer is said to have descended a level. Before terminating. the subroutine may call on itself or another subroutine. o 15 ~ me o mr m. v,'W,,"!'rm t WW!!I'W ""rrn:rr:r""",'lJ"lfiM'M¥:iWbM,i(j, R 15 Again Hl is pushed down, saving the point at which processing was suspended in the subroutine, and processing continues at a lower level. When a subroutine terminates. Hl is popped up and the routine one level up resumes action. A combination of the manner in which Hl is used and the ability of the working cells to keep the contents of routines on different levels from becoming mixed, allows a subroutine to call on itself. The instructions in the IPL computer are kept in lists. The p, Q, and SYMB of a cell make up the instruction and the LINK indicates the next instruction. The IPL computer follows instructions from cell to cell o down a list rather than executing instructions sequentially in memory. I This allows routines to be manipulated with the list processing subroutines. It is conceivable that a main routine could construct a subroutine using list processing subroutines, execute that subroutine, then erase it, that i~ return all of its cells to the available space list. In communicating information to a subroutine. a special cell HO, the communication cell, is used. The symbols required as inputs by the subroutine are placed in HO using the push down operation. The subroutine accepts these inputs, removing them from HO, and before terminating, places all output symbols in HO where they o are recovered by the maintroutine. 15 9 16 0 In addition to producing output symbols, some subroutines produce a yes or no answer. "+" / For this purpose a cell called the test cell, H5, is provided. cell may be in one of two states, '1 The test or, "_"., and an instruction is provided to allow conditional branches or transfers within the program on the basis of the state of the test cell. There are only eight basic instructions in IPL-V, most of the processing being done by the numerous subroutines. Two instructions are used for placing symbols in HO, one instruction instructions each for calling on subroutines, two for removing symbols from HO, one instruction for popping up cells or pushing down cells, and one instruction for conditional branching on the status of t he test cell. The P digit determines the type of instruction and SYMB contains a symbol, the name of a cell, or the name of a subroutine, depending on the context. The Q digit is used in connection with SYMB for three levels of addressing. For more complete information concerning instructions and a sample routine, see the appendi x. ,The external form of IPL-V is quite simple. Lists, instructions or data, are written vertically on the coding sheet. Each line represents a cell and space is provided to indicate the name of the cell and the P, Q, SYMB, and o " LINK of the ce 11 . I f ali nk i s 1eft b,l ank, the ce 11 i s 16 u 1 me •••• t tt I '$rr'·,,' r t : TlMU?E,r:w'MJI• •n '!!pt'.' • P"','?,-,,'·', 'J _ ! : :.." ' _ 17 o assumed to link to the cellon the following line of the coding sheet and the name of the following cell may also be left blank if its memory location is unimportant. Thus to create the list T4->1 + -4----,>1 z28 ,_ _--L--.! ~->I z29 ,01 we write on the coding sheet NAME T4 PQ SYMB LINK + z28 Z29 0 o o 161 S 7" c III The University of Oregon IPL-V system for the IBM 1620 computer, developed and written by the author and John D. MacDonald, was designed with two objectives in mind. It was intended first as an educational device to acquaint students with list processing and symbol manipulation problems, and second as a system for checking out IPL-V programs before running them on larger computers. In view of the educational aim, operating speed was sometimes sacrificed for operating ease and additional safeguards. Because of the speed and size of the 1620, the system was never intended as a production tool. o The 1620 system is based on the specifications of IPL-V set forth in the manualS and is fully compatible with those specifications, though not all options are available on the 1620 system. Operating on any 1620 equipped with card I/O, indirect addressing, automatic divide, and special instructions, the system provides approximately 640 IPL cells at run time with a 20K memory. An additional 1,660 cells are available with each additional 20K of memory. The system operates at approx- imately 80 IPL instructions per second and is equipped Ssee reference [3]. ,.,., U -18- 162 trW o Mts rtt ¢ _ _ _ h__ ti t rrr = t r 'Z • • , =t .. !lmm','!",', m. : r. . !'Mn. ''11 19 with all tracing and monitoring features specified by the manual. These features include operator or program controlled trace with output on any unit, automatic trapping on error conditions, and flexibility in trap recovery. The system consists of three decks, the assembler, subroutines, and the interpreter, which are loaded in that order with the source deck placed between the assembler and subroutines. The assembler loads into the lower portion of memory and assembles the source deck directly into the upper portion, producing an assembly listing on option. o Next the subroutine deck i.s read by the assembler and those subroutines called for are loaded into memory. After the last card of the subroutine deck has been read, the interpreter loads into the lower portion of memory, occupying the space previously occupied by the assembler; the computer halts; and execution begins when START is pressed. The internal form of an IPL cell is a twelve digit field with an odd address. From low to high address the cell contains the P, Q, SYMB (five digits), and LINK (five digits). Provisions are made for writing additional subroutines in SPS and including them in the source deck. It is also possible to reserve blocks of space in the 1620 memory for use by other systems. Methods 163 20 of setting_ up linkage between systems are described in the documentation. The documentation is in the form of an appendix to the'manua1 6 with cross references. A master copy of the documentation is maintained on cards for easy editing and reproduction. During the summer and fall of 1963, a preliminary version of the system was written. This version was distributed to approximately twenty participating users for field testing and was used in a one term seminar in IPL-V programming at the University of Oregon. Students in this seminar used the system for problems such as analyzing poetical structure, construction of Farey sequences of numbers, calculation of all closed paths in a planar graph, and construction of a machine for playing Hex. The system has also been used for map coloring and analytic differentiation. The preliminary version does not contain block handling, auxillary storage, read/write, floating point, save for restart, or post mortem dump routines. During the summer of 1964, a final version will be written, which will include all features except auxillary storage processes. The final version will be submitted to the 1620 Users Group's General Program Library for distribution. 6See reference [3]. ~~~-~--~---~~--- - -- - ..---~----..-....-.--.-------.~- "- _._-_.- ..._._._._---------------------- 0, "WWfT '·eW."i'HiliM'IW,.&W',uW,wirsX W [ IlI'WW'I'PMmiiNlliliiM'ihW'MWMI\!fW·WP.iMiifiW t15t ?ft'YM''tlM''¥W.f:'MIiIW"WtWW"'PM'Ttf" '¥M'BfUlK¥WJN''I!! 2' J t ,* !P'N',!! em:: eM I •••• * r _ = , ....... Appendix The IPL-V Instruction The Q digit of an instruction operates on the SYMB to produce a transformed symbol S as follows: Q=O S is SYMB. Q=l S is the symbol contained in the cell whose name is SYMB. Q=2 S is the symbol in the cell whose name is contained in the cell named SYMB. For example, if we have the following cells in memory, Al -~I T4 101 T4->1 J8 101 c and the SYMB of the instruction contains Al. the Q digit produces the following transformations: .fQ SYMB oAt S Al T4 J8 1 A1 2 Al The transformed symbol S is stored in a register; the SYMB portion of the original instruction is never altered in memory. After the transformed symbol has been obtained the P digit determines the action as follows: o p=o call on the subroutine whose fir s tin s t r uc t ion i sin ce 1 1 S. P=l push down HO and place a copy of the symbol S in HO. -21- 165 22 P=2 copy the symbol in HO into cell S, then pop up HO. P=3 pop up cell S. p=4 push down cell S. P=5 same as P=l except HO is not pushed down first. p=6 same as P=2 except HO is not popped up afterward. P=7 if H5 is -, transfer to cell S for the next instruction. if H5 is +, continue. Sample Problem As an example of how the instructions are used, we will write a short subroutine below. It will be necessary to understand the operation of two of the J routines. J2 accepts two inputs in HO. symbol. Each input is a J2 compares the symbols and sets H5 "+" if they are equal and "-", if not. J2 leaves no symbols as output in HO and the two input symbols are no longer in HO after J2 terminates. J60 accepts one input which is the name of a cell on a list. If that cell is the last cellon the list, J60 sets H5 "-" and leaves the input as output. If the ce 11 i s not the 1as t ce 11 on the 1 i s t, J60 places the name of the following ce 11 in HO and sets H5 "+". We now code the routine E4. E4 i s a routine which eva 1u.a tes a function of X at a given point. More clearly, E4 accepts a symbol representing a given point, 1 6 ti =.'_ _ _. .11 IiIIII/IP"W!IIIiIIIi!'lPFIiIIiI'rlllllmp_r$_!_"_!~t.t.t ___S._IIIIIIITDi_,._,,,!,.,'t_nnllililf'21'Ig!1!_ _ t =.'••".','.'_m_n.,-__lIIIiI_t.___ · __..._&IIIIIIIIIII. . .iIIiIIII....._ _ _.. $118·_ _ · . . ._ _ _ _ _ tt 23 o say A, and a second s ymbo 1 assumed to list representing a function of X. be the name of a For example, the list Fl below: Fl X ~t~ L 0 G ( B / X ) II "xlog (~) x E4 then evaluates the function at A by replacing every 0 occurrence of the symbol X on the list by the symbol A to yield the list: Fl A ~I, o L 0 G ( B / A ) "alog(~)" a 0 In addition since E4 should leave no output i n HO. E4 does not set H5 as part of its output, the status of H5 should be the same after execution of E4 as before. But E4 must call on J2, which does reset H5. For this reason, it will be necessary to push down H5 at the beginning of E4 to save its status, then to pop it up at the end to restore its status. WO and WI will also be needed. o Two storage cells, It is assumed that the routine which called on E4 input the name of the function '"' 24 list first, then the symbol representing the point. A little study and liberal use of a black board as a simulator will make the operation of E4 clear. The symbols 9-1,9-2, and 9-3 are called local symbols and are used for internal branching within the routine. Name PQ SYMB E4 40 H5 40 wo 40 Wl 20 WO 9 -1 1 2 HO 10 00 70 60 11 9-2 9-3 LINK 21 00 70 30 30 30 30 X J2 9-2 Wl WO W1 J60 9-3 WO W1 H5 HO 9-1 0 Comments Preserve H5 Preserve WO Preserve W1 Move "point" to WO Input symbol in list cell Input X Compare symbols Go to 9-2 if not equal Copy list cell address in Wl Input point symbol Move point symbol to list cell Find next list cell If no list cell, clean up Restore WO Restore Wl Restore H5 Pop up HO, terminate o 160 r * o Selected References IPL-V, Primary Sources [1 ] Dupchak. Robert, TIFi..: Teach Informatioil Processing Language, The RA:\lO Cor~()ratlont'--"'Rt-:il-3819-PR, --October 1963. [2 ] Newell, Allen, "Doc 'Tlentation of IPL-Vt', Comm. ACtvl, vol. 6, No.3, March 1963. pp. 86-8~. [3 ] Newell. Allen, et al., Information Processing Language-V Manua 1, (Second Edl tl on-) , Prenti ceHall, Inc .. "Englewood C'iiffs, New Jersey., January 1964. IPL-V. Applications 0, [4 ] Newell, Allen, A Guide to the G2neral Problem-Solver Program GPS-2-2, The RAND torporatlon, RM-3337PR, February 1963. [5 ] Newell, Allen, and H. A. Simon, tIGPS, A Program that Simulates Human Thought," Lernende Automaten, H. Billings (ed.), Oldenbourg, Munlch, 1961. [6 ] Newe 1 1, All en, and H. A. Sima n , " The Log i c The 0 r y Machine: A Complex Information Processing System", IRE Trans. Info. Theory, Vol. IT-2, No.3, September 1956, pp. 61-79. [7 ] S i mo n, H.A., " Ex per i me n t s wi t h a He uri s tic Com p i 1e r , " J. ACM , vol. 10, No.4, 0 c to b e r 196 3, p p. 493 - 506 • [8 ] Stefferud, Einar, The Logic Theory Machine: A Model Heuristic Program, The RAND Corporatlon, RM-3731-CC, June 1963. f·,.II General Informati on on Lists o [9 ] Baecker, H. D., "Happed List Structures," Comm. ACM, vol. 6, No.8, August 1963, pp. 435-4~ [10 ] Banerji, R. B., "The Description List of Concepts," Comm. ACM, vol. 5, No.8, August 1962, pp. 426-432. -25- 1 6 !J 26 [11] Bow1den, H. J., "A List-Type Storage Technique for Alphanumeric Information,Jr Comm. ACM, Vol. 6, No.8, August 1963. pp. 433-434. [ 12 ] Weizenbaum, J .. "Knotted List Structures." Comm. ACM, Vol. 5, No.3, March 1962, pp. 161-165. Other List-Processing Languages .[13] Bobrow, O. G., and Bertram Raphael, A Comparison of List-Processing Computer Languages. Comm. AeM , V0 I. I, No. 4 , Apr 1 I I 964 , p p . 2 3 1 - 240 • [t4] Cooper, O. C., and H. Whitfield. "ALP: An Autocode List-Processing Language," Comp. J .• Vol. 5. No.1. April 1962. pp 28-32. [15] Gelernter, H., J. R. Hansen, and C. L. Gerberich. "A FORTRAN-Compiled List-Processing Language," J. ACM , Vol. 7, No.2, Apr i l l 960 p p . 87 - 10 1 • t [16] Green, 8. F., Jr. "Computer Languages for Symbol Manipulation," IRE Trans. on Human Factors in Electronics, Vol. HFE-2, No. I, March 1961, pp. 3-8. [17] Mccarthy, J., et al., LISP 1.5 Programmer's Manual, MIT Computatlon Center and Research Laboratory of Electronics, Cambridge, Massachusetts, 1962. [18] Weizenbaum, J., "Symmetric List Processor," Comm. ACM, Vol. 6, No.9, September 1963. pp. 524- 57+4. [19] Introduction to COMIT Programming, Research Laboratory of Electronlcs and MIT Computation Center, MIT Press, Cambridge. Massachusetts, 1961 • tririe - = * -- = ttth er t - - _r * .»Ht o PETROLEUM EXPLORATION AND PRODUCTION APPLICATION FOR THE IBM 1620 AND PLOTTER By Jack L. Morrison Oil Information Center University of Oklahoma Research Institute Norman, Oklahoma Delivered at: IBM 1620 Users Group Western Region Meeting Denver, Colorado June 17-19, 1964 o 17 1 It was refreshing to h~r Dr. Edward N. Brandt, of the Un iversity of Oklahoma Medical School Biostatistical Laboratory I say in his keynote address that the problems dealing with computers in the field of medicine are such that they are basically related to and parallel the problems which are encountered in the oil industry. Dr. Brandt also related that the use of computers in medicine has required that the users better define their problems, which gives them a better understanding of the overall situation. The same can be said about the use of computers in the oil industry. In the next 15-20 minutes, I plan to tell you a little about the Oil Information o Center which is an integral part of the University of Oklahoma Research Institute. I will discuss the Oi I Information Center: 1. Why and how it was established 2. The goa Is and objectives 3. How it is connected with the University computer usage genera I Iy and the IBM 1620 specifically 4. What we are presently doing, 'and 5. Where we are going I. GENESIS OF THE OIL INFORMATION CENTER' Two independent oil men in Oklahoma, Mr. Ward Merrick, Ardmore, and Mr .. Howard McCasland, Iv\ack Oil Company I Duncan, were concerned about three 172 ttrtzrtttr «Mtttnrn = rUt t he • • t ,'m! apparently unrelated situations and problem areas in Oklahoma. lhese three probl em areas were: 1. No attempt had ever been made to gather groups of oi l.field related information on a J ibrary basis. 20 The Oklahoma Corporation ·Commission needed an assist in, some of their data processing problems and engineering ca leu lations 0 3. The computers at the University of Oklahoma were not being utilized as much as could be reasonably expected by local industries, particularly the oil industry. The concern of these two independent oil operators led .them to the concept of the Oi I Information Center and as a direct result they furn ishedthe impetus by o suppl ying financia I assistance through the medium of their personal foundations ~ A two-year budget was set up for the in itia I phase of this Center. One obvious obiective of the Oi I Information Center was that sooner or later it must become self-supporting from earned income. It was felt by all concerned that these problem areas just mentioned would be the strong nucleus upon which the objective of self-support would be reached. After a series of conferences, oil industry executives and University people agreed that the log ica I central location for libraries of oil information wou Id be on the campus at the University of Oklahoma. The categories of information which seemed desirable to collect were electric logs, scout tickets, drillstem tests, sample logs, o 2 1·73 and Oklahoma Corporation Commission completion forms~ The University of Oklahoma has been famous for years in the quality and quantity of graduates pointed toward the oil industry. The University has probably turned out as many petroleum geologists, petroleum geophysicists and petroleum engineers as any university in the United States. The Oil and Gas Conservation Department of the Oklahoma Corporation. Commission needed assistance with some data processing problems . They wished to work directly with a group who could help them in their work, on whose integrity they could rely and in whom they could have confidence. The Oil Information Center devised a plan to prepare computer programs to assist with some of these problems, and Commission representatives gladly accepted this plan. II. OPERATIONS OF OIL INFORMATION CENTER A. Introduction to University Relations The actual operation of the Oi II nformation Center is concerned with various areas of effort. A ma jor area is connected with university activities. These are: 1. The graduate program of the University of Oklahoma 2. The Oklahoma Geologic,a I Survey, The University of Oklahoma Schools of Geology and Petroleum Engineering 3 'fJ,u!rmrpp, 'M"W"'EiJJ""".! WE""Wi; 0 ', . " 3. Conducting seminars on oil related topics 4. Attracting people in the oil industry to the campus Geology Graduate Student In checking the records I found no evidence indicating that any geology graduate student had used the computers or plotter to assist them in their master's thesis work. I sought out someone who might be ·interested in using the computer and found a Humble Oil geologist, on leave from his company to do master's work, and who was wi II ing to work with me. Since the geologist was not a programmer, arrangements were made for his programs to be written for him and through the cooperation of the Computer Lab his key punching was accomplished. This graduate student's thesis was on the geology of an oi I field in North Central Texas. His study of the electric logs on each well furnished him with formation tops, well elevations, etc. for his study of 25 different formations. With this information punched into cards he was ready to use the 1620 and plotter to prepare his isopach and subsea calculations and his many maps. The computer program as written was general enough that calculations could be made for isopach.thicknesses, subsea formation tops, and sandshc;J Ie-I imestone ratios. This is an example of what can be done in working with graduate students and we hope to encourage others a long these lines. o 4 175 -~-~-~----------.-- "I 1 ,I I I: o· Oklahoma Geological Survey and University of Oklahoma Schools of Geology and Petroleum Engineering The Oil Information Center has attempted to work closely with the Oklahoma Geological Survey and the University of Oklahoma Schools of Geology and Petroleum Engineering. The I ibraries of oi I field information being gathered by the Oil Information Center are a valuable complement to the Core and Sample Libraries now existing at the University of Oklahoma. The Geological Survey uses the electric logs, sample logs, drillstem tests, etc. in their statewide geologic investigations. The Schools of Geology and Petroleum Engineering can use the same information as teaching aids. c Conducted Symposiums An important acti vity in the university phase of our operation is the conduction of symposiums. The Oil Information Center, in conjunction with our libraries of informaHon and computer services, has conducted two symposiums on the campus. One was related to our Drillstem Test library to which we were able to get good industry speakers from a II over the Southwest. The second symposium was directly connected with the Mid-Continent Well Data System in Oklahoma City. In addition to the speakers .at this meeting, the Oi I Information· Center in cooperation with the Un'iversity of Oklahoma Computer lab demonstrated an information retrieval program. I $hall discuss this demonstra- 01 5 176 - ntt t - s' , em s m o tion in more detail in a few minutes. These symposiums have been extremely helpful in our relationsh ip with oi I industry people, particularly on the operating level. The sharing of new ideas and approaches is always helpful. Bring People to the Campus Directly through the efforts of the Oil Information Center a large number of people have been directed to or through the University of Oklahoma campus. Our seminar on drillstem testing attracted 148 people for two days ot meetings. The Mid-Continent Well Data System Symposium was for one day and was attended by 65 peop Ie. o Major oil company and consulting geologists from Tulsa, Ardmore, Norman, Ada, and Oklahoma City have been to the Oil Information Center libraries for various reasons. Maior oil company representatives have also been to our computer installations using our computer and plotter services. Others have investigated the services which we have to offer in order to determine how this information could be beneficially used by their company. Industry Effort To the best of my knowledge, this is the first industry wide effort of information gathering undertaken by the University of Oklahoma. Acceptance of ,the oil Iibraries could well lead to the establishment of the gathering of information in o other fields of endeavor. 6 17 i P""S'''"E''hl'' Oi I Industry in Oklahoma With the advent of oil industry data retrieval pilot studies in West Texas, the Oil Information Center found it advisable to conduct their own pilot project on the digitizing of scout tickets and a retrieval program to recover this information. The Autwinefield in Kay County, Oklahoma, was chosen for this study for several reasons. The field has more than one producing zone; it produces both oil and water; both major oi I compan ies and independent 0 iI operators have wells in the field. Scout tickets were received on 122 wells which included some surrounding dry holes, and the information was keypunched to our predetermined format. A computer program was written for our 1410 to retrieve certain information from these cards. The program was written to gather certain usable groups of information: 1. list the wells which cored the Red Fork formation, 2~ List the wells and the detailed results of all drillstem tests in the Red Fork formation, 3. list the casing programs in each well, 4. List the formation tops from some wells, 5. list each well that penetrated the Mississippi formation, and 6. List the details of the acid and fracture treatments on each producing Red Fork well. These are some of the categories of information chosen to be retrieved for this demonstration. This information is typical of that which is used by the exploration geologist and the petroleum eng ineer in some of their everyday problems. 7 c 170 iIIIr_x ... :_ =""'".'_.__ ••••-.'_ _ _ _•_____.'...t _ _ r_milllilillll'.·_ _ _.'.'_ _:.'1'"."'_ _ _ 1.... • _ _ _......_ _ _'•• "·_ ..'.a:_tt __ ._ttllllllllit_ _ _ _ _ WiF O· , ,~i! Oklahoma Corporation Commission Preparation of Oklahoma Guymon-Hugoton Gas Allowable Schedule: Due to the large amount of paper work which they process, and .their general work load, Gas Conservation Department personnel ·often were two or three months late in the preparation and distribution of the Guymon-Hugoton Gas Allowable Schedu Ie. ' By the time the operators of the well and the purchasers of .the gas received the schedules they were practically of no value. The Oi I Information Center worked as .1 iaison between Corporation Commission engineers and the Computer Lab programmer so that a computer program could o be written to calculate the monthly gas allowable for each well in the field. When Corporation Commission personnel prepared this gas allowable schedule on a desk calculator, they required approximately 70-75 manhours per month. After an estimated five hours of keypunching and keyverifying per month, the IBM 1410 makes these calculations to prepare this gas allowable schedule in 0.4 hours per month. Calculate one-point back pressure test: An Oklahoma Corporation Commission statewide rule makes it mandatory for all allocated gas wells to annually report a one-point back pressure test. This information is used in assigning per we II gas allowables for the following year. o 8 1 '7 ~ 1'1' o· An estimated 1,800 - 2,000 of these tests are filed with the Commission each year and the Gas Engineer is required to check each of the calculations. The Gas Engineer informed me that with no interruptions he could check five or six of these calculations per hour. This meant that two or two and one-half man-months per year was spent in checking these previously calculated tests. An O.U. Computer Lab programmer wrote a program for our 1410 to make these calculations. The 1410 processes these tests in 4.25 hours, which is a significant dollar saving estimated at 3-1/2: 1. This Gas Engineer is now freed to do more productive and original work for the Commission, which represents the true saving. c B. Introduction to· Commerc ia I Appl ications Our other major effort is the industrial commercial activities. We have worked directly with: 1. t-Aa jor oil companies 2. Independent oil operators 3. Oil-field service companies 4. Petroleum consultants In mid-1963 IBM released a group of programs from their 1620 library, which are called the Petroleum Package. These programs were written by experienced petroleum engineers, geophysicists, and geologists for a rather wide range of c 9 bEt • = -_. :.$ • M'.' II' It o commonly encountered exploration and engineering problems. The engineering programs deal with primary oi I recovery, secondary recovery, economic eva luations, casing design, gas production rates, flash calculations, etc. The exploration programs deal mainly with geophysics, but are also related to map contouring, electric log analysis, dipmeter calculations, map preparation, etc. In the past ten years petroleum oriented companies have become more dollar conscious and overall economics have played an ever increasing part in top management decisions. Computers are being used more and more to funnel detailed geophysical, geological and petroleum engineering information to these top management people for their perusal in making their decisions. In the recent past it was not feasible to make many groups of calculations in the fields of geophysics, geology and petroleum engineering. These calculations were known applications and approaches to their problems but were too detailed and too time consuming for the engineer or geologist toiustify spending the time from his oth~r doily duties. With the advent of computers, it became more realistic to consider making some of these calculations. Also, in the pQst, the necessary data to make these calculations were not gathered knowing that they would never be used. Such is not the case now, and it should be pointed out that the gathering of these data in many cases makes for a more efficient operation on all levels. o 10 181 "UW:,.":" I I I. I In several application areas the use of digital computers is becoming more valuable as magnetic tape recording devices are used.inthe field. Some of .these instances are: 1. Electric logs (and their companion logs) 2. Dipmeter surveys 3. Geophysical field surveys Many of the large oi I field service companies are instal.ling magnetic tape recording devices in their field trucks. This will lead to a more detailed study of data now being received but not efficiently used. However, most of the commercial work which we have done in our 1620 Lab ·c\ L/ . is related to geophysical problems. The reason is rather obvious when the users were questioned. In many-instances geophysic ists were not making certain known approaches to their problems because of the number of manhours requ ired to prepare the data I make the ca Icu lations and plot certa in information. The use of computers and digital plotters now makes it more practical to better utilize data gathered in the field by geophysical crews. As some of you know, a reflection seismograph crew costs an oil company between $15,000 to $60,000 per month depending on the overall services rendered and the field equipment involved. As in most any other service operation, reflection seismograph field crews can and do have certain problems. c 11 182 'W!"'YUr";;"ii*)'fliMdW'ijlij'¥f'''WitlitHWWI'''!J!J"'!im,r!!p' nj,.t!LTtPlfM WNU JT PPUZ"''W:!'HilSl''*'HtWMfllfw'Mi!iiQWlWHliiUi'W ""M'''rern.'tltl·l1' P'III ,,'Yi!'f"WbiWlN'Mt¥f "P!lU7ll1'' Ii 1M lf '?YI!!TOOW1fHMrW'!¥'.,fiR'WII" h ,",m pp'." If the field data are being processed on computers as work progresses, the errors can quite easi Iy be rectified. However, if there is a large time lag between the error and its discovery, it may not be so easy to make the necessary adiustments. A geophysical group of a maior oil company in Oklahoma City has been our largest user of commercial time on our 1620 and plotter. This District office is responsible for the geophysical work in all of Oklahoma, all of Kansas, the Texas Panhandle, North Central Texas and the northern 2/3 of Arkansas. In addition to the reflection seismograph field crews gathering new data, they o are continually reviewing old seismic records previousl y shot by themselves or by other companies. One geophysicist pointed out the following,relative to the information gathered from 300 shot .... poi nts. The time requ ired to hand ca Icu late and hand plot this data frqm 300 shot-points would be an estimated two man-months. To use computers, this same amount of work would require an experienced geophysicist one week, another week to key-punch, one to one and one-half hours on the 1620 for calculation, and five and one-half to six hours on the 1620 dnd online plotter. This represents a vast saving of time as well as money. One geophysic ist pointed out that the use of our 1620 computer on their reflec- o tion seismograph field data makes it possible for them to better uti! ize the 12 183 information which ~ be gathered from seismic records. He said that they can now prepare ten to twelve useful sub-surface maps where previously they were fortunate if they were able to get five to six maps from a set of seismic records. Dan Merriom of the Kansas Geological Survey and John Harbaugh of Stanford University through their joint effort developed a computer program to assist in the location of mineral deposits. (1) Based on certain known geologicQI land/or geophysical information and certain mathematical computations trend surfaces are fitted so that the sum of the squared deviations is the least possible value. The trend surface analysis may be used to: 1 . Predict projected depths to geological units within an area, c! 2. Del ineate unconformities or changes in structural patterns, and, 3. Extend better "geologic guesses" into adjacent unknown areas of no control. Close agreement exists between loca I structural features and trend-surface residuals. The residual maps were found to stress or emphasize trend relationships not otherwise clearly observed from original data and to emphasize the local component of the structural pattern by essentially removing the regional component or regional dip. Inasmuch as in many regions the oil and gas producing areas are systematica fly assoc iated with structura I features, there is the possibi I ity that a study of the residuals will indicate previously overlooked areas favorable for additional oi I exploration. 13 o if!"'!!! • 7 'BU'fflffUa!l!" wY',lf!ltJ('U,,"tE W"'."lI"W'W'f.mr!1!"rrpmU!!ywwllt15JWfW5fMfH' M','t'?T'!Z1l'fltI'[ " W t • MWNT'III'P'!F"tJPT!T1M',,,,, t " t t.rt t. err s nnw ** trW o The Oil Information· Center plans to take advantage of the existence of this program but we plan to rewrite the program to use the IBM 1620 and plotter rather than using the printer to prepare the map. John P. Dowds, a successful petroleum consultant in Oklahoma City, has worked on the laws of probabi Iities and the appl ication of statistical methods to help analyze the problem of obtaining commercial oil or gas production. Dowds, in a recent paper, stressed that "exploration geologists and geophysicists need to become statistically minded and to think of locating oil and gas fields as a problem in applied possibilities. II (2) o Dowds uses entropy for his mathematical model to learn of favorable trends and patterns in searching for logical locations for drilling new oil or gas exploration wells. Dowds determined a long time ago that his calculations were too difficult and the number of these calculations required were too many to be done by hand. An Oil Information Center programmer recently wrote programs to Dowds' formulae for his entrop), calculations. These are now being run on our 1620 and plotter. The final output to be studied for purposes of exploration is a series of contour maps. Dowds is representing a large independent oil operation in Oklahoma City in their search for sizable oil or gas reserves. o 14 185 James M. Forgotson, Jr., research geologist with Pan American Petroleum Corporation in Tulsa, said in a recent Oil and Gas Journal article that the use of electronic computers to evaluate electric logs is very practical. He said, "The speed with which these computations can be performed makes the analysis of many z ones or formations in thousands of wells practical. II Forgotsonwent on to point. out that IIwithout the aid of the computer, approximately eight manhours are required to calculate shaliness, saturation ratio, and favorability criterion for four ~ones in one well. II He also made ar interesting co mpa.ri son stating that "with the use of computers approximately one and one-half man-months would be required to process four zones in 1,000 wells while without the use of computers o fifty-·four man-months would be consumed. II (3) III. SUMMARY The Oil Information Center is serving a useful purpose to the University of Oklahoma, to the Oklahoma Corporation Commission, and to the oil industry in general in Oklahoma. With the 1410, 1620 and the plotter now in the University of Oklahoma Computer Lab, we are able to offer computer services to: 1. Maior oil companies 2. Independent oil operators 3. Consultant geologists and petroleum engineers o 4. Oil field service companies 15 18 G o Using the 1620 Petroleum Package of programs has proven successful up to a point even though the large maiority of commercial time which we are able to sell has been to companies who have written their own progrdms. C,:' REFERENCES ,j 1. Merriam, D. and Harbaugh, J. "Balgol Program for Trend-Surface Mapping", distributed by the Kansas Geological Survey (Special Distribution Publication #3). 2. Dowds, John P. "Application of Information Theory in Establishing Oil Field Trends", presented in June 1963 at Stanford University during the 3rd Annua I Conference on Computers in the Minera I Industries. 3. Forgotson, J.M., Jr. II How Computers Help Find Oil II, Oil and Gas Journa I, March 18, 1963. o 16 187 o A CONTROL SYSTEM APPROACH TO AUTOMATIC JET ENGINE TESTING c 1620 User's Group western Region June 17,18,19 - 1964 Aubrey D. Wood IBM Systems Engineer Oklahoma City, Oklahoma rtr,bttbtW tit. r $. **bt $. t .rt n tt j 'dO tt It $ _ = 1m tl. • ttr t tr_ TABLE OF CONTENTS I. INTRODUCTION TO THE PROBLEM A. B. II. HISTORY OF JET ENGINE TESTING PRESENT TEST PROBLEMS 1. Present Test Techniques 2. Testing Techniques 3 . Instrumentation 4. Human Errors 5. Rerun Statistics 6. Capacity 7. Safety SOLUTION TO THE PROBLEM A. INTRODUCTION B. PREVIOUS WORK 1. Data Logging 2. Research and Development - Jet Engine Test 3. Industrial Testing Systems - Discrete Process C. AUTOMATIC JET ENGINE TEST CONTROL SYSTEM 1. System Design Requirements 2. Control System and Interface Description III. ECONOMIC JUSTIFICATIONS A. TANGIBLE B. INTANGIBLE IV. SUMMARY AND CONCLUSIONS o 189 $ t Wi A. History of Jet Engine Testing: .After the first jet propelled airplane was captured from Germany by the United states in World War II, development of the jet type aircraft has proceeded in rapid fire fashion. The first truly great use of the jet airplane came about as a result of the Korean War. In a few short years since the early 1950's, the development and production of the jet engine has proceeded at an amazing rate. With the production of the first jet also came problems in the maintenance and overhaul of these complex, high thrust engines. At the beginning, especially during the Korean War, maintenance and repair was carried out in the remote airstrip locations and centralized repair facilities using the out-moded piston engine repair and test facilities. The piston engines had not required the highly substantial and instrumented test facilities that the newer high thrust jet engines were requiring; so, many of the first tests were performed in a crude makeshift manner. In the initial stages, many of the repair personnel became engine test personnel. Because the jet engine development had proceeded in a hurried fashion, adequate testing procedures were lacking; so many of the first test cell personnel found themselves preparing their own through pooling, interchanging and accumulating their experiences. Many of the basic principles of these early testing technical procedures are still in use today. Also, the great majority of today's test cells are modified piston engine of low thrust jet engine test cells and their instrumentation leaves a lot to be desired. Much of the instrumentation was installed on a "guess and try" basis. CJ Since the early 1950's the production rate and number of jet engines in the air has risen considerably. With these increases also came increases in the number of engines to be overhauled and repaired. The test facilities in many instances have been updated with new instruments. The engine manufacturers have also been allowed time to adequately prepare better testing procedures. Even with all of these improvements there still remains two pressing problems. They are: (1) the large number of engines awaiting the testing facility and (2) the advent of the higher thrust (turbo fan, J75, etc.) jet engine has again outdated the test facilities. B. Present Test Problems: Because of the rapid expansion of the test facilities to accomodate the increased workload of jet engines and the complexities of the higher thrust engines, many problems arose in acquiring an adequate balance between the production and quality control functions. These problems are presented in the following sections. They are grouped into areas in order to present a detailed view of each. It should, however, be noted that the problems actually overlap into other areas and even overlap each other. Many times a 0 JiIH' $rt o m"F.'l"?r,wmMWIIZJ n ¢F 1 P' 1==_.1_ .nt: tstr _tr H rt*tt particular problem arises because of testing techniques, instrumentation, and the facilities being used. 1. Present Testing Methods In order to fully understand the problems associated with the present testing techniques, the following is submitted as a general discussion of the overall testing procedure. The typical jet engine test cell has two or three men assigned to it. During the initial installation phases two men perform all necessary physical connections. This will include steps (a) and (b) of the test procedure. During the running/testing of the engine, one of these men will control the throttle and instrumentation necessary to run the engine and make recordings while the other man makes the balance of the necessary recordings at the appropriate times and places. A third man acts as an inspector. His job is to observe the readings being made and perform a reasonableness check on certain limits to see if recording errors have been made. He also takes observed readings and corrects them to a standard day (sea level or other) condition for comparison with the technical order specifications. On the final analysis he either accepts or rejects the engine based upon its performance within the limits and specifications of the manufacturer's technical order. As the engine is routed to test from final assembly it is complete as required by technical order to the final piece of safety wire. a. Dres sdown Upon receiving the engine at the test area, numerous steps are necessary in order to prepare the engine for testing. The first step is checking for possible external damage which might have occurred during transportation. The engine normally is assigned to a particular test cell prior to dressing for test. Special plugs, fittings, and some harness have to be removed in order to install test equipment. Special test harness is installed in order to obtain individual thermocouple readings for temperature spread checks. Various pressure taps are installed throughout the engine in order to obtain internal air, oil and fuel pressures. Engines are so designed that internal pressures must meet certain limits. If engine internal ratios are below values outlined by engine manufacturer, it becomes necessary to change some specific internal clearance in order to obtain required ratios. o The next step is to install a workhorse tail cone or afterburner. Altogether, there are approximately ten test fittings and adapters that must be installed in addition to temperature harness. Engine mounting adapters and bellmouth adapter rings are installed. Finally the engine oil tank is filled to capacity. This about completes the initial dressing. If the engine is designed for an after-burner, then an AB is attached. There are additional functions of preparations to be performed after the engine enters the test cell. 2 191 b. Preliminary Check After installation of the engine in the test stand, it is necessary to perform some inspections at particular times. This will include such inspections as freedom of compressor rotation and making sure no foreign objects are present in the compressor inlet. It is necessary to accomplish this type inspection prior to installing the bellmouth and inlet screen. If an inspection is performed after installation of the bellmouth, it is quite easy to overlook some small item which might result in compressor damage. c. Preliminary Shakedown After engine is properly secured in the test stand with all pressure and temperature connections, attached, a complete shakedown is accomplished by a quality inspector. This shakedown is necessary to pick up anything which may have been overlooked during the engine installation. d. Functional Component Check The next step is a functional component check out. This consists of selecting the main fuel control emergency system, afterburner system, and anti-icing valves for functional operation. These checks are necessary prior to starting the engine in order to replace such items that may be faulty. C: e. Dry Run Prior to starting the engine a dry run is performed in order to flush preservative oil from the fuel components, pressure fuel and oil system. Leaks are sometimes found during this check. Afterwards, the dry run oil system is replenished and the pressurizing valve sense line reconnected. f. Running Prior to Acceptance The engine is then ready for a start. After the engine has started and reached idle R. P. M. a complete shakedown is made to check for air, oil and fuel leaks. If no abnormal conditions are found, power is advanced toward top power and preliminary checks are made on oil pressure, E. G. T. and vibration. g. Performance Runs After the preliminary run has been completed, the engine is ready for a performance test run. This test run consists of numerous functions in order to test the basic engine and its attached components as a complete assembly. o 3 192 ! o ! Other checks that follow during the actual performance test are acceleration checks, simulated afterburner runs, emergency system runs, oil consumption check and performance calclilations. The test run begins with an initial power advance after start to approximately nine thousand RPM. This is necessary in order to obtain specific data for test run calculations and warm up the engine oil. The engine oil must be heated to actual operating temperature in order to obtain valid consumption during test run. Oil temperature must be noted at the time oil level is checked on a sight gauge and again at completion of the test run. Oil temperature at the time of the final check must be within ± 2 degrees F. of the initial temperature. Oil consumption is actually determined by visually observing a sight gauge. This sight gauge is calibrated to the engine oil tank and actually seeks the oil level within the engine tank. The oil level sight gauge is marked with ten increments to the inch and each increment represents a specific amount of oil. The data collected during the initial warm up period is used to determine the exact power position required for various test runs. Four power runs ranging between seventy-five per cent and take off are required in order to help determine the quality of the engine. Test run power positions are determined by charts representing given thrust positions. All data from such charts represent standard day conditions biased for temperature variation. Actual thrust requirements are subtracted from points corresponding to various power positions by using compressor inlet temperatures. Once having obtained required corrected thrust output, this data must be converted to actual time conditions. This correction is a function of present time condition variations from a standard day and test cell correction factors. Each individual run has a time duration of five to twenty minutes depending upon the position of power. Recordings of internal pressures from compressor inlet to turbine discharge are made. Temperatures of air inlet, oil, fuel and turbine discharge air are logged. Other recordings such as fuel flow, thrust, turbine discharge pressure, RPM and vibration are necessary. o All data logged directly related with the functional operation of the engine must be corrected to a standard day condition. This data is also corrected for compressor inlet temperature, barometric pressure and test cell correction factors. There are approximately 175 calculations performed during the test run. Thirty-five points are plotted on special graphs in order to determine if any maximum limit; has been exceeded. Also plot points are necessary in ot;der to determine minimum RPM required to obtain guaranteed rated thrust. Other correction factors which are necessary pertain to the emergency fuel flow and cooling air ratio. Other checks of the emergency system consist of acceleration procedures and engine starts. Such steps are necessary in order to determine if the emergency system has the ability to operate properly and take over engine operation in the event the main system fail. Cooling air ratio is a necessary factor in order to determine if a sufficient 4 amount of air is being furnished to the hot section parts. If the air ratio is below values outlined by the engine manufacturer, damage could occur to some parts . . h. Simulated AB Runs After completing the necessary performance checks, the engine afterburner system is simulated. The complete afterburner system is subjected to all functions of operation without actual firing. The method used is simply to rout afterfurner regulator fuel back to the pump inlet. The ignitor valve will fire, nozzle control will function and afterburner regulator will meter fuel. This system actually is quite practical insofar as all fuel is returned to the inlet supply. 2. Testing Techniques Jet Engine Testing has many problems associated with the techniques encountered using the present manual methods. Some of these problems can be directly associated with human capabilities and reactions during the test cycle. These represent man's inability to cope with the complex situations and the split-second decisions at a speed and with the accuracy required for maintaining a high quality test procedure. other problems can be attributed to inaccuracies in the existing mechanical and electrical means of transmitting test data to the test cell personnel from its primary source on the engine. These problems are created because a primary signal in the form of an electrical pulse, voltage or current, pressures, and temperatures must be converted to a mechanical means of display for use by the test operator. a. Standard Tests A standard test is defined as one in which the test procedure for each type, model, and series engine is conducted in the same manner each time it is conducted, e. g. all data are gathered the same, analyzed the same, and all decisions are made under the same rules without variance. This does not mean that the magnitude of each number in the recorded data will be the same each time, but the manner and intervals at which the recordings are made remain constant. If an engine is tested and found to be acceptable under one set of ambient conditions, it should also be acceptable when tested under another set of changed ambient conditions. The procedure for testing after overhaul contains the necessary charts and calculations to correct all recordings to a standard day condition; thus, all data should be acceptable under the standard test limits each time it is taken, if it is acceptable at anyone of the times. 5 I #Sttri##tC_M o Even though the testing instructions gives a description of the major procedures to be followed in testing a jet engine, it would become an insurmountable task to specify to the test cell personnel all the exact steps to be taken during the test. Located at the test facility are many different operators and inspectors, (quality personnel). Because each man is capable of thinking and making individual decisions, he will conduct a jet engine test in a different manner. Because the technical order allows the variations in the manner in which a major test step may be conducted, each operator will not perform each step the same. This situation as well as inconsistencies in the test cell instrumentation will create many different techniques in testing and a possible multiple variations on the acceptance or rejection of an engine under varying conditions. Not all of the problems associated with the Jet Engine Test can be completely removed by achieving the standard test alone. However, in the process of achieving this standard test many of the "ills" of the present method of testing would have to be eliminated. The achievement of a standard test can only be realized after correction of the problems in the forthcoming sections. c b. Correlation of Test Cells In most test facilities there are two or more test cells. In order to obtain a standard set of test data on a engine test in one or more of these cells, it is necessary to inter-correlate the cells. Either a "gold plated" or standard engine that has been tested in the manufacturer's cells is tested in the production cells. This process is commonly known as calibrating a cell. It involves running the standard engine in the production cell, comparing the data gathered with the instrument recordings made in the manufacturer's cells. This will produce a correlation or correction factor to be used with each cell. The correlation of one cell may require from five to eight hours to completelonger if trouble is encountered. Trouble is common. Difficulties arise from changing cell ambient conditions (air temperature, humidity, etc.) inaccuracies of data from readout mechanisms, changing of test cell personnel, etc. The accuracy of the data acquired in a final test phase will be directly dependent upon the degree of accuracy obtained in calculations of each cells correlation factor. Not only are the inaccuracies involved a problem, but there are extra manhours, fuel costs, and engine wear characteristics encurred. o Rather than correlating every ninety days as is now required, X (average) and R (range or deviation) charts of all instrument reading deviations from those readings 6 195 of the production correlator engine would· produce cell correction factors. This would allow a constant updating of the cell correction or correlating factor as well as indicating trending abnormalities that may be developing. Using the present techniques of testing jet engines; it is impossible to gather sufficient data, calculate the X and R's of the recordings, and do the correlating. The data recordings must be gathered and analyzed over a sufficient period of time to detect trending conditions. This usually involves such things as EPR's, EGT's, Nl and N2 speeds and their average and range deviations from the standard engine recordings. Even if it were possible to gather the data, the magnitude of the calculation and analysis is enormous and would require many manhours. c. Penalty Runs It may become desirable after either a major test or a test segment completion, to conduct a penalty run. The penalty run would involve running a small segment of the test, several test segments, or the complete test. After the completed test and the performance calculations have been made, the engine results could indicate an off specification; thus, requiring the need for a recheck of the calculations and test recordings. Many times when a borderline situation exists, the inspector will call for I the same recheck. Because of the inconsistency in testing methods, calculations, and decisions, the inspector may feel it necessary to repeat a portion of the test in order to gather additional data for analysis, or verification of calculations and recordings. Even when a penalty run is made, conditions may exist (the need for simultaneous readings) that cause the data accuracy to be insufficient, e. g., it is impossible to obtain simultaneous recordings under the manual methods. Because the operator and inspector know of the inconsistencies that exist, several extra minutes or hours along with many extra gallons of fuel may be consumed in conducting the penalty run in order to obtain suffieiently accurate data for a correct test. c 3. Instrumentation There are many and varied problems in the instrumentation areas. The sensing elements on most instruments are reliable and accurate. However, the actual readout mechanism is very difficult to keep within the calibration limits. Because most readout mechanisms present problems of nonlinearity in changing from one setting to another, time and manpower must be spent on a periodic (usually monthly or bimonthly) basis to insure accurate calibration. Many tirnes an instrument can become erradic in its reading and the test cell personnel not become aware of it until a new calibration is made. In the mean time many "good" engines have been rejected and many possible "rejects" are flying or in storage. 7 196 C =- - o t t n 4. Human Error Throughout the test procedure recordings are being made on a second timing intervals. Many of these readings should be made simultaneously, but because of the human inability to observe and record on a "split" second basis many of the reading~ will change by large increments before they can all be recorded. This is especially true during acceleration and deacceleration of the engine. Because all the instruments are not located at a 90 0 angle with the eyes of the man making the recording and because many of his recordings are made at a fast rate, it has been found that many recordings have been made with large errors (sometimes a completely gross transition error- is made). A 5 lb. pressure error or 5% temperature error is enough in some readings to reject a "good" engine or accept a "bad" engine. 5. Rerun Statistics o If after the sequence of test events, calculations and plotting of data the engine does not perform according to the technical specifications it is not always rejected and sent to the rework area immediately. After a series of checks on the calculations made by himself, gross range errors on readings,or minor detectable instrument error, the inspector will apply his knowledge in conjunction with the trouble shooting points listed in the TO to diagnosing the area of trouble in the engine. These diagnostics will then be sent back with the engine to the rework area (overhaul line). If he feels that some element of doubt is present in a reading or calculation, portions of the test or the complete test may be performed again. This re-running may consist of re-trimming the engine, re-running the performance runs, or giving an AB function check. Many times the ability to diagnose the problem area relys solely upon the experience and background of the inspector in charge of the test. The majority of the inspection personnel have not gained this type of experience. Because of this inexperience, many of the engines may be re-run or rejected needlessly. If the engine must be re-run several times in order to find the source of trouble, large quantities of time and fuel are consumed. If we consider the price of a complete overhaul of an engine ranging from $12,000 to $15,000, the needless reject of a good engine or the improper diagnostic of an engine for overhaul becomes an expensive waste. o Because of the advance in design of the jet type engine year by year, itbecomes a large task to keep the test personnel updated on the new techniques accompanying the advance design engines. During the period of time when the modernization of the cell and training of personnel are being done, many costly errors are made. 8 If we consider a facility that tests 2, 000 engines per year, the annual fuel bill will be approximately $550,000 per year. It has been estimated that 40 per cent of this fuel bill can be attributed to running reworked engines or performing a portion of a test over again (because of improper readings or calcuation errors). The preceding sections describe in general the testing procedures and some of its existing problems. Do not be "misled" by the seemingly simple test procedures described. There are many things not covered in as minute a detail as possible; also not mentioned are the many splitsecond decisions that must be made during the course of the test and at times when possible malfunctions occur. 6. Capacity In cases of national emergency, or increased workload responsibility, the need for increased test capacity in the high thrust cells could develop into it major production "bottleneck. " Pressure could be relieved in these situations by creating extra shifts of men to handle testing and facilities maintenance; however, the increased utilization of the cells under the present test time and procedures would increase many fold the manhours of ~aintenance as well as cause the quality of the engines released to the field to be inferior because of this increased pressure. In either case, the cost of an increased workload under these conditions, can become enormous. 7. Safety There are several events that could take place to endanger the lives of personnel working in the test cell while an engine is running. No "concrete" solution will be found to completely remove all these danger areas. The technical order regulations specify where and at what time personnel may be in the cells while the engine is running. Because of unusual circumstances, the rules are many times "bent" to fit the situation. In many of these cases, danger may be at its peak. Examination of the possible dangers of these situations reveals that there is a possibility of fuel leaks and thus flash fire while trimming. The bleed valve may also dump excess air overboard while decelerating. The force of this air can be enough to knock a man off his feet. There are also dangers from any engine part or accessories not being securely fastened and thus breaking away. 9 198 ~---.-- ..-- ._------ -- ----.--~--~----- ... --- -~~~~--.--.---~~-- -.~-"---------~-""'-- .. m o eM« _itr. II. SOLUTION TO THE PROBLEM A. Introduction Some of the problems existing in testing a turbo jet engine have been discussed in the first section of this paper. Not all of the intangible problems were brought out, but inference was made to them. The forthcoming discussion is submitted as a possible approach to the solution of many of the problems that act as a plague to the efficient and correct testing of a jet engine. There are many alternatives to the degree of automation that can be applied by the use of a computer in a jet test cell. The primary problem rests on two factors: (1) What degree of control should the system have and (2) Whether the system should be a primary "slave" to the operator or the operator a "slave" to the computer. The one chosen discusses a completely closed loop operation (In this instance, the running of the test including start-up shut-down via an IBM 1710 Control System, related hardware and any special features). The advantages and disavantages of operating in a manual and open-loop mode as compared to the chosen approach are discussed. o A great majority of the following information has been derived by working with prospective customers in the jet engine test area; however, due to reasons which will not be discussed, customers' names will not be mentioned. * B. Previous Work 1. Data Logging One of the first attempts at applying an on-line device for the logging and reduction of engine test data was tried by the U. S. Naval Airforce. A special device for these purposes was built by Gilmore Industries (3) to perform such a function. The primary design of this device was for gathering piston type engine data. Many of these were later modified to receive data from test cells geared for jet engines. The data logger was usually located in a prototype cell where certain special test runs could be made. The data logger was primarily an analog type sensing device. Its primary readouts were instrument faces, graphical x-y plotting and type writer data that had o * Contact author for further information. 10 -------.-.---,---~-:--- been convert~d by an analog-to-digital converter to a scaled digital form. The number of channels or sensing and readout elements depended upon the elaborateness of the model ordered. The acceptance of this system was "poor" especially in commercial installations (where a few are found war surplus) where the price/performance ratio was much too great. 0 This piece of equipment contained the same hinderances as the analog computer does. No logical ability coupled with an "exponential" increase in price for flexibility, plus inadequate readout accuracy. Enough of this type of gear to log data in one cell often times cost as much as the digital computer components to control multi-cells. 2. Research and Development Jet Engine Testing One of the first companies to apply a computer to the role of gathering and reducing test data was Pratt and Whitney. The computer is·an IBM 1410 with a special interface (Analog-to-Digital Converter) to take data gathered during tests conducted for research purposes. The system acts as a data monitor. It logs and reduces data only during the time the engine is in the performance run phases. Special instrumentation has been added to detect malfunction of components at high temperatures and fast speeds. After one test has been conducted, the instrument leads are then automatically connected to an engine awaiting test in another cell. o Because the purpose of this system was to do only a data logging and data reduction job, no further effort has been made to perform a close-loop function. Cases of research and development do not readily adapt themselves to a close-loop operation. There are many times when extraordinary or special tests need to be conducted which would not be compatible with the programs that had been written for test. There are also under development in the NASA Space Program the adaptation of fast general purpose computer to missile checkout. This program like all other programs in jet engine control is in its infancy. 3. Industrial Testing Systems - Discrete Process Industry has entered an era in which the processing of production and product performance information must be incorporated as a part of the manufacturing oper'~,, tion. As the profit squeeze continues along with the need for increased production~ the cost of manufacturing the product must be reduced to maintain or improve the profit position. Much has been done and is being done to reduce the cost of making the product through advances in technology and by automation. However, the costly operation of quality assurance which continues to receive more demanding tasks is not keeping abreast of its production counterpart. To parallel the giant step made in manufacturing through automation, the quality assurance program in industry has 11 2 tj 0 CI H' _7"22'1 o 'P'!" IEt_ tr '1 tnm.w· made and continues to make drastic advances through in-process test and inspection systems. Some of the industrial testing applications using industrial process control systems are: 1. Space Vehicles - Analog to digital converter used in logging, reducing and analyzing data on space vehicles in environmental chambers. 2. Potentiometers - Final testing of potentiometers. 3. Automobiles - On -line quality control to determine defects in assembly as they happen. 4. Aerospace Nose Cones - FM Tape playback of data telemetered from missiles. 5. Nuclear Research - On -line recording of information from a spark chamber. 6. Atomic Powered Naval Ships - On-line measurements and computation of shielding experiments. o In general Industrial Testing with control systems controls plant test procedure, analyzes product test data and contributes to production test equipment the capabilities of: 1. Testing dynamically at prod~ction speeds. 2. Correlating the test data for each product. 3. Determining the classification of each produced unit based upon specification. 4. Sorting product unit after final test. 5. Storing test data for future analysis. 6. Initiating reports during production runs. 7. Checking and calibrating of test equipment during production runs. 8. Scheduling produced product. 9. Determining critical trends as they develop. 1 The prece~ing paragraphs have shown the development of automatic control systems in the continuous process industries and manufacturing operations involving discrete processes. In each case one of the main objectives is increasing the quality of the end product. It can also be clearly seen that automatic testing is not an idea with unproven results but the missing link between production and quality. o 1. IBM Application Brief, No. K20-1725 12 ·20 o C. Automatic Jet Engine Test Control System 1. System Design Requirements In the preceding sections of this report the various phases of the actual jet engine test were discussed in moderate detail. These are functions performed by the operator, recorder, and inspector. The following describes the functions the control system will perform in regard to the various test phases. The functions are necessary to deliver a high quality engine with minimum cost. a. Control of the Independent Variables to Set Up and Sequence Tests The jet engine control system will select various test phases for an individual type, model and serial number engine -use information gathered from the engine in the test cell, such as pressures, temperatures, flows, etc., and determine appropriate test sequences and procedures taken from the Technical Order to send control signals to the engine in the cell. By designing the test phases as a series of logical steps, the system will use each test phase as a sub-program and execute the over-all series of sub-programs under control of a master monitor routine. b. Data Acquisition and Control c Each instrument pick-up will be connected to a transducer which will be connected to a transducer which will be connected to a multiplexer and terminal unit which will be connected to an analog-to-digital converter. The analog-todigital converter will provide a digital voltage to the control system main frame. The main frame will scan all instrument leads for each pressure, temperature, flow, etc and convert these by the use of equations into meaningful engineering values. These values will then be used to control the system. The system will also convert a digital value to an analog voltage for control of the throttle, trimmer and other relay switches in order to control the speed, thrust, fuel flow, and other controllable variables. c. Calculation of Performance Parameters After gathering all data (instrument readings), one of the test phases will correct all data to a standard day (usually sea level) condition in order that all parameters may be compared against the T. o. limits for trimming and reject status. o 13 202 St.' o d. t.o_ Operator Guide for Engine Adjustment Such things as warning messages, trim guides, test status, etc, will be logged for the operator. Any transducer reading will be available upon operator demand. Any time the engine must be stopped or shut down by the control system, a message will be logged on the typewriter giving the reason, a complete diagnostic, and recommendations for repair or rework. e. Automatic Instrument Calibration This can be done by either or both of the following: 1. Comparison of a known standard signal with the transducer output from this signal. 2. o f. Comparison of the transducer output to other related signals. This will, in essence, tell if the signal is abnormal (too high, too low, or fluctuating). From this an automatic calibration can be done. This will insure against catastrophic results from a faulty transducer. Check Calibration of Installed Engine Transducers Many pick-ups are installed on the engine during dress-down, thermocouples, tachometers, etc. It is possible for one of these to be faulty (disconnecttion or off speCification in the thermocouple not detected during test). By using the calibrate feature, control system abnormalities may be detected before the actual test. g. Conduct Penalty Runs After the major test phases are completed and the acceptability of the engine is ascertained, it may be necessary to re-conduct portions of the major test or call upon special penalty run procedures to be executed. This need arises when certain T. o. limits have been exceeded or an engine has been accepted on a marginal condition. This will insure correctness.of data and calculations as well as insuring that an out-of-limit condition was not a transient. The ability to automatically select and execute these routines under control of the automatic system will improve both speed and accuracy of the over-all test. . h. o Engine Diagnostics During the running of a sub-portion of the test or aft~r completion of the major test, conditions may arise that will indicate off specifications in the engine 14 1III J,I I, or one of its components. By gathering data at high speeds, using past historical data on engine rejects, failure incidents and rework data, and building a series of logical steps or a mathematical model of certain sections of the engine, it will be possible to determine the exact cause of the abnormality and make recommendations for repair. There will be a learning process by the system. is gathered, the logical model will improve. o As more and better data The ultimate aim of the system is to furnish complete re-work information to the engine penalty line. In many cases, this will save time and prevent unnecessary rework of an engine. i. Logging and/or Punch-Out of Test Data and Engine Data After a major test has been completed, all instrument readings, calculations, and diagnostics remarks will be stored on the disk storage unit. The operator in the cell control room will execute a request to the central control system room via the manual entry control. The control system then will print a completed log or run sheet giving the three items above for each test phase. The log may be used by the operator to select penalty or re-runs if it appears that a componm t or recording is marginal to the limit. Several carbon copies may be produced so that copies may be sent to all authorized personnel. An engine data plate card will be printed to accompany the engine and a military run data card punched for Quality Analysis. j. Store Test Programs and Parameters for All Type, Model and Serial Number Engines The control system will use a mass random access unit for storage of the test parameters and limits for all engine models, types and serial number that will be tested. This type storage insures immediate access to all types of engine programs for complete asynchronous testing and control for the test facility. Mass storage will allow the system to be open ended ~or expansion to future cells. By the use of this mass storage, a better and more complete engine diagnostic can be performed (as pointed out in the previous section). The system will be designed to allow the updating of all engine technical orders on a daily basis. As a secondary function, statistical data will be stored for analysis. By storing summary data, critical trends can be detected early. All causes for rejects or defects can be stored by type, model, and serial number. Summary data will be quickly available upon management request. 15 0, t 0 ', I .1 k. tr. t&rtMtzd . t+rl Detect Emergency and Unsafe Conditions and Take Appropriate Action The fast instrument scanning speed of the control system permits dangerous trends to be detected in many of the instrument readings and appropriate corrective action to be initiated to prevent occurence of out-of-limit conditions. In out-oflimit situations, the system will quickly bring the test and engine to a halt to prevent serious damage. One of the most important things to consider when designing the actual control system is achieving a high degree of reliability. Two types of failures can occur. The failures and corrective actions are: Type I - Transient Failures These are internal system transmission errors and occur on a transient basis. In this case, the system will record the failure and try twice more to perform the operation. The recording will be used by maintenance engineers for regular preventive maintenance (once per week). A transient type error will usually be eliminated in three attempts. Type IT - Complete Component Failure o In this case, the system will try to by-pass the bad component switching to manual control or bring the engine to a safe stop. A by-pass procedure will be incorporated for' emergency action. 1. Quality Analysis 1) The system will use store data to perform reliability calculations for engine and individual components. 2) The quality analysis will produce data assurance for a better test engine. m. Production 1) Scheduling - Using advanced techniques such as linear programming, a master plan will be prepared for scheduling the cells. 2) Planning - Better methods of machine and manpower utilization can be prepared. 2. Control System and Interface Description o To approach the problem of determining the necessary hardware, one must keep three factors in mind. They are (1) design functions as determined in the meeting of section 1 (2) instrumentation-present and future, (3) and layout of the basic test cell. 16 --~------ .. ---.,.-- If we notice the basic test cell layout as shown in illustration I, it shows the location of the control room as being between two test cells. If there are more than two cells (there are usually several more) then it is logical there will be two or more control rooms. Because a typical control system will control more than two cells, it will be necessary to locate the computer in either a remote location or in the rear of one of the test cells. When this is done, there arises necessity for remote communication devices. 0 ,' , Attention should be drawn to the design function to operate in conjunction with this communications device. Whether the operator or the central computer system is the "slave". It will be necessary to place a device for the operator (inspector, etc.) to select the particular test function he wishes to perform. It will also be necessary for him to get return information from the instrument readings, pertinent calculations and emergency or troubleshooting messages. Many times it will be necessary for the test cell foreman to have information concerning phases of test of engines in each cell in order to coordinate the overall movement of engines in test. He will also need access to stored statistical information pertaining to reject, re-run and other engine test functions. Many times upper level management will inquire of the cell foreman on these statistics. Things that could be available on an inquiry basis would be: 1. 2. 3. 4. 5. Number of rejects/month on a certain model number. Major cause of rejects. What was done for correction. Ranges and standard deviations from set standard operating limits. Etc. Using the system described in illustration 2 and treating the requirement as 8 test cells, each component and its function will be discussed. a. Central Computer (1620 Model IT) Because of the speed needed to accomplish the sampling of the necessary instrument leads" making all necessary calculations, actually sending output signals, for control and receiving feedback input signals for correction, the 1620 Model IT with a 60,000 position memory was chosen. The 1620 as the heart of the 1710 Control System contains the necessary machine instructions and programming systems (Executive System-e. g., monitor) to operate in conjunction with an asynchronous test system design. b. Auxiliary storage (2-1311 Disc Files) Even though the majority of the skeleton test functions are the same for all engines, there still remains different test parameters for each engine model. Each of thes~ parameters must be stored for immediate access. Because no central computer memory would be large enough to contain all test program phases, these must 17 0 , ~~~ o--~ 0 N ~ 1712 D 1712 n 1712 m STORAGE WORK TABLE 1625 J 1711 1620 1622 I~I I:;; 1443 o J I~ I 26 [ n I Fa. I + c o DESK - SCALE: 1/4-= .' 1712 IV. V • VI • VIII ; CARD FILE -- DESK CONTROL ROOM I I also be stored for immediate access as they are called by the skeleton control program. Also contained in auxiliary storage would be necessary diagnostic routines available upon request, as well as emergency limit and correction routines. A second disc file would be used as in intermediate store area for input/ output information, if all input/output devices are busy and they would be available to store quality and production control data gained as a by-product of each test. This data would be available upon inquiry from management. c. Interface Equipment (1711 and 1712's) In order to attach all necessary points for 8 cell (see Appendix B) and con-' vert the analog (electrical) signal into a digital form in a sufficient time period, the analog to digital converter (1711) has the ability to convert 200 pOints/second. In order to handle all necessary analog input points, analog output points, contact indicating and operating relays for an eight (8) test cell facility, it is necessary to have three (3) multiplexing and terminal units (1712's) to the system. d. Test Cell Input/Output Gear (1713, 1715, 1717) Located adjacent to each instrument control panel will be an IBM 1713, 1715 and 1717. The operator will have the option with the IBM 1713 manual entry devicethrough a set of coded instructions -to dial in either the command for a complete test or portions of a test. The command will be dialed through the use of twelve (12) rotary knobs with zero (0) to nine (9) selection ability. An enter key will be hit, the information will go via the SIOC channel and interrupt the computer, the computer will read the rotary knobs and start the processing. e. Interface All instruments that furnish an electric signal of a standard form will be sent via shielded cabling to the 1712 multiplexing unit, all non standard (pulsed, etc) and pressure type signals will be transformed via transducers (in the test cell control room) to an electrical form and ~e~t to the central system complex. All pickup signals from the engine are easily adjusted to the standard 1710 signals; however, more specialized servos must be bought or designed to control the throttle and trimming mechanism. There are several types of stepping motors or feedback systems on the market today that can handle these tasks. c 19 o All existing cell instruments will remain intact as man;ual back-up for the system. Through a specially designed panel, the operator will be allowed the option of switching to either automatic or manual system at any time. m. ECONOMIC JUSTIFICATIONS The justifications for considering the "Control Systems Approach to Jet Engine Testing" can be broken into tangible, intangible and possible savings categories. The justifications can vary depending upon the application. Some of each are listed as follows: A. Tangible 1. Increased Engine Throughput: This can be accomplished by a. Simplifying the testing procedure. b. Decreasing delay in such things as trimming and shakedown. c. Operator Guide Print-Out for prompt emergency and testing actions. o The best time estimate for engine throughput with no major hindrances is 5 hours 55 minutes. As previously mentioned, the average throughput is approximately eight (8) hours for an engine with time running up to twelve (12) hours if there are several re-runs or persistent trouble exists. The control system would increase the capability of the cells to take on added workload without added facilities. This need would arise in wartime emergency for federal customers and with added contractual obligations for both commercial and federal. 2. Reduced Manpower Requirement/Engine This would free inspection' and operating personnel for a greater engine throughput. One operator would be substantial for testing procedures, where the present system utilizes an inspector and two (2) shops or production personnel. 3. Avoiding Re-Run of Engines o a. By eliminating bad instrument calibration--erroneous transducer signal. b. Bad instrument reading--can be eliminated. The signal will originate completely at the transducer and eliminate' the nonlinearity of the instrument . read-out mechanism. Operator error in reading will also be eliminated, e. g., simultaneous reading of instruments. 21 4. Decreased Ftlel Costs This can be saved with automatic trim procedures and avoidances of excess penalty runs. 5. Decreased Calibration Costs By automatically calibrating the transducers the computer will give a correction or tare factor for the back-up reading devices. The time between calibrations will decrease. The maintenance costs will correspondingly decrease. B. Intangible·s 1. Better Engine Quality a. Better checked out engines through more certain detection of off-specification units. b. Simultaneous recording of the instruments, thus insuring proper data for checking limit parameters. c. Consistent methods of testing, thus insuring proper acceptance or rejection of an engine. 2. Decreased Re-Work Costs Through Better Diagnostics As mentioned previously, the system with its on-line mass storage can furnish pinpoint diagnostics to eliminate complete overhaul for minor abnormalities or defects. 3. Data .Assurance This assurance can be derived through getting simultaneous readings or reliability in instrument calibration and will result in better customer (the pilot or branch of the armed services) satisfaction. 4. Increased Safety for Personnel, Equipment and Property 5. Increased Readiness Program on First Line Aircraft 6. Reduced Paperwork Handling Complete and accurate unit performance logging. Here we have better customer satisfaction through hard copy records. Diagnostics are automatically printed to be sent back for re-work.· -, 22 2i 0 ·, 1m _N* _r tt .. b IV. SUMMARY AND CONCLUSIONS This paper has attempted to discuss one particular approach to the application of a digital computer to the closed loop control of a jet engine test cell. As has been pointed out, there are many approaches to consider in designing a system for a particular application. In summary, the things that must be considered are repeated and listed as follows: 1. Degree of Control Desired-Open or Closed Loop 2. How Much the Operator is "Slave" to the Computer or Vise Versa 3. How Many Test Cells Must Be Controlled Simultaneously 4. The Functions that the Customer Wishes the System to Perform The justifications for a control system can be varied, depending upon what the customer wishes to accomplish; however, the "state of the art" of automatic control in jet engine testing is in its infancy and there are many justifications in all cases. o It should be pointed out that the general approaches and ideas used are appli- cable to many other industries and are not limited to jet engine testing. 23 o BIBLIOGRAPHY 1. o Industrial Testing Systems at the IBM Components Division Poughkeepsie, New York. Application Brief, No. K20-1725 2. Sales Guide Industrial Testing Systems IBM Internal 3. Introduction to Control Systems General Information Manual F26-5577-0 24 o o GOOO/iE4R GOODYEAR AEROSPACE CORPOR,\TION ARI10NA DIVI~ION lITCHF IELD PARK ARllONA GENERALIZED C"'·, FILTF~ NETWORK I,IJ ALe STEADY STATE ANALYSIS PROGRAM by D. Ho O'Herren AAP-18911 o May 1, 1964 I~l l i1 II I o TABLE OF CONTENTS I Generalized Filter Network Alc Stea~ State Analysis Program • • • • • • • • • • • • • • • • • • • 1 II Program Input Data • • • • • • • • • • • • • • • • • • 6 III Sample Problema 8 • • • • • • • • • • • • • • • • • • • -. 8 a. Sample Problem 1 • • • b. Sample Problem 2 • • • 11 o. Sample Problem 3 • • • • • • • • • • • • • • • • 16 d. Input Data Problem 1 • • • • • • • • • • • • • 0 22 e. Output Problem 1 • • • • • • • • • 23 to Input Data Problem 2 • • • • • • • • • • • • • • 24 g. Output Problem 2 • • • • • • • • • • • • • • • • 2, h. Input Data Problem 3 • • • • • • • • • • • • • 27 10 Output Problem 3 • • • • • • • • • • • • • • • ·• • • • • • • • • • ·• ·• • • • • • • • • • • • • · . .' . IV APPENDIX A v Li sting of Fo rtran II Source Program • • • 0 • • • • • ·• ·• 0 ·• • • • • 0 • • • • • • • • 0 • • • • , ~, ,II 1,1 l \'.lI 28 38 c 2 -1".i· 'II -_...._ - - - - - - - ._.-.-.._ - - t o .t _m GENERALIZED FILTER NE'IWORK Ale STEADY STATE ANALYSIS PROORAM This program has· heen written to make possible comprehensive surveys of theoretical filter designs. It opens up a more sophisticated range of filters to theoretical consideration and evaluation. The program' input is general enough that almost any filter network consisting of cascaded inverted-L or symmetrical lattice sections may be handled easily. The minimum machine requirements are a 1620 with La K core storage, auto divide, and indirect addressing. The source language is Fortran II. There are 6 au bprograms plus the mainline program. ~ Filter design has been speeded in recent years with the advent of tables of normalized low-pass filter element values*. Even if these tables are used, this program allows the desiener to compute the effects of component tolerances, finite Q's, and mismatched terminations. These introductory remarks have centered around. .filter design, rut it will be apparent that the program is useful for analyzin~ any RLe network, e.g., amplitude or phase equalizers. The filter designer needs to know how a proposed design will perform over a particular range of frequencies before making recommendations to those who will implement the design. Manual 0. 1. o PROGRAiVi 0.0 221 1 1 1 +.lOOOOOOOE-U5+.00000000E-99+.10000000E+04 +.OOOOOOOOE-99+.20000000E'::05 1 2 1 +.lOOOOOOOE-07 +.OOOOOOOOE-99+.00000000E-99+.10000000E+06 +.OOOOOOOOE-99+.1000nOOOt-O~ 1 1 0 +.OOOOOOOOE-99+.00000000E-99+.5000COOOf+02 o t)"e-- 2 cJ t.t GENERAL NETWORK ANALYSIS PROGRAM 0 SAMPLE PROBLEM 1 FOR GENERAL FILTER NETWORK ANALYSIS lOR:: 50.000 S~CTIONS= ORDER K= 1 , C( 1 C( 2 t 1 1 , , , 1 C( 3 C (. 2 2 2 K= 0 1= 1 , , , t 2 t 1 1 , 3 , F 1.0000 1.0010 2.0000 2.0010 3.0000 3.0010 4.0000 4.0010 5.0000 5.0010 6.0000 6.0010 7.0000 1.0010 8.0000 8.0010 9.0000 9.0010 10.0000 10.0010 MQ:c .1 .10000000E-05 .OOOOOOOOE-99 1 .00OOOOOOE-99 .20000000E-05 R :: R = .lOOOOOOOE+04 .OOOOOOOOE-99 :& .OOOOOOOOE-99 .OOOOOOOOE-99 .lOOOOOOOE-05 R •. R = R = .OOOOOOOOE-99 .10000000E+06 .OOOOOOOOE-99 L = .00OOOOOOE-99 R .50000000E+02 = = L l 1 Mp: 1 1 ) = 1 1 ) :: ) ;I 2 MO= .lOOOOOOOE-07 .OOOOOOOOE-99 .OOOOOOOOE-99 1 :: L = L L 1 I= 3 lI: 1 Neyel::: MP= = ) ) 315 ORDER C( 1 1 I= 2 C( lIS 2 29APR64 0.000 VOl- 1.000 LETTR= 2 2IS ORDER K= VORa PROGRAM MP= 1 1 ) :: i MQ= .OOOOOOOOE-99 DB 4.9381 4.9450 10.1053 10.7105 15.4715 15.4819 19.5824 19.5862 23.2046 23.2981 26.4500 26.4531 29.3887 29.3915 32.0716 32.0741 34.5311 34.5395 36.8156 36.8118 BETA 76.462 76.513 114.502 114.531 138.912 138.993 156.885 156.901 170.717 170.729 181.741 181.751 190.725 190.133 1-98.113 198.180 204.434 204.440 209.160 209.765 0 ZREAL 5.2953 5.2851 1.1714 1.1760 .4043 .4039 .1 711 .1709 .0829 .0828 .0443 .0443 .0256 .0256 .0151 .0157 .0101 .0101 .0068 .00b8 II: ZIMAG -9.2583 -9.2400 4.1843 4.1947 13.2695 13.2711 20.9877 20.9950 28.1290 28.1360 34.9788 34.9855 41.6625 41.6691 48.2434 48.2500 54.1565 54.7630 61.2225 61.2289 T(F) .14120277E-06 .81361108E-01 .57249998E-07 .43333332E-07 .34055554E-07 .27527777E-07 .22638888E-07 .18916666E-07 .15999999E-07 .13666666E-07 o -23- 2 ;j ,'j SAMPLE PROBLEM 2 NETWORK ANALYSIS PROGRAM 50. 2. 100. 211 +.10000000E+05+.S0000000E+Ol 100. IMAY64 1. 0.0 o • 1 +.10000000E+OS+.50000000E+Ol 1 2 +.30000000E-I0 +.20000000E-07 +.OOOOOOOOE-99+.10000000E-06 1 +.lOOOOOOOE+05+.50000000E+Ol 1 2 +.30000000E-09 +.40000000E-06 +.OOOOOOOOE-99+.15000000E-06 1, 1 0 +.OOOOOOOOE-99+.00000000E-99+.10000000E+03 o GENERAL NETWORK ANALYSIS PROGRAM SAMPLE PROBLEM 2 NETWORK ANALYSIS PROGRAM 1MAY64 ZOR= 100.000 VOR= 1.000 VOI= 0.000 RAO: .lOOOOOOOE+05 RBO= .50000000E+Ol SECTIONS= R( 1, K= C( 1 C (2 C( 4 R(l, 1 , 2 K= C< 2 LETTR= 1 NCYCL= .10000000E+05 R( 2, 1 ) t: Ir 1 MP= 1 MQ= 2 1 , 1 ) = .30000000E:-IO L::: 1 , 1 ) = .20000000 E-07 L = 1 , 1 ) = .OOOOOOOOE-99 L:: ): .10000000E+05 R(2, 2 ) 1= 1 MP= 1 MQ= 2 2 , 1 ):a: .30000000E-09 l::: 2 , 1 ) = .40000000E-06 L = 2 , 1 ) = .OOOOOOOOE-99 L z:: 1 ) 2 1 C( 2 , C( 4 ORDER K= 315 1 1= 1 3 F o 50.0000 50.001Q 52.0000 52.0010 54.0000 54.0010 56.0000 56.0010 58.0000 58.0010 60.0000 60.0010 62.0000 62.0010 64.0000 64.0010 66.0000 66.0010 68.0000 68.0010 70.0000 70.0010 72.0000 72.0010 14.0000 14.0010 76.0000 76.0010 78.0000 = MP= 1 MQ= .OOOOOOOOE-99 DB 21.6168 21.6170 21.9569 21.9511 22.2846 22.2848 22.6008 22.6009 22.9061 22.9063 23.2015 23.2016 23.4874 23.4875 23.7646 23.7647 24.0334 24.0336 24.2946 24.2947 24.5484 24.5485 24.79:>3 24.7954 25.0356 25.0358 25.2698 25.2699 25.498,2 BETA 89.216 89.216 89.613 89.613 89.994 89.994 90.361 90.361 90.716 90.716 91.060 91.060 91.393 91.394 91.718 91.718 92.033 92.034 92.341 92.342 92.642 92.643 92.937 92.931 93. 225 93.226 93.508 93.509 93.786 0 L = 1 ORDERS .50000000E+Ol = C( 1 , 3 , 1 ) o 1 = .OOOOOOOOE-99 .OOOOOOOOE-99 R::: .OOOOOOOOE-99 R = .OOOOOOOOE-99 .OOOOOOOOE-99 .10000000E-06 R 1 ORDERS .50000000E+01 = R:: .OOOOOOOOE-99 R = .OOOOOOOOE-~9 .OOOOOOOOE-99 .00000000E-99 .150000001:-06 R = .OOOOOOOOE-99 .OOOOOOOOE-99 R = .lOOOOOOOE+03 ZREAL 5.2596 5.2596 5.2426 5.2425 5.2274 5.2274 5.2138 5.2138 5.2016 5.2016 5.1906 5.1906 5.1806 5.1806 5.1715 5.1715 5.1633 5.1633 5.1558 5.1557 5.14A8 5.1-488 5.1425 5.1425 5.1367 5.1367 5.1313 5.1313 5.1263 ZIMAG -4.8518 -4.8517 -4.6662 -4.6661 -4.4943 -4.4943 -4.3347 -4.3346 -4.1859 -4.1859 -4.0471 -4.0470 -3.9172 -3.9171 -3.7953 -3.7953 -3.6809 -3.6808 -3.5731 -3.5731 -3.4715' -3.4714 ,-3.3755 '-3.3755 -3.2847 -3.2847 -3.1987 -3.1986 -3.1170 T(F ) .55555554E-09 .55833331t-09 .52499998l:::-09 .52777776E-09 .472222201::-09 .472222201:-09 .47222220E-09 .41666665E-09 .41666665E-09 .44444443c-09 .416666651:-09 .41666665E-09 .38888887E-09 .38888887E-09 -25- 78.0010 80.0000 80.0010 82.0000 82.0010 84.0000 84.0010 86.0000 86.0010 8S.0000 88.0010 90.0000 90.0010 92.0000 92.0010 94.0000 94.0010 96.0000 96.0010 98.0000 98.0010 100.0000 100.0010 25.4983 25.7209 25.7210 25.9384 25.9385 26.1509 26.1510 26.3585 26.3586 26.5616 26.5617 26.1604 26.7605 26.9549 26.9550 27·1455 27.1456 27.3323 21.3324 27.5154 21.5155 27.6950 27.6951 93.786 94.059 94.059 94.32R 94.328 94.593 94.593 94.853 94.853 95.110 95.110 95.364 5.1263 5.1211 5.1217 5.1174 5.1174 5.1134 5.1134 5.1097 5.1097 5.1062 5.1062 5.1030 95.364 5.1030 95.614 95.615 95.862 95.862 96.107 96.107 96.349 96.349 96.588 96.588 5.1000 5.1000 5.0971 5.0971 5.0944 5.0944 5.0919 5.0919 5.0896 5.0896 -3.1170 -3.0395 -3.0394 -2.9657 -2.9657 -2.8954 -2.8954 -2.8284 -2.8284 -2.7645 -2.1644 -2.7033 -2.7033 -2.6449 -2.6449 -2.5889 -2.5889 -2.5353 -2.5352 -2.4838 -2.4838 -2.4344 -2.4344 .38888887E-09 0 .36111110E-09 .36111110E-09 .36111110E-09 .38888887E-09 .36111110E-09 .36111110E-09 .36111110E-09 .33333332E-09 .361111101:-09 .33333332E-09 .33333332E-09 o 01 239 SAMPLE PROBLEM 3 NETWORK ANALYSIS PROGRAM O?· 5. 2 1 1 2 o 2 100. 100. 2 1 +.lOOOOOOOE-05 +.OOOOOOOOE-99+.00000000E-99+.10000000E+04 3 1 3 +.lOOOOOOOE-ll+.OOOOOOOOE-99+.10000000E+07 +.30000000E-06+.20000000E-12+.10000000E+04 +.lOOOOOOOE-ll +.OOOOOOOOE-99+.10000000E-05+.100000QOE+03 +.lOOOOOOOE-12+.00000000E-99+.50000000E+02 +.OOOOOOOOE~99+.20000000E-12+.10000000E+03 o 1 +.50000000E-OB+.IOOOOOOOE-07+.10000000E+03 1 3 0 +.OOOOOOOOE-99+.50000000E-06 +.OOOOOOOOE-99+.00000000E-99+.10000000E+03 +.lOOOOOOOE-05+.00000000E-99+.1000QOOOE+Ob o o 4MAY64 1• 0.0 "~"~!!W\'~--' .. ~ .. -.. '.~~;"'::':";.. ~' .. ";"~,;:':;' -.",..... ,.. , ..."",,"............. =:.--..:..-~..::..:.-.;..---- 'I GENERAL ANALYSIS NETWORK PROGRAM SAMPLE PROBLEM 3 NETWORK ANALYSIS ~OR= VORa 100.000 SECTIONS= lIS ORDER Kc t 1 t 1 ORDER K= , , , , , , 2 , K= C( 2 C( 4 f K= 2 C( 2 t 2 t 2 t 2 2 2 t , , C( 1 C( 3 5 t cc 1 1 1 1 :: ) :: :: ) ) :: 2 ) J:: ) :: ) :: 1 1= 1 3 3 3 F 5.0000 5.0010 10.0000 10.0010 15.0000 15.0010 20.0000 20.0010 25.0000 25.0010 30.0000 30.0010 35.0000 35.0010 40.0000 40.0010 45.0000 45.0010 50.0000 , , , R R :: = .OOOOOOOOE-99 .10000000E+04 R :: R :: R = R ':: .10000000E+07 .10000000E+04 .OOOOOOOOE-99 .10000000E+03 .OOOOOOOOE-99 .20000000E-12 R :: R = .50000000E+02 .10000000[+03 .lOOOOOOOE-07 R .lOOOOOOOE+03 .OOOOOOOOE-99 .OOOOOOOOE-99 1 ) 3 = = :: L l 2: .1OOOOOOOE-ll L = .OOOOOOOOE-99 0 MQ= MP= .lOOOOOOOE-12 .OOOOOOOOE-99 L :: 0 2 L = L MQ= ;: L = MQ= 3 .OOOOOOOOE-99 .OOOOOOOOE-99 .10000000E-05 0 DB 180.7194 180.7228 192.0992 192.1008 198.3916 198.3926 202.1228 202.7236 206.0672 206.0678 208.8217 208.8282 211.1986 211.1991 213.2865 213.2869 215.1560 215.1563 216.8501 .B ETA 180.838 180.838 182.003 182.004 182.824 182.825 183.368 183.368 183.685 183.685 183."835 183.835 183.869 183.869 183.'828 183.828 183.741 18:3.741 183.628 .OOOOOOOOE-99 .200000001:-12 .OOOOOOOOE-99 .10000000E-05 1 .50000000E-08 MP= 1 ) = 1 ) MQ::: 1 .1OOOOOOOE-ll .30aOOOOOE-06 MP= 3 3 3IS 3 2 1 L = L = MP= ) 1= 2 :: 1 , , 1= , 2 , 2 2 0.000 3 I:: ORDER K- 1 = ) ) MQ= 1 .10000000E-05 .OOOOOOOOE-99 MP= 1 1 2IS 2 C( 1 C( 2 C( 4 C( 6 Neyel:: 2 0 4MAY64 1 Ie: 1 C( 1 C( 2 VOI= 1.000 lETTR:: 2 PROGRAM L L L :: 0 == == = .50000000E-06 .OOOOOOOOE-99 .OOOOOOOOE-99 ZREAL 1065.2886 1065.2887 1066.2497 1066.2501 1067.7597 1067.7601 1069.6027 1069.6031 1071.5953 1071.5957 1073.5931 1073.5935 1075.4981 1015.49B5 1077.2542 1077.2546 1078.8370 1078.8373 1080.2430 R = R :: R :: ZIMAG .2987 .2999 4.5705 4.5712 7.4989 1.4994 9.6923 9.6927 11.2RIO 11.2812 12.3601 12.3602 13.0270 13.0271 13.3159 .OOOOOOOOE-99 .10000000E+03 .10000000E+06 T(F) .77777775E-09 .55555554l:-09 .38888887E-09 !" .22222221E-09 .13888888E-09 .55555554E-I0 .OOOOOOOOE-99 13.3760 -.27777777E-IO 13.4893 13.4893 13.4347 -.55555554E-IO 241 0 _ 0 50.0010 55.0000 55.0010 60.0000 60.0010 65.0000 65.0010 70.0000 70.0010 75.0000 75.0010 80.0000 80.0010 85.0000 85.0010 90.0000 90.0010 95.0000 95.0010 100.0000 100.0010 216.8505 218.3995 218.3998 219.8269 219.8272 221.1500 221.1502 222.3827 222.3829 223.5363 223.5365 224.6202 224.6204 225.6422 225.6424 226.6089 226.6091 227.5257 221.5259 228.3976 228.3977 183.628 183.501 183.501 183.369 183.369 183.237 183.237 183.108 183.108 182.985 182.985 182.867 182.867 182.757 182.757 182.653 182.653 182.555 182.555 182.464 182.464 .. htrn 1080.2433 1081.4807 1081.4R09 1082.5648 1082.5651 1083.5125 1083.5127 1084.3404 1084.3405 1085.0647 1085.0649 1085.6991 1085.6998 1086.2579 1086.2580 1086.7501 1086.7502 10207.1857 1087.1857 1087.5724 1087.5725 13.4347 13.2648 13.2647 13.0190 13.0189 12'.7260 12.7260 12.4067 12.4066 12.0753 12.0752 11.7419 11.7418 11.4133 11.4132 11.0938 11.0937 10.1862 10.7862 10.4922 lO.'t922 -.55555554E-10 -.55555554E-IO -.5555~554t:-1O -.83333331t-1O -.555555541::-10 -.55555554E-10 -.5555~554t-1O -.55555554E-1O -.55555S54t=-lO -.55555554E-1O -.55555554E-IO o M4 _ _• • ::;,= ----.---~-.--~-'~~- -- o APPENDIX A The following paragraphs will outlin~ the equations programmed for the vari.ous subprograms. These assume steady state conditions on the imaginary axis (5· jel) with linear, lumped, bilateral, passive elements. The program processes basic sections starting with that section nearest the output termination. Regardless of section type, this section will have ZA and ZB impedances •. The DRPTZ subprogram calculates the impedances and admittances for a single order as in Figqre 8 for both the ZA and ZB impedance. The number of parallel branches, or resonators 8S they are called in the Westinghouse report, allowed is a function of available core storage. As written, three resonators per order are allowed although this could be increased by reallocating storage, i .•. , reducing the number of orders fer basic section impedance or the number of basic sections. The impedance for a single branch or resonator i8: 1 1 Zl • l1. + j001:L + ~ • ~ + j(~ - COCi) o 1 • R.. --~ YRLl • 2 + (Ca) R2 1 + (tAl ~ - Ll - 1 ~ 4-)2 VI Ca) )2 o ~ o I I t I I I I LJ~ FI(;{Tf.' B .. " Sl~0LF Onnfn M~ I I I t I I f I onrFn 2 o I := 2 ~( Z(2) - ..c.--- :L 1) - ~ Z(~) Z(3) - o onrrRS &2 1 OPlwr 3 -31- Em MiJSUSSUi a ., IlliX) 1$ lIIffttit'dHillti' _____ .. _._._. ___ ...... "..".'"'·'·"'''.r.·''.'''''.''"" ....... _,,~··_~r;,;.J..oo""""'"~.......,~. 'WitH 'rlN&' ttl """WJ.tIii:o i " .... _.'-' • Summing admittances for all k parallel branches in an order gives: k L mI;. • YRLi i-l k YIMT - y T L i-1 YIM i • The DRPTZ subprogram stores values of ZRL-. and ZIM.r impedances. control is then given to the ORDER SUb~ra.m. Program The ORDER subprogram first stores the impedance values just computed by DRPTZ as subSCripted impedances Z(I), with the order counter I set at 1. Control is returned to the main line which determines if either the ZA or ZB basic section impedance consists of more than one order. If not, control It there are more than one order in either ZA or ZB' the counter is incremented to 2 and con trol 18 returned to mpTZ. is given to the proper recursion subprogram. The DRPTZ subprogram then zeros the variables YRLr and YIHr and repeats the process described above using the element values for the second orders of the appropriate basic section impedances. Control is then given to the OHDER subprogram. The ORDER subproeram determines i f the order counter I equals 2 or 3. If I is 2, ORDER simply sums the impedances ZT just computed by Dlli)TZ with those previously stored as subscripted ·impedances when I equalled 1. illustrates the situation. Figure 9 C o Z(l) • ZTRL(l) + jZTIM(l) Z(2) • Z(l) +ZT • Z'ffiL(l) + jZTIM(l) + ZFLT + jZ~ ZTP~(2) • ZTRL(l) + ZRL T ZTIM(2) • ZTIM(l) + -ZTRL(2) + jZTIM(2) ZIMT If the order eounter I is.3, the ORDER subproBram has the 51 tuation in Figure 10. IT for the DRPTZ again has comj::u ted the impedance ZT and admittance third order. Y • YRLT + jYIM.r (for Order 3) T Y(3) • 1/Z(3) • • ~ IT· ZTRL(2} + ZTRL(2) ~ jZTIM(2) + IRLr + jZTIM(2) + IT + j y~ where o y(3) • i ~ [ZTRL(2) + A(YRL T)] +j r(YIM.r) - ZTIM(2)] } Let B • ZTRL(2) + A(YB~) C - A(YIMT) - ZTIM(2) Z(3) • 1/Y(3). A B+ 3c • A(B - jC) B2 + c2 Control is returned. to the main line. If there are no further orders to be processed in the particular basic section impedances, control is eiven to the appropriate recursion subprogram LADDER or LATTICE. The LADDER subprQgram has the a1 tuation in Figure 11. computation proceeds from right to lett. As n;:;ntioned previously The first step is to pick up the complex values of impedance just comp.tted by the DRPTZ and ORDER subprograms, and the voltage and impedance existing at. the being considered. 4C) au tput of the particular section If the first section to the left of the output termination is being processed, these latter two values are Vo and R. 0 In the general case t he7 are V _ and ZK_lo K1 VK_1 • ~-1 + jV~_l -)J~ $ V· 1... "" I\TlT Vk - ZBk SFCTIO~ k ~ TFiN Zk- c V_ k.-. -- Z"k -. ... _ _ _ _ ...._ _ _............_ _..6 ..- - fIGCnr. 11 - C \SC.\DEO l..u~nF.n SFCTln~s 7,\;_1 - . FICtlW l~ v·1 - - SY\~~FT!lIC'L l.~TTlr.f. srcflo~ V -- ZT Z "" -. fHTPF 11 • F II TF!~ ~FTWOnK T~nv.I'f~TIO'\l Z ",' n I ~prT 2 ''{ ~40 ZT -341,1, I'i:'I' 1,'1 Note that Z~ and Z\ are equal to the appropriate Z(I) values determined in DRPl'Z and ORDER where I is the maximum order of the kth section. The LATTICE subprogram has the situation presented in Figure 12. It will be noted that s,ymmetrioal resistive pi sections are placed at each end ot the basic s.ymmetrical lattice section to permit a definite 4:) amount or attenuation or isolation between sections. 8trective~ This feature can be eliminated by inserting zero values tor the series e1ernents ZAx R B and very large values tor the shunt RA elements. Again and Z~ have been COMplted by the DRPTZ and ORDER subprogra..~ and are equal to the appropriate Z(I) values where I is the ma.,, • ".aaUiM.::I;;: 1M ., •. --.-~."--~~.--~~~~---- 14, 3X2HI=t 14, 3X3HMP=, 112 FORMAT (2HK=, 14, 3X3HMQ=, 14) IREC • MP(K,I) L =1 9 IF (IREC-l) 7, 8, 8 8 INOX = 2*L-l c C READ 103, C(INDX,K,I), AL( INDX,K,I), R( INDX,K,I) READ C(FARAOS),l(HENRIES),AND R(OHMS) VALUES OF ZA TYPE IMPEDANCE C C WHICH IS PARALLEL ARM FOR INVERTED-L SECTIONS AND IS SERIES C ARM FOR LATTICE SECTIONS. c c PUNCH 101, INDX, K, 107 FORMAT ( 2HC ( , 1£14.8, 2X4HR I, C(INDX,K,I), AL(INDX,K,I), R(INOX,K,I) ,, 12, 2H :: , 12, 5H ) 12, 2H " =, E14.8, 2X4HL =• EI4.8) L = l+l GO T09 7 IREe = MQ(K,I) L = 1 10 IF (IREC-l) 2, 11, 11 11 INOX= 2*L C C REA D 1 0 3 , C ( 1 NO X, K, I ), Al ,( I NO X, K , I ), R ( I NDX, K ,I ) C READ C(FARADS),L(HENR,IES~fANO R(OHMS) OF IB TYPE IMPEDANCE v~HICH C IS SERIES ARM FOR INVERTEO-L SECTIONS AND IS PARALLEL ARM FOR C LATTICE SECTIONS. 255 -L2- "--~-"-.---- o ........- ......._ .............. _...... .._-_..... -_ .. .. I ~I' fw,r"'''''WlffiiiM!.ltMiJeSli:/iZiW"W"Y Hlp"'",,!" 5 t, • r. oc PUNCH 107, INOX, K, I, C(INDX,K,I), AL(INDX,K,I), R(INDX,K,I) L :: L+1 GO TO 10 2 CONTINUE FCOUN KKLL = FMIN =1 JFREQ = 1 " NSECT 20 ZRL = NSECT-l lOR c 101=0. 0 ZIM = 0.0 VRL = VOR = VOl IREC = 0 VIM K = 1 I z: 1 N :x 1 GO TO (42,43),lETTR c C CALCULATE DROP FOR PI PAD PRECEDING FIRST SECTION. c 42 CALL LATTIS 43 IREC &: F 10 F2 I ! N 2 = (FCOUN*1.OE+6)*PI2 &: =1 F*F 256 il ,--j I i K := 0 1 = 25 lORD IOROR (K) =1 I 22 CALL QRPTZ CALL ORDER IF ( IORO-I ) 3, 3, 21 ,21 = I 1+1 GO TO 22 3 GO TO (36.,.37},lETTR 36 CALL LATTIS GO TO 38 37 CALL LAODER 38 IF(NSECT-K) 27, 21, 24 24 K C = K+l GO TO 25 21 IF(NCYCL-N) 30, 30, 29 29 N K :; N+l = 1 GO TO 25 30 CONTINUE GO TO (99,44), LETTR 99 ZRLN == ZPRL lIMN = ZPIM GO TO ( 97 ,98 ) t KKLL 44 ZRLN &: ZRL lIMN :: ZIM 257 GO TO (97,98), KKLl 91 PUNCH 109 , I -.j!~ --- - - -- ~--- ------- - -- .. 0 I • he KKLL 0C = rrt. .rttsttssri 2 C C 98 K ::: SET C NSECT+l K AND I VALUES PERTAINING TO INPUT TERMINAL C BE EQUAL TO NSECT+l, I C IA TYPE. = 1, IMPEDANCE. K WILL WHERE THE IMPEDANCE IS CONSIDER~D C I =1 CALL T ERMIN 32 GO TO (33,34). JFREQ C =2 33 JFREQ CALL BETAD8 C CALCULATE DB AND BETA. c c BETAl = BETA C C ZRLN AND lIMN ARE IMPEDANCE VALUES OF INSERTED 'NETWORK WITH OUTPUT TERMINATION ZOR. THESE INCLUDE PI PADS FOR LATTICE SECTIONS. PUNCH 108, FCOUN, DB, BETA, ZRLN, lIMN c C INCREMENT F BY 1000 CYCLES AND REPEAT ENTIRE PROCEDURE IN ORDER TO OBTAIN TIME DELAY. C Q 258 FCOUN = FCOUN+O.OOl -45- _cuaciU,M ',1 /II II GO TO 20 34 CALL BETADB T = (BETA-BETAl)*CONST JFREQ :: 1 PUNCH 108, FCOUN, DB, BETA, ZRLN, lIMN, T c C DECREMENT FREQUENCY BY 1000 CYCLES. c FCOUN = FCOUN-O.OOI C C IS FREQUENCY AT MAXIMUM VALUE. c IF(FMAX-FCOUN) 12, 12, 35 C () c NO. INCREMENT BY AMOUNT AT INPUT AND REPEAT CALCULATIONS. c 35 FCOUN = FCDUN+FDEL GO TO 20 C C YES. GO TO START TO READ IN NEXT COMPLETE FILTER PRObLEM. C 108 FORMAT (F9.4, 2XF9.4, 2XFIO.3, 2XFIO.4, 2XFIO.4, 2XE14.8) I 109 FORMAT C/4XIHF t lOX2HDB, 8X4HBETA, 8X5HZREAL, 7X5HZIMAG,9X4HT(F)/) END 25U ------- -~-~-- 0 HfU'W"W#"ttWliIllW'w'W'rrM'Y"''PZF oc - tr 1U .'!$= , ! '" SUBROUTINE BETADB CALCULATES BETA(RADIANS) AND DB FOR FREQUENCY DIMENSION C(6,5,3), R(6,5,3), AL(6,5,3}, MP(5,3), MQ(5,3), IRA(5), RB{S), IORDR(5)t ZAKRl(3), ZBKRL(3), ZAKIM(3), ZBKIM(3) COMMON C, R, AL, MP, MQ, RAt RB, ZAKRL, ZBKRL, ZAKIM, ZBKIM, YARL, lYBRL, YAIM, YBIM, ZARL, ZBRL, ZAIM, 2 VPR L , VP 1M, Z R l t Z 1M, VRL, V 1M, K, ZBIH, F, F2, IREe, ZPRL, ZPIM, I, I P UNC, R A0, . R B0, 0 EN, L, 3INDX, REAL, AIMGt RlOEN, ANMIN, RLNUM, DENIM, A, B, QUAN, IORDR, 4LETTR, VAR1, lORD, ZARLK, ZBRLK, ZAIMK, ZBIMK,NSECT, NT COMMON XK, RK, lOR, lOI, VOR, VOl, BETA, DB, XIMAG PI = 3.14159265 PI2 = 6.2831853 CONY :: lSO.O/PI 4() BETA = (ATANF(VPIM/VPRL»*CONV c C MAKE BETA POSITIVE ANGLE BETWEEN 0 AND 360 DEGREES. c IF(VPRL) 1, 2, 2 1 BETA = BETA+lSO.O GO TO 4 2 IF(BETA) 3, 4, 4 = BETA+360.0 3 BETA = ZOR*VPRL-ZOI*VPIM XIMAG = VPRL*ZOI+ZOR*VPIM XNUM = SQRTF(RNUM*RNUM+XIMAG*XIMAG) 4 RNUM RLDEN o = RK*VOR+ZOR*VOR-XK*VOI-ZOI*VOI XIMAD = XK*VOR+ZOI*VOR+RK*VOI+ZOR*VOI DEN 26U = SQRTFlRLOEN*RLDEN+XIMAO*XIMAD) I; , " DB = . "_.~ __ .".~ .• ".,,_w . .~_ _ _ . 2.0*4.3429448*LOGF(XNUM IDEN) o RETURN END o 261 01I. I -48. rt o , C nttr rrt SUBROUTINE LATTI$ RECURSION FORMULAE FOR SYMMETRICAL LATTICE SECTION c CALCULATES lRL, lIM, ZPRL, ZPIM, VKL, VIM, VPRL, VPIM DIMENSION C(6,5,3), R(6,5,3), AL(6,5,3), MP(5,]}, MQ(5,3), lRA(S), RB(5), IORDR(S), ZAKRL(3), lBKRL(3), ZAKIM(3), ZBKIM(31 COMMON C, R, AL, 1 Y BR L, Y AI tvl , YBIl"l , VP I 1"1, l R L, 2 VP R L , MP, MQ, Z AR L , RA, RB, ZAKRl, lBKRL, ZAKIM, ZBKltII, YARL, Z ARL, ZA 1M, Z 1M, VRL, V I ~" K, 7 B 1M, F, F 2, I, IRE C, ZP R l , I P U(\j C, RA0, R 80, DEN, l P If-1 , L, 3INDX, REAL, AIMG, RLDEN, ANMIN, RLNUM, DENIM, A, B, QUAN, IORnR, 4LETTR, VARl, lORD, ZARLK, ZBRlK, ZAIMK, lBIMK,N~FCT, NT COMMON XK, RK, lOR, lOI, VDR, VOl, BETA, DB, XIMAG o REAL = I.O/RAO AIHG = RBO C C IF FIKST SECTION, GO TO 4022 TO CALCULATE INTIIAL PI PAD D~OP. C COME BACK TO CALCULATE FIRST SECTION AND FOLLOWING PI C PAD DROP. c IF(IREC> 4021, 4022, 4021 4021 ZARLK ::: ZAKR L ( I ) = = ZBRLK ZAIMK l BKR L ( I ) ZAKIM( I) ZBIMK = Z BK 1M ( I ) 0 REAL = Z AR LK+Z BR LK AIMG ;;: ZAIMK+ZBIMK RLNUM = ZPRL*REAL- ZPIM*AIMG ') ,') ') ;;.ij:.", RLNUM o RLNUM+2.0*(ZARLK*ZBRLK-ZAIMK*ZBIMK) RlDEN = 2.0* lPRL+REAL -u9- II.! 1 I'! c ANMIN = = ANMIN = ANMIN+2.0*(ZARlK*ZBIMK+ZAIMK*ZBRLKJ DENIM 2.0* ZPIM+AIMG ZPIM*REAL+ ZPRL*AIMG DEN = 1.O/(RLDEN*RlDEN+DENIM*DENIM) ZRL = (RLNUM*RLOEN+DENIM*ANMIN>*DEN = (RLDEN*ANMIN-RLNUM*OENIM)*OEN lIM / = ZBRLK-ZARlK = ZBIMK-ZAIMK REAL AIMG = = RlOEN DENIM DEN ZPRL#REAL- ZPIM*AIMG ZPIM*REAL+ ZPRL*AIMG = 1.O/(RlDEN*RLDEN+DENIM*OENIM) A = VPRL*RLNUM-VPIM*ANMIN B = VPIM*RLNUM+VPRl*ANMIN VRl : (A*RLOEN+DENIM*B)*OEN = (RLDEN*B-A*DENIM)*DEN REAL = l.O/RA(K) VIM = RB(K) AIMG 4022 DEN l.O/(ZRl*ZRL+ZIM*ZIM) c RLNUM = 1.O+REAL*AIMG+AIMG*ZRl*DEN ANMIN = RLOEN = 2.0*REAL+AIMG*REAL*REAL+(ZRl*REAL*AIMG+ZRL>*DEN DENIM = DEN = 1.O/(RLDEN*RLDEN+DENIM*DENIM) AIMG*ZIM*OEN (REAl*AIMG*ZIM+ZIM}*DEN ZPRl ~ (RLNUM*RlDEN+OENIM*ANMIN)*DEN ZPIM = (-RLDEN*ANMIN+RLNUM*DENIMJ*OEN = 1.0/ (ZRl*ZRL + ZIM*ZIM) VPRL = VRL+VRL*REAL*AIMG+(VRl*AIMG*ZRL+VIM*ZIM*AIMG)*OEN VPIM = VIM+VIM*AIMG*REAL+(VIM*AIMG*ZRL-VRL*ZIM*AIMG)*OEN DEN o 263 I I I II Iii I: 0 1 ,.,j W $ rim Mrirzw 4002 RETURN END o o -51 . . ==:WJ£iiiSiliii1JS:S:S(t!kiil$ X(, ti ; Q ' ... ,- ..•".-.•.•"...• ,_._I..• _"- . . . -~.~'-······"·····- o SUBROUTINE ORDER C COMPUTES TOTAL COMPLEX IMPEDANCE FOR BASIC SECTION IMPEDANCES C ZA AND ZS FROM INDIVIDUAL ORDERS. lRA(S), RB(5), IORDR(5), ZAKRL(3), lBKRL(3), ZAKIM(311 ZBKIM(3) COMMON C, R, AL, MP, MQ, RA, RB, ZAKRL, ZBKRL, ZAKIM, ZBKIM, YARL, lYBRl, VAIM, YBIM, ZARL, ZBRL, ZAIM, Z81M, F, F2, IREe, ZPRl, ZPIM, 2 VP RL, VP 1M, ZR L, Z 1M, VRL, V 1M, K, I, I P UNC, RA0, R B0, DEN, L, 3INDX, REAL, AIMG, RLDEN, ANMIN, RLNUM, DENIM, A, B, QUAN, IORDR, 4lETTR, VARl, tORD, ZARLK, lBRLK, ZAIMK, ZBIMK,NSECT, NT COMMON XK, RK, lOR, lOI, VDR, VOl, BETA, DB, XIMAG IVAL ;: 1-1 IF(IVAL) 5001, 5002, 5001 5002 ZAKRL(I) = ZARl IAKIM(I) :;: ZAIM lBKRL(I) ::: ZBRL Z8K 1M ( I) ::: Z B 1M GO TO 5006 5001 IF(I-(I/2)*2} 5007, 5004, 5007 5004 ZAKRl(I) = ZARL+lAKRL(IVAl) lAKIM{I) = ZAIM+ZAKIM(IVAl) ZBKRL(I) = ZBKIM(ll = lBIM+ZBKIM( IVAl} ZBRL+ZBKRl(IVAL) GO TO 5006 5007 RlNUM = ZAKRL(IVAL) DENIM = ZAKIM(IVAL) QUAN =RLNUM*RLNUM+DENIM*OENIM B = YARL*QUAN+RlNUM ? I'fII {~ r' V t.t .. 52- o 4C) A = YAIM*QUAN-DENIM DEN ~ 1.O/(8*B+A*A) ZAKRL(I) = S*QUAN*OEN ZAKIM(I) = -A*QUAN*OEN RLNUM • ZBKRL(IVAL) DENIM. lBKIM(IVAL) QUAN =RlNUM*RLNUM+OENIM*OENIM B = YBRL*QUAN+RlNUM A • YBIM*QUAN-OENIM DEN = 1.O/(S*B+A*Al ZBKRL(I) • B*QUAN*OEN Z8KIM(I) • -A*QUAN*OEN 5006 RETURN END o 266 -53- L .£itau a &*144,4$1.4 o SUBROUTINE TERMIN C CALCULATES IMPEDANCE AND VOLTAGE AT INPUT TO I~PUT TERMINATION. DIMENSION C(6,5,3), RC6j5,3), AL(6,5,3), MP(5,3t, "0(5,3), iRA.'), R8(S), IOROR(5), ZAKRl(3), ZBKRL(3), ZAKIM(3J, IBKIM13) C, R, AL, HP, MQ, RAt Ra, ZAKRl, ZBKRL, ZAKIM, ZBKIM, YARL, tOMMO~ lYBRl, VAIM, YBIM, ZARL, ZBRl, ZAIM, 18IM, F, F2, IREe, ZPRL, ZPIM, 2VPRL, VPIM, IRL, lIM, VRl, VIM, K, I, IPUNC, RAO, RBO, DEN, L, 31NDX, REAL, AIMG, RLOEN, ANMIN, RlNUM, QENIM, A, 8, QUAN,. IORDR, 4LETTR, VARl, lORD, ZARLK, Z8RLK, ZAIMK, ZBIMK,NSECT, NT COMMON XK, RK, lOR, lOl, VOR, VOl, BETA, DB, XIMAG GO TO (5,6), LETTR 5 ZTRL = ZPRL ZTIM = ZP 1M GO TO 1 6 ZTRL 1TIM ZRL :I = ZIM c C USE ORPTZ TO CALCULATE INPUT TERMINATION IMPEDANCE. C 7 CAll DRPTZ Z8TRL • lT~l+ZARL = ZTIM+ZAIM zaTIM RI( II: ZARL XK == lAIH REAL &I: VPRl*ZBTRL-VPIM*ZBTIM XIHAG- VPIM*Z6TRL+VPRl*ZBTiM OEN • ZTRl*ZTRL+ZTIM*ZTIM 267 I 1 1 F eDEN) 1, 2, 1 -54- \.\ .• \ I' z&s 4C) , HI en 2 PUNCH 101 101 FORMAT (41HERROR .DENO" I-NATOR IN TERM IN eQUAL TO ZERO) PAuse c C CALCULATE INPUT VOLTAGE REQUIRED TO PRODUCE SPECIFIED OUTPUT C TERMINATION VOLTAGE. c 1 VKRL a (REAl*ZTRL+XIMAG*ZTIM)/OEN VKIM = (ZTRL.XIMAG-ZTIM~REAL)/OEN VPRL z; VKRl VPIM I: VKIM RETURN END o -ssdAiM ., .•...,•..,.,'.••"•••.~......c.-..._ _, o SUBROUTINE LADDER lRA(S), RB(S), IORDR(SJ, ZAKRl(3), ZBKRl(3), ZAKIM(3), ZBKIM(3) COMMON C, R, ALt MP. HQ, RA, RB, ZAKRl, lBKRL, ZAKIM, ZBKIM, YARl, lY8RL, YAIM, YBIM, ZARl,ZBRl, ZAIM, ZBIM, F, FZ, tREe, ZPRL, ZPIM, 2VPRl, VPIM, ZRL, lIM, VRL, VIM, K, It IPUNC, RAO, RBO, DEN, L, 3INOX, REAL, AIMG, RlDEN, ANMIN, RLNUM, DENIM, A, S, QUAN, IOROR, 4LETTR, VAR1, lORD, ZARLK, Z8RLK, ZAIMK, ZSIMK,NSECT, NT COMMON XK, RK, lOR, 101, VOR, VOl, BETA, DB, XIMAG C COMPUTES RECURSION EQUATIONS FOR LADDER SECTIONS. C CALCULATES ZRL, lIM, ZPRL, ZPIM, VRL, VIM, VPRL, VPIM. ZARlK • ZAKRL(I) ZAfMK = ZAKIM(I) ZBRlK = IBKRL(I) o ZBIMK = ZBKIM(I) VPRL :II: VPIM = VIM RlOEN VRL = ZARlK+ZRL DENIM • lAIMK+ZIM DEN = RLDEN*RLDEN+DENIM*DENIM RlNUM XIMAG = ZARlK*ZRL-ZAIMK*ZIM = ZAIMK*ZRL+ZIM*ZARlK RNUM = RlNUM*RlOEN+XIMAG*OENIM XMAG =XIMAG*RLDEN-RLNUM*OENIM IF(OEN) 1, 2, 1 2 PUNCH 101 101 FORMAT (32HERROR,ZERO DENOMINATOR IN LADDER) 2 6 ~..l PAUSE -56- l~ !I - --~-~----- sr c:; 1 ZPRL • RNUM/DEN ZPIM , ZRL ZIM DEN XMAG/DEN 2& = ZPRL+ZBRLK = ZPIM+ZBIMK = ZPRL*ZPRL+ZPIM*ZPIM IF(DEN) = 1.O+(Z8RlK*ZPRL+ZBIMK*ZPIM)/DEN 3 REAL = XIMAG VRL 3, 2, 3 (ZBIMK*ZPRL-ZPIM*ZBRLK)/OEN = VPRl*REAL-VPIM*XIMAG = VPRL*XIMAG+VPIM*REAL VPRL = VRL VIM VPIM C" ) o = VIM RETURN END 270 -57- auZS4iSUili h:.uti Zi4 ; _ C SU8ROUTINE ORPTZ CA~tULATES IMPEOANCE AND ADMITTANCE FOR A TWO-TERMINAL IMPEDANCE c iOROER). C IE. YARL, VAl", rBRl, Y8IM, ZARL, ZAIM, IBRL, ZBIM 1£. YARl, VAIM, VBRl, OIMENSION C(6,5.3), Y~IM, Rt~",3), ZARL, ZAIM, ZBRl, 181M AL(6,5,3), MP(5,3), MQCS,3), lRA(5), R8(S,t, IORDRt5), ZAKRL(3', lBKRL(3), ZAKIMC3!, ZBKIM(3) COMMON C,R9 AL, HP, MQ, RAt RB, ZAKRL, ZBKRL, ZAKIM, Z6KIM, YARl, lY8Rl, YAIM, YBI", ZARL, ZBRL, ZAIM, lSIN, F, F2, IREe, ZPRL, ZPIM. 2VPRL, VPIM, ZRl, lIM, VR~, VIM, K, I, IPUNC, RAO, RBO, DEN, L, 3JNDX, REAL. AIMG, RlO£N,ANMIN, RlNUM, DENIM, A, 8, QUAN, IOROR, 4LETTR, VAR1, lORD, ZARlK, Z8RLK, ZAIMK, 18IMKtNSECT, NT COMMON lK, RKt lOR, lOI, VOR, VOl, SETA, DB, IVAL ~IMAG MPCK,!) I: G 'tARt- 0.0 .,AIM 0.0 c L • 1 11 IF CIVAL-L) 9, 10, 10 10 INDX III .L* 2-1 VARl • R'INDX,K~I) QUAN F2*tCINOX,K,I) c IF CQUAN) S, 4, 3 4 QUAN = AL(INOX,K,I)*F TO GO J QUAN ~ ~ ·(Al(INDX,Kfl)-l~O/OUAN)*F OEN • 1.O/(VAR1*VAR1+QU4N*QUAN) YARL = YARL+VAR1.OEN YAJM • yAIM-QUAN*DEN 1. c L+l 271 -$8- tt o .. *= GO TO 11 ~ 9 OEN VARL*VARL + YAIM*VAIM I F (D EN) 12 t 1 3 ,. 12 = l.O/DEN 12 DEN 13 ZARl a YARL*OEN ZAIM II: -YAIM*OEN IVAL = MQ(K,I) YBRL :: 0.0 YSIM ::: 0.0 l 1 lit 16 IF (IVAL-L) 14, 15, 15 15 INDX o = 2*L VARI ::: R(INDX,K,I) = F2*C(INDX,K,I) QUAN IF(QUAN) 6, 7, 6 7 QUAN : AL(INDX,K,I)*F GO TO 8 6 QUAN =(Al(INDX,K,I)-l.O/QUAN)*F 8 DEN R 1.O/(VAR1*VAR1+QUAN*QUAN) YBRL ::: VBRl+VAR1*OEN Y81M ::: YBIM-QUAN*OEN l = L+l GO TO 16 14 OEN = Y8RL~YBRL + VBIM*VBIM IF (DEN) 17, 18, 17 o 17 DEN = 1.O/DEN 18 ZBRL ::: Y8RL.OEN ZSIM ::: -YBIM*OEN 272 -59- o RETURN END o 273 o I -60- il I: 3.' o 1620 USERS GROUP WESTERN REGION MEETING June o 18, 1964 FORTRAN II - DEBUGGING TECHNIQUES AND AIDS Leon P. Goldberg Technical Staff Princeton University o .=_.UiMi,a:eit: ~I I I I ! o FORTPAN II fLo A.--t \'t"\C:r !. \'\'r"\v,\)'i'; i\K\":. NON RET-sOCABLE SUBROUTINES .Address Subroutine na.me Function 01510 SHe r/o r/o rio r/o r/o r/o rio 01768 03158 03182 01418 01574 00986 01022 01058 01800 02052 02152 02380 03280 03300 06020 06052 06528 07316 07348 07416 07440 07484 07570· 07604 07698 .07932 . 08152 . . 08586 09044 09356 09504 09528 09740 .09808 09856 09952 10000 Q(,""(to Otz.Sf.:, Oo+~5 COi'.fPLT RJi...TY RAPT Men SLlillH HATY ltlAPT \-lACD HTYPE REDO fu,~ ITYPE· FTYPE ETYPE XTlPE ATYPE H.4.TRIX FXA FXSR FXS F.XH FXD F'LDR RSGl{ FLOAT FIX ·FIXI FAX! FAX8 FT1FAC I/O r/o Irollertth conversion Multiple field specs. Multiple parenthesized specs. I specification F specification E specification X specification A specification Reading arrays I+J -(J-l<-K)+I I-J r-t:-J r/J l/(r/J) -I or -A A==I· I=A I**J A~~-I A~"*B FAC to A FSB FAD FSBR F1vrP A+B -(P.*B)+C A-X-B FD AlB FDVR TOFAC C A-B I/(A/B) A to FAC T~\<':.t "?~~?EJ\\ (\\~~) t=t\~ 0 275 'bin o :r. - up. INTERACTION OF tOMMON AND EQUIVALENCE IN UNDIMENSIONED VARIABLES. COfvlMON X, Y, Z EQUIVALENCE (X,A),(Y,B,C),(Z,L) C A=4.15273 LJ.1 8=2."k A R=X+2. S=B+R Z=X+Y+S STOP END TURN SW 1 ON FOR SYMBOL TABLE, PRESS START 11043 41527301 110 51 2~1(100Qj01 59999 X 59999 A 59991 59991 59991 Y B C 59983 S9983 Z 11~159 11067 L R s END OF PASS I o 276 c COMMON STORAGE IN DIMENSIONED VARIABLES. Y(2 ,5), Z(3 ,4,5 ) CO/lIMON x, Y STOP o1t1E NS ION X(5)-, END TURNSW 1 ON FOR SYM~OL 59959 . X 59999 59859 y -5991~9 11045 Z 11635 END OF PASS- I TA8[E,PR~SS o START Oi 277 1et EXAMPLES TO DEMONSTRATE THE COMPLEX INTERACTION OF EQUIVALENCE AN·D COMMON STORAGE ASSlGNMENTIN DIMENSIONED VARIABLES. DIMENSIONX(5), Y(2,5), Z(3,4,S) COMMON X . EQUIVALENCE (Y,Z) STOP END TURN SW 1 ON fOR SYMBOL .TABLE, PRESS START X ~9999 59959 Y 11135 11045 Z 11635 11045 END OF PASS .1 ENTER o PRESS START DIMENSIONX(5), ·Y(2.,5), Z(3,4,5) COMMON X EQU I VALENCE. (X(S), Y(10) ,Z(60» STOP END TURNSW 1 ON FOR SYMBOLTABLE,~?~ESS START SOURCEPROGRAM~ X ~9~99 Y S9999 ~9409 Z 59999 END OF P'ASS' I 39959 59909 ENTER SOURCE PROGRAM,' PRESS START . DIMENSION XeS), Y(2,S), Z(3,4 5) EQU I VALENCE· . (y( 10) ,X(5},Z(60~) STOP END TURN SWl ON 'FOR SYMBOL TABLE, PRESS START T104SZ11635 11545 Y 11635 11595 X 11635 END OF PASS I ' o 27~ • _ _ Ail. C EXAMPLES TO DEMONSTRATE THE COMPLEX INTERACTION OF EQUIVALENCE AND' COt~i''10N STORAGE ASSfGNt~ENT !t4 DfHENS!ONED VARIABLES. D f ME ~!S ION X ( 5 ), Y ( 2 , 5), Z (3 ,4, 5 ) EQU r VA LE f\ CE ( Z ( 6 ~1 ) ,X (5);; Y( 1 0 ) ) C 0. STOP END TURN SW 1 ON FOR SYMBOL TABLE, PRESS START Z 11635 X T1635 Y 11635 . 11045 T1595 T1545 Er~D OF PASS I ENTER SQURCEPROGRAM, PRESS START DIMENSION X(5)~ Y(2,5), Z(3,4,5) c0 r !! TF BNF CF TF A(v1 TF 8NF CF TF AM TF BI\JF CF TF Alv1 BT BT BT BT BT BT B 5 XPGM-l,5,010 A,XPGH-l,Olll *+36,A,()l A, ,0 A,A,Olll XPGlvi-I , 5, 010 B, XPG~'i-l ,0 III ~~+36,B,01 B, ,0 B,B,0111 XPGf 1-1,5,010 C,XPGt"i-1,0111 o v ~,"+36,C,Ol C, ,0 C,C,0111 XPGM-l,5,010 D, XPG~,'i - 1 , III *+36,0,01 D, ,0 0,0,0111 XPG("'1-1 , 1 , 010 TOFAC,A,l FAD,B,l t=RrvlFAC, C, 1 TOFAC,A,l FSB,B,l FRfviFAC, 0, 1 XPGM-l,,06 ° I I' FLAG CONVENTION ••••••••••• ° IF P IS RELOCATABLE, FLAG OPERAND IS IF Q IS RELOCATABLE, FLAG OPERAND IS 1 NORMAL FLAG OPERANDS ARE STILL IN EFFECT FOR IMMEDIATE AND INDIRECT ADDRESSING. FLAGS ARE USED OVER THE OPERATION CODE TO DENOTE RELOCATION TO THE LOADER. THESE FLAGS DO NOT ALTER THE OPERATION OF THE INSTRUCTION. 0 (,\ ,1;1 T EST c C 10 P i-< UC; R Af-/I FUR 1,4 Lt· 3 [) R If,! T F 1<' P un, S p.1 (X) \I s. x. U I f'/I EJ\j S I U hi X ( 5 00 ) , Y ( ~ 0 0 ) P R II"lT 1 0 F CJ k i·1 A T ( 1 H 1 ) T=O. DU 1 I=1,200 X ( I ) =T Y(I)=SINF(T) T=T+.01-::-3.14159 1 C 0 i\! T I [\! U C PAUSE CALL PLOT (X, 10., U., 5, Y, 1., -1.,10,2(0) STUP E 1\1 [) o o _ :a:tii2i4iIiJ: C FURTkt\f\i SUBRUUTli\]E FOR 1443 t>RII\ITER PLDTTI/\jG, C C B Y L. o H[j F F [i! i\ ,\J, GUGG E i\J H F I i'v1 L I~ BS • SUBRUUTINE PLUT(X,XMAX,XMIN,NX,Y,YMAX,YMIN,NY,N) DIMENSION OUMMY(2),OUTPUT(102),X(2),Y(2) XCHtd~=. 20 YCHAk=.71 CHAR=.14 BL ,Lj,I\lK=O. XNU=100. NOX=XNU+1. YL AS EL=Yt'lAX DX=(XMAX-XMIN)/XNO DY=(YMAX-YMIN)/50. MOVE MAX DOWN BY ONE-HALF bOX •••• YYfll,AX=YHAX+. 5{~OY XXr'HI\J=Xfv1Ij\J-. 5~~DX C C KY=O i\IX l=/\')X+ 1 DO 1 1=1,51 C CALL INIT(OUTPUT,SLANK) DO III II2=I,NUX III OUTPUT ( 112) =BLAhli< C CALL GRIO(OUTPUT,DX,OY,NX,NY,KY,I,XCHAR,YCHAk,IND) IND=O IF(I-1-50*KY/NY)211,222,211 222 OD 332 JJ=l,NClX 332 OUTPUT(JJ)=XCHAR 1 ~JD= 1 211 Lr·44 C 2221 3331 1121 H 10 2 11 3 12 1 KY=KY+l DO 444 JJ=l,NXl 12=( (JJ-l)~~(NOX-l) )/NX OUTPUT(12+1)=YCHAk ZI=I UP=YYMAX-(ZI-l.)*OY OUWI\J=UP-DY CALL FINDY(X,Y,UP,OOWN,OUTPUT,N,DX,Dy,XMAX,XMIN,CHAR) DO 1121 IF=l,!'\') IF(Y(IF)-UP)2221,112l,ll2l IF(Y(IF)-DUWN)112l,333l,333l CUNT I j\jUE JJ=(X(IF)-XXMIN)/UX JJ=JJ+l DUTPUT (J"J) =CHAR. CONTINUE IF (U\jD) 10, 10, 11 PRINT 2, (OUTPUT(J) ,J=l,f\!UX) FURMAT(12X,50Al,5lAl) GO TU 12 PRINT 3,YLABEL,(OUTPUT(J),J=l,NOX) FORMAT(lX,EI0.3,lX,50Al,51Al) YLABEL=YLABEL-DY C(Ji'\JTIi\JUE RETURN o EI"~O 11 ----.---.---~.-.- J".rp*rrIlUIZ'N" .. liz M FDRTRAI\J I I S P S S lJ BR0 UTI [\1 F. S, L. H[) F Fiv) A'\1, GlJ GGEf\I HE I 1\1 l 1\ b S • 0 1 ASS EfVI Bl Y Af\J D F I [\1 ALP HAS e UF S P S SUB S. FIJ R Fj\\ I I • 1) USE 1620/1710 SPS TO ASSEMBLE AND COMPRESS THE SPS PROGRAM. 2) ReMUVE THE FIRST TWO (2) AND THE LAST SEVEN (7) CARDS FROM THE COM P f<. ESSE D DEC K• (T HIS Dn ES i\1 0 T I f\I CLI J DF THE Th' n Rl A. 1\1 K CAR 0 SAT THE ENO OF THE DECK) 3) AOD HEADER CARD AS NO.1. 4) ADD TRAILER CARD TO END OF DF.CK. 5) CORRECT DSA'S , IF ANY,USED IN THF SPS PROGRAM,nTHERWISE, GO TO 7. 6) PUNCH A FlAGG~D ZERO IN CUlUMN 62 OF All ORJECT DECK CAR[lS PRODUCED BY DSA'S IN SPS PROGRAM. 7) CHECK FOR RElOCATABlE CONSTANTS, IF NONE, THEN GO TO 9. 8) PUNCH A FLAGGED 1 IN COLUMN 62 OF All CONSTANTS NOT TO BE RELOCATED. 9) PUN CH NE\Ai CAR D I\i O. I 1\1 TRAI l ERe ARn T[) C(1 1\1 TIN lJ F S FqUE 1'-1 C Ir'\! G• 10) THE DECK CAN NOW BE USED WITH A FORTRAN CALL STATEMENT. THE HEADER CARD ••••••••••••••••••• ••••• COlS. 1-12 C\ COlS. COlS. COlS. COlS. COlS. COlS. 13-20 . 21-22 23-24 25-62 63 64-80 SUBROUTINE NAME IN TWO-DIGIT ALPHANUMERIC FORM WITH FLAG OVER HIGH ORDER DIGIT AND RIGHT JUSTIFIED. BLANK FF, lENGTH OF FLOATING MANTISSA, FLAG OVER HIGH ORDER DIGIT. KK, lENGTH OF FIXED MANTISSA, FLAG OVER HIGH ORDER DIGIT. BLANK RECORD NARK (0-2-e) BLANK, EXCEPT FLAG IN COL. 7A THE TRAILER CARD •••••••••••••••• •••••• COlS. 1~62 COlS. 63 COlS. 64-80 BLANK FLAGGED 1 BLANK, EXCEPT FOR CARD NO. IN COL. 78-80. o _au Ztili.l2UC&fU:; a:iII $II (J&\l Jf • -- - '••;.,:",:..j~ ... .:i:.~~~"~, ----------,,-'---~".''-~ 1:\ , PLOT FOR 1443 PRINTER, L. HOFFMAN GUGGENHEIM LABS. * * * I I ! AN EXAfvlPLE OF AN SPS SUBROUTINE FOR FN I I . ~- 11036 11040 11045 11050 11055 11060 11065 11070 11075 11080 11085 11092 11094 11106 11118 11130 11142 11154 11166 11178 11190 11202 11214 11226 11238 11250 11262 11274 11286 11298 11310 11322 11334 11346 11358 11370 11382 11394 11406 11418 11430 11442 11454 11466 11478 11490 11502 11514 11526 11538 11550 11562 11574 11586 11598 11610 4/24 AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO 00005 00005 00005 00005 00005 00005 00005 00005 00005 00005 00007 J1 KO tvlfvl L3 KO Jl KO lvJM L3 KO J1 KO MM L3 KO Jl KO ~~111J L3 KO J1 KG fvlM L3 KO Jl KO M~I L3 KO J1 KO MM L3 KO Jl KO MM 'L3 KO J1 KO MM l3 11093 11045 11154 11045 11045 11093 11050 11214 11050 11050 11093 11055 11274 11055 11055 11093 11060 11334 11060 11060 11093 11065 11394 11065 11065 11093 11070 11454 11070' 11070 11093 11075 11514 11075 11075 11093 11080 11574 11080 11080 11093 11085 11634 11085 l. 000-5 1109L 11045 00000 1104N 000-5 1109L 11050 00000 1105000-5 1109L 11055 00000 1105N 000-5 1109L 11060 00000 1106000-5 1109L 11065 00000 1106N 000-5 1109L 11070 00000 1107000-5 1109L 11075 00000 11071\J 000-5 1109L 11080 00000 1108000-5 1109l 11085 00000 AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTLJ AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO AUTO PLOT X XHAX XMIN NX Y Yfl-1AX YfVl IN NY N * START DORG OS OS OS OS OS DS OS OS OS OS OS AM TF BNF CF TF AM TF BNF CF TF A~'l TF BNF CF TF A~1 TF BNF CF TF AM TF BNF CF TF AM TF BNF CF TF AT·1 TF BNF CF TF At4 TF BNF CF TF AM TF BNF CF 11036 5 5 5 5 5 5 5 5 5 5 7 END OF ARGUMENT ADDRESSES/ START-l,5,010 X ,START-1,01ll ,01 *+36,X ,,0 X ,X X ,0111 START-1,5,010 X(v;A X ,START-1,0111 *+36, XfvJAX ,01 XiVlAX , ,0 XfvlAX , Xf'1AX ,0111 START-l,5,01O X/\'I I I" ,START-1,Ol11 *+36,Xr'1!N ,01 XMIN , ,0 Xivi IN , Xf'vl IN ,0111 START-l,5,010 NX ,START-1,0111 ,01 *+36,NX , ,0 NX ,NX NX ,0111 START-1,5,010 Y ,START-1,0111 ,01 *+36,Y , ,0 Y ,Y Y ,0111 START-1,5,010 Yt-1AX ,START-I,0111 *+36,YMAX ,0 1 YMAX , ,0 yrv1AX , Yf~AX ,0 111 START-1,5,010 yrvJIN ,START-l,0111 *+36, yr-1 IN ,0 1 Yjl/lIN ,,0 YMIN ,Yf'1IN ,0111 START-1,5,010 NY ,S TAR T-1., III ,01 *+36,NY , ,0 NY ,NY NY ,0111 START-1,5,010 ,START-1,0111 N ,01 *+36,N 307 N 0 Ie ° ,,° 0 1 ....-..---..-~~~~~..............--.-------~~-~--- 11622 11631.;· 0 1 11646 1165 i j- i< II Jl 1,:9 T~ 1\: !\. tJTl.l l . ! 'i ST/J.L~.T-l,?,UlU 12 {) 11.;· UUUUU /\UT l., /\ IZ Ii t.1 i· i; , H Di) l~. c; .;:--3 S Y:·it'.iIL , ".J j I JT (l .~:~ FF 3/17 3/17 KK 1)5 , :) O()Ll'OS f= t.',C !)S , it K ~) 01402 016:32 (lU()()O uoouu LrJ 2."3 Lf/2. 3 DS US ,1(')';7 U2336 UU()f.)O U2112 01772 0135H OO()UU (JOO()O uouoo 03u()2 OUO()() 02 Lj·30 uoooo 02336 ()OOOO 00000 00000 U3'+3 i + i)S DS ,2336 ,2117 ,1772 i)S , FIX 't/23 4/"2.3 4/23 US ,3 OC) 2 DS ,2430 ,2336 US DS DS DS DS L-:/23 CflHPLT DS ,56~)O uooo:.:; 4/23 Fj'"T 00003 4/23 DC DC DC DC DC DC ~ , 5 -( C) 4 3,002 5,7,')50 3,010 4/23 4/23 4/23 00003 00005 00012 U0102 00001 00005 Lt 3/13 3/22 3/22 3/25 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 OOOOLt 00003 OU005 00005 00(:)10 00010 00010 00005 00008 00002 00010 00010 00010 00010 00001 00005 00008 00002 00008 00002 00008 00002 00005 00010 01314 /23 "-/2'+ -l}*~:.-::--~ **eJ~~--::- J2038 3/22 3/22 3/22 3/22 3/22 3/25 3/22 DC DC UUf'.' OUT D.t·.C OUTPUT D.l\ S Df-IC KEDO I JJ 112 12 IF ZI UP o[l\tI}i\1 COUNT 1-11-\ L F 1 HALF YYHAX DY 5,62HO 12, 102 1, • DS 4 3 DS 5 5 FF FF FF DS 5 DC DC B,5()OOOOOO 2,00 FF FF FF FF 1 KK DC 8,10000000 DC 2,03 8,10000000 2,01 [) i\l E 1 DC ClI'JE DC DC F1 FIFTY KY XXfv1 IN 2,03 5,57 C)L'r 3,OOZ K'/i'- OS DS OX OS YLABEL OS I h!\) DS NUX OS Xf\IOl Xf\!(l ,6066 DS DS DS OS DS US DC OS DS AR.OUND BTn H I:: ~~. , 131 /j- 4/23 ~·/Z3 ST;\i':TS ,321H , ::iOIH 00005 i\ !'iI ) Cll , 3 Lt3 <';- DS DS 00003 00002 00005 r.-: ,30 1: 6 TY S\'IC oouoo i\UL 1 .:)? -, ('".:. I·.! /\ 00000 1P 1)5 FLDi\T Fi·:F/.I.C .FSfl , FlI.D FSbl-( Fi·.ip FDV FDVr< Tf"lF t\c T , iLI-U2 R S(;I\ 4/2.3 4/23 OOOOU IlbrA C r- F)(l) 41 Z:j L1·I 2 3 LI-/23 5()12 1.1-/23 00000 00000 0131 LI0502£3 06066 0565U 11658 11661 11666 11669 11671 11676 11679 116b7 11711 11<,)15 11920 11924 11 '127 11932 11937 11947 11957 11967 11972 11980 11982 11992 12002 12012 12022 12023 12028 12036 12038 12046 '.2048 12056 12058 12063 12073 1207'+ !.~/:~ 3 1.1-/23 Tf<,:~ US ,(1 , 1 () "3/1-( 030<,;-6 U321d 0 i\UTi) U()U-;~ UUUUO (Juno:) UUUUU OUU l):; OJ 11 U h\i 1109~~, /~ OCOlU ,0111 110d:J 8,50000000 2,02 KK FF TOFAC, Xi\![), 17 308 j: ·"" ....... .! • '"I t;: .......~" .• ~~~", ...•• ............".... ....... ~." ,~''".~, lilil "I ~I~I 12086 12098 12110 12122 12134 12146 12158 12170 1P 17 K6 2P 1P 2P 1P 1P 02430 01772, 12028 01314 01402 03002 03218 01402 J2048 000-0 00485 11070 J2022 11075 J2058 J2002 12182 12194 12206 12218 2P 2P 1P 1P 01314 03002, 03218 01402 11050 11055 J2038 J2012 12230 12242 12254 12266 12278 12290 12302 1P 2P 1P 1P 1P 2P 1P 03046 02336 01402 01314 03046 02430 01402 J1982 11055 J2073 J2002' J1982 11070 J1992 12314 12326 12338 12350 12362 12374 12386 12398 12410 12422 12434 "J6 J6 J6 JO KJ KJ J6 J1 KM M6 J5 12063 11920 11927 11972 11972 11972 1197K 11927 12028 12350 12023 0-000 00-01 00-01 J1711 11927 11927 000-0 000-1 11927 01300 00000 12446 J3 12458 32 12470 26 12482 ' 2P 12494 12 12506 2J 12518 14 12530 M7 12063 00095 00485 0'1652 00485 00485 00485 12662 -00500000 00099 11080 000-1 11920 0-000 01200 12542 12554 12566 12578 12590 12602 12614 12'626 J6 11924 JO 11972 KJ- - 11972 KJ 11972 J6 1197K Jl 11924 KM 12028 M6 12554 00-01 J1711 11924 11924 OOOKO 000-1 11924 01300- 1263~8 J5 J1 12023 12063 00001 000-1 12662 J6 12674 20 12686' 12 11924 00485 00485 00-00 12028. 000-1 12650 3/22 3/22 3/2,2 3/22 3/24 3/22 3/22 3/24 3/22 3/22 3/22 3/22 ' 3/24 3/22 3/25 3/25 3/25 3/25 3/22 3/22 3/24 3/22 3/22 3/22 3/25 3/22 3/22 3/22 3/22 3/22 3/24 3,/24 3/22 3/22 3/22 3/23 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/25 3/22 3/22 3/22 3/22 3/22 3/24' 3/24 3/22 3/22 3/22 3/22 3/22 3/22 3/25 BlM BTM TF ,BT BTM BT BTM . BTM * * * FAD,ONE,17 FIX;0,10 NOX,FAC,O TOFAC,YMAX,l TRACE,YLABELj17 FSB,YMIN,l FDV,FIFTY,17 TRACE,DY,17 BT BT 8TM BTM TOFAC,XMAX,l FSB,XMIN,l FDV,XNO,17 TRACE,DX,17 BTM 8T BTM BTM BTM BT BTM FMP,HALF,17 FSBR,XMIN,l TRACE,XXMIN,17 TOFAC,DY,17 FMP,HALF,17 FAD,YMAX,l TRACE,YYMAX,17 0 KY,0,08 1,1,09 112,1,09 COUNT,OUTPUT,017 COUNT,II2,01 COUNT,II2,01 COUNT,0,0610 112,1,010 NOX,112,01 'c BNN RTN111,,0 TDM IND,Q,O GR I D•••' •••••• KY,50,0711 MM SF 99-KK+1 FAC,99 TF FXD,NY,l BT FAC,1,10 SM FAC,I,1 A FAC,0,8 eM BNE 1211,,0 TFM TFM RTN1 TFM RTN111 TFM A A TFM AM * *1222 TFM RTN332 TFM A A TFM AM C BNN * *1211 TOM AM TFM RTN444 TF SM 0 JJ,1,09 COUNl,OUTPHT,017 COUNT,JJ,Ol COUNT,JJ,Ol COUNT,20,0610 JJ,1,010 NOX,JJ,Ol RTN332,,0 INO,l,O KY,1,010 JJ,0,09 FAC,NOX,l FAC,1,10 3U9 0 ·"W"ri'flftnml'Bwn'''" t 0 7 12698 2L 12710 32 12722 26 12734 . 2P 12746 JO 12758 Kl 12770 Kl 12782 Jl 12794 J6 12806 J1 12818 KM 12830 M6 12842 12854 12866 12878 12890 12902 12914 12926 12938 12950 12962 0 0 - tt 00485 00095 00485 01652 11932 11932' 11932 11932 1193K 11924 110612674 20 00485 33 00483 32 00481 17 02112 1P 01402 1P '03002 1P 03046 IP 02336 1P 01402 1P 03002 1P 01402 11924 00000 00099 11060 J1711 00485 00485 000-2 OOOPI 000-1 11924' 01300 11920 00000 00000 000-0 J1941 J2048 J2002 J1992 J1957 J2002 J1967 12974 12986 12998 13010 13022 13034 13046 13058 1.3070 13082 13094 13106 13114 13114 13126 13138 13150 13162 13174 13186 13198 13210 13222 13234 , 13246 13258 13270 13282 13294 13306 13318 J6 J3 2J 27 IP M4 J3 2J 27 IP M4 M9 11937 00-01 11937 OOOJO 00099 11065 01314 00099 03002 JI957 13246 00483 11937 OOOJO 00099 11065 01314 . 00099 03002· J1967 13114 00483 13246 00000 J3 2J 27 1P IP 17 11 13. 32· 1J 16 Jl KM M6 ML L9 M9 L9 11937 OOQ99 01314 03002 03218 01772 00485 00485 00095 00099 0009R 11937 1108N 12986 13318 11687 13366 11687 OOOJO 11045 00099 J2073 J2012 000-0 000-1 -0002 0'0000 J1711 000J4 000-1 11937 01300 12023 00900 00000 00901 13330 13342 13354 13366 IP IP 17 IP 05028 06066 05650 01314 J1653 J2022 000-0 J2022 3/22 3/23 3/25 3/22 3/22 3/22 3/25 3/22 3/22 3/22 3/24 3/24 3/22' 3/22 3/24 3/24 3/22 3/22 3/22 3/22 3/22 3/24 3/22 3/24 3/22 3/22 3/22 3/22 3/22 3/22 3/25 .3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 3/22 * * M, SF TF BT TFM A A AM TFM AM C BNN FAC,JJ,1 99-KK+l FAC,99 FXD,NX,l 12 ,OUTPUT, 0 1 7 12,FAC,0 12,FAC,0 12,2,010 12,71,0610 JJ,1,010 NX,JJ,016 RTN444, ,0 TF CF SF BTM BTM BTM BTM BTM BTM BTM BTM FAC,I,l FAC-2 FAC-KK+1 FLOAT,0,10 TRACE,ZI,17 FSB,ONE,17 FMP,DY,17 FSBR,YYMAX,17 TRACE,UP,17 FSB,DY,17 TRACE,DOWN,17 IF,I,09 IF,FF,010 99,Y,1 TOFAC,99 FSB,UP,17 11121,FAC-2,0 IF,FF,010 99,Y,1 A TOFAC,99 . BT BTM FSB,DOWN,17 BNF 13331,FAC-2,0 11121,,0 B DORG *-3 IF,FF,OlO MM 99,X,,1 A TOFAC,99 BT BTM FSB,XXMIN,17 BTM FDV,DX,17 BTM FIX,0,10 FAC,l,10 AM FAC,2,7 MM SF 95· 99,OUTPUT,17 AM TFM 99,14,610 IF,I,010 AM C, . N, IF,016, BNN RTN121"O Il1,IND,Ol BO OUTPUT-24,00900,O WA 112,,0 B OUTPUT-24,00901,0 WA ADD YLABEL OUTPUT • • • • • BTM WATY,FMT-5,17 BTM SWC,YLABEL,17 BTM COMPLT,0,10 310 BIM . TOFAC,YLABEL,17 TFM RTN121 MM A BT BTM BNF 12221 MM 13331 3/22 ·3/25 3/22 3/22 3/22 3/22 3/22, 3/22 3/22 3/22., . I 11-2-1 3/24 3/24 3/22 4/2 110 3/13 4/2 III 3/13 * 4/24 4/2 4/2 3/13 112 U_ =Uki:UJiitCeaUtii.ilii¢ _ _ _ _ _ _,, _ _ 13378 13390 13402 13414 13426 13438 00000 ~~,, 1P 1P Jl J4 M7 M9 _______ 03002 01402 11920 "11920 12338 1109L J2002 J2022 000-1 OOONI 01100 00000 ~~ _ _ _ _ _ _ _ _ _""""''''''''''''''''''''''M't''''''"'-........ ,,"t"""",,'··t"""'· .. 7..::=-"··-·~~~~ 3/13 3/13 3/13 3/13 3/13 3/13 3/13 'BTM - FSB, DY, 17 BTM TRACE,YLABEL,17 AM 1,1,010 1,51,010 eM BNP -RTN1,-,0 START-1,,06 ~ DEND 0 c i II I 31 -----~~---~~-~---~-----~-- o j\J Ui"li: o k I eLI S T I 1\1 G UF Ut) J EC T f.lt: C K 0 F S P S P L UT S (JEHU JUT I h! F F CJ f{ F i\1 L. H U F F H!-\ 1\; , (; LJ C; GE I\! H E p1 II• L A r~ S • OUO 0 575356630000000008050 OOOl}OO OOOUOOOO 00000 0 OUO UOOOOOOOOOOUOOR 00 OUU UOOOU OUOO 001. 11110930000526110451109344111541104533110450000026110451104~k0011109411154000002 111109300005261105011093441121411050331105000000261105Ol105UkOOll11541l214()00003 1 1 11 093 000 05 26 11 0 5 5 11 () 9 344 1 1 2 7 if 1 1 0 5 5 3 3 11 0 ') ') 00 () 002 6 1 1 0 5 5 1 1. 0 ::> ~) ROO 1 1 1 ? 1 '-I. 1 1 2"7 4 000 () 0 4 1111093000052611060110934411334110603311060000002 611 0601106 UkOO 111 2741133 LtUOOOO 1) 1 III 093 00005 26 1 1 06 5 1 1 09 3 44 11 39 411 0 6 5 3 3 1 1 06 5 000002 6 1 1 06 5 1 1 () 6 :; ROO 11 1 3 3 L~ 1 1 39 1+ 0000 () 6 111109300005261107011093441145411070331107000000261107011C)70R001113941145400U007 1111 09300005 26 11075 110934411 5 1 4110 7533 11 0750000026 11 075 1 1 () 7 5 ROO 1 11/+ 5 4 11 5 1 L~ 0 0000 H o 11110930000526110801109344115741108033110R000000261108OllOeOROOll151411574000009 11110930000526110851109344116341108533110A500000261108511085R0011157411634000U10 111109300002491207400000kOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0000000001116341165bUOOOll 0579400207850010030579400206280ROOOOOOOOOOOOOOOOOOOOOO00000001011165411685000012 OOOOOOOOOOOOOOOOOOOOOOOOROOOOOOOOOOOOOOOOOOOOOOOOOOOOO00000001011168611710000013 OROOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO00000001011191411916000014 5000000000ROOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO()000001011197311983000015 l00000000310000000015000000002ROOOOOOOOOOOOOOOOOOOOOOO00000001011202912059000016 170131412038170243012048170177200000261202800485270131411070R0011207412134000017 170140212022270300211075170321812058170140212002270131411050R0011213412194000018 o 312 270300211055170321812038170140212012170304611982270233611055R0011219412254000019 1701402120731701314120021703046119b2270243011070170140211992R0011225412314000020 o 161206300000161192000001161192700001161197211711211197211927R0011231412374000021 211197211927161197200000111192700001241202811927461235001300R0011237412434000022 151202300000131206300050320009500000260048500099270165211080R0011243412494000023 1200485000U1210048511920140048500000471266201200161192400001R0011249412554000024 161197211711211197211924211197211924161197200020111192400001ROOl1255412614000025 241202811924461255401300151202300001111206300001161192400000R0011261412674000026 26004851202d120048500001230048511924320009500000260048500099ROOl1267412734000027 2701652110b0161193211711211193200485211193200485111193200002R0011273412794000028 16119 32000711111924000U 124110601192446126740 13002600lt·8 511920kOO 112794128 540000 29 33004830000032004H100000170211200000170140211947170300212048R0011285412914000030 1703046121)U2170233611992170140211957170300212002170140211967R0011291412974000031 16119370000113119370001021U009911065270131400099170300211957R0011297413034000032 441324600483131193700010210009911065270131400099170300211967R0011303413094000033 441311400483491324600000kOOOOOOOOOOOOOOOOOOOOOOOOOOOOO00000000011309413118000034 1 3 1 1 9 370 () 0 102 10 () 09 Y11 0 L~ :> 270 1 3 14000 I.) 91 -, 03002 1 2073 1 7 032 18 120 1 2 ROO 11 3 1141 3 1 7 400003 5 1701772000UOI1004e~000011300485000U2320009500000110009911711R0011317413234000036 160009900U1411119370000124110H51193746129A601300431331R12023R0011323413294000037 391i687009U04913~66000U03YI168-'00901170502811653170606612022ROOl1329413354000038 313 0 o 170565000000170131412022170300212002170140212022111192000001R0011335413414000039 141192000051471233801100491109300000 ROOOOOO 0000000000000 0000000 113 f t 14134500000 'to 00000000000000000000000000000000000000000000000000000000000000100000000000000041 o o AUTOf\'iATIC C C C C I Ii'JI<.AGE GENERATOR FOR SPS SUBS. FOR FN I I. o L. HUFFMAN, GUGGENHEIM LABS. *1205 C 27 1 2 4 3 S P S - F j\J I I LIN 1<./..\ GE A1\1 [) C0 [\! S TAf\.I TAl JT ni'vl ATIC GE~,j ER;~\ TOR •••••• DIMENSION VAR(20) RI?'AD 1, N FORteIAT(I5} READ 2,(VAR{I},I=I,N) FORlviAT (A6 ) TYPE 3 FORMAT(22H TYPE SUBROUTINE NAME. ) A CCE P T 2, S j\! A f,l E I F ( Si\J A/"'1 E ) 5 , 4 , 5 5 PUNCH 6 PRINT 6 6 FORMAT{5HAUTU ,6X,9HDURGl1036) PUNCH 7, Si\iA/'JIE PR I 1\IT 7, SNAI"iE 7 FORMAT(5HAUTO ,A6,5HDS 5) DO 8 1=1, I\J P lH\j C H 9, VAR ( I ) PRINT 9,VAR{I) 9 FORMAT(5HAUTO 8 COi\JTIi\lUE ,A6,5hDS PUf\ICH 10 PRINT 10 10 FORivlAT (5HAUTO ,6X,5HDS PUNCH 30 PRINT 30 PUNCH 31 PRINT 31 PUNCH 32 PRINT 32 PUNCH 45 PR Ii'll 45 PUNCH 38 PRINT 38 PUNCH 33 PR !f\lT 33 PUNCH 34 PRINT 34 PUNCH 37 PRINT 37 PUI\lCH 36 PRINT 36 PUNCH 35 PR lr'~T 35 PUNCH 41 PRINT 41 PUNCH 40 PRINT 40 PUNCH 39 PRINT 39 PUNCH 42 PRINT 42 PUNCH 43 PRINT 43 PUNCH 44 PRINT 44 5) 7) 0 0 315 ------~--~.-~~~~--- . - - - - - ----------------- --------_._------_. --- 0 0 I'; "~ PUNCH 46 PRINT 46 PUNCH 49 PRINT 49 PUNCH 48 PRINT 48 PUNCH 47 PRINT 47 ,4HDS ,6H,10 30 FORfvlA T { 5HAUTO ,6HFF ,4HDS ,6H,5 31 FORfvlAT (5HAUTU ,6HKK ,4HDS ,6H,485 32 FORIV1AT ( 5HAUTU ,6HFAC ,4HDS ,6H,1314 ,6HTOFAC FORlvlAT {5HAUTO 45 38 FORfvlA T ( 5HAUTO ,6HFI'IFAC ,4HDS ,6H,1358 33 FORfv1AT (5 HAUTO ,6HTRACE ,4HDS ,6H,1402 ,4HDS ,6H,1652 34 FORMAT(5HAUTO ,6HFXD ,4HDS ,6H,1772 37 FORfv1AT {5 HAUTO ,6HFIX ,4HDS ,6H,2112 ,6HFLOAT 36 FORIViA T ( 5HAUTO ,6H,2336 ,4HDS ,6HRSGN FORt-iAT (5 HAUTO 35 41 FOR~'lA T { 5HAUTO ,6HFSBR ,4HDS ,6H,2336 ,4HDS , 6H, 2L~30 40 FORI-/iAT (5HAUTO ,6HFAD ,4HDS ,6H,3002 ,6HFSB FORfvlA T ( 5HAUTO 39 ,4HDS ,6H,3046 42 FORMAT(5HAUTO ,6HFl'iiP ,4HDS ,6H,3218 43 FORfvlAT ( 5HAUTO ,6HFDV ,4HDS ,6H,3434 ,6HFDVR FORfvlAT (5HAUTO 44 46 FOR/viAT ( 5HAUTO , 6 H\"J ATY ,4HDS ,6H,4920 ,4HDS ,6H,5028 49 FORjvlAT (5HAUTO ,6HPRA ,4HDS 6H,5650 , 6HCOfvlPL T 48 FORfvlAT ( 5HAUTO ,6H,6066 ,4HDS ,6HSWC 47 FORMAT(5HAUTO PUNCH 11 PRINT 11 11 FORMAT(5X,IH*,12X,25HEND OF ARGUI'~Ef\lT ADDRESSES/5X,lH*,14X,14HSTART I I I LINKAGE. ) START=.6263415963 BLANK=O. DO 12 1=1, N IF(I-l)13,14,13 14 PUNCH 15,START PRINT 15,START 15 FORIVIA T ( 5HAUTO , A6, 4HAlvi ,13HSTART-1,5,010) GO TO 16 13 PUNCH 15,BLANK PRINT 15,BLANK 16 PUNCH 17, VAR ( I ) PRINT 17,VAR(I) 17 FORMAT(5HAUTO ,6X,4HTF ,A6,13H,START-l,Olll) PUNCH 18,VAR(I) PRINT ·18,VAR( I) 18 FORI"lAT (5HAUTO ,6X,4HBbJF ,5H*+36"A6,3H,Ol) PUNCH 19,VAR(I) PRINT 19,VAR(I) 19 FOR~1AT ( 5HAUTO ,6X,4HCF ,A6,3H"O) PUNCH 20,VAR(I),VAR(I) PRINT 20,VAR(I),VAR(I) 20 FORMAT{5HAUTO ,6X,4HTF ,~6,lH"A6,5H,OlI1) 12 CONTINUE M=N/2 M=M·~2 0 I F ( fVl-N ) 2 1 , 22 , 2 1 22 PUNCH 23 PRINT 23 23 FORMAT(5HAUTO , 6X, 4HA~1 ,13HSTART-l,I,OlO) 316 GO TO 24 21 PUNCH 25 PRINT 25 25 FORMAT(5HAUTO ,6X,4HAM ,13HSTART-l,2,010) 24 PUNCH 26 PRINT 26 26 FORMAT(5HAUTO ,6X,4HB ,9HAROUND"O/5HAUTO ,bX,4HDORG,3H*-3/ 261 5X,1H*,12X,36HSYMBOL TABLE AND CONSTANTS HERE ••• ) GO TO 27 END _DATA FOR AUTO-LINK. 9 X XMAX XMIN NX Y YMAX YMIN NY N 317 ---~ o - - - - - - - - - - - - - ------------- ----------------------- 1,,. t tr t trm.tm. If - A SURVEY OF THE BEGINNING PROGRAMMING COURSE o Clarence B. Germain College of St. Thomas February 20, 1964 Last Fall, a questionaire was sent to the 280 schools which are members of the qSERS Group. 175 schools responded. The results are tabulated on the following pages. 1. No allowance has been made for non-respondents. This does bias the results. 2. Since the survey covers only schools baving 1620's, the figures for the end of 1964 do not reflect the influence of schools which will acquire their first 1620 during the year. 3. A suprising number of respondents gave incoaistent answers; e.g., they indicated floating-point hardware, 'but not divie hardware, or they indicated that 35% of their students run their own SPS programs, while they taught SPS only to 20% of their students. 4. Figures for index registers, binary capabilities, and the 1627 plotter may not be indicative since the questionaire was circulated too soon after announcement of these features. 5. Average enrollment in the beginning programming courses in 170 students per' school per year. 6. Many of the Mod.el II 1620' s will supplement existing Model I' s, not replace them. 7. Relatively few schools indicated any plans to obtain the 1443 printer. 8. The disk units willmore than double in popularity during 1964 with 1/3 of all schools having at least one disk unit by the end of the year. 0 9. While 3% of the schools offered no course involving Fortran, 35% of the students were taught more than one version of Fortran. 10. At the end of 1963, 51% of the schools had the hardware necessary to run Fortran IIj by the end of 1964, this figure will rise to 5910. 11. 85i of the students get tlhands on experience in running their own programs on tl the computer. This percentage is about the same regardless of what programming systems (SPS, GOTRAN, etc.) are taught. 12. Jim Moore's Multi-Trace, 1.4.003, was the most commonly mentioned trace program taught to students. However, 85% of the schools indicated that they used no trace program in their courses. 13., The figures for textbooks are for use in at least one course. Many schools use more than one text in a course. 31% of the schools use only IBM publications as texts. While a wide variety of texts, many unrelated to either Fortran or the 1620, are in use, only four commercial texts and a half-dozen IBM publications are used with any frequency. Of the non-programming type texts, numerical analysis books, particularly Stanton's, were most "often mentioned. 14. The textbook percentages in no way indicate sales of books; these figures are quite different from the percentages shown here and were not a 'part of this study. o 318 =_ _i_eaZilisa':;;iiUOU\i iU&lliiiii X.i ,iI AMP;; RESPONSES OF 175 SCHOOLS TO A SEPTEMBER 1963 QUESTIONAIRE Results are given as a percentage of the number of schools replying to the questionaire. Probable errors do not' exceed ±3% except for items marked with an asterisk (*) where the probable error is less than ±eJfo. Results are given for the end of 1963 and for the end of 1964. Changes for 1964 are only for equipment now on order. Slight discrepancies in the percentages are due to rounding. 1620 Model: 1964 One 3 31 3 Special Features, ~4ode1 II (1964) Automatic Floating-Point 65* Index Registers 0* Binary Capabilities 5* 1 1 31 14 31 14 1 1 3 3 1 II 11 Special Features, Model I AFP, Div, IDA, Edit AFP, Div, IDA AFP, Div, Edit AFP, Div Div, IDA, Edit Div, IDA Div, Edit Div IDA, Edit IDA Edit ~o special features Summary: Autamat~c Floating-Point Automatic Divide Indirect Addressing Additional (Edit) Instructions Number of 1620's in the school: 8910 I 31 o 1 3 o 3 13 13 Type of Courses Offered: 35 Both credit and non-credit Non-credit courses only Credit ',courses only No answer or no courses 64 82 82 5 Installations with Printer (1964) No disk 23* 1 disk 15* 2 disks 54* 0·:' 3 disks 4 disks 8* 1 82 82 95% ~o 1 34 0' 5 36 13 41 o 64 Departments which offer 'courses: Storage: 20K core, 40K core, 60K core, GOK core, 4'oK core, 60K core, no disk no disk no disk disk disk disk Input-Output: Paper Tape only Paper Tape and Cards Cards only Magnetic Tape Paper Tape Cards, 1622-1 Cards, 1622-2 Cards, RPQ to read 800 cpm 1443 Printer Disk, one or more 1627 Plotter 1710 48 21 17 5 4 5 4 10 86 4 13 83 13 3 14 4 2 38 18 13 12 9 9 4 10 86 4 14 81 16 3 8 31 Engineering Education Mathematics Business Other 40 1 45 31 40 Subjects Taught: Machine Language Operation of the Computer SPS GOTRAN FORTRAN with FORMAT FORTRAN II or II-D FORGO, etc. Use of same library trace Block Diagramming Monitor I 32 66 29 17 47 33 35 13 63 9 4 3 1- 319 Disks: No disk 1 disk 2 disks 3 disks 4 disks Hardware necessary to run: Fortran II only Fortran II and II-D Fortran II-D only 86 68 8 5 20 11 1 1 o 31 9 5 Students are expected to write and run their own programs using: SPS II GOTRAN o FORTRAN with FORMAT FORTRAN Pre-Compiler FORTRAN II 25 15 43 28 21 29 19 11 Required or recommended texts: IBM Publications 14 1620 Reference Manual 1110 Reference Manual BPS Reference Manual 4 49 GOTRAN Reference Manual 1620 FORTRAN Reference Manual 1620 FORTRAN II Bulletin FORTRAN General Information Manual 22 61 38 23 12 15 1620 Program Writing and Testing Bulletin Introduction to IBM Data Processing Systems Programming and Block Diagramming Techniques 12 Commercial Publications Germain--Programming the IBM 1620 Leeson-Dimitry--Basic Programming Concepts and the IBM 1620 Computer Gruenberger-McCracken---Introduction to Electronic Computers Guide to FORTRAN Programming Organick-A FORTRAN Primer Colman-Smallwood--Computer Language Smith-Johnson--FORTRAN Autotester McCracke~A 27 39 6 38 38 6 3 o 320 . aam:iSa,,;;;: Utah State University Logan, Utah o FORl'RAN ''TEACH'' PROBLEMS by Wendell L. Pope These problems are designed to be of assistance in introducing the neophyte to 1ORI'RAN. Problem sets and programs to check them are provided for arithmetic statements, subscripted variables, fixed and floating point variables. functions and control statements, loops and input-output~ The problems do not require that a student be able to write a complete program. They provide a means of acquainting him with the characteristics of FORTRAN in easy stages and help to bridge the gap between the introduc'tion to computing and the writing of a compl4te program. The student's statements are checked for correctness by imbedding them in the appropriate checking program. They are checked for compilation errors by the FORGO processor, and for accuracy by the checking program itself. This is done by comparing the values computed by the student's statements to a predetermined set of "correct" values. For wrong anSwers, the number of the problem and the value computed are output, for right,answers only the number of the problem is output. 321 C FLOW CHART - TEACH Problem Checking Program Read values to . be used in this problem set 1 Read the "right answers for this problem set f Execute studen~ statements evaluating ~ for each value of i l o ~ ~" r-__~==~~~~ea~c~h~v~a~l~u~e~o~f__i ______________________________. I, No Is X. Store'i as wrong l. Yes Store i as right Punch i, X. and l. the right answer I Increment i ________ & test for exit i iExit ~re any prob-\---...-.:J. Punch the numbers 0 lems right? of the correct problems Are any prob,lems wrong? I o lNo B _J Yes_..J ~nch the numbers of the incorrect prob- 1..J..ems.... ~ __ _ 322 • Ama_LiualUiS;' !I dUI ; C TEACH PROBLEM NO 1 PLACE STUDENT HEADER CARD IN FRONT OF THIS CARD DIMENSION NRITE(10),NWRNG(10),RIGHT(lO),X(10) READ,A,B,C,D,E,F,G,H,(RIGHTCI),I=1,10) C INSERT STUDENT STATEMENTS BEHIND THIS CARD K=O L=O DO 9950 1=1,10 IF(X(I)-RIGHTCI» 9912,9914,9912 o 9912 L=L+l NWRNG(L)=! PUNCH 9920,I,X(I),RIGHTCI) GO TO 9950 1914 K=K+l NRITECK)=I 9950 CONTINUE IFCK) 9922,9924,9922 9922 PUNCH 9923,(NRITECI),I=1,K) 9923 FORMAT(6H RIGHT,10I5) 0'24 IF(L) 9925,9~26,9925 9925 PUNCH 9927,(NWRNGCI),I=1,L) 9926 STOP 9920 FORMAT(9H PROB NOo I3,4X,9HYOUR ANS=E1608,4X,10HRIGHT ANS= E1608) 9927 FORMAT(6H WRONG, 1015) END 10.341296 10.345599 8.6867569 ~400683394 15.683097 .0034784067 .OOOj4329602 1.1234567 ~10664629E+02 0.41901625 0.84447317E+02 0.12039480E+02 .65303219E+09 0010259388E+01 Oo14580430F+Ol Oo87982136F+Ol ~46184027E+02 O.11729093E+03 323 o h $ tt 0 ', '11'i ttt *t irittnf tritrit . C TEACH PROBLEM NO 2 PLACE STUDENT HEADER CARD IN FRONT OF THIS CARD DIMENSION NRITE( 5) ,NWRNG( 5) ,RIGHT( 5) ,XC 5) ,R(4) ,ft,(4,4) READ , K, L , ( ( A( I ~ J ) ,J = 1 , 4 ) , I =1 ,4) , ( 8 ( I ) , I = 1 , 4), ( RIGHT ( I ) , I = 1 , ~ } INSERT STUDENT STATEMENTS BEHIND THIS CAR~ C K=O L=O DO 9919 1::1,5 IF(X(I)-RIGHT(I») 9912,9914,9912 9912 L=L+1 N\~RNG(L)=I PUNCH 9920,I,X(I),RIGHT(I) GO TO 9919 9914 K=K+l NRITF(K)=I 9919 CONTINUE IF(K) 9922,9924,9922 9922 PUNCH 9923,{NRITE(I)~1=1,K) 9923 FORMAT(5HRIGHT,10I5) 9924 IF(L) 9925,9926,9925 9925 PUNCH 9927,(NWRNG(I),I=1,Ll 9926 STOP 9927 FORMAT(5HWRONG,lOI5) 9920 FORMAT(8HPROB NO. 13,4X,9HYOUR ANS= E1608,4X,}OHRIGHT ANS= E1608) END 2 '3 2.3964587 6 0 0247685 o 3 G 6241346 4 0 1024678 7.3214680 8.3469201 5 0 4673001 410125807 401357653 503751468 9 0 3704368 4913189 0 2 503422c)87 6e0347'1'12 10.437695 770104270 3.3524569 7.3107~86 -1~3579430 086225934 50312 l 0 3687946 So3420769 60046-2 0 5347962 4 C TEACH PROBLEM NO 3 PLACE STUDENT HEADER CARD IN FRONT OF ,THIS CARD DIMENSION NRITE(10) ,NWRNG( 10) ,RIGHT! 10, ,X(10) ,8(4) READ,K,L,(BCI),I=1,4),(RIGHT(I),I=I,5) C INSERT STUDENT STATEMENTS BeHIND THIS CARD K=O 0 . ' L=O DO 9919 1=1,5 IF(X(I)-RIGHTCI» 9912,Q914,9912 9912 L=L+1 NWRNGCL)=I ,PUNCH 9920,I,X(I),RIGHT(I) GO TO 9919 9914 K=K+l NRITE{K)=I 9919 CONTINUE IFCK) 9922,9924,9922 9922 PUNCH 9923,(~RITE(I)'I=1,K) 9923 FORMAT(5HRIGHT,lOI5) 9924 IFCL) 9925,9926,9925 9925 PUNCH 9927,(NWRNG(I),I=1,L) 9926 STOP 9927 FORMAT(5HWRONG,10I5) 9920 FORMAT(8HPROB NO. 13,4X,9HYOUR ANS= E16~8,4X,lOHRIGHT ANS= E1608) END 2 3 157~09495 -103579430 9 0 3704368 10 0 437695 -1 0 3579430 150915392 205 0 01975 1100 ~2Q5347962 o 325 ---- --.~.-.--~- .. ------------~-- ...... ~-------~- C TEACH PROBLEM NO 4 PLACE STUDENT HEADER CARD IN FRONT OF THIS CARD C TEACH PROBLEM NO 4 PLACE STUDENT HEADER CARD IN FRONT OF THIS C DIMENSION NRITE(5),NWRNG(5),RIGHT (5),X(5) READ,A,R,C,O,E,F,G,H,K,(RIGHTCI),I=1,5) C INSERT STUDENT STATEMENTS ~FHIND THIS CARD K=O L=O DO 9919 !=1,5 IF (X(I)-RIGHT(I) )9912,9914,9912 9912 L=L+l NWRNG(L)=I PUNCH 9920,I,X(I),RIGHT(I) GO TO 9919 9914 K=K+l NRITECK)=I 9919 CONTINUE IF(K)9922,9924,9922 9922 PUNCH 9923,(NRITE(I),I=1,K) 9923 FORMAT(SHRIGHT,10I5) 9924 IF(L)992S,9926,9925 9925 PUNCH 9927,(NWRNG(I),I=1,L) 9926 STOP 9927 FORMAT(5HWRONG,10I5) 9920 FORMAT(SHPROB NOo I3,4X,9HYOUR ANS= E1608,4X,lQHRIGHT ANS= END 1.2457369 203580123 308609756 407602541 503025768 602047536 700367524 803205689 2 1 01161260 103001178 078878076 204429843 ~o648008P' E1608~ I 10 I :=a::iUSXUSll£ii :au,::: :H;; : !.il: ; .. &14 KZIS;; ( ' " , , ".. , ........'""'... ,~~=-"- C TEACH PROBLEM NO 5 PLACE STUDENT HEADER CARD IN FRONT OF THIS CARD DIMENSION NRITE(S),NWRNG(5)9RIGHT(5),X(S) REAO,A,8,C,D,E,(RIGHT(I),I=1,5) C INSERT STUDENT STATEMENTS BEHIND THIS CARD K=O o L=O DO 9919 1=1,5 IF (X(I}-RIGHT(I»9912,9914,9912 9912 L=L+1 NWRNGCL)=I PUNCH 9920,I,X(I)~RIGHT(I) GO TO 9919 9914 K=K+l NRITECK)=I 9919 CONTINUE IF(K)9922,9924,Q922 9922 PUNCH 9923,(NRITE B > A > H > C > F > G > 0, compute X(9) =A+ + B+ C+ E + F + G+ H to obtain the .most a.ceuracy. o 329 'iU •• "rf1' mr TEACH PROBLEM SET 2 Subscripted Variables o Assume the arrays A = andB = are in storage. Write and keypunch correct FORTRAN statements to evaluate each of the following expressions. 4 2. 1. X(2) = E a ° b. j=l l J J 4. x(4) where {b ! denotes the integer portion of b l • l o TEACH PROBLEM SET 3 Fixed point, Floating point, and subscripted variables Assume that values of k and Land B are in storage • = .. • b n Write and keypunch correct FORTRAN statements to evaluate each of the following expressions. 4 '1. Xl = E b~]. 2. ~ i=l 3. 0 4 4. X4 X3 = . E1 bok. ].-]. l.= 5. X5 = ~ + kL +. =b k 3 -5 = ~ b~ i=1 '- k3 .;. 2 330 cai.lUi AUIiIZUiJik!iih I. Ii.: , Li, #; -------_. _--.--------_ . -.-- ~~~-~--~-~------~-----------""""~~~!""""---."""'-.-.'="'_ .. . , . .-._._. . _--_ .. _ - - - _ .. _ .. _. . TEACH PROBLEM SET 4 Functions and Control Statements Write (and keypunch) statements to evaD.uate ~, X , X and X5 according 2 4 tQ the instructions in the flow chart below. Assume A, B, C, D, E, F an~ K to be defined. o i \ A = e ~A<'O Test A .... A>D .~ VA Xl = " A = 0 ~JL ~= 0 ,ir "/ '*" ~ ...... ~- ,r.,.- X = Arctan(A+B) 2 . i3 'v' = VCos(C-D~ "V X4 = In(EoF) !-- K = 1 Test K K =3 X4 == ln~F-E>' K=2 '~ ~4 = ,II It In(E+F) -"" i/" "' 'V' X5 = I \+X2+X3+X4 , 1 c 331 .. - ..---~-.-.---.--- ..-.----.-..--~----------- -~----------- .------------ - - . -- TEACH PRO BLEM SET 5 Loops Write (and keypunch) statements to evaluate Xl' X2 , X3 , X4 and X5 in ·the exercises below. 3 3 1. 2 3 Xl -- 3 + 6 + 9 + 12 2. Xi 2 + • • • + 99 • c =A D + B = 2, 3, 4, 5. i "EoX. v 1-1 TEACH PROBLEM SET 6 Input - Output (without formats) ~ad o A,B,e & D Punch A,B,C & D i =1 ..• '-'f" ~II Read Z ,1,/ I~ = AZ3+BZ2+CZ+D ,~ o Punch Z, X.J. ,,(~ Is i < 5? ...];No . Yes ) i + 1 ... i 332 :::aaama"" stli iiililMii .SW , \ .Ji 0 LOAD-ANn-GO SPS WITH MONITOR CONTROL Kenneth M. Lochner and Glenn R. Ingram c o 333 urw'n" ",!!"n P' t tr t .'.& WEtt' r' MSC ASSEMBLY SYSTEM: I.) INTRODUCTION This paper will discuss a monitore(l ~ load-and-co type assembler developed at Montana state College and the conditions that prompt.en tts development. It .is a report on work done by Ken Lochner, 'formerly of the rJSC Computing Center, ant.l SOO!l to aSSUr1C d\1t.ie~::; as chief programmer at the Dartmouth Complltation Center, in the actual writing of the processor. To s~ggest some of the background reasons for this processor, it is well to admit that I am a relative newcomer to 1620 ranks in completing my second academic year with a 1620, after leaving a 709 :Lnstallation. Anyone who follows this path finds himself wondering why in the world he did, and then develops a feeling somewhat akin to tbe fellow who had a job with a circus. This particular job consisted of followinr; behind the animals during a parade, and cleaning the street with his little shovel. After an espeCially trying day, he complained so bitterly that his wife asked, "If it's so bad, why don't you quit and get another job?" The man replied, 1I1llfhat! And get out of show bu~)i ness?" If t.he analogy isn't exact, it may be suggesti.ve that some things could be cleaner in the 1620 tent. IT.) BACKGRQUND AND To indic ur, reliak1ility of the test, the xank t percentile l"ank:. z""scorS;l and t-s·r:orefor 0 ll each ;Jtuder.lt_ o ·,,4 ..· 343 EQUAtIONS USED o Alpha-3 = Alpha-three: 1 ncr"3 Alpha-four: Alpha-4 Arithmetic Mean: X= Pe rcentileRank: PR = Reliability of Test: Rt = = 'n 2£ (Xi - X)3 i=1 1 no-4 100 Cn ... rank) ·n 1 Z - X (1 - X) crZ (I - 1) Standard Deviation: o T -Scor.e(Sipa Score); T-Scor.e = 50 + Z -Scor.e (Standard Score.): Z-Score = Le~end: 10'(Z score) Xi - X· I = Total Possible Points n =.Number taking the test N n if nZ 30 n-l if n< 30 Rt = Reliability of the test Standard.Deviation X = Arithmetic Mean Xi Individual Score. = = = o _UA& 'Ustail::Z1l ( a ¢# ¥4 II 1 John E. Frue~d, Modern Elemen.tary Stati!~~£.~.~fE-;ngle·wood Cliffs, 1963p, p. (indicated a.bove)" Z V·'lc.or t H . N011, .1.ld~:rOo.Lh,,~,lOU ... ".'., "' .. ~., .... { . .,•.. '0 ..J"-J·~;'U" 1' t1 ,"... 1::,.t._C ""i ·n":'.l 11.;'t ";"'~'I"'" -1"'~~ .. ,.a.. ~ ,"'d,·a ..,r.",'.I(.",:ne,.; ..;~ p. (indicated ab()ve;·.. ':b OJ> - _ _ ..."......-._.!,...,,~..l.,j.o. ................::Ro, ...... ~ /I. >OQo.' ...... h;6J• • ............ _" _ -.~"", ••~"" ••" " , . " " . _ . _ ......... ,..."-~ ... _ " ... ~u.• _~_" ....... _ _ .......... , ...... , ..... -........, :~, ! ./,..,!OS"(.) 'i:." ,.}- .. 'Vi _r o COMMENTS ........... ..... __ enveen:a·,c;~J~:ted group's m~. . and a#,<,apecl(ied J·aY':;'.core. value. The'mean is o. O~:&n.d. the.taIldar.:d. 'is": :f~ 0., ae.tion J"~re~ple, 8Uppf).~:;;a.:.tudeDt had a.c~r. of 49~li~f.i. to be, cOIXlll&red wi~ his cl.·elJ.. :~~en that the mean aDd 8tandat.~, Clevia~ion are 40. ~ and 6. 0,,) re.pectivelY~> we find hi. z-.core is '(49 -40)t~' = 1. '5, whi.~)l· co;t'r•• ponda:" at-score 0(,:$'0 + 1'0 (1.5) :: 65~ In otb.~ w~'rds. this 8tud.~ti, • :o~~ ,is 1. 5'~tandard de~#.O.8 abO~ the mean. :Thi.~y:~18o beiJl~~.t.d to ~••n ~t.--ite 8corftdbe~!'tba~abc)ut 93 per cent of the'l~up~ •. o . -8- I, . PREPARATION 'OF DATA O" J Data Carda: Each IBM data. input· card (except the header' card) must be prepared in the followbig manner: Columns Field Student Number Name (see Exceptions, below) Score (fioating-pointwlth de¢~ in columD.: , 32: may have one P~c.e.·., to right of . .' "ecimal in columD.3:~l. 1-5 7-Z7 28-33 0 Oa 40 46-47 55-56 0 .. 64-65 o. 72.·73 O. . . a~ader tb:f) Card;: The first card of the data nust contain in columns 2.8-33" total poss.~ble points. CC)lu.mns 40-73 (,re punched aS8ta~d above. Sortlns: o The data cards are then sorted in descending order', according to' columns Z8-33.After sorting is completed, the header card is thell placed, on top of theciec~j it-is the first data care. . . LimatiODS: The program wH'i' handle ~ lJ!aximum of SOO iDdiftdualtest a,cOr••• Exceptions: The program is so design(d that column,s l--Z1of the data inpUt cards may beleit blank wi~out affectillg the rest of the prog,\am. o aasMi!JiiUtiA I,ll i.i.iltti ii.UQn $ ~.t;.,i III :1 OUTPUT \ Printed. Output; The arithmetin mean. standard deviation, alpha-three, alpha .. lour, and the reliability of the test are printed out first (in that order.) This Information is printed regarene. 8. of the switch settings. o The reinainiag irdormatic)D. Ulay or may Dot be printed out. depending on the switch settinl•. The heading, etudeatllUmhe:r and name. score, number•. rank, percentile rank. z-eco%'e. -and t-score is printed out neXt. Final1 y, e«c.h ~~De of tyJjeCl output contains ~e following information "in the indicated or01:; student ni'libber. name, .core, a "coW'lter" number, rank, percentile rarike IS-score, and f.score .. Students receiving the ·same score are grouped together; the type.writer double spaces between groups.- When the last card ~8 - been proceesed, the_ word STOP is printed. Pun«;hed Output: The first seven cards contain the ~_rithmetic mean, 8tancm.rd deviatioD. alpb8.-three~_ alpha-four; the reliability of the test, and two blank , cards. 'This lnformation is punched in COlwnllS 1-35. on The He,adlng: Name, Sere. ~(),1 Rnk,Ptrk, Z-Scre. and T-Scre is punched the eighth ca~d. 0 The remaming cal-dscontain the following infornu~tiot\ Field 'Student Number C91n~Da 1-5 _1-Z7 ~ama 28~33 38-40 Be'ore A "counte r" numbe r 44 .. 49 Rank 54-59 Pe rcentile Rank Z ...Score T-Score 63 .. 68 71-76 Sorting PUnched Outl!ut: The punched output cards Jnay be sorted alphabetically by sorting in ~8,<:ending order-according to columns 1-5. (This assumes that the student nuinbers are assigned in sequence 'when the last names are in - :alphabetical-order). The first eight ca.rds of the output will be rejected by the sorter because columns l-S-are blank; however, save these cards to process through the Alphabetic Interpreter. o J I -10- 349 I -j o .Alphabetic Inte rpreting of Punch~d Output: The information contained in the punched output cards may be printed at the top_ of each card by means of the IBM 548 Alphabeticlnterpreter. The alphameric interpreter board should be wired as follows: Wire Read Brushes The Print Entri.es to 1 ... 29 30 ... 32 1-35 38-40 44 .. 49 54-59 41··~46 63,..68 48·,53 71 ...16 5~/,,60 34 ..,3Z o -11 ... 35U I I ') :AEAIi :ail LiCSi!6t!U;;;:, $ a .!I MACHINE REQUlREMENTS Equlpmeat Specificadoria::: IBM Sto~a.e .l\eqwrel1JeDta: . ~OK" 16'20, l6ZZ" , , 0 " ,' .. Source Lang'la,e:" .Mit FORTRAN Special Features: None c 0, I. ""-lZ~ 351' o Step-by-Step Procedure: A.sUme that ·the object deck i8 compiled, and at Then· ~.~ - hand. ~ad 1. Clear 1"62,Z Card z. Load A£i~F~rtran in the 1622.. 3. Check S~tche8,' Puach. load.~. object clec~. Sw~tCh Afi~ .F9rti'an Subroutin•• STO~ PARITY.: ,Check Switch ... 1/0 Chec:~ Switch OVERFLOW Checlt and the STOP . PROGRAM 4.lf»re •• RESET and Reader, ..,OAD,·. 5. Set Program Switches :fol' desired option. Switch: .. a, ON for PUNCHED o.tpat. . SwitCh Z'and 3 OFF for PRINT ltD output. Switch Z and 3 O~ fO,-",PBlNTJD and PUNCHED output. Switch 1 :and 4: 6.Pre~8 7.· S.TART Place data carda' in the 162.Z an4pre88 READER START. a.Press 9. OF~ • ~UNCH STARTif . neCe8$ary. Typewri~:t· prints aD.~lor 16ZZ punch••• .. . ' ~. 10. Repeat Step 7. (i.e. two pa$see are·.%equired). II. Typewriter printsaud/or' 16ZZ punche'8. prOc••• ~d. the V:'0rd S·TOP 18 typed.' When th.la;.t .,*1'4 bas beea. o I :s:aiJiUiUZ &t4Iititi:iltiiiU J M.4 1M, Mi ;sa : ----.-.. . --.-"-~. --.-.-,-~~-----~--------"'"""""'''~''.,"~="~= Expected Stops: 1£ the data .is not sorted in descending o~der, a "card out of sequ.ence" message ~ill be pl"intedout and the. program will bl"anch to the STOP command. The restart procedure is then neces sary. The computer will automatic'a1ly'stop if more than 500 scores are' read in. o Resta~t Procedure: Press INSTANT STOP p RESET, INSERT g RELEASE, START Repeat the step-bY"'step pro~edul"e for program exe«=ution. 0 o ,,,1<'1- -~--~- ..... -----.. .. - ...--",-~------ 353 -- ----"",--~~----------------".---.-- .... F.low Chart; tor . Teet Score. AnalYsis Format o Specification tor. I/O Set Counters Equal to . Zero -----~ Count. the Number Persons Taking ot the Test Yes S'i'O.P o Read a Raw. Score PRINT Error Compute Arithmet1 o Mean -15..._IliM"::::": ; ; (( II NMM $Ii ¥I@ . ZS $ = .. ,.;. .:,."'"'-.-""'. . -" ;;;..;;,. .. _ _ _ _ _ _ ..... ~.~ •• _ . _.~ _ _ _ _ _ _ _ _ _ _ _ _ _ ~~~! . . . .__ . . . ..~.". . __ • .u."""'"~._~."""'._ ~~~ _ _ _ _ _. Find the Deviat1on' ot Each Soore From o the Mean . F.ind the SuP.l of the Squared., . Cubed., and the Fourth Power of the Deviations From the l-1ean Use n tor Oalcu.lat1on of No_ .-( ...a-_ _ _ _ . Is the Number ot Persons ?aking the Tes~ ~ 30? Yes Use ~;-='--ooE!iIIooI (n-l) tor Calculation of CalcUlate the Standard Deviation, Alpha-3, Alpha-h, and· the tieliability .t>UlICH, Arithmetic Hean, Standard >--~----li~ Deviation,. A~lpha~3, Alp~4, Reliability POUCH Head;~g' . Off . ott Read Total Possible l:'oints 355· -16- 0 ----:_ _ _ _ _-BII o sad .studsilt Number, Name. Raw ·Score· . No Compute Rank, .. Percentile Rank" . z,..score, T-score PUHctrj: St~ent Hlmlber. >-_On_·_____~ NalIle,ScQ~J:. No, Rank, Oft Pereenti1$' ·Rank.. ·· .' ~scorei.·:·T~score . .. ' ott: o !O ~RlNT, student Number; , N~~:Sc~re~ No~ Rank, .t li- Percentile .Rank, z,;..sco~~ T~sCore *Where ·If .~ua;ts th8.. number· of p~ons . rt' talctrig the test,. -11- 356 ,...:anawl liIat $. !:. :; \ uu i #i II i o 357 ---~~~-------- \,:~ III -18-----------------~------------- \~ I ---_.------ .._--- t"MNtii ,*W*'Hrh " 'Ii ffiW"nm 1!!1F7.,'!!, P' II En j t r - i . - PRINT 889,AMEAN T0794 10818 o T0818 PRINT 778 PR .NT 900~ STDV sAlFA3 T0842 PRINT 778 PRINT 901,AlFA4,REl T0938 PRINT 778 T0962 IF (SENSE SWITCH 2)39 42 10982 39 PUNCH 390 10902 0 11006 T1026 f1050 41 IF (SENSE SW ITCH 42 PR fNT 778 3)l~2j)44 ' PRINT 442 Tl'074PR~NT 443 PRINT 776 T1122 44 OLD-l000 0 0 11098 T1146 . READ 100, TPTS& Zl 92:2, Z3, Z4f) ZS DO 83 N-l NMBR T1230 T1242 100,SCRE(N)"l1ltZ2~Z3&14r'ZS READ J-N T1350 Tl3.74 IF IN-NHBR)Sl f)63 99 T,1.442 51JF SCRE(N)~..SCREtN+l) )16,,52962 T1554 S2 IF J-l)99~54,S3 T1622 53 IF (SCRE(N.... t) -SCRE(N» 16 1'641)54 T1734 54 SAVEN:-N T1758 55 N-N+l 11794 IF (N ...NMBR)S7 ;,58 99 ' 11862 57 IF (SCRE(N)~"SCRElN+t» 161'55 ,58 O il T19.14 58 TH ISH-H TI998 ADD-(TH'SN-SAVEN)/2 0 g 12046 12082 T2090 T2158 I12222970 4 12302 12326 SRANK-SAVEN+ADD GO TO 64 62 IF (J-1)99t)66~63 63 IF (SCRE(N.-l)-SCRE(N»16,,64 fJ 66 64. RANKaSRANK GO TO 61 66 RANK-N 67 ~J T2350PtAtlK- - (( Ttf1BR,,4tANK)*1 0( 0) /TNMBR 0 12410 ZSCRE-(SCRE(N)eAMEAN)/STDV ' 12482 TSCRE-ZSCRE*10 0+50 0 T2530 If (SENSE SWITCH 2)73,,75 12550 73 PUNCH 100,SCRE(N)"N"RANK~PRANKl)ZSCRE9TSCRE T2658 IF' (S£NSE SWITCH 3)7Sf)82 o T2618 1S 0 tF(OlD--SCRE(N})16~76f>78 12766 76PR iNT 12790 . GO TO 12,798 18'PR tNT 12822 19 PR 't~T 716 79 718 l00f>SCRE(N) vN&RANKe>PRANK.,ZSCRE, TSCRE 12930 82 OLD-SCRE(N) 12918 83 CONTINUE 1301'4 99 STOP T3022 , o END END OF COMPILATION 1354113810 35~ .. 19- . , aa:sosUiUU Ii lit: .' .ii IA. q . ::44#4.1 " "L,II A.LFA 3 o = alpha~;.th:t;e~ .ALFA 4 NMBR = total number of p~rsons taldn.g the test PRANK RE.L = relia'bilityofthe test SAlfA 3 = E.(Xi-Xb 3 i=l n SAFA4 seRE = individtlal test score (raw score} .~ SDEV {Xi ~ Xj o SSCRE STJ)V = stanciat'd d3viation STDV 2 =~ STDV 3 ::: 0- 3 STDV 4: = 0-4 TNl.-iBR = total ;tlumber of persons TOTAL = TNM:6R - 1.:0 TPTS if TNMBR .:~> 30 = TNMBR = total possible points or tdal number of test items, ::: 0- talt~n.p' if TNMBR the test. ~ 30 which ever is greater. o 359 -zo- :, ',I·' /I " TSCRE 0 VAR Z'3CRE Zl - ZS r_ oW _. T=SCRE (Xi .~, X~ Z·,.SC()RE~ = DurIlmy Var.ia.bh~s o o 36H II .1 • tI " llit! ; :, Z!ii J. z .Aa:WiF ARITI+lETtC MEAN IS STANDARD OEV ~/rr ~ON 821) 256 VS ".,c,828 Al,PHA 3 IS ALPHA L~ IS REL lAB fL·ITY OF TES1' 11 () 09 f4 3 695 0 ,S I) 890 o 'wuerm"!!!', !I'R!!JUINU' "tiM j= trttittt#± "II ECHARD Wt-~ RA ND~\lJ~ 6!}~~65 PASGE WANDA ElA~NE 6!.i~9U J·i)I\TR iCK GERRY W~\YNr~ (f725 W~lDER RICHARD l'fNN 55475 ~tC .COy CAROL RUTH 2l}31 0 ~- 22977 DOUGHERTY CHARLES \i:. 25383 EMMER1" R.OB E~T tARt. 76 00 16 O 16 0 Q 28 29 29050 29;;.50 '4 o;J;t ~t~.. S - 240358 24Q358 0 30 3t 29':050 29()50 74,,0 3? ~2()OO t1~948 70 0 ;,> 33~SO ~4o ~O2 16 0 0 ~.':t 21~\:)358 "~o563 44,,360 440360 440 36;J 44()360 . c '1 ~,.lJ. 420558 ~1~104 38 a 952 38 952 38 051 563 e3 0 ~63 • ".8 Q.. - e 563 ~~ "i.!..t~. 4iJQ!II 10()O 34 33,)50 l'~o *02 ,,·1 el Oil 69 0 35 35<~OO tO o 256 ,,~·1 69130 RAUSCH MICHAEL 73936 SANDERS JAMES W 64 0 0 36 50 36. 50 6 u iJ..l0 60410 ~R,~645 ..,~1 c\645 330544 33 (,5 1}4 62091 NEWHARD NANCY FAYE 28065 FOLTA JAMES VI tl CE NT STOP 62 0 .0 36 37 -,;8 ;.~ 38 00 2 0;) 4:'6bit <'£>1 C'825 310741 48.0 39 39.00 0000 c·~3o 087 19,,122 02025ARMSTEA[$ ROB Eft l' LEE 0 0 64 0 0 0 Q 0 «: 194 0 0 o o 362 LLI Lilli II _iliSiiid:':: iULIOUkt, . # i ; $; .. 4U4 .i . ~I I o A PROPOSAL FO,R AN ADDITION IN GRADE REPORTING PROCEDURES TO ALLOW'FOR , AUTOMATIC PROCESSING OF PROBA'rION STUDE'NTS Indiana State Colle ge Computer Center 'May S. 1964 363 o j8 h"IbI'HWti'i%H¥'i i'NWT,!'Jii,'Mijw'trnrump"" "1"1,,"'8 ' ***# FLOW OF C"': I'l PROCEDURE o I II II IliiiutM; : Ii : ( aM ;;Uti., II Explanation ,of Flow Chart o A All card files are merged together in alphabetical sequence by student number. B Computer writes grade reports and updates student index file. Probation cards for failing students. Store used cards. C Grade reports' to students and school officials. D Output: New student index cards and probation report writing cards. E New index cards go back into file for report use. F Probation report cards are sorted by school. G Probation reports are printed and sent to the deans of the schools. H Probation cards are stored. o 365 -~.---.-.-------~-~~~- ----.-.---~~---~-.--- 1 J r 00 Student Grade Card File :- 0 0 Student Index Card File Student Index Exception Card File 8 8 Comment Card File Student Name and Address Card grade card o ~~ort Ilrtm ort updated index card o Students and c. c. to concerned.school authori"'-------I n ex cards index card 1ndex card index card ew Student index ca rd file o Page 1 r 1111 , UI, ii: : til; ¥ 44M IXU ::UU'.A " robation Repor Card FUe o 'Schoolof Education· n School of Liberal Ar '. o Business 1620 Computer 0: L Page 2 3·6 7 O:~ II',!' CARD. JUSTIFICATION o o I iUI.am;;;: ii; Ii.: .. 36.~. ; $lIIUi :.mad. II I I I Explanation of Card File After completion of Step A (shown by the flow chart), the input to the computer consists of multiple card groups, one per student. It is the purpose of this report to show the need for including each type of card. o The speed and efficiency of any data processing procedure are largely dependant upon the volume of data to be processed. Because card volume is of such importance, it is to the users advantage to keep it at a minimwn. The five types of input data cards necessary for the student academic progress reports follow. They are: Card 1 - Student Clas s Grade Card These cards enter the flow through the registration line. They are the yellow striped cards the student submits for each clas s he attempts. After registration, the cards are held until the instructors turn in their grades. Each grade, with the respective grade points, is entered on the correct grade card. The result of this activity leaves a workable file of all work completed on punched cards. Card 2 - Student Index Card A continuous file of student index cards is maintained by the Computer Center. This file records the complete scholastic history and present status of each student. It is this card that records the amount of college work completed, with the grade average earned for this work - on a cumulative basis, and also on a single semester basis. Such information as where the student lives, what social organization does he belong to, the number of credit hours transferred in from· other colleges, his fir~t two major areas of study, his minor area of study, and the sex, is all recorded in a numerical code on this card. It is the student index card that facilitates all reports on academic progressjfrom a report showing the current and cumulative index of each girl living on the third floor of Reeve Hall, to a report of the numbe rs of hours all Education majors carried any given semester. Because this card is so vital to our work, and because the grade report contains both cwnulative and current credit hours, grade points· earned, and grade point ratio, it is necessary for this card to be re-computc::d at each semester's end. o 36!J Card 3 - - Student Requ~red Index Exception Card The purpose of this card file is to automate the detailed processing of studentt; having scholastic problems. Because the student's academic progress is of upmost concern to the college, careful monitoring and guidance techniques are essential. The inclusion of this card greatly facilitates much of the detailed analysis work necessary. The student index card allows the proper school authorities to care-' fully supervise the progress of a 'student. By submitting a probation form to the Computer Center, a school official can stipulate exactly what scholastic level of achievement must be met. This is done by simply stating what grade point ratio the student must earn, either on a cumulative, or semester basis. This information is then entered into the student's card group and allows the compute r to analyze the student's work accordingly. If the student fails to m.eet this requirement, the computer will generate a card from which a complete scholastic report can be written and sent to the appropriate official. The card is labeled "exception" because, in the absence of such a card, the computer will use the. standard required index schedule to analyze the student. See Probation Scaling. o Card 4 -- Student Name and Address Card The name and address card allows for automatic addressing of the grade report. Carld 5 - - Comment Card The comment card allows a school official a maximum of two lines (68 characters) of comment on the student grade report. Through the use of a comment code, the comment may be printed only if the student fails to meet specified grade conditions. It is also possible' to print the comment under any conditions. See Comment Printing. o 37H • I II I 'I dSMm:a:us::ti .: ,a: .UU.l;II.1i Output Data Cards o Two types of cards are generated by the computer. The first, an updated student index card, replaces the input student index card (see card 2 - student index card). The second, a probation report card (card 6), is punched for every student falling below certain ininimum grade average requIrements. The purpose of this 'card is to allow a file of cards to be maintained on all probation students. A more complete description of this will be found under Probation Report. Comment Printing The grade processing procedure utilizes two kinds of comments. The first type, those entered on the comment card, may be worded as the school official desires. However, because the probation report sent to the school authorities should explain what kind of comment was made, and who authorized the comment, the inclusion of an authority code, and of a comment classification code is necessary. The classification of comments is as follows (without respect to wording): Code 1 2 Description Place on Probation Contact the Dean of your school Contact the Re gis trar Withdrawn from school for scholastic reasons Withhold Permit to Register 3 4 5 6 7 8 9 Machine generated probation comment. The authority codes are as follows: Description Registrar Dean of Students Business Office Dean of the School of Ed. Dean of the School of Lib. Arts Code 1 2 3 4 5 o 371 -------.--~- -- --_ ...._ . _ - - - - - - - ~--~-------------~------~-~----- - ----~- - - ---- .. ---..... -~ ..----- 6 7 8 Computer Center 9 It should be noted .that the presence of a comment does not necessarily mean a student is on probation. A special punch over the comment code will suppress printing if the required index is met or exceeded. Thus, a code 1 and this special punch would read J, a Z would read K, 3 - L, etc.. The second form of comment, machine-generated, are those printed on the grade report to notify the student he is being placed on probation, or that he is being removed from probation status. These comments will be indentifie d by a code 9. o 0, ",...? tl • - - -=MiAZi&£ilMltiitiU i ilL i.# . 4 .$* $ =¥i4Aii i_A I Probation Scaling o As each student i~dex card enters the computer during grade processing, the probation code found in card column 47 will be examined. A zero in this c~lumn indicates the student is not on probation this semester, any other digit means that the student was on probation last semester, i. e., a 3 would indicate a probation student for the 3 preceding semesters. If the student again fails to earn a satisfactory grade .index, the probation code will be incremented by 1. If the student earns a satisfactory average, the probation code will be made zero. As each student'~ grades are proces:sed and the new cumulative hours and grade points are brought up to date, the computer will, in the absence of a required index exception card, scale the cumulative hours and find the required grade -point average. The scale is as follows: Cumulative Hours o - 16 17 - 32 33- 45 Required Grade -Point Average 1.00 1.25 1.50 1.BO 46 - 60 61 + 2.00 Any student who does not meet or exceed this scale will be automatically placed on probation. The probation code for such students will be incremented and a "Probation Report Card" (card 6) will be generated. The ~omputer will notify such students of this condition 1;>y - in the absence of a comment card - printing on the grade report. See comment printing. Special Provisions If a "Required Index Exception Card" (card 3) is present, their grade -point. average will be scaled as specified by this card. c 373 I )'+1:')'1:* Hritiod .m e o CARD FORMAT o o I alliliOU SII;:, . U1Z,AMAM 1$·· -I i. • 1.$ $ a ------.---.~.~- ....... ,•..--"---'--'-'---.:.. • •. _. "'-'E'-'~~'- - t"" j), ~ Card 1 CJ Student Class Grade Card Card Column Description Student Number School Name Clas s ification Curriculum Number in Class Department Course Number Section Number Course Description Course Number Time Clas s Meets Days Class Meets Semeste r Code Grade Grade Points Hours of Credit Code 1 1-5 6 7-27 28 29 30-32 33-36 37-40 41-42 43-57 58-61 62-65 66'-70 71-72 73-74 75-77 79 80 0 o iI 375 - -------- -- - - - - - - - - - - - - - - - - ITC" Card 2 Student Index Card Card Column 1-5 6 7-27 28-31 32-35 36-38 39-40 41-43 44-46 47 48 49 50-51 58-61 62-63 64-65 66 67-69 70 71 72-73 74-75 76-77 78 80 Description Student Number School Name Cumulative Hours Cumulative Points Cumulative Grade Point Ratio Semester Hours Seme ste r Points Semeste r Grade Point Ratio Probation Code Grade Point Ratio Hours Housing Total Hours Toward Graduation Hall House Social Organization Semester Code New or Transfer Students Teaching or Non-Teaching Minor Second Major First Major Sex Code 2 o 376 1 I _iilUS JliJ! II . : t itA: \" Z;;d HI .. N$#$i i'~",I '! Card, 3 o Student Required Index Exception Card Card Column 1-5 6 7-27 28 29~31 32-34 35-37 80 Description Student Number School Name Authority Required Cumulative Index to be Earned Required Serne,ster Index to be Earned, Semester Code Code 4 o o 377 PO Card Format o Card 4 Student Name and Address Card Description Student Numbe r School Name Street Address City and State Code 1 Card Column 1-5 6 7-27 28-51 52-72 80 o 37~ 71d1 I MtttJilitl#lU $$1.: t. AP ; ;; :ati¥ . t4i1PQ Q.L Card 5 o Comment Card Card Column 1-5 6 7 8 9-11 12-45 46-79 80 Description Student Number School Authority Comment Code Semester Code First Line of Comment Second Line of Comment Code 5 o o t "' tttttt Card Coding For Schools o Card Column 6 Of All Cards School Education Liberal Arts Business NUl"sing Other Code I 2 3 4 5 Card Coding For Classification Card Column 49 Student Index Card Card Column 75 of Probation Report Card 0" Code I 2 3 4 5 Clas s ification Freshman Sophomore Junior Senior Other Cumulative Number of Hours o - 27 28 - 56 57 - 85 86 - 124 o 38U aaz:saaiitJiza: $ ;, : :':61 Z 4; au M g;;1t ,II' l t ,~ o FORMS o ~! I 1, o 38 1 ~, 00.- Probation and Comment Request Form ,~ To: Computer Center / 1964 Date: Please C1 Place on Probation CJ Withhold the Permit to Register CJ Make the Following Comment only, Mr. Mrs. Miss Student Name ----------------- a Student Number If a CJ cumulative current grade-point ratio __ e _ _ __ Do Not Fill This Out For A Comment Only is not earned. Please make the following comment o 1-- --- f on the Grade Report Comment Code CJ if the student is plaeed on probation, regardless of the scholastic achievement. First Line of Comment 34 Characte rs only - Include Blank Spaces Second Line of Comment Classification of Comment CJ AUTHORITY CODE Signature of Authority o o o ~-.~----=- o o o ~ Probation Report 00 C'~ After .grade processing the probation report cards will be used to generate a report of all students currently on probation. The format of this report is as follows: CD g ~ ~ ~ ~ .a 88 §. .~ Student Name Total Hrs Required Toward Index '. <1Hours~nts,GPR-,Ho,Yll__,E911lts,_G.ElLr.-O.ts.Ji.1..,.--rCumlC.ur, .~ :d Q) ~ Student Number . .. i U .~ ~ 0 .., " ~ N Q) j ~ 2u 3] Po4 til g~ 0 ~ INDIANA STATE COLLEGE COMPUTER CENTER ,00 ~ 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 99 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 1 2 3 4 5 6 7 • 9 10 11 12 13 14 15 16 11 18 19 20 21 22 2324 25 26 27 28 29 30 31 3233 34 35 36 37 38 3940 41 42.43 44 45 48 47 48 49 50 51 52 53 SrVOENT ~ U NvriBEff STUOENT NAME : f f ,N 1 a Will'll." ·~ t, • • Nvl'l"~ CLASJ I ~ j,. 55 5& 57 58 59 10 11 &2 63 64 65 16 61 sa It 70 11 S 'COII/.16 DEl'r. ~ . •c. C04lIlI£ CDwrsr OESCIf''''''N f NII"".IA I ,..0 9~999999999999999999a9 « M TI/~l G G ,. R J ~• I DAi$ I I II _9999 S CLASS ME¥T.5 tOll6 n 7374 1578 77 181• • ~ D • p 0 J ~ .s I HII 0 III 9 9 999 9999 9 9 9 9 9 9 99.9 9 9 9 9 99 999 999 9 9 9 9 9 919 9 99999 99 99 999 9 _9 1 234 5 6 1 • • 10 11 12 13 14 15 16 11 11 ,.20 21 22 2324 25 26 27 2821 ~ 3132 33 34 35 36 3738 39 40 41 42 143 44 45 48 47 48 48 50 51 52 53 54 55 56 57 ~511011 ~13~1S ~'7 .. lt70 1172 7374 7.,. n 11 ~. Grade 9999'9999999999999999999999999999999999999999999999999999999999999999999999999999 1 2 3 4 5 • 1 • • 10 11 12 13 14 15 1& 17 11 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 48 47 48 48 50 51 52 53 54 55 51 51 51 SIlO 11 U 13 14 IS II .7 .... 70 71 72 73 74 7S ,. 77 71 71 • 5rUDEN1' 5TVJ)£NT NAME 5TftE£T t CITY ADDRESS NUMDEff i STATE t. 99999 9999999999999999999999 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9' 9 999999999999999999989999 99999 12145 • 1 • • 10 11 12 13 14 15 16 17 11 11 20 21 22 23 24 25 26 27 .2I3O~3233~~36n38.~~42~44a.U484850~ ------- L-______________ ~ _________ ~5354U5lP.HIOI1UI3I465I1U ~ • • 70nn73U7S 78777171_ -- ----------------~-- Address 99999999999999999999999999999999999999999999999999999999999999999999199999999999 1 2 3 4 5 • 7 • • 10 11 12 13 14 15 11 17 11 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 3& 31 31 39 40 41 42 43 44 45 • 47 48 49 50 51 52 53 54 55 51 57 51 59 10 11 &2 63 14 IS II 81 .... 70 71 SrUD£NT Nvl'l'." f ~ t STUDENT NAME b CIIMIII-ArlII4 ~ : Ho.·' '.IN" RA". } I '( I I I CI/lflfElYr r,rA" H«5. /\'HrJ II.". H.... Iflftll"~" IN~'IC CUM evlt. l' I I I Il , f IIr l~ M M 5 C if. ~ t ~ ~!tllt R 1 "" n 73 74 7S 71 77 71.71 • £ I It 1 I G. • c 99999 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9:9 9 9 919 919 999 999 9 9 9 9 9 9:9 9:9 9 9:9 9 9 999 9 9 9 9 9 99 9 9 9 999 9 9 9 9 9 I I I i 1 2 3 4 5 • 7 • 9 1011 12 13 14'15 16 17 18 192021 22 23 24 25 26 2728 29 30 31:32 33 34 35!36 37~38 • 40 41 4243 44 45 48 47 48 41 50151 52:53 54 55156 575159 80 61 62 63 &4 65 16 87 II It 70 71 7273 74 ~5 78 n 7t 711. Probation Report o o o ~~---- -- - ~-~ ----- ---.........------- ----,....,- o o o :L~ i: INDIANA STATE COLLEGE COMPUTER CENTER 999 999 9 9 9 9 9 9 9 9 9 9 9 9 9 9 ~ 9 9 9 9 9 9 9 9 9 999 999 9 9 999 9 9 9 9 9 9 9 9 9 9 9'9 9 9 9 9 9 9 9 a 9 9 g 9 9 9 9 9~9 9 999 9 9 9 9:9 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 11 18 1920 21 22 23 24 25 26 2728 29 30 31 3233 34 35 36 37 38 39 5 STUDENT ~ : N"M8ER STUDENT CtlMtlI-A1IV£ . NAME HOVp..s I L POINT3 . I 999999 9 ~ 9 t9 9 9 999 99 9 9 999 9 9 9 9 9 9 919 99 9:9 ~ 41 42 43 44 454& 47 48 49 so 51 52 53 CU/ViENT S()IfT RATIO HItS. fOlNrs RATiD I I ~ I COP~.5 t~· fr 'U 55 56 57 58 59 60 61 62 63 64 65 66 6768 69 70 71 72.73747576 n 7. 7. 80 TOTAL ~ H H A 0 ~ f l }/OtJiU L ~ ~ .1 L E JEjVf. ~r I M v:l~~ ~ I. I COM i = It /S£ ;'NII M I ~ ~ X : .' : ~I 9~ 9 999 919 919 ~ ~ ~'/9 9999999 9 9~ 9 999;99999 g 999999991 l' 1 1 I ~ ME: 1 2 3 4 5 6 7 8 9 1011 1213 14 15 16 1118 1920 21 22 23 24 25 26 27 28 29 30131 32 33 34135 36 37 38 39 40 41 42 43 44145 4& 47 48 49 so 51 525354 55 56 5758 59 601&1 62 63 64 65 66 67 .. 69 70 71 7273747578 n I c 5 1 I I 78 10 Index 999999 9 99 9 9 9 99 999 999 9 9 9 9 9 9 9 99 9 9 9 9 9 9 9 9 9 999999999999999999999999999999999999999999 1 2 3 4 5 & 7 • 9 10 11 12 13 14 15 16 11 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 4& 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 &7 &8 • 10 71 72 73 .74 75 78 n 7878 10 STUDENT f A ItSlllllllt> --IIID.)c 1 ~ CVM~ It t/i.Ar,vL IN' JE/1. II STUDENT 0 &. NAME c,*- INPUT C",,/! rXCEPTION II CARD ~ I 3: ~ y 99999 9 9 9 9 9 9 9 9 9 9 9 9 99 9 9 9 9 9 9 9 9 999 9 999 999 9 999 9 9 9 9 9 9 9 999 9 9 9 9 9 999 999 9 9 9 9 999 9 9 9 9 9 9 9 9 9 9 9 1 12345 6 1 • 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 21 ~ 29 3G w. 323334 353631 3839~~~4344454&~48495O~~53545556~585960~U6364e66V.a1071nnUn78nn7t -_ .. -- - --------~------~---- 80' ~I Exception 999999999999 9 9 9 9 9 9 9 9 99999999999999999 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9.9 9 99999 9 9 9 9 999999999999:99 1 2 3 4 5 & 7 • 9 10 11 12 13 14 15 1& 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 4& 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 678. a 70 71 72 73 74751. . SrI/DENT ~ ~T ~ SEIrt, ,, (ODE #I 8~ 0 '- A J. n7. 78 10 I~ SECOND LINE OF COMMENT FIRST LINE OF COMMENT •" c0 41 ~ y 5 99999 9 9 9 9 9 9 9999999999999999~999999999~9999999 999 999 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 99 9 9 9 9 9 9 99 9 12345 6 7 8 9 1011 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 2B 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 4& 4748 49 50 51 5253 54 55 56 5758 59 60 61626364 65 66 67 &8. 70 717273741576 n 7~.78 ~ --- - - -- -- - ----------- Comment ~=---- - o COST STUDY PROGRAM ABSTRACTS Programs 1, 2) and 3 o 38u nU.:WiSe,,::;: ! : i: ;; $, : . .: MU#ti U $\ om suaa:¥ . .lI ;;ap;;tR4; ·~----------~""""'--~"""""",."=,-,.-.,, ....... . (1.) PROGRAM 1 Description: This program extends the enrolhnent and class hour figures in the 2 cards. The input, the 2 cards sorted by Depa;rtment Number, is read from the 1622 card reader. The computer multiplies the. enrollment figure times ~"1e clas s hour figure. The entire 2 card is reproduced for output - containing the extended amount in card columns 26-31. o When a change in Department Number occurs, the machine will genera. te a 3 card. The refore, there should be one 3 card for eve ry department entered. The 3 card will contain: The Year, Dept. No.·, Total Student Class Figure, and the Total FTE Staff Figure. Both totals are from the 2 cards. Error Conditions: As the file goes through the I'n.achinc, a sequence check for equal group:> is performed .. If a 2 card is out of order an error message will be typed or. the typewriter ("error in Dept. sequence. "). B~cause the comput'er' recognizes an error condition on an "Not Equal and Not High" compare break, the department to which the error card belongs may have already passed completely through. In such a case it will be ne'cessary to: 1. Adjust the 3 card for that department. . 2. Rerun the department ~with the card included. Irrespective of. which method is used, the run must be started from the firs t depa. rtm ent for which the re is no 3 card. . Switch Settings: I/O - Stop Parity - Stop Overflow - Stop Console Switches: Sw. 1 Not Sw. 2 Not Sw.3 See ·Sw. 4 Not Used Used Ope'rating, Sugges tions Used r, \ o / -------~---~---.-- ---------- ~---.---------~-------...-------~---- . - - - - - . " - - - - .. -~--- .. -.. ~--- I : (2) 0 ,'\ PROGRAM 2 ,i Des cription: This program is used to pro-rate the number two cards. After Program 1 is complete, the Department Total cards (3 cards) have two total expense amounts keypunched in them. These cards are then placed back in the file of 2 cards in front of their respective department files. This program (Program 2) then accepts that file as data. The program a'ccepts the data in sequential order (by Dept. No. - 3 card first, followe.d by the two cards) and pro -rate s the two amounts from each 3 card into each two card. The two amounts to be allocated are found in card columns 19-24 and 25-30 of each 3 card. The first amount is pro-rated by weighting the student class hours figure on each 2 card against the total student class hour amount on its respective 3 card. This is then applied to the first total (stored in the machine) for each 2 card 1 s share. The second amount is prorated on a similar basis, i. e., FTE. figures are weighted. ~he remainders are carried into the following dividend, allowing the last card to zero-balance the anlount to be allocated and the amount allocated. C 1 :1 Error: Conditions: The program tests for six error conditions during processing. If an error condition is detected, the computer will type "Error Nil and halt. The six error conditions follow: Error 1 2 3 4 5 6 Sequence - 3 card Sequence - 2 card Amount allocated on student hours doesn It ze ro balance Amount allocated on FTE doesnlt zero balance Student hours from 2 cards do J'lot equal tot~l from 3 'card .FTE percentages from 2 cards do not equal total from 3 card On any of the above conditions, except I an.d 2, it will be necessary to start from the last departxnent not com pIe tel y proces sed. o 38~ ¥P j 4 alL. , AU? AiP .. I (3 ) PROGRAM 2 o Switch Settings: I/O - Stop Parity - Stop Overflow - Stop Console Switches: S ...v.l Not Used Sw.2 Not Used Sw.3 Must be same setting as in Program 1 Sw.4 Not Used I I 1 o 38~ 1 I ( 4) PROG'RAM 3 Des cl"iption: This program allocates the Pl"O -rated totals on the two cards to a 7 card for each course. The output fron1. Program 2 is sorted into course within departlnent order (card columns 5-10) with the 7 cards then being mel"ged in. If the data is in correct order, there will be one seven card in i1 0nt of every group of 2 cards. 4 The program accepts its data through the card reader. The cornpu-;;(:r will read all the data cards for an entire department, then punch a new 7 card with the sum of the three expense totals from the two cards (FTE Expense + Salary Expense + Student C:ass Hours Expense) allocated to the six different grade levels (Freshmen, Sophomore, Junior, Senior, Special, and Graduate). The total enrollment of each class is weighted against the remainder from each division is carried to the following dividend to zerobalance the last card of each group. The 7 card generated, then, has the complete cost of the course allocated to each class of students. o Error Conditions: The program is extremely lilnited as to the number of error conditions it can check. The 7 cards are presumed to have been crossfooted, and the 2 cards have been through the machine checks of Program l' and 2. Three error ~onditions might arise. They are: 1. An urunatched two card 2. A sequence error 3. The 7 card allocation did not ze ro balance - This would happen if a 2 card was missing. On any error condition, the typewrite will describe the error and halt. It is then necessary to start over, from the last course 7 card not punc;hed. Switch Settings: I/O - Stop Parity - Stop Ove rflow - Stop r Console Switch Settings: Not Used o _au a ; "JUt;;;: iik , . 4 $. Ii ( am Zdd( ii: I ~1 (5) I I OPERATING SUGGESTIONS o 1. Before running any program, clear the machine. This is done by: pressing "Reset", then "Insert" on the computer console. Type in 260000800009, then pres"S the R-S key. Wait about 1/8 of a second and depress "Instant. Stop". Then press "Reset". 2. After clearing the machine, place one of the object decks in the read hopper and press "Load" (This is the yellow button). After the machine stops reading, press the green "Reader Start" to read in the last card. The machine will be in the manual mode at this point. Load the data in the ·read ;hopper, and blank cards in the punch hoppe r and press Start to begin processing. 3. If programs one and two are used with swltch 3 off, there 'will be a duplication of output. That is, Program 1 extends the two cards with switch 3 off. This may be avoided by using the programs with s\vitch 3 on (Programs 1 and 2). This method will allow just Program 2 to generate the 2 ~ards. Program 1 \vill then punch only the 3 cards and Program 2 will extend the 2 cards as it pro-rates the:m. Irrespective of the switch setting used, it must be the same for both programs. See the following flow chart for a m.ore complete description. o o 391 \,~ Ii I"~ (6) Equipment Specifications 2.0K 162.0 Computer 1622. Card/Punch Unit Additional Feature s TNS and TNF Ins tructions >:c Indirect Addressing Not Used ):;These instructions greatly fac:ilitate numeric conversion, but if not avaiiable, they "may be siInulated with several transmit digit instructions. As an example of this substitution, Program 1, line 01090 (TNS in+lO, Now, Dept. No. 3-6) might be changed to: " C TD TD TD TD SF Now, in+lO Now - 1, in+8 Now - 2., in+6 Now - 3, in+4 Now - 3 This is·because a number such as 0036 will be_.+ead as 70707376. o 392 ••• :a"a:sualm12:: .JIM:: ::;;! 11M' .!.:. M 4 AI. : . 4\ t\ ; .. .il.4OZUi Ii O( 11M .1 ;~I 'I I o F LOW CHART FOR S,\VITCH 3 SETTINGS FOR PROGRAMS 1 AND 2 3 cards and extended 2 cards Just 3 cards 1 (3 lJ-kn \ f3 !i I I I I 6- ------0---1 I ~ Program 2 i 1620 {r-2.s.....:.-.. (? _ _ _ _ _ !~ ~ 'I (2 ! . Pro -rate d and extended 2 cards ! I 19n U Off;;;...,FPro-rated 2 cards j The output from Program 1 with Switch 3 ON This then becomes the input for Prqgram 2. ~ I is merged in the input flle. 393 0 I lo~ 11 ~ ~ ,"=> 9 9 9 9 9 9 9 999 9 9 999 9 9 9 9 9 999 9 9 9 9 999 9 9 99 9 9 99 9 9 9 9 9 9 9'9 9 9 999 9 9 9 9 9 9 9 9 9 g 9 9 9 9 9 99 9 9 9 999 9 9 !. 1 2 3 4 5 & 1 • 9 10 11 12 13 H IS 16 11 18 19 20 21 22 23 2425 2S 21 28 P£"7I No. No. CQU L5l<" l~l9 919 ,17 t-'--t~ ADD MODI FI CA TION TO CALL NEXT PROGRAM MODIFY 1/0 INSTRUCTION FOR DISK OPERATION LOAD PROG. INTO CORE IN NORMAL MANNER TRANSFER PROG. FROM CORE TO DISK CALL PROG. BACK TO CORE WHEN NEEDED CHECK OUTPUT TO BE SURE PROG. WAS NOT ALTERED BY MODIFICATION o )'#triitttsW ht .... o for call at any time it is needed. It is suggested that the output of the test run be compared with the original test data to point out any erroneous changes which have altered the logic of the program. We will now go back and examine the individual blocks of this flowchart (Slide 1) in more detail. If it were decided that our program, when finished, should call in a second phase or program we could simply replace the final halt with the two instructions needed to call the next program from the disk. In Slide 2 we have replaced the halt at 10588 with an Op Code 34 followed by an Op Code 36. These will transfer the next program from the disk to core. This will work fine if the fir st instruction of the new program is located at o 10612. This is because the instruction at 10612 will be the next instruction executed. From this, we can conclude that the Op Code 34 and 36 instructions to call the second program must be located just in front of the address at which the first instruction in the second program will be loaded. This is illustrated in Slide 2 where the Op Code 26 at location 10612 would be the first instruction executed in the second program. To further illustrate this point, we will take the case of an SPS Program which has a halt at location 0 followed by a branch, in Loc 12, to the origin or beginning of the program. Here we have replaced the halt at 10588 in Slide 3 with a 49, or branch, to 19976. 19976, we have placed our Op Code 34 and 36. At location The next instruction executed after the Op Code 36 will be the instructiop. brought out from o location 0 which is normally the first instruction in an SPS Program. I ~ 17 o MODIFICA TION TO CALL NEXT PROGRAM CORE LOC. OP CODE I I P&Q ADDRESS • I I 10588 48 I I , 00000 00000 I FINAL INSTRUCTION IN UNMODIFIED PROGRAM C' I 10588 10600 34 36 10612 26 I I I 19962 00701 19962 00702 FIRST MODI FI ED INSTRUCTION I I I •I o # IJ ~-·t ".~.. (j , o MODIFICATION TO CALL NEXT SPS PROGRAM OP CODE CORE LOC. C " o --- - 10576 10588 ..... ~ LAST - 19962 19976 19988 1000 34 36 00000 00012 41 49 P&Q ADDRESS ~ I I I 49 - - I INSTRUCTION IN PROG. 19976 j, I I I I I I I I 00200 00000 19962 00701 19962 00702 0000000000 01700 _.... .._... --------~~------------------_ '." " ... ",;,- o Now that the new program is in core, this is also the first instruction of our second phase or second program. This takes advantage of the wrap around feature which allows us to execute the instruction at location 0 J after the instruction located at 19988 is executed. This also takes advantage of the fact that very seldom are instructions located in the upper end of core in the 19900 area. This allows us to use this area. Let us now examine the block labeled "Modify I/O Instructions for Disk". Slide 4 is an enlargement of this block. The first decision block asks,'!Al-ethe're 300 to 500 core locations available?" This figure, 300~500 is rather broad and varies depending on the number of read or write statements in the program which are to be modified. If we have this number of core positions available, then we go to the next decision block which asks, "Does the program read into or write from an area beginning with an odd address?" If this answer is "yes", then our modifications will require an additional 200 locations. These additional 200 locations are used to program the transfer of data so that the disk works out of or into an area beginning with an even address. If this is not done, the first or last digit of the data will be lost in a disk transfer. The transfer of data to core is made prior to a write disk or following a read disk instruction usi ng a transmit record so that record marks and special characters would also be transferred; thus not altering the data in any respect. Once we find that sufficient core storage is available, we proceed to the writing of the modification program. o I - 18 ---------~----~--~---- --~ o o o 1/0 MODIFICATIONS SLIDE # 1 YES SIDE OF WILL PROG. USE DISK FOR DATA. YES SUFFICIENT CORE LOCATIONS ARE AVAILABLE TO PROCEED WITH MODI FICATIONS. ~ ~....... USE TWO ROUTINES, ONE OVERLAYING THE OTHER. THESE 100 LOC. WILL BE FOR CONTROL OF ROUTINE IN. CORE. USE TWO OVERLAYED PROGRAMS WITH EACH, AT ITS COMPLETION, CALLING THE OTHER IN. 1/0 MODIFICATIONS YES USE FALSE B.TM· 16 ADR. OF RETURN TO PROG. 49 TO DISK I/O FOR EACH EXIT. NO USE UNCONDITIONAL BRANCH (49) TO MODIFICATIONS FOR DISK OPERATION AND A (49) BACK TO ORIGINAL PROGRAM. OK TO USE BTM TO GO TO DISK MODIFICATIONS FROM ORIGINAL PROGRAM. NO PROGRAM MUST KEEP COUNT OR MODIFY LAST CARD TO STOP READING. WRITE PROGRAM CONSIDERING THE ABOVE TO READ, WRITE DATA AND KEEP TRACK OF DATA LOC. ON DISK. NO ON SLIDE #1 . CLEAR ALL FLAGS FROM DATA PRIOR TO WRITING DATA ON DISK . ~ -" .~~ o ('j LOAD PROG. o tri.o.M'# ,'S o Ifs in the first block in Slide 4, we had found that sufficient core was not available, we would have gone out the no side to the next decision box which asks, "Are there at least 100 locations available?" These 100 locations will be necessary for a control program which will control a section of core storage calling either the I/O routine or the main program whenever either is needed. In other words, a portion of core will be shared by both the I/O program and the program which normally resides in this area. This way we can actually use a program which fills practically all of core and still have available the additional programming necessary to take care of reading and writing on the disk. o cannot be found. There are very few cases where 100 locations In many cases, an output error message may be modified or abbreviated . . The locations acquired in this manner may be used for control purposes. The locations from 0 area may sometimes be used. = 80 in the product By dumping the program out on the typewriter, there may be other areas which will become evident. In this way, sufficient area may be found to contain the instructions necessary to call the alternate program and control whichever program is in core at the present time. If we reach a situation where fewer than 100 cores are available, and if the logic of the program will allow, the best solution would be to have each of the two overlayed programs, a t its completion, call its counterpart in on top of the existing program. In this way, you are alternating back and forth and each program, when executed, will automatically call the next. o 442 I = 19 o The next decision block asks, "Are branch and transmit instructions used to go from the original program to its "read a card" or "punch a card" routine?" If the answer to this is "ye s" we will not be able to use branch and transmit type instructions to branch our disk routines. And, in the case of this, we go out the left side of this block to the next block which asks, "Is there more than one place where the original program branches to its read or write routines?" The point here is: If branch and transmit instructions had been used by the original program to go into its routine which, after modification, we will be branching from~ we may not use a branch and transmit again prior to reaching the branch back. Thus, in cases where the branch and transmit instruction is used by the original pr~gram to branch to its I/O routine, we must use a 49 'type branch to branch to disk routine. c! But, if the original program had not used a branch and transmit instruction, we maY9 in turn, use branch and transmit instructions to get to and from the disk routine. The next block says, "Does the program recognize the last card of data?" With card operation, when the last data card has been read, the card reader will stop, but the disk will continue to read sectors beyond the last data unless provisions are made to sense this last data. If the answer to this decision block is "no", then some provision must be made so that the program will not read the disk completely never knowing when it has finished the last card. This may be accomplished in two ways: o I = 20 S'ttt'=_ t= o 1. A count can be maintained of the number of cards stored on the disk when the data was originally written on the disk. This count can be checked as the data is read back for the last position. 20 Or, the position following the last data segment on the disk could be loaded with a special indicator which the next, or following, program will recognize as the last data area. In the case of AUTOSPOT AND AUTOMAP, the program already made provisions for the last card by placing a fini card at the end of the data a s it is written. o This fini card contained a 99 and was recognized by the following program as terminating the data. From here, we go to the next decision block which checks for input in alphanumeric form. Again, in this block we run into a uniqueness of the disk which in some cases would be an advantage~but which we must watch for. When we read the disk or write disk in alpha, flags are transferred with the data. cards. data. This is contrary to reading or writing on Thus, we must make provisions to remove the flags left in the The programs with which we are working assumed that there would be no flags in the input data and went on to set flags in the input area which were later used for data transfer. The extra flags left by the disk can cause serious errors if allowed to remain. To correct this, we used a clear flag instruction to clear al1 flags from the 80 positions of data prior to writing on the disk. o In this way, no flags were read back from the disk into the input area. I ~ 21 o Coming down through the flowchart (Slide 1)1I we have now reached the point which says load program into core in normal manner. The modifications which we have described up until now may be inserted into the original program deck in two ways: 1. We can modify the original deck prior to loading the program into core by repunching the necessary cards. 2. We can load the program in with no modifications and then write a "trailer program" which will load the modifications on top of the normal program. Either method is satisfactory. Now it becomes necessary to transfer the complete modified program onto the disk for recall at a later time. Again, there are a ((l\.\, ~I number of ways in which this can be done. 1. This can be done by placing the two neces sary disk instructions into the input area. After we had loaded the program with the modifications, we would branch to these instructions which in turn would load the program on the disk. When the program is read back from the disk these two instructions (34 and 38) would still be in the input area but the assumption is that the first data read into this area would be read over these instructions and they would have no affect on the program. 2. If we had used the trailer program to load in our modifications after the initial program had been already loaded into core, we would have included these two I - 22 o tt $' j"TI o instructions. Again, either way is satisfactory. The main point is to get the program on the disk with the modifications. Now all that remains is to have the program called in from the disk. Here, we may use the same philosophy which we had used when we had one program calling the following program. The main thing to remember here is that the Op Code 36 instruction must be located just in front of the first instruction to be executed in our next program. SPS programs, we used the call routine illustrated on Slide 5. an Op Code 41 followed by an Op Code 34, 16 and 49. into location O. The Op Code 41 will do nothing. In We have here The program is read We go to the 34 which will seek the disk address which we specified in the control word located at 44.. The 16 transmits immediately the 36 to location 0 and 1. We then branch back to 0 and exe cute this instruction which will now be a 36 or "read a disk". The new program will be read in and the next instruction executed, after the instruction located at location 0, will be the instruction located at 12 which, in the case of SPS program, will be a b ranch to the or igin of the pro gram. In the case of non=SPS programs where the branch is not located in position 12, we may use the program similar to the one in the second part of Slide 50 Again, this program is read into location 0: the first instruction is a 34, Ii seek the disk". The second instruction, an Op Code 26, will transfer the Op Code 36, instruction 9 located at 0046, to a location just in front of where the next program will start after it is read in. The third instruction Op Code 49 will branch to and execute o I = 23 ,-----~--~~~---...........- -= . ------" .....", ..."',.""'-.-"' ...... ""..."""" ... ---~~~ .. -~---=."-'''''''-' ........ '--'"-= ..-'='---..=--' .. -=."'=-."-"""""'--"".. ---"=------""'-', ........ - =~~ o TWO CALL PROGRAMS TO CALL AND START PROGRAM ON DISK CORE LOC. OP CODE 00000 00012 00024 00036 00044 41 34 16 49 00000 00012 00024 00032 34 26 49 00046 00058 36 , I • CONTENTS OF P&Q ADDRESS 00044 00702 00044 00701 00001 00036 00000 105000 200 00000 (DISK CONTROL WORD) rr"t IV; 00032 00701 START ·1 00057 START·12 105000 200 00000 (DISK CONTROL WORD) 00032 00702 o , .~ .... L1'1 , - - .. -----------------.---.--~~ .... ~-~---~~ hli 1 r $ t ±bi . zHziWtirt - ! 'II r II o the Op Code 36 instruction. The Op Code 36 instruction will be executed reading in the new program and the following instruction which will be the first instruction in the new program will be the next one executed. The final block on Slide 1 points out the advisability of checking the finished program by comparing its output with the output from the program prior to any modifications. In this way, we can be relatively sure we Ive not altered the main philosophy of the program in any way. I have included in the appendix a typical set of modifications for your reference. I might add one precaution modified programs get on the Monitor Disk. ~- DO NOT let any of your Probably Monitor would have to be reloaded and your program most likely would not run any way. o In concluding, let me say that I hope I have brought to your attention an area of disk operation which has receiveq. very little publicity in the past. You must realize that there are very definite limitations to the use of the disk with programs modified in this manner. The program must be in a complete core image and if programs are linked together with each calling the next, the sequence is restricted and there can be little deviation without rewriting the modifications. But, for programs which will run in the same sequence, or for a single program that is run very often, a considerable savings can result. The resulting program is fast, economical and easy to operate. I realize we have covered some rather technical material here in a rather short time. o Therefore, I invite your questions either now or this afternoon during our workshop when we hope to s it down with you I ~ 24 and help you modify your post-processors or any other programs which you have to modify. Please bring a copy of your program listing and find out the last location that your program uses in core. Anytime in the near future that I maybe of as si stance, please feel free to contact me through the Huntsville Branch Office. our workshop this afternoon. Now, are there any questions? I - - - - --_._-- _.. _ - - _.. __. __. __•..._ ... 25 ~ _ _._--_. ._ - - - - --- I hope to see many 6f·y~u in o dr. t zib • tt o II Appendix 1 OF Code Reference Table and Disc Word Explanation o o o CODE MNEMONIC TYPE OPERATION 11 AM Add Immediate 12 SM Subtract Immediate 13 MM Multiply Immediate 14 CM Compare Immediate 15 TDM Transmit Digit Immediate 16 TFM Transmit Field Immediate 17 BTM Branch and Transmit Immediate 21 A Add 22 S Subtract 23 M Multiply 24 C Compare 25 TD Transmit Digit 26 TF Transmit Field 27 BT Branch and Transmit 31 TR Transmit Record 32 SF Set Flag 33 CF Clear Flag 34 SK Seek (0 :: x~7xl) 34 K Control 35 DN Dump Numerically * 36 RN Read Numerically 37 RA Read Alphamerically * 38 WN Write Numerically 39 WA Write Alphamerically o o 451 11-1 .-~--.- •...-.. -~.-.-.-----~~ jM • un 0 r t··· r "b. .±it'st. ttibtHb#ttttt 1 CODE TYPE OPERATION MNEMONIC 41 NOP No Operation 42 BB Branch Back 43 BD B ranch Digit 44 BNF Branch No Flag 45 BNR Branch No Record Mark 46 BI Branch Indicator 47 BNI B ranch No Indicator 48 H Halt 49 B Branch 55 BNG Branch No Group Mark C) *Read- Write disk modifiers on next page. o 11-2 o DISK CONTROL FIELD In order to read from or write on the disk there are four things that must be known. (1) The se are; The disk drive number if more than one drive is attached to the system. (2) The five position disk sector address. (3) The number of sectors to be written or read. (4) The starting core location. The disk control field incorporates all four of the above items into a 14 position field. Thus: F l' F 2 ' F 3' F 4 ' F 5 S 6' S7' S 8 M 9 , M 1¢' M 11 ' M 12' M 1 3 The disk drive number is located in F~. This drive code number varies with the number of drives attached to the system. used. For drive ¢ a I is For drive 1 a 3 is used. A sector on the disk is equal to 1¢¢ positions of core storage. are 2¢, ¢¢¢ sectors on each disk. from ¢~¢¢¢ ~. 19999. These sectors are numbered sequentially The disk control field F 1 ~ F5 contains the sector address. This sector address determines where~ on the disk, the write or read will start. Next is the number of sectors to be read or written. S6 ~ S8. There This is located in The maximum number of sectors that can be read ·or written is 2~¢ and the minimum number is ¢¢l. The method for reading or writing fewer than 1¢¢ core locations is explained on the next page in "R ead~ Write Disk o Modifiers". ----.--~ ...-- (' ,.... (1\ if ~) tj rl ! - HtttrifflH o M9 ~ M 13 contains the core location of the leftmost position of the data transferred to or from the disk. This core location must be an even number. In a seek, read? or write disk instruction the "P" address is the core location of the leftmost position of the disk control field. This leftmost position must be in an even location. The "Q" address of the disk instructions contains ~7 in Q8 and Q9 and a modifier in Qll. modifier in read~write The modifier in all seek instructions is a 1. instructions is explained in "Read- Write Disk Modifiers" . o 11-4 The o READ- WRITE DISK MODIFIERS All read-write disk instructions must have a "Q" address of where M is the modifier. "x~7xM" The modifier determines whether or not a group mark (*) will have any effect on the data being transferred. The write disk instruction (38) with a modifier of ~ will be determined after the first group mark encountered in core has been transferred to the disk. If no group mark is encountered the instruction will be terminated when the sector count has been decremented to ~~~. The read disk instruction (36) with a modifier of ~ will be terminated after the first group mark encountered on the disk has been transferred into core or, if no group mark is encountered, when the sector has been decremented to ~~j1. The read or write disk instruction with a modifier of 2 will treat the group mark as data and transfer data until the sector count has been decremented to _~j1. o 11-5 o III Appendix 2 Machine Language Modifications to AUTOSPOT, AUTOMAP and Milwaukee Matic Post- Processor o Note: The post-processor included here is Users Group Library number 10.4. 004 - the "Computer Routines for the Milwaukee Matic Solid State Contro:ITed Machining Centers". o 456 AUTOMAP PHASE I Statement number 1 is a five position field for the indirect address which shows from where to start the transmis sion of the record (statement #2). This is done because there are two write statements in the main program and each writes from a different location. As each of the two locations are odd numbered, they must be moved to an even location and, since the only locations left are 11~~~~ - ~~~8~, these will be used. Statements #2 and #3 move the data from the odd numbered program output area into even numbered locations. Statement #4 writes the output data on the disk. Statements 5, 6, and 7 check indicators; address check,wrong-Iength record/ read back check and write check respectively and, if either the address check or write check indicator is on. a branch to a "seek" instruction (statement flIp) and then a branch (statement #l~) to the write instruction is made. C If the WLR/RBC indicator is on a branch to the next instruction is made simply to turn off the WLR/RBC console light. This is done due to the fact that this indicator is turned on each time a record wi th length unequal to 100 character multiples is read or written. Statement 8 adds one (1) to the sector address. Return to the main program is accomplished by a branch back (statement #9). Statement H12 is the write output data disk word. Statements 14 - 19 type the message "FIN!" to indicate the end of phase 1 and to set up the calling of the next program. 457 111-1 o mm trt±tt***" o ** j" "Ms Statements rim 2~ r _ 25 are changes to the main program. disk output area. 2~ and 22 clear the 21 and 23 branch to the write disk routine and transmit the starting address into the area reserved by statement #1. a group mark after the 8~th Statements 26 _ 31 dump Statement 25 insures position to terminate the write disk instruction. the program on the disk in a core image. Statements 32 _ 39 load the modifications, read the first program loader card and branch to continue loading the main program. o o 111-2 0 AUTOMAPPHl WRITE DISK STATEMENT NUMBER CORE LOCATION OPCODE P·ADDRESS QADDRESS 1 19802 00 0000 2 19808 31 00000 19807 ~ 19820 26 00079 15390 4 19832 38 19914 00700 5 19844 46 19894 03600 6 19856 46 19868 03700 7 19868 46 19894 00700 8 19880 11 19919 00001 9 19892 42 10 19894 34 19914 00701 11 19906 49 19832 0 12 19914 10 00000 01000 13 19926 00 14 19928 10 42002 00000 15 19940 00 46495 .6490r 16 19952 39 19943 00100 17 19964 34 19928 00701 18 19976 48 00000 00000 19 19988 36 19928 00702 - C: 0 , ~ ,; . 111-3 45H \'ff tt 0 ht tdbtitrttittrii AUTOMAP PHI LOADER AND CHANGES TO MAIN PROGRAM 1 STATEMENT NUMBER 0 o CORE LOCATION OPCODE PADDRESS Q ADDRESS 20 02504 31 00000 15982 21 02516 17 19808 16063 22 11086 31 00000 15982 23 11098 17 19808 16065 24 11182 49 19952 00000 25 16062 :I: 26 15402 34 15440 00701 27 15414 16 00004 -41000 28 15426 38 15440 00702 29 15438 48 30 15440 10 40002 00000 31 15452 00 32 00000 36 00080 00500 33 00012 36 15402 00500 34· 00024 36 19802 00500 35 00036 36 19882 00500 36 00048 36 19962 00500 37 00060 49 00080 38 00080 36 00000 39 00092 49 00000 111-4 - 00500 PHASE I SPECIFICATION SECTION OF 0 AUTOSPOT - MODIFICATIONS FOR DISK OPERATION LOAD 00000 00012 00024 00036 00048 00060 00072 PROGRAM 36 19522 36 19680 36 19840 36 19600 36 19760 36 19920 49 19626 19522 19534 19546 31 19648 05819 49 19976 05914 10500020000000 00500 00500 00500 00500 00500 00500 9 Load the modifications into co re Save data for next phase Branch to ftc all next program" Disk control for this phase FALSE BRANCH & TRANSMIT 19560 10520014904948 Disk control next phase 19574 15 19969 00009 Change 42 to a 49 19586 16 19821 [0701 Transfer "write from" address 19598 49 19822 0 Go to entry of program 19606 15 19969 00002 Change 49 to 42 19618 49 10660 0 Return to program Ci MODIFICATIONS TO PROGRAM 19526 36 00000 00500 Reset location 0 to 80 19538 26 10654 19785 Modify unit instruction to branch to unit on disk routine 19550 26 07755 19759 19562 26 09039 19771 19574 26 06368 19778 Modify end of program 19586 15 05911 0000:t Used to save jata for machining section 19698 32 07751 00000 19710 32 09035 00000 19722 34 19546 00701 Write program on disk 19734 38 19546 00702 19746 48 19748 27 19822 02753 19760 17 19822 10877 Modified instructions to be inserted in program 19772 49 19976 19779 49 19574 19786 19798 19810 19822 19834 11 11 19821 19816 25 19626 14 19816 47 19878 00001 00001 00000 19785 01200 Increase address and transmit qigit C' ,.J Has all data been moved? , oj l.! lj III-5 --------------------------------- - ----- ---_.------------_._--._----------------- i 1 wa.. *t ±it'. . . tif& )"" tit""' " ... "!"! " PHASE I SPECIFICATION SECTION OF 0 AUTOSPOT - MODIFICATIONS FOR DISK OPERATION 19846 19858 19870 19878 19890 19898 19912 19924 19936 19948 19956 19968 19976 19988 00000 16 19840 19878 16 19816 19626 49 19912 0 16 19840 19786 49 19810 0 10000000119626 38 19898 00702 47 19956 03600 34 19898 00701 49 19912 011 19903 00001 42 19606 0 34 19560 00701 36 19560 00702 Prepares this section for return pass 1st pass set-up Write data on disk Rewrite if we had an address check Increase sector address by one Return to Program C all next program o o 1II-6 · PHASE I MACHINE SECTION Same as Phase I specification with these exceptions 19522 41 00000 00000 19534 49 19976 04948 Branch to end of program 19546 10520014919746 Disk address for this phase 19560 10540020000000 Disk address for next phase 19626 19638 19650 19662 19674 19686 19698 19700 19712 19720 41 00000 26 15929 26 07595 32 15913 34 19546 38 19546 48 49 19574 49 19534 19746 31 19758 49 00500 19711 Modify write instruction 19719 Modify end of program 00000 00701 00702 Load this program on disk 00000 05819 19648 04984 0 To transmit record left by specification section of Phase 1 (~: ~ o 111-7 11 tie tdrr't o ..... =&W [ MODIFICATION TO PHASE 2 OF AUTOSPOT TO RUN FROM DISK PATCH PROGRAM 19466 26 13225 12033 19478 11 13225 00010 19490 26 13225 12055 19502 11 13225 00004 19514 49 11226 0 ·19522 34 19560 00701 19534 49 19976 06850 19546 10540020000000 19560 10535001001900 19574 bbbbbbbbbb 0 Program patch that was in the way moved to here End of this phase.;.. seek for next phase Disk control word this section Disk control word next section MODIFICATION TO ORIGINAL PROGRAM 19584 36 00000 00500 Read in last cards of modification 19596 36 19680 00500 19608 26 01762 19712 Modify "read a card" instruction 19620 26 03219 19705 Modify error routine 19632 41 00000 00000 NoOp 19644 26 11780 19720 Modify "write a card" instructioq 19656 26 12024 19727 Modify end to call next program 19668 34 19546 00701 19680 38 19546 00702 Write these modifications all on the disk 19692 48 19694 46 19466 01200 19706 49 197420 19714 49 19912 4919522 READ DISK DATA 19728 10000000119600 19742 36 19728 00702 Read a card from disk & seek if necessary then go back and read again 19754 47 19786 03600 19766 34 19728 00701 19778 49 19742 0 19786 11 19733 OOOQl Increase sector address by one 0 TRANSFER FIELD JUST 19798 25 13056 19600 19810 14 19809 19679 19822 46 19866 01200 19834 11 19809 00001 19846 11 19804 00001 19858 49 19798 0 READ TO EVEN LOCATION Transfer data to area beginning with an even address Check for last transfer Increase disc location by one 46 Repeat nI-B i~ MODIFICATIONS TO PHASE 2 of AUTOSPOT TO RUN FROM DISK o TRANSFER FIELD -JUST READ TO EVEN LOCATION 19866 16 19809 19600 Housekeep because all data is transferred now 19878 16 19804 13056 Return to program 19890 49 01768 0 - WRITE DATA ON DISK 19898 10200000113136 19912 38 19898 00702 19924 47 19956 -03600 19936 34 19898 00701 19948 49 19912 0 19956 11 19903 00001 19968 49 11786 0 19976 19988 36 49 19560 02318 Write data on disk If addree; check - seek first then write data Increase sectoraddress by one Return to program 00702 End of this phase - read in control program for next phase 00000 Load program to load in modifications into core 00000 00012 00024 00036 00048 00060 00072 36 36 36 36 36 36 49 19466 19546 19626 19760 19840 19920 19584 00500 00500 00500 00500 00500 00500 0 Load modifications into core and branch to the first modificatid{;! o 111-9 • htrttritt"itt***· .. MODIFICATIONS TO AUTOSPOT PHASE 3 FOR DISK OPERATION 0 READ DATA FROM THE DISK 18312 36 01984 00702 Read data from disk 18324 18336 18348 18356 18368 47 34 49 11 25 18380 14 18392 46 18404 n 18416 11 18428 49 18436 16 18448 16 18460 49 18356 01984 18312 01989 13411 03600 00701 0 00001 18626 If not correct cylinder seek and go back to read Add one to sector address Transfer data to data read in area which start with an odd address. Use transmit digit 80 times. 18379 18705 18436 01200 18374 00001 18379 00001 18368 0 Housekeep transfer data routine and return to program 18374 13411 18379 18626 13334 00000 CALL OVERLAYED ROUTINE "FLOAT" & CONTROL WORDS 18472 16 19964 f0292 If program branches to FSIN or FCOS set at to call FlDAT and go to 18786 0 18484 18496 18508 18520 18fi34 18546 18560 49 18786 00000 16 19964 10376 49 18786 00000 00000000000000 41 00000 01700 10560020000000 1053500100190000 Orgin of program 01700 Disk address this program Disk address of next program ROUTINE TO MODIFY ORIGINAL PHASE 3 PROGRAM BEFORE IT IS PLACED ON DISK ALSO 18626 to 18705 ARE USED-FOR A TRANSFER OF DATA AREA 18576 36 00000 00500 Read in a- card and branch to it 18588 49 18596 26 18608 26 18620 36 18632 36 18644 16 18656 16 18668 16 18680 36 18692 36 18704 34 00000 13801 13329 13298 13620 10322 10406 12882 13272 00000 18546 0 00063 00071 00500 00500 19966 02590 02570 00500 00500 00701 Modify write data instruction Modify read data instruction Read in another card Read Modification into 13620 Modification to original program to call float Read modifications into 13272 Read ori.ginal information into 06000 Write complete program disk 0 , 111-10 .', Lj- \) :- h MODIFICATIONS TO AUl'OSPOT. PHASE 3 FOR DISK OPERATION o ROUTINE TO MODIFY ORIGINAL PHASE 3 PROGRAM BEFORE IT IS PLACED ON DISK ALSO 18626 TO 18705 ARE USED FOR A TRANSFER OF DATA AREA 18716 38 18546 00702 18728 34 13280 00701 Write "read/write data on disk" routine on disk 18740 38 13280 00702 18752 48 20 - O's IF PROGRAM NEEDS FATN CALL IN FIDAT ROUTINE 18774· 16 19964 12852 18786 34 00000.00701 Seek float and go to 19946 which will read It in 18798 49 19946 0 WHITE OUTPUT DATA ON DISK FOR NEXT PROGRAM 18806 16 18841 13847 Housekeep transmit digit 18818 16 1.8836 18626 18830 25 18626 13847 Transmit digit 80 times 18842 14 18836 18705 Have we transmitted digit 80 times? 18854 46 18912 01200 18862 11 18836 00001 Increase count· or transmit digit and write data 18878 11 18841 00001 18890 49 18830 0 18902 00000000000000 18912 38 19986 00702 Write data on disk 18924 47 18956 0360() 18936 34 19986 00701 If at wrong cylinder seek and rewrite 18948 49 18912 0 18956 11 19991 00001 Increase sector count by one and return to program 1~968 49 13806 0 18980 0) INSTRUCTION FOR IDCATION 0 TO 80 00000 49 01700 000· Branch to origin of program 00010 44 18312 18966 00022 36 13280 00702 If the read/write disk routine is in core go to R/W if not call it in 00034 49 18312 0 00042 44 18806 18966 00054 36 13280 00702 00066 49 18806 otn042M3 00080 INSTRUCTION FOR LOCATION 02528 TO 02610 02528 39 02559 00100 02540 26 17043 16793 Error message - no tool card: 02552 49 02634 0 02560 55 56634 30F "NO Te" III-II -.--~~~.----------- ---_.. _-_ .. _-- - -~ - -----~-- - -~-------~------ 0 1 II'S mtirttrimr T fts#S !' .• p J'I' MODIFICATIONS TO AUTOS POT PHASE 3 FOR DISK OPERATION 02570 44 02582 49 18774 18668 18966 0 02590 44 02602 49 02610 18496 18966 17676 0 Check to see is "float" routine is in core - if not call it in INSTRUCTIONS FOR LOCATION 19946 to 00000 19946 36 13628 00602 Call in float and branch to proper location 19958 19966 49 00000 0 44: 18472 18966 Check to see that float is in core if not prepare to call it in 19978 19986 49 17708 0 10200000118626 Control word for read data from disk 00000 o o Ill-12 MODIFICATIONS TO AUTOSPOT PHASE 3 FOR DISK OPERATION 0 Load cards for phase 3 modifications 00000 00012 00024 00036 00048 00056. 34 00056 00701 38 00056 00702 36 18552 00500 36 18632 00500 49 18576'"0 10536700718312 00000 00012 00024 00036 00048 00060 00072 00080 36 36 36 36 36 36 49 18312 18392 18472 18712 18792 18872 18576 00500 Load in s Ix modification cards 00500 00500 00500 00500 00500 0 Call in next card 00000 00012 00024 00036 00048 00056 00064 00072 36 36 36 36 49 49 18930 19892 02530 01936 18596 00042 00010 00500 00500 00500 00500 0 0 0 49 Load "flmt" on to disk Load in two modification cards Call in next card Control word for float Load in 4 modification cards 0 Return to Modification program Modification for write a card Modification for read a card o 111-13 'm rtt'tr'tz .t. . - . II ! T I W _ TRAILER CONTROL PROGRAM FOR AUTOSPOT DISK CONTROL WORDS 01900 01914 01928 01942 01956 01970 xX XXXX2 0000000 Control word to call post process~r 10560020000000 Disk control word to call phase 3 10200000110000 Disk control word to punch out put 10000000000000 Disk control word for last data written by last phase 10200010010000 Disk control word for write - X .per data 10000010010000 Disk control word for read trace per data TRANSFER DATA FROM SECTOR 100000 TO 102000 SO POST PROCESSOR WILL FIND DATA 01984 01996 02008 02020 02032 02044 02056 02068 02080 o 26 34 36 34 38 11 11 24 47 01947 01970 01970 01956 01956 01975 01961 01975 01996 19991 00701 00702 Transfer data, 10,000 location at a time 00701 00702 00100 Increase sector address for transfer by 100 00100 01947 If more data still - so back & transfer again 01100 CHECK FOR CARD OUTPUT 02092 34 02104 39 02116 34 02128 48 02140 47 00000 00102 02375 00100 Type out instructions 00000 00102 00000 00000 02224·00100 Check switch PUNCH OUTPUT ON CARDS FROM DISK 02152 34 01928 00701 Read from the disk one card at a time & punch it 02164 36 01928 02176 38 10000 02188 11 01933 02200 24 01933 02212 47 02152 00702 00400 00001 01947 If not finished, get next card 01100 CALL IN NEXT PROGRAM 02224 26 01927 01913 Enter here from phase 3 02236 34 00000 00102 Enter here from phase 2 02248 39 02441 00100 Set up switches for next program 0 02260 02272 02284 02296 02304 02316 48 34 26 49 36 00 00000 01914 19999 19988 01914 00000 00701 02317 C all in next program from disk and branch to its start 0 00702 111-14 470 c TRAILER CONTROL PROGRAM FOR AUTOS POT CHECK SWITCH 3 02318 34 00000 02330 39 02645 02342 48 00000 02354 46 02236 02366 4!l 02092 PRINT AREA 02374 62 66007 . 02386 55 00465 02398 43 41594 02410 64 63215 02422 45 63006 02434 59 630ts 02446 64 57006 02458 46 56590 02470 67 63005 02482 47 20594 02494 63 00626 02506 63 ot 02540 10 00 00102 Check switch 3 for either phase 3 call or card output· 00100 00000 00300 0 10056 65900 40056 94562 26341 24563 26600 05545 75956 56245 34159 SW 10 TYPE OUT AREA: NbFORb CARDbo UT!RES ETbSTA RT/SET UPbSWb FORbNE XTbPRO G-RESE TbSTAR G TI LOAD CORE & DISK 02544 41 02220 00500 02556 36 02300 00500 Call the remainder of the program into core and load the whole program disk with a correct halt at 02628 02568 36 02380 00500 02580 36 02460 00500 02592 36 02620 00500 02604 34 02630 00701 02616 36 02630 00702 02628 48 02630 10535001001900 02644 62 66007 30056 SWb3bO 02656 55 00465 65900 NbFORb 02668 57 48007 32159 PHb31R 02680 45 62456 30062 ESETbS 02692 63 41596 301 TART LOAD 00000 00012 00024 00036 00048 ROUTINE -TO LOAD PROGRAM INTO CORE 36 01900 00500 36 01980 00500 36 02000 00500 36 02140 00500 36 02220 00500 0 Ii 7 III -15 .----~----- ....--.. -----.---------~- .. ----~------ 1 1 o TRAILER CONTROL PROGRAM FOR AUTOSPOT WAD ROUTINE. - TO WAD PROGRAM INTO CORE 00060 36 02540 00500 00072 49 02544 0 o 111-16 472 II ~II I 0 A UTOMAP PHASE II This phase is loaded on the disk in t,-vo sections. This is done because all core locations are taken and the "read in" area is defined as "DC" rather than "OS". Statements 1 - 14 load the first section and 15 - 23 load the second section. Statements 24 - 31 are changes to the main program. Statement 24 branches to the read disk routine and statement 25 adds one (1) to the read sector address upon returning to the main program. Statement 26 transmits the field of numerical blanks to location ¢~¢¢~ rather than to the output area. branches to the write disk routine. "END" typeout to a "no op". Number 27 Number 28 changes the halt after the Statements 29 - 31 change record marks to group marks to insure termination of the write disk instruction after the transfer of 8¢ characters. Statements 32 - 42 are the read disk routine. The program branches to the read instruction (#33) and if an address check or write check occurs a branch to the seek instruction (#32) is made. statement 37 checks for a "FINI" code. Upon completion of the read operation, Upon finding a "FINI", statement 39 sets up a branch to end of program routine. Statement 4¢ branches to the main program. Statements 43 - 56 are the write disk instructions. The program branches to statement 44, which gives the option of either; (1) putting the output on the disk or (2) punching it in cards. Statements 45 and 46 move the output from the odd numbered core location to an even location. III-17 Statement 47 writes on the li 7 3 o o disk. Statements 48, 49, and if necessary. 5~ check indicators and seeks (statement 4~) Statement 51 adds one (1) to the sector address and statement 52 returns to the main program. Statement 53 in the punch statement and statement 54 returns to the main program. The "End of Job" message is contained in statements 57, 58, and 59. Statements 62, 63, 64 and 65 type "End of Job" and call the next program. o o 111-18 0 AUTOMAP PH2 CORE TO DISK STATEMENT NUMBER CORE LOCATION OP CODE PADDRESS QADDRESS 1 00000 36 19640 00500 2 00012 36 19720 00500 3 00024 36 00080 00500 4 00036 36 19900 00500 5 00048 49 00080 6 00080 36 00000 7 00092 49 00000 00500 R-W DISK C' _ Il~,.: CARDS - 8 19900 16 00004 41000 9 19912 16 00009 00000 10 199~4 34 19950 00701 11 19936 38 19950 00702 12 19948 48 13 19950 10 42001 98000 14 19962 00 15 00000 36 15000 00500 16 00012 36 19800 00500 17 00024 36 19840 00500 111-19 - o llu""tmM'pNtI.i $." _*_.. 'rid... 'UU t tit, - • "j AUTOMAP PH2 CORE TO DISK STATEMENT NUMBER CORE LOCATION OP CODE P ADDRESS 00036 18 19 36 19920 49 15000 Q ADDRESS 00500 20 15000 34 15026 00701 21 15012 38 15026 00702 22 15024 48 23 15026 10 43980 02198 00 FROM R-W C\ i 0 DISK CARDS 24 01714 49 19652 00000 25 01726 11 19749 00001 26 01854 31 00000 11372 27 01866 49 19770 00000 28 07872 41 29 11290 ~ 30 11371 1= 31 04603 ..~. 32 19640 34 19744 00701 33 19652 36 19744 00700 34 1~664 46 19640 03600 35 19676 46 19688 03700 III-20 - [i 76 !. 0 U AUTOMAP PH2 CORE TO DISK 1 STATEMENT NUMBER 0 CORE LOCATION OP CODE P ADDRESS Q ADDRESS 57 19908 00 57487 20062 58 19920 63 41596 30043 59 19932 56 550;!:: 60 19938 10 44000 72000 61 19950 00 62 19952 39 19909 00100 63 19964 48 00000 00000 64 19976 34 19938 00701 65 19988 36 19938 00702 'I,. 10 1II-22 ~ ____ ~ __ ~, .~ _ _ _"""""",_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .IfIiJ,"._""""'W".'=:i'."''''~''I\; ___ ''''''''''''""''''*"_" _-=""""""= h""",'-"","",,",,'',""""","'"'"""'""......... c POST PROCESSOR PHASE I Statements 1 - 7 are changes to the main program. Statement 1 branches on indicator (equal) when the "FINI" card is read to initialize the starting sector address. Statement 2 branches to the read disk routine. to the write disk routine. rather than Phase 3. Statement 3 branches Statement 4 branches to set up the calling of Phase 2 Statement 5 branches to call the next program. Statement 6 changes the message from "Reload G. P. Output" to "Starting Pass Two." Statement 7 changes another message. The old message was, "Use Phase 2, Contouring", the new message is "Calling Ph 2, Contouring". Statements 8 - 12 load the program on the disk. If-""," ~I Statements 13 - 21 load the changes into core. Statements 22 - 31 are the read disk routines. Number 22 is a two position field to receive the transmission from the BTM entry_ The group mark on the disk is the 81st character and this program only has 8; positions defined for the read in area so that the first character beyond the read in area must be saved. Statement 24 accomplishes this. Statement 25 then reads disk. Statement 26 turns off WLB/RBC console light, and statement 27 returns the digit moved by statement 24. Statement 29 returns to the main program. Statements 34 - 44 are the write disk routine. Again in order to get all 8_ positions on the disk the 81st character must be moved ('36) anp a group mark placed in the 81st position ('37). Statement 38 writes on the disk, statement 39 111-23 - - - - - _ . _ - - - - - _.. --_ .. _ - - - - - 479 o '7 tt o • -turns off the WLR/RBC light and statement 4P replaces the digit moved. Statement 41 adds one (1) to the sector address and statement 41 returns to the main program. Statements 43 - 48 set up the program to call Phase 2 rather than Phase 3 if desired. Statements 49 - 54 call the next program. o o III~24 ...M . \;6"\;'/-'· .... · _ ..... "._.,. , .. ··:...It.;~.:.:.;w~oIil...;..;...j,..;~_·_"c • i:li'l I I POST PROCESSOR'PHl C READ-WRITE DISK STATEMENT NUMBER CORE LOCATION OP CODE PADDRESS Q ADDRESS 1 11132 46 19784 01200 2 11556 17 19696 00000 3 19512 17 19806 00000 4 19596 17 19908 00000 5 19620 17 ' 19952 00000 6 02033 STARTING PASS TWO (Alphamerically Coded) 7 01985 CALLING PH 2 (Alphamerical1y Coded) - - CORE TO DISK (;\1 8 01770 34 01808 00701 9 01782 16 00004 41000 10 01794 38 01808 00702 11 01806 48 12 01808 10 58002 00000 00 WADER 13 00000 36 19694 00500 14 00012 36 19774 00500 15 00024 36 19854 00500 1II-25 ~~~-- _ _ _ _ _ _ I_~_. __ -----------~--.-.------.. - ..----.-~ ______..____ ._~.. ____ ~ ____. _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ o , 8 (1 L! 0 POST PROCESSOR PH! READ-WRITE DISK STATEMENT NUMBER 0 CORE LOCATION OP CODE PADDRESS Q ADDRESS 16 00036 36 19934 00500 17 00048 36 01770 -00500 18 00060 36 00080 00500 19 00072 49 00080 20 00080 36 00000 21 00092 49 00000 22 19694 00 23 19696 34 19770 00701 24 19708 25 19703 01850 25 19720 36 19770 00700 26 19732 46 19744 03700 27 19744 25 01850 19703 28 19756 11 19775- 00001 29 19768 42 30 19770 10 20000 -01017 31 19782 70 32 19784 16 19775 02000 33 19796 49 11496 0 34 19804 00 0 00500 - III- 26 ,f l..! 8 ')_ 0 POST PROCESSOR PHI READ-WRITE DISK STATEMENT NUMBER CORE LOCATION OPCODE PADDRESS QADDRESS 35 19806 34 19892 00701 36 19818 25 19837 10086 37 19830 15 10086 OOOOF" 38 19842 38 19892 00700 39 19854 46 19866 03700 40 19866 25 10086 19837 41 19878 11 19897 00001 42 19890 42 43 19892 10 44 19904 06 45 19906 00 46 19908 16 19939 06000 47 19920 39 01985 00100 48 19932 42 49 19934 10 62002 00000 50 19946 00 0000 51 19952 39 01931 00100 52 19964 48 00000 00000 53 19976 34 19934 00701 54 - 00000 01100 ~~ '" - 19988 36 IIT-27 - - 19934 00702 li83 0 I . 'nrittritr ' . o *. POST PROCESSOR PHASE II Statements 1 - 16 are the read disk routine. The main program branches to the read statement (1#4), then checks indicators (statements 5, 6, and 7). If the disk address check or read check indicator is on a branch to a seek (#13) is made. Following a correct transfer from disk to core the input data is transferred to the odd input address. Statements 8 - 12 are needed for this. Upon completion of transfer of the 8_th character, statements 13 and 14 initialize the transmit digit instruction (##8). Statement 15 adds one (I) to the sector address and ##16 returns to the main program. Statements 17 - 29 are the write disk routine. Since there are only 8~ positions defined as an output area, the 8lst position must be saved (#121) in order to set a group mark (#l22) to terminate the read instruction of the next program. Following this is the write disk instruction (1123), indicator checking instructions (statements 24, 25, and 26) and a branch to a seek (1119), if necessary. After the seek the digit is transmitted to the 81st position (112_> before returning to statement 21. Upon completion of the transfer from core to disk, the 81st digit is replaced (1I27), one (I) is added to the sector address (1#28) and statement 29 returns to the main program • . Because Phase 2 is not always used and only two areas are defined on the disk for input-output, it is necessary to move the data output by Phase 2 so that input for programs to come will be properly oriented. Statements 34 - 60 do this. o 111-28 .. ~ Statements 3~ - 33 move the output exchange statements to the high end of core, out of the way of incoming data. 0 On completion of the exchange, statements 61 - 7~ set up the call of the next program. Statements 71 - 73 are changes to the main program. read routine. #71 branches to the #72 branches to the write routine and 73 branches to end of job routine. Statements 74 - 79 load the program on the disk. Statements 8~ - 94 load the changes into core. o III-29 --.----------------~~~~ : $ mtb *iriririrt mzt 0 POST PROCESSOR PH2 READ-WRITE DISK STATEMENT NUMBER CI CORE LOCATION OP CODE PADDRESS Q ADDRESS 00000 01150 34 10720 00701 10746 36 10720 00700 5 10758 46 10734 03600 6 10770 46 10782 03700 7 10782 46 10734 00600 8 10794 25 02365 15000 9 10806 11 10800 00001 10 10818 11 10805 00001 11 10830 14 10805 15080 12 10842 47 10794 01200 13 10854 16 10800 02365 14 10866 16 10805 15000 15 10878 11 10725 00001 16 10890 49 01844 0 17 10898 10 20000 01024 18 10910 ' 70 19 .10912 34 10898 00701 1 10720 10 2 10732 00 3 10734 4 - " I - 0 III-3D 48\) 0 POST PROCESSOR PH2 READ WRITE DISK STATEMENT NUMBER CORE LOCATION . OP CODE PADDRESS Q ADDRESS 20 10924 15 02550 00000 21 10936 25 10935 02550 22 10948 15 02550 OOO~ 23 10960 38 10898 00700 24 10972 46 10912 03600 25 10984 46 10996 03700 26 10996 46 10912 00700 27 11008 25 02550 10935 28 11020 11 10903 00001 29 11032 49 03264 0 30 11040 31 19618 11084 31 11052 31 19938 11404 32 11064 39 03475 00100 33 11076 49 19626 0 34 11084 (19618) 00 0 35 11087 (19621) 00 000 36 11092 (19626) 26 19625 10903 37 11104 (19638) 12 19625 02100 38 11116 (19650) 11 19620 00001 39 11128 (19662) 14 19625 00000 1II-31 ----~~---.-----~----- .- - C; C) ! 48 'l --------._- .... _-------- -------"._----------- - · '11'11 POST PROCESSOR PH2 0 READ WRITE DISK STATEMENT NUMBER 0 0 CORE LOCATION OP CODE PADDRESS Q ADDRESS 40 11140 (19674) 47 19734 01300 41 11152 (19686) 12 19625 00100 42 11164 (19698) 11 19620 00001 43 11176 (19710) 14 19625 00000 44 11188 (19722) 46 19686 01100 45 11200 34 19888 00701 46 11212 (19746) 36 19888 00702 47 11224 (19758) 46 19734 00600 48 11236 (19770) 34 19874 00701 49 11248 (19782) 38 19874 00702 50 11260 (19794) 46 19770 00700 51 11272 (19806) 11 19879 00100 52 11284 (19818) 11 19893 00100 53 11296 (19830) 12 19620 00001 54 11308 (19842) 14 19620 00000 55 11320 (19854) 47 19734 .01200 56 11332 (19866) 49 19952 0 57 11340 (19874) 10 00001 00005 58 11352 (19886) 00 59 113q4 (19888) 10 20001 00005 60 11366 (19900) 00 ( 19734) 111-32 - [i 8b 0 POST PROCESSOR PH2 READ WRITE DISK STATEMENT NUMBER CORE LOCATION OP CODE PADDRESS Q ADDRESS 61 11368 (19902) 00 59456 24563 62 11380 (19914) 00 62634 15963 63 11392 (19926) 23 00574 8730;7!- 64 11404 (19938) 10 62002 00000 65 11416 (19950) 00 66 11418 (19952) 39 19901 00100 67 11430 (19964) 48 00000 00000 68 11442 (19976) 34 19938 00701 69 11454 (19988) 36 19938 00702 70 11466 C ..1" CHANGE TO MAIN PROGRAM 71 01832 49 10758 00000 72 03252 49 10936 00400 73 03372 49 11040 00000 74 15000 34 15038 00701 75 15012 16 00004 41000 76 15024 38 15038 00702 77 15036 48 CORE TO DISK - 111-33 0 48 ~J --------~--- ---~~-- ----~- _ "Ii ±"" rit ri de&" t "bit" "".. 'ibritarictbhftp"-' 0 If! " POST PROCESSOR PH2 READ WRITE DISK STATEMENT NUMBER C CORE LOCATION OP CODE P ADDRESS Q ADDRESS 60001 20000 36 00080 00500 00012 36 00160 00500 82 00024 36 15000 00500 83 00036 36 10720 00500 84 00048 36 10800 00500 85 00060 36 10880 00500 86 00072 36 10960 00500 87 00084 36 11040 00500 88 00096 36 11120 00500 89 00108 36 11200 00500 90 00120 36 11280 00500 91 00132 36 11360 00500 92 00144 36 11440 00500 93 00156 36 00000 00500 94 00168 49 00000 78 15038 10 79 15050 00 80 00000 81 o 1II-34 490 POST PROCESSOR PHASE III Statements 1 - 18 are the read routine. The main program branches to the read instruction (#5) and transmits the starting core location into the disk word (II 3 and 4). Statements 6, 7, and 8 check indicators and branch to "seek" (#1) if necessary. Statements 9 - 14 transfer the input data to the odd input address. Statem ents 15 and 16 initialize statement 9. Statement 18 returns to the main program. Statements 19 - 32 are the output routine. The main program branches to #23, checks indicator (program switch 2) and if it is on punches a card (#31) then returns to the main program (#32). If program switch 2 is off the program will write the output on the disk and return to the main program (#3~). o Statements 33 - 35 are !'fill in zeros". Statements 38 - 4¢ are the end of job message. Statements 41 - 45 type end of job and call the next program. Statements 46 - 51 are changes to the main program. Statement 46 "branches and transmits" to the read routine and 47, 48; 49, and 51 branch and transmit to the write routine. Statement 5¢ branches to the end of job routine. Statements 52 - 57 transfer a core image to the disk. o Statements 58 - 67 load the changes into core. III-35 49 m ! • r,.,."· . r OJ POST PROCESSOR PH3 READ WRITE DISK SATEMENT NUMBER 0 o CORE LOCATION OP CORE PADDRESS Q ADDRESS 1 19524 34 19544 00701 2 19536 49 19558 0 3 19544 10 00000 01000 4 19556 00 5 19558 36 19544 00700 6 19570 46 19582 03700 7 19582 46 19524 03600 8 19594 46 19524 00600 9 19606 25 03653 00000 10 19618 14 19612 03732 11 19630 46 19674 01200 12 19642 11 19612 00001 13 19654 11 19617 00001 14 19666 49 19606 0 15 19674 16 19612 03653 16 19686 16 19617 00000 17 19698 11 19549 00001 18 19710 42 19 19712 34 19732 00701 20 19724 49 19746 o 1II-36 - 492 POST PROCESSOR PH3 0 READ WRITE DISK STATEMENT NUMBER CORE LOCATION OP CORE . 21 19732 10 22 19744 78 23 19746 24 PADDRESS Q ADDRESS 20000 02039 46 19832 00200 19758 15 04139 OOOQ.L 25 19770 38 19732 00700 26 19782 46 19794 03700 27 19794 46 19712 03600 28 19806 46 19712 00700 29 19818 11 19737 00002 30 19830 42 31 19832 39 03979 00400 32 19844 42 00000 00000 33 19856 00 00000 00000 34 19868 00 00000 00000 35 19880 00 00000 000 36 19890 10 64001 70000 37 19902 00 38 19904 59 45624 56300 39 19916 62 63415 96323 40 19928 00 43565 56501=" 41 19940 39 03085 00100 42 19952 39 19901 00100 0 III-37 --.--~~"-.~.----- - - ~----,~- .. - - - - - .. ""-~-----.--- 0 4~3 "'!!Jw!f"!II'!!lmS"'!!!!I,?"!!!i'VUSII",,' _ tr ... 1 o &d , IT' POST PROCESSOR PH3 READ WRITE DISK STATEMENT NUMBER CORE LOCATION OP CORE P ADDRESS Q ADDRESS 43 19964 48 00000 00000 44 19976 34 19890 00701 45 19988 36 19890 00702 CHANGE TO MAIN PROGRAM 0 - 46 01940 17 19558 00000 47 02536 17 19746 03978 48 11754 17 19746 03978 49 12546 17 19746 03978 50 12570 49 19940 00000 51 13894 17 19746 03978 52 03654 34 03692 00701 53 0:J666 16 00004 41000 54 03678 38 03692 00702 55 03690 48 !;() 036H2 10 {)2002 00000 57 0:.n04 00 CORE TO DISK 0 III - 3H C' POST PROCESSOR PH3 LOADER STATEMENT NUMBER CORE LOCATION OP CORE PADDRESS QADDRESS 58 00000 36 00080 00500 59 00012 36 03654 00500 60 00024 36 19524 00500 61 00036 36 19604 00500 62 00048 36 19684 00500 63 00060 36 19764 00500 64 00072 36 19844 00500 65 00084 36 19924 00500 66 00096 36 00000 00500 67 00108 49 00000 C) o 111-39 • If! o IV Appendix 3 Sample Problem o o 4U6 ---_.-......_...__ ._---------- PACE ".-.- .. -·-·-·.. ·-..-·--··--·-.. ·---LiiCff·-QlJ-rnsfliUE'YfTiIoniN'"s,-----------------4 .. ,,'i I iiis j>[j r -• 001 - - -..- (l1l2 UO) - ~ IT> , r'MilEiCl'o ~-'fij'rr' ...... - - - - - - - - - - - - - - - - - - - - . . . . . . jlltIlI1SPOT PHASr} - - . - - - - . --- I'IIIISI:? Sf.III~S hlJrPlIT Of. ._._ PtlmTliU;llltl) ..,kitivni{fi'[7C'7i0ii5i\1iO:------------...,...--------;...--~--.-I rm CIII!!) I~II"JT/()V'PU' siH i.iFt"·piHNl"OUfl)ur -5J=-t'ilAst? ~,WI 004 PU:'II S TAIH fO CONTINUE ON ERROtt S lllPS 00'. ';'i(ltliitI bo~ UNTi.H:jiffmTf)~ft'· ..fVP@····' ....-...-...----.--~---:----------------I .. --.---".-.--- ... - - - - f())(o)J----..· ...----ri:v:I(i'trr"ft:tlt·lr~fJriJlriIL;.nN~"rfli8jETfRrNi!'(=-')i1rril~Nhrilii01i()rL-=-I:.:..A:;:H:l::..E::..-_ _ _ _ _ _ _ _ _ _ _ _ _.......:-----_-......:.-~----.....-----t I 1.P.t{UlO pATtERN "'tilE txCHOE[l I.l{l{f)I{', PA IlEltN t.[NGH' E'XCEEDF.f} 00'1 oejii Oll'l ·-ERifrrA'~·"P'inF:lm'tTIitlfAfllI}.rn-tr!=fjfij-------"·----·---··-"-"-·-""----------- Ui, I - 'iiiiHli {Ii I.HI N[Xr 11(2 f RROR6 PA I fERN OPERATION r.RRO~i'" ,'tJ:t/ om- --.--.. --..-.... - ----·----....,.---·--------------oI11Hr-..--·--·---·--··----..---.--..:..-:..-------~ 01100 46 01132 00200 (.IIIW5 --r·nr·--·-··Tm'rJij~l'rT'r·::-;·l~.r.t~,:-;,:n~.,.N1JiIir-------------------*n-rr-T.Hm~-*~~----~----------~ t ::~ ;':OIH' ~ ~i.r ::: CAI((lS ~:~:~11' .P.---.--.-..P..... -" ..- ._..___._._P·.. _......· _____ ·_· __ ..·_·__ gg~1-··~. -0.1.8~.()-- ~~~O.C?__ ~~~~~~ir-L~~~:.!.!.~~.~.~-.!-'!!'.!.!.!-.-_---- '''-'--' ..... -----~--·-~H~~· jj·YHg~·~~~·~~·- . ·/ .- I l40 !tCfll{() ~~ICD irA t- l:~ i~F ~~.~~r;!~~?~{M;·~~~AJ!.D---·----. --.. ·-···,·-··--· . ·-·- ......-gH3~"H·¥~~~~·HH=--·-· I 142 (AfUJI-T'I" ,R[AD FIRST CUI) 01 /(,o---u; ffi20Ho'i6 -.\. CIIIWX· .. ~ l::~~"-"P-~~if' ~~\~~iAl~P': 1~1!~~--l~!.I!!!..!'ortiTER--.-1 1.,/ AM --'''1 j'ti'iifF.mlP 1 III,: 1 III l ~"RCIIl 1~~~ 1 40 19"Ll') :] .'!1'1H4f'T~"I:l:"l::I>+---------l....I.:.-I 1J11S6 3b_lJ9'i& 99'H19 INLOC,CAf(ui-", '1 '1' -- ..............- . INLnc, 1, IV COMj,r;r.O~tr COMPl,COMl'lI;lO . .. .. --------.--..-- ·-·oUl,c<{;;"oPfn .. olToo-·---·-----·--------- .-.----.---.---.-.-----... . ·~tHI;~,~:-·~6.!~I!'~P.!.,!.L---.....-------.--.-... ?()(I - - ·..·2'·illf) 1 07.0 AM COM!') ,."lU B ;[I(CIII . H·~·U~~· y~~;t-- . . ···-·----- !l UZ20 000-1 0[A40 [6 12152 Jzt!i4 01852 lit 12152 J2112 1&7. rFI1 CM 1311 ··fif.,'ii. ··-.. -----·--,,----· ..{U~~ 01628 ....... . ...." .•-.--- ...-- ..·.. -gl-~·~~·~~ Mgi~ M~~~·--01900 11 12152 000-2 . . . . ---- ~ 01912 It'I Oles.! 00000 01?l9 (JIIIH• • -', ....- ......- .. -----01 920"n--f2iiI6 12'025-·..··-:---------- i o2i cFINn 'fR- ·ilurpijf:..;i,'=r;iNi;iJT-I.' ..---·--·---·-..· · · I 027 2 O}" 7. U24 ijNF sr' fjufilrn:..;;,c; .- ... ·---.. ··-·-·-·..--·-·--.. --~ ..---..··orij44 .. 'j~T29n .&l6,tF)NO Ot732 44 019h8 01920 cr: CFINfl i7. 0/6 O?U fi(I~I; i ' : " f ' - ' "-'-"'" .-.-. 01956 3J 01920 00000 01')(,0 11 IlltO', -·-::O,.;;:o.;:o~o---------------01980 49 01700 00000 til tiC I NPIJT-',(., Gilrl.i 01 .,~PtJf-4('d<' .. 0, frST - ..-··T·o;;r;··----nTM· 1I1'IIT,,7 2 0';1 1 on LI rRVl(. } 0<;3 1 0';4 ~ ~~.~ fio.m. . B;~[ ·m!F NFxr . ..... - .. --.---.. --.-.... -'6fqlij"" -- ......... - ........ --.-.-.---.-.-.-----.--..- - - - lO, f[ST _.F.ll~ .(~IJ..!~ ... _. ?(', . .-----.------.----- ijoi)oo~----··----··-·-· 019118 4·6020'36··61200·14 12026 000K6 .. ·.. - .. ·-·--.. ----· ..--- .-........- ..--------_ .....- .. --.--- -.-. 02000 ... - .... -.. Hl~ STOP 07012 14 12026 000L9 02024 1f1 on12 oi200 CONI ~~·t~~: ':~~nl~H'" rq'-r:.n~.. P~!J.~~~_.__ -......-- -.-----.. - ... - -. .. -.- ... -.. g~g!~ tl gng~ ·u~n . ·. .--··-- -- .. -- .. - ..... -- ..... - ......-.... -~ c; --- -------------------------------------.-~ ----------------------------.-.-~-~., ----------------------------------------------------------------------------------------------~ ... ... PG tIN LABEL OP l. oo;t ;; ijt;ii l 1lr;1) 1 058 7. (1'38 - . 2 O'>1i MAJer. 2 05U 2 olH\ . 2 or,8 ---~'(RR 2 o~a ~ 0<;8 2 00;0 2 (11)8 2 O~,A 2 ot,acor;rr 2069 2080 2 030 "T"Tfu 7. OIJO 010 Z "~f (FINn nr FA CFINO Clf' . TNPUFltl';Tr;-nr;rrsr~rm .-.-..--...--.--... --- .. --- .... -.-. 02096 49 019211. 00000 07.103 -·lr~liJ4l7t-no'_lfOOO::.o---- rURfIW-· ..···-·---·-·-·- .. -----·-·-- . '-''''oH21iT1f 120JO-OOoJr-·-..- -..··-------·----.... ··..-·· J~I(G ..1~P'~!~2.2....~~.'--V.:~ . ...... __ ... - ..------ ... - .--- .. -.-.---- ---.-.. -- .. -.- --.--.... g~g~ !~ ~n~~ ··.:~~~~~,.;;§~6----------------- ~TVPE_;TN,..Ur.:.·4l;·; ;S,wrUC'TVJJE'''--' - 2 144 2 14(, TF rr:w 02340 44 02436 13213 2 148 2 "14!J 2 ISO 2 153 1 15', 2 l'>~. 2 1"/1 SIx 2 I SCi- - ... ------ 023n"l2 D264lJOOO(J'- ...- .....--- .. ·PFlllGr.2" - - SCNCT,PFLAGt~,6,~E5[r (NT~ C~ 1~I'UT-46"ln TF rvpr,INPur-4t,.5AVl INjiul 1,6,02,(1) -rnI\Gt:If';~1f oENt11 .. -"---T'~-"'--n or (H .. --" CIIDE trfJTUr:-~6;or.'TO' SIX rrTnUm,INJ'''T-f?, ;Shvr TOIll NUHS'fR 01 INPIJT-46,O'" [0 "i'iJ" cr pIp fnLM"ll,INI'UT-)(.;,SAV T(JIIL MOO ... .... ...-- R"' _ _ _ " _._._.-._",.- ._-. • •• - . - - _ . _ . - •••• .--..--.---.. ---~g~~~.:;..:~~:'~f~:--!-~~~~r~;-;"t~~~g;;g~~";'6'-------------""'"""---U2'011, 13 oZ4~!i4j 12W~6 000-5 o;?j,iJr'j(ilJO 2 oa8 --"'2' 08? 2 121) CKOOtl Z UO 2 140 [VEN "CHJlfJ(l "-'-" "--'-"--"'-'--'- . --_.- .... - .. -- ...-.. _-_ ... - 17.026 'oUiJon - .. ----.,,- ... -.". 02316 32 01920 00000 CKIJOD,PFlflGf.ll sr .' t- 02231 CFINO 1\1) i ... -.------..- .--.-- -g.;~gg {~ ~g:~~J~~~~ .-4 II"JF 20M .. .. _._ .. l_ ... _.......... _ _ _ {FINO SF 1. on) .. - - " - 1 - on!.? 16 oro,"'i}.:-2~1.;;.tl:;.'-----------------T-1 .. -----.-~.. --.-----.- "---"CJZ"2Jrt-4 -. f"lol6 CM' --T>lPlIT.::1thnr'HJ; TF5T' F=t1R·NF"lrwmnR'SfCT· .. ·· 02744 46 02751 BE "'Coot'·PllrCK .......... -.- .. - ........... --------------------.,,-- "-'" -".... ------'IlH--' -- -1f121j~- ~6 0'l06'8 Ol2~O 4' 01100 8NF Nf-Xr,PfLAG 02280 j3 li2~2 CI= "1-1.4(. 022.92 44 0237.8 f)f~F .1; If" INPtlT-I." -cr-' ·-mrUF4"6-·----......--- -----.. - .... - - - . -.. -.. -.-.--.-.---.--...... -...... 0'2"JOlt "n 1'202[, T~' • I 02140 46 01920 01200 ..--....---...-.-.. --....... -.---.-- . - . - -..... -.-.. -...... ---.-.... -.- .. -.--- ...-.. -.... oilfbH Obi,22"Jztl4" . F II i -----.--------...- '--' ----.. -.-,- 07.116 41 01920 01200 01164 16 OOlt4,) ,12040 2 Olll (iffo .. -. INPUt-3?, r)(fY-ll 2 080 ~ . . .... cr Il I NO OORG --4 CM'-..··HWlJf..:7i4i-,TI'f·- .. · ---- ..... --- .. - - .. "1 '0 5U----- B 2 Jj8 ... g~!g: f~ ~~·~~;~-o~lg~g~~~;~~--------------~----~-------- . -.----.-------... ·~~~~ . 1·~-~~ri~~gMg~------ - . -' .-' -----.. OZ532 46 02580 01200 - ·.... 02-.;,.·4 '21j 12134 120H g~;i~ ... -'-'- 01~80 ---ri2ljii~ :0 Ii .~~~?~)~~-~~g~~~~~~~~--------------------------------26 12177 12036 49iH'j2ri OOtJOO 1'1 't IT ( [ n OP !:,~~I~~~O.~_ ... _. _______ .________________- - - -______ ._. _______ L_Il~~_!!l~!~l'_CT IO~_. _ _ ._ _ _ _ _ _ _ _ _ _ _ __ I)fI«lJ ·-4 025'19 SF PFLAG&lO., ,rlIR SA'IE 02"17. 32 AH pIA~.d3.lo 02~96 11 2 182 11M PFLAG&4,l,lO 02732 11 1]266 000-1 2 lSb I)OI(G .-4 Z 18" pAtck SF. ann 32 11262 00000 . - ··~r---·pH1\G·i:n;·;';·pAfffJm(ilffRmlrn---------·--·--·---___onOOfill2lT'o110ri'Oino;no--..,.-------.......;.----- o 2 119 3212 00000 j2~O 600-] IF. StNtT.PT~R •• hOO«ESS OF SCAN COUNT fOR rHIS PATTERN 0270H 2~ 13255 13260 -. AM·-·--lJIAW;rrr; I"o--· ----.----- ------.--- ..--------.---:---.---.-.-.----oTIZOn-l·ffiooooq---------------- 2 180 . ··2 fiif 2· 184·---· -U---- T\TEN--·---·- ---.-----~- .. ---·-----··-------··----·------·02f41tW(5"14i;o-ooOOO 02151 PF[AG NExr."o[ RFC TYPE 03008 4b 01100 01200 INPUT-46,11,lO· 03020 14 12026 OOOJ1 ---cFI"N·jf···-·-----·---·----· ---------------------.-.- - .-.- ---------oJ03r4-60i9io-QIzoo--------------1 100 FA INPUT-J6,TXTY-11 03044 16 00469 -]019 ·3 1 0 · 0 - -.----- .. -. -.-.---.- - -.----..------ .- - .. -.-.---------------. --··----------030·5"6-16-00· i.-4!)-J2036 .... - - . .,.-----. - - - - . - - - - - . - - - - - ) 100 0)0611 49 00422 Ji1!4 J 110 FA iNpUT-26,fXTY-l 03080 16 00469 -3115 3 068 1 OH9 TRANS j Ol}ij } 110 "j ffO· OE eM. ·B·~ --_..- ~-~.". -~ ---------------- --_._-- ..- --~- . 0]092 16 0044~ J2046 .. -···---·---:-----tfjlo4 -49 '"OCf4'2-t- --.-- . - ---,- --.. ~J212~·'·-· ---~-- ----~~-.- . .- ~--.---- --.-------- o• B tFINO \30-·-----·- Tii'iRG -.-;:4·------·--------------------·-··eM INPUl-46,4'1,lU RL TRyJo l 120 03116 49 01920 00000 3 140 MIX 3 147 03124 14 12026 000.H9. 03136 47 03232 0[300 03148 14 12026 OOONI .j. 3 14~ CM INPUT-46,~1,10,rEST FOH OAIMETER 3 150 CII INI'IH-46,r,'>,10 . 3 1',8 ERRl 131M EI~SUB,7100,u,WRITE 'J ·-r ii/;-· ojf2J---------· . - - --------o-n6047>oJ27i4--0u00 _031!~ 14 _1202~~0~05 ._______.______________. ________ - 3 1 ' . ) 2 " - -ill··- tFTNo---·--------:--·-------·------·--·----- ------:--orr8 /,tt'f 0192"0 O[:WO" 3 \54 C,I Wi'UT-46,'H,lO,TF.ST FOR TOOL CARD 03196 14.12026 OOQU ~6 J lr,l) 1If' r,jt/ 000-4 12_o33_ _ _ _ .____ ._ _ _ _ _ _ _.______ _ ·1'41\p1,·17r;1O····-· .--------.- .. --.---------------..... ---,-.----------.- -Wi!J4 n T3U~ OboJO TA~pr,IN~UT-11,6,SAVE lIP ANGLE 19966 26 1322N 12055 8 E~SUI\t;3b",OUrpUT TOOL CARD 19990_~_?._!_~?2.6__ ~~_~~ __ . _ _ -.----.----'--.----..- - - - ·i)iTU;-·c·;lT(l&Tl··-·-·· .-.---- -..- - - . -.. -.---- . ..,-. - - - - - -..... --------. --·--"--'012"3"2 ~. ~~~. !R_Y ~_~._._~~. . .~~;~r.j.~-~!.~~!-!!'-~-~.~ _~_~~~ 5TAT -----.-- - - ---.---.----~ -----~ii·~_i~;~~~~--~--'3 18" H -·----TT90 tn)t{/.'; 4 040 EQU~l " thO E~~l",ILLfGAL CnDE _-I, 0]263 C.., lYPF,1,9,n:Sr FOR CSINK OR SPDRILL 03~64 14 12_13Q_.~0.-0! _ _ _. __ .___._._.-'-'-_. _____ ~ ·AE- .. ··filij·Glr-:-------------·-------·--------· ··-----·--·-;----:·----··o.n:I6-4t;-·OjTIzOifoo-. . 4 060 4010 4 012 ANGLE eM It OUr 4 066 IHM··EiGllll'7Z·00;ll;rOiJ1.. 03288 14 12130 00-09 TYPE,00'!,9 ---, -. -- ·--··--------·-----03300 41 ·d19Zo--oI2i)lr---·--·-· - ... ----.-- ..-.--..--------.-.- ~Nt-·cf-lN1i· TFM 03312· 16 13230 J2191 hO~[S,'BLE-~~6 -·-4(f111~rmTTrvr_·~l)i{rr.li\Tf;,..:_-'1r79":,-,C's7"CTAnN-r!'r-01'l":,~~~.lnO'i1'O'ITL--';;NmOr-4 080 AL -&~4 . o OUA I, ,)()O t i.lr· /VI 4 120 . - .-... .~g-:~----rr-TfiL fA,qq, .Sflvl (1INV,_Hfr!) 4 Ill) FI) TliElf1,lWll 4 HO--:··-·4 IJO " 134 4 140 4 [40 4 140 ----------------------~O~jn3rrz~3230 JlnllHsSI(\i~----·--------·--·- TOLNO,AORES,ll ANGUYC-·· -- .-. I\ll,~rS, 14, 10 ---·-it-092---··-T----SUiCIli. 4 094 onRG --4 It 0')6 ·ANt-lnt ti~r l'iIiRrs-;T(j,rfl;t"(i~pUn: Dtl'TP. 4 170 HI -AIlR[S,RCONV 'I --- ----.------.--- 03256 49 03220 00000 Tr-· HfCTli;mliV'iJ-FSI~ BUFFN,TH[T~ JZ7Z8 03336 47 03360 01300 ... -----.,.---·OH4!1n- nrlflr OP"2lr(}--- --.------.-:--.---~.------~j360 24 121]4 1]23-·----·----lf3Tfi ·460:f"404-0I200-------~·--··---- -.-------.--. 03384 II 13230 000J4 OJ396 49 03324 00000 03403 . ________-'-. __ ..__ .__ .. ---.-.--.-.-----.--.--. ------·-··-----oJr.04·rr-UHoo"OoJlr--· -~- . , ---------------g~:-}~--}~-g~~~:-1j;~~------------- --.-----------.----- A~G"L£:------ 03440 49· 01262 J3409 OJ452 26 12154 00099 03464 16 00469 -3499 -_·--_·-oJz,'11i 26~~(jI16o 121!i4- ~- .-- - - - OJ4aa 49 01422 J2144 ----------0)1500'2.6 12154 U(fU99·--03512 16 14210 -3547 ~ --~-- ---- ---------- . -.. -------.---.----- -·--~~~·a-i-. ~ -HH;- iHH--------~--·---·- -.-.- ------..4-9-.ti.- .... I',. , IN ll\t.lr.t. '''-.''- : "!.!rrN,I~"HIJ fl' It "10 I. f'OO " I , .. OJ'I1t 16 00469 -56\9 h,',J';'" .......... - ..·-...... ·..~~~~-·n-tm~g=A:;· ..............----....... -.-........ S 0(.1) flllI'.l; .-.. criNo",hllfpOf UEPft\ 6n46 '" 01920 00000 ~~... .~.~~~?- ;·r';Tif;-fHrJ!ID{"'mmlrlf 'tYRllfll fnrf nr"---' -'" . .-...... 10 OItO 'UIl ·i\li"··~mj{r-.-pT-mT.nf·-- CF ...... lf 10 ilit 'i:W 10 US 411 031116 )] &3272 00000 ... - ........ _- . - .... _--............ -- 'anzif' U··Dz·l)O sP, I~PUt-41, ·(tvOK'~.pU'v.;:r-· .... -.... - ...........- ... _......... COHr?t6,2, HI, .HUrl( fsr lfV POINTER _&20,COMpqr,fJ, li, rEsT PAT l[viL TFHl .. '8'~' 038t111 26 118,)U 12030 . ----"-0'194'1"'-'''''- 0]963 0)964 16 IlZ?} 000-[ 0)Q16 lit 12121 000-1 . ..... ·--·-jJjijiiii.. ·UtiHoooiioo·-- FOR OUTPUT ·N[X'---··· ...........- .... - .. -..........................- - -........ -.. .. .. t~!~rat!t:¥hti;!IJ)CATtu·liiiS PAffriiij~-. - .. --.. AH CO~P9r.",02, 10, aNon POINTER FORWARO fI/lfF .t:ltJ,rArop"SWlTtH r-nR PAftfRH GENUlfltJN CF PATOP . I', 050 I i'052 if· .. ·· GNO;COMP9&'i,-;i r;SiVE . ___ .•._J i • ___ .. _J OJ91t1t 16 13277 OOOJI ·..· ...... 01956"19..·01·9'16· MOO,)"'-' .... - ..... -... PLEV,l,IO :~ ~!g Pl~I~1». ~~.· 1)96' IZ89if 0)91Z 11 lllt14 000-2 o.J914 43 f))'l",,"'iii;i1rOJ936 49 03964 ooono . RrYPf,l~IO,lrST noo'::" :H:~ ~~~¥ "iJiioo'n" .......... - .............. '--- IfH PlEV.l1~10,S[r PAT lEV INDltAr"R H- .... TF.VOiC·....- -'-'-" - .... -.---........ --- ..-----. 1)1 'lUi .-'. ft 03864 11 11286 J202S . ...... ---..--. -.... - '''-''lIJiiflj-'4J' Ol'i1;;"b2f6 PArHO,I~PUT-"l."SAV[ PATTllllN NO --jiii'i(iC;irATiffi;~HICAn"lftn"pn"iNpiffAitn" Al) JFH CM n:;~ AOU or . GH.,:RUfO" I"OICnOR ',STORE IIII.H[R GEN(RU inN PAHeMN R ......fjiji((; ·.:::4.. -·--------....... ----·..- -...---- ...·-..·· 10 118 io 1 ( 00099 CfJ'69T-16"fITlioTfo16 . --;;..~---. I, I r, 0">1t uon 0)668 Z6 fN"in:~J6,.1.f;7~....Of15lTf~Hl{-·-..··-··..- - - ..·--....·-- .._mft-~~~~ ~.~~:: .. Of·O -Tii·Tt6-' to 111 10 nit .---.. . ... . ._____ ._ . . . . _._______ _ :JI·IJ)'!.~-:-~.~!.!.I}!,~.~:~. _ _ .. _.. __ ._____ ._._ ... _...... ~- ____ •__ ..... _....... 0)680 16 00469 -111'\ f:1l It'· (11M Hi . ~·~~:l ll-g~t~· iU~ 0 . { 'I ---:.--------~ OJb!j6 1t9 ot422 J2.144 - .. fT:~·..,N"l"-H),UOtl·l·l i.i~b· 10 0'1<; . '1 ij tl(;O . ' OJ5462& bUM tzll»' IIJI'I/ I -. l.ft, ~ 1411 ........._ .. ---- - •. - - - -..... -.-._... - •. - .... --- .-.---.--.- - ....... PI . s ~:g. .:: 010 .--·t~ T~}: nn: ·iUU-.. . . ·-·_·-..- j ...... -.- .... - .....- - - -...-..... - . . . . . - - - - - - . - . -.... ~;;~!; ... -.! 1~~.. .:. ;~g. . 10 (HO l(J' (In() ~- III" '{II'IIIS (lr 4 l',e) 4 1';0 It 160 It IbO ---_._._----------- '---"'--- --_ .. _... ..._-_._---_._ ...._... ....... . ....... _......... -.-.---~~~y~ -n·un: H~:g 04024 It 04036 1t4 . 0<\048 33 ........ ·-..0;;060·26 GEN~RAHtiiUv"iNoiCAfUif""'-' , ......... _J Hit'" 000-2 04012 0114" 03148 00000 .Us}'!; . i.i4iH o. .. p.~u:,-,.~ ~A~'~~._ . . Q.P....__ OP;...E:;.;R..:.;A.....;t""iO;..;:S"-_ _ _ _ _ _.. ____________ h oiJo OtlR.G --It 1", 100 llNF LOA02.PFUG&8" IS A P~T DEFltfED ON A PAl ·U'l(fl;·_ ..·· ....... B1Jf..·..nsmr;noo .. I•• 1tSf FOR PA, tEN ('UOR ._...... _--.-.. IS 120 LOIlf)Z n llti... · .. ·.......·.. ·- l'i 14() ~"1S0 15 160 Sf o\lbJ 04104 "It 04128 11210 ..-.. ···.... ··--cr~n6TrTn~uiJIi;;·6(J· ..... _-- - ... PFt.AG&8, .. SrC LEV SNITCH 'rr"'" 'JrOfe2 ,COMp·>tl" "" PlOC2,O),10 rr: 'tHO.nOel 04128 32 U270 00000 ,lnlO SECr-EN "PERU I tIN BOf'FEn--------..·--· --lfn~(r '2li'H't'q~-l In4 . -.---- .. 04152 11 121,)'; (!On-) • __ • _ _ . _ 04164 i" illi/)iJll7'l''--,------·-----04116 21 12000 1219M . I!. l1iJ ..... ~ ... -....·Slf-·-,en-~.'Sl'TTOmnr1lJ'·-mtrnn~ ..·-- ---·---·--···-lI1t·nul·'lZ'..l7.80ir 000- ... 1'i 112 1'i 1'4 -'J1r~ l~O' Itt JM' -1';-'10 ~H PlotZ,I,IO . OI'IIU' .-It . 0 RliI4t.l 04267 'CKl.1f2' ... RRF'--"'t·I'STRtl"Frl~rr.trrn·"FllR""1AO tENp. r .-.-.--- - ...................... -----~1U··~ .. O~'.Jj'1. TJ27l1: 81) .UO,CoKP9&6, tl 01t280 It) Olt30U IHIJt b4ni 49 d4j4lt dOUGb 15 IRQ OI)RG .-It 16 lqf)··· ...· ...... T!{,.. -l'tl)Cl-\, pLOCz-4 t'tRT')'JUr[tv OlE(( iiUf't=US IS 1t)0 sr PFLAGU,,,SET 3RD GEN PlT SlurCH IS 440 n 01t200 11 1279'.> 000-1 flNJt .-U,PlOC2, ll, IHOEX SCA.. POINTER 042"8 1t5 04211, 12'1t)H fi --'PlftJAP:cn-....... - .. _-..---·---.. • - ..-------..---.-- -~- .. -....... - .. '-'1R2U"U ~·O 11 DOutJo IS 1-16 i ~ Btl 140 PCNT2,PlOC2,11 . n - .. bnR'2::r;l'[litl'fn·;nvr·'tJ1'U-;WIW;Plnm... iJI"~lUf·l'J(IJfS'· .. ---lIUIZ-..n·T2'iior·llAIi . IF PUltQ,SUltF1.I;Z· 01t7.21t 2.6 12S'J0 12806 ___ ._. ________ ._ .. _. AM p[(ttz,hlb 04236 Il 11lijr,...;oc;;u""o;.:--r,------ .. IS' nJ 1" 15' A 1t';O 11'.... TClAiJlt'T'T;t'1frrUl,r-1Rn ·LEV'UFrr-·--------- -~ .... - .. -nORr; .-4 04299 • -....--- -..... - ..... ---·-lJrtmr-n"'lH~fi" IUn"-'" 04112 32 13h9 00000 .-.; "----cr4Jn' 1;9 "O'lt'ii?f tJOfitift04111 . . .-. 'r;-T61rt'J{Sllr"'nr-l:miU7''-~()Mr'ltf', ii, t Est FOR. S '"ptE PI, 041)2 4J oltli8 h4tM 15 1t'0 RINGI lOj 41ICY' . 04344 11 11414 000-3. .. fF"" 'PltAT'ttOR1"lU-"'-- .--., - --,...., - ....- .-..... -.... - .... - ... ~ .................. -... . "--···fiJ-n-n- r:zUl"111fl~"" ... . IS "90 . APTCtn'COIIIP9&6.1l AM COMP9&6,OJ,10 I n ·-;OO·SM... · . -PrtRT'1ff'T(J;~rr"[E"tTR"·1'1F·,....r·"" .. -----.-.....-............. is 505 P' IHttH ,P'tNI t.S[f PAT liMn FOR MEVERSE OPEItATIIlN 04368 21 12182 llltl" · .. ·· ... ·tJ·U8'O·12.. ·U1'IfZ'·D1JD-=a- .. 01t392 26 12181 12182 .~ ._ .... __ .___ ....__ •_ _ _.. _._ ........ ·_M ... _____ ._:..._. __ - •. _ ..____. -. .•. .•. --:--•. _- ._ ... _ .. _ - _._. .. ~ilM'HW'twtittt"WtWiN'WNW!!l,,!!T'''' ri .: __ ..... p.~_._~.~~__ ~~Il.~J_ .. ., !'~_._f.J!'_~~Nf):; _____. _ _ _ _._____________•________I:.OCN Olf INSTRUtttdfllS o 5 -. -... ;~. ~·~Z .';;1: . . ----··--m~l_l; gt1::·.:a3Ai-·---------- - ~:~~~. ~~M - '-ci;~~ 1600!2.!~.!._~t~~ROR ________._ - - - - . - - - - - - - - . - -~~ ;;~'{f ~ln: ·- . ·l;-·~l~----·--~-"-·~}~H;_!::~·~~ND GF.Jr-l'U-· 15_ 911 --4 ---------.-- - DOR\i ----r; 'HR 'FK OP[Q2,,10 048H . . 04.8(216 12802004-" . ·n ~~~.- -···U-·--Rf~A.~···--···-··------··~-·-----------:---·---~-------·-&l;·i~--~~l"~h---··--~----· ----_______.___ .___ ... .. t~ ·~~!··Rmw . ~~~~-.!?-----.--...-.-.----.---------.---...--------..l:! q~r, CM O. -'1j'l2 ~~.:;~ ~!.!~oo,!~.---- -:-.-----...- - - - - - OPt:IU 31t 0110 31tD' lifN PAT MR2 OU56 JM 86 2851 OOot.6 .- . N~~~ . -.----- ---··t~f:-·-. ~~~-C.!.t~}---.-.------· -··-----·-·--g·l::;-j}-g:;~·ho::;a .----.,. . . ~--------.--~ t~ ~~~. ·--··&t·-··-·{·~-~~J.:·:U.!~.!!.'!!.I~~L.!-Q..~.~~~--lliJ!.~. ~!!--- ---- ...- --···-·---·----:-~t;~-ll1~Fo-~~~"~--·-··-·· . --..:.-----..:...------·-·----.. 'F fF 15 'Ill -[5'""'H1 10; 'l'B '114 FGJX,fRANJ-21 FG1VdIUNJ-il. EXEtl' 0"928 26 U481 I·Z861 041J40 26 1349lUH 049'2 49 0'028 00000 ,-i; -...-- ·-·iY1iHi·.-::it-----··· ..-.-----.-.. ----------------.. t~ ~~~ ll!CJ···_·~[·-- . ····l~~~~::g,l.?J.t~--- ··-····--·--------·---·---··----------··---·-----·~~;l-l·;-~~:-:H~i·-··---·-·~··--·-------~--·-It \5 .,40 tF rf --g'-q;;r-:.1'i 942 h ·)ItJ--- - .. o 0 ------:----.----.--.----.·-oitt;!iif--·-~-·~-····--··-·-·---··---··--····-·- 1 FGlX,tR4N2-21 FGJY.IRAlf2-U EXEt) -----.-------.. -.. -.----.-.. -.. ---.- .. -- ~O(fR·G···.:;.ii·-····----·----·-·-·----:---· 04984 26 3481 1:1818 041J'I6 26J"91Uzi 0500849 05028 00000 O~015--···-·····-·-------·---- -._.. • Pl;.:.I .. !~... !-.A!!~L . !l-"_.~~,-~_~~ _ _ _ _ _ _ . _ _ _.---------------l:.Q!:.~-Q..P--L~.!RUCtIOMS-.-._._.____________ 15 'J~1t lO!;) 05016 31 13412 12191 rR FGJX-Q.ST-2G, ,SAVE 1ST PT OF PAT 80 r)(2-;O.(lPf~3.,HST FORJRD G{M INVERT 15 961J Rtll lOC 1 05088 46 05208 OUOO l~ 8 -&,,0 05112 4905152 00000 o~1l9 ' 05120 16 05163 J28~1 .. -I ~ ji;'l4" t U h t.'J'; ,)61. XrCr-CJr--,-yp1:=r.2-.TOtTITfFlJRITRI':""RTIT·----------------lrS61rn-rnno-oo=r-;zI-~--UtlF. EX2&24 0~.040 ft1 051M OUO~ 80 EXl.IRrOT-1t6,.Mlll INStR-tREtit FOR INVERT MMi It) M4b iZbi 15 %1 05064 43 Oo;i20 ilast· . .-.- -.--CllJOU·--f(-:1.ln}"·-mJDt'------·---· - . - - - - - CJ4'---'(1pr:Rz,J1J;n-;-;-rntrnI 2!'40 GEN T'fImr-- 15· 1,,1'1-'· .. p; 910- . ··----'nr-·-EXnntnlftn-----------------------·-~tiO-·l1i-·o5TU_JlI1)... z - - - - . -- ..-----. "11 ----n- 1" 12851 -Ub -. ~~~~t·l.~·l~~_I_~ __ . --- .---- --.....- .. ---------.--..--.-----.-.~~~. !~~~~r_~~---------. IIfH OfltAN"to.eOMPUfE 2NO «;f.N DIFF tRANS 05424 11.:09120000-o---------.. ----. . Cl-t ··OP(RJ,·:J4.HI;nn-Ft1R~"&c_.r·INVERS101f----:---··-·---·----·-------1);4]&-[412851 OIOH. nH lotU 051t48 46 05412 onoo 1)1 oSli6b it 01S44 Ua5l INVkI.(Jp::iB e'4 TRAN1-Jl.13.10.feST FOR 3RD GEN ROUTlON· 0~412 14 128"i1_00.0J' _____ . ____ ._. __ _ lI"Ir -11Kf4- - ------------·---·--·-~-----~-----·-:--·----·iB1oi1~·1J'j51f() ~l2l)O _ . _ _ _ _ _ . _ . _ ._ _ • __ ••• _ _ _ _ . _ . _ _ _ ••••• _ _ _ _ _ _ _ ... _ _ . . . . . . . _ _ _ _ _ _ _ _ _ • • _ _ _ _ •• _ . - .... _ ••• ______ 0 •• __• . ..- '-' - - _. -. -.-. ...._.....•.. 16 II' lOCI'.,IHlt,I.,I\.%l. r 1:"1,. H.I li ' l i H . . l"l~~ lOU.Ii liT 6 J .16 1:1 I 6 DlUiC 11 . .... INI'!JT'·'I(•• Vt,10,TESTf'nn INVERSION HIT .... lOCii, .... R~E 11.7 l~~'«li: i'r 16 I~(, 16P;~;· 16 (ri'" lOUr) --'-p;--ro;"(-- l~ l~': ---... --.-.--- .. -..-.... INVRf.INrUT-It~ ""F lot h, f 1{\Gr- CnUIl IITM COMP9t~,~fCNT SF I ' ~()'* If 'i"~. 11 'U4 17 ;'1l4 ~Uy~ 4~ 5!~0 ~~oog 1)':i816 11 00516 000-0 0~1I>'8 It 1l/,B4 -0000 0~~40 24 t1414 12181 ..----.-------. xM"iinr .. - .. ······· ...... --...- .. -.......... IH",f.:[SEf SWITCllfS ON lAST (JfAi\W··--·- .. ·-_· .. ·· .. -.. ... ~(InG ~~:'6:- 17 201 [X[C 11 202 ........ 17 20J l' 70 i • 0'31(,0 14 12032 OOOJt 0581i) 01200 MHZ 41 .. t . t,r. 05152 49 05816 (lOOOO ....... '''''6H~~ .. " ' " .... -.. . . ... g:~g~ . ~; .¥g~~ H~~~··- . ·. ··· .............. -.-.---.---- 165 16 164 .. ·or .. P.! 'fi~sij2' .1.1~ ... PIIT g~~:~ 12 49 olt112 00000 0':iH9(' lit 12129 000-2 .iXii . . - - ... ' l j i ) ' j 0 8 ,,7 "05956 'oliolf 'lI~f 01 INf>'Jr-',I>, 1/,.lO, H,ST FOR Mill INVERSION IITM CM "'IUR,INI'Uf- /,6 IN"'UT-',6d /"iOolNV[RSION TEst BII -&J7 fli"j'fifviiv or "'" 0597.0 14 1207.6 OOOL4 . ......... - .. ·-d5'ni4i.'06000 o1l60 . ... -., ........ -.-..... .... J 1 '0236 J2026 0~?5f> 14 12026 0 0 0 l 4 - - - - - - - - - - - - - - - - 05168 46 06000 01100 05944 iNVRr. i"NI;Ur:.:."l,·';· ..·-·-.. - -- ..- ......-... ... --~-~ .. --.-----:------... '05ijOO"27 '01'3't /, 12026 --.-.----.~ .. ------~----.--------------~-.--- ....... P G .. ~.! ~ LA i~F:~ . ...!l~.....g:~.E R.o\Np_S_'_____ .___ ... ___. __ .. __ .. _ .... ,._ .. _ _ ._._. _..._ .._._. __.. __ ... . . _"!..Q~ '17 205 itt 'O~620' bUbo 051164 17. 09)48 00000 oiJiil6' j ] o91io606d6 Tvpr-l,l,lO. ([SI ~.Q~.I,J.~~.MgL._. CM .... - ...... --- ....---.. - - - - - - - ..gH~g·~}·~~~~i~~-···- 1\ LOtHI .... ljnifG .::4·-·--· ............ -- ...... -- ....------ .. Ci~ INPUT-40, 11.11), rF S f Fon ROT W/(J TRANS aNt n' D03Fu(joJl- o.,m-47t05'14;i;"o-T13r23-;6r:lr---------------- H· ·{lftH;~~t'~.l~·~~· . -.... -..--.. -----...--........... TRhN •• I.f). F HlhL n~flti:;f()I~~'" T/ ON OPtJl •• ,OUT,.."t Tlils PoiNT --ht;-Ht--- 0~6RQ 21 01~44 12076 ......... --.- "05(;'i"2 O~7Q4 ~1 Q~7~O 01200 H TM Ii, 'US o~6('li·41. LUt20 t~-m lnc 1 II 16 166 0'.,M4 17 096[4 000-0 05656 14 12026 OOOl ... '(569) bIllio' " .-.-.-.' '.'" 'lNPUi';:ijo·~·i·.r;lO;·H"srf.oifiWf"wnl(lNs"·· ~r·· ~!;g~~i~~ilj¥~}ri. . '1;, t,,1 ~~~! 1-} -H&tl gA·~3g_-··..·-·...-..~- -----.--0']632 'tl) 05816 01200 .........-. ..... .U }~~ , g U R G .~~~19 .___.16 0559' ..-.- .....- ..---.-.--.. CN BT O'j516 2"1 t0096 12861 -or:'.>-r;~8~1l1l'"'j4i19riOn5r<5r32c;l":-iI'O~O~007r.oir---------------- . IllHII;;/"Lo,coArtitE JAb ct:N bIn TRflNS 16 146 0~5~'7 47 05!i'16 oUoo ...-. -(f5S(,iiij4 - (ji;"5l'o U26,,---.. ·-·-·-··---·--····-------- lOC III Bf: It; '145 LUEU' 'CM'-- . 1.1ll. I 'I . o·,i;ijlf [ii"1roT oooll . fliMB-n Ilfr~ 16 141 {I) ':>leo, -_------O.'- 0">')3' 49 0551$6 06000 ()'l~i~'l ,./, il~II/J ,. i L, Ii; [(Y,"f(sl F",'jji' 11t1)-GEN,Uirli71f 'fRANS I fie II ....... _..... _........... Uk i;;, I R i ,;r . .~ ~:( '. I~M; f :;·ij(r;l~TO ..----·-··-·.. 1f, 1:15 .t~ ___ .. _. _________ llJt II 1 3261 2"7-lOOi6- !iiU·-·----_··--------0~5~·2ii'i.OH2ti7riO;iTri6_R8T2_j,tORiO~9~'r--_ _ . -··--------·o!)~ijli i 'Hii.iJ-i,I,.j.. o·,To~·frsTTffif"3if()"-(;""£1f"T>TrrTRANs·-·-· .... --•.---.. - .. - - ·ij"5"5in-t.. -n026·OOOmr·--··-·-'·' .. -.-... -.. - - - - - - - . - - - c;~ --rr'U6 1<, 137 16 LIt; _------------------054964409520 II:;T RI1!.!!.L:.'''!S~~-1 011;(1; .-/, U t~~ ....._.•.•....•. si ell, illMH· i······· .... -n;-:l ... , R 6 12'j Oil;!'.• . "t,'fg li'jt; 14 . t ~ 16 In WI!" Ii) 118 11"11 6 ll~:.: '111 r 1 1t _-_.•_--_ R TRYRO -.- ..---- ... -.---.-.-. • rt ...Q!.. .!.~.~t.~ !!~!!~~._. ____... __..._.. " ._.._... ___ ." _. ___ . ___. 059~2 49 06092 00000 11' 20b ....- .. -- lifThc·;::t;·-------------·--··-.. ·· ..·······-·· ...........-.... --.-------.-.--. 05'I'J9···-···---··-·-··-..- - · · · - · - - · -.. · - - . - - -..- -..- - . - - . - - . 11 208 tRYRV eM INPUT-46,36,10,TEST FOR REveRSAL OuOOO 14 12026 00016 [1 209 tRvRO nNE sr 11 210 ~ Obd12 41 06012 0[200 REVRS,INPUT-46,,36 n'n-.r -..... - ... trrn·iJr>tJT;·;-;uuTrur-oorf;1ITRr-·---·-·-·-- ........ .... --. 11 212 tFM ourrut-~o,,10,REs£r GOTO OR STOP SWITCH Hz r:r c. ... ···C-m:rI"jt6;pn'frn~lJ!ID'f·rl~l- .. r>ol fIT' ----....... -...-.-... --... --. .. 17 214 RII ER36-J2 [7 ztt; t\ fltVftV&Zf,,,,GtT NEXt POiNT OORG --4 10 21b 11 21'1 T!tVilll ·C"R· .... mPUl::-4o.r.r;TO;-Tr.sT-FTIlf"1ml··gTfIfliN~·-· --- ............. -........ g H~ ~~~···l~~hf'2';f~fG~·;·;i\NG[E 11 210 BT l ' 226 LA~T1 or ··---T17l5·---- (Ji 17 .11 17 11 ZU 11 23~ Z7.9 210' l31 ·..--·1"....,-32 I1 11 ;:.> JIt ~l~ ",REtURN Tn MAIN RourlNE INPUT-40,11,lO,TEST 06200 49 ~OOOO 00000 0(.2U7 ~OT WIO TR~NS ·Tl'rrr-:-··------·.. ·-·· ..·---··-·-·-··- .-- ..-.... --- .--..-.. -.-.--.-. lASTl~17,TRIGF ~lctr;TIHSur=Jij ..-- .. -.- - ..-- ... -.. ..... "--"-' - ....-- .... -· .. ·t LII.Tl-17 OOKIJ ....:'i .. - . . 11 ?.ttl 11 Z'd' 11 244 ....·-177r,P;11 746 E4 11 2.4' 1'- 2 ,8 ' 11 249 El 11 7~0 '--Tr'7"5T CM """- FUR LINE HILL ..-.,. -... '" . ~OWJ6 .. tn.lJPER2 MlllR,OPENl I I ' ,,;;;>1\" ............. · (JIIR:; tr.M _-I. t4c;l t ',rfW\If-4(, II (4 . .1:.12.:·_···· ... .... nT INVRI,UPEM? 11 ?'ll 1\ NoWl OfH(\.i Ij:lr "1\ HI .-1, 062.11, 06288 06JO'0 O',H? 43 06356 12026 H rli!O~ 00014 46 06420 O"l...,I':':'O""O,..--_ _ _. 01,32(, 1(, 06347 J2fl02 06J,6 17 lu23h J2BU2 06J55 I 0(,3'56' : n; .~':! 06341 Jlol(,' ~L-i b6J1S 2q.:.,.3.~'il7~ /,lt4(""'!~48~21)' 20tOt.LO-4l"'" GEN INVERf u .~ 0 ".j vo '" \I " 06400 21 0754~ 12802 ()(,i~I'0641') J . U64M QOOQO" 06420 14 17U0;""OO~ O(,t,}l 41 0(,46', 01200 .. ti(,I,i,i..- 11 Cilil/ii, 000-0 IJPtM2,.i6,l/J.ff'if INvEHrltN NOWl.i Hrvif<;;.To· ... ~---- .... - - - -....- ...--~- ~~~~-~~- I I 0636B 4q 06336 00000 till 17 1"1 11 1.'>1 ..' .. __._. _-i. 17 7~1 17 2';[ 1326-1 · o ; ; H a . 41 Di,j';;. 00000' 11 751 [2 -TT7~Nm!l(,-vr O~23Z 44 06140 ·(}61.44 21 .10696 12b~2'" 06256 49 .o6l40 00000 06264 14 12129 000-2 1,7 nHn 01'-00' . Ill) . [l,fNI'UT-4" .. rE:STFfJ~ MILL INVtRSHlN Of. · ..l'1pt~2j31j;Tij;If:'Sl·FfiiCmiYi';nl r'4ILtfNVt:Rf'"" All trM HfM 06208 l4 iZOn 000J1 ilb22d 4T MfS2oflCHr .. ·· . 062(') TVPE-l.2.10,f[~r ij/.JE· /:'2 .... Unih; c~, OPFR2.J4,lo,.n~T FOR 2Nn 11 ?',l - U6lli4 17 IHH4 -(iiJOtr 0611(, 24 11414 12181 XRATf.r;j·-;nrXrflijHlr-·..···-·..--··-·---·--..- · · · - · · · ..... ····(JHII8 47 04{,2il' {JUOO" ~ 11 240 EXEC2 ~41 o61'to 21 O"f6J!i'TIitFi" C0I1P9&6,IHUH"TF.ST FOR lAST PT THIS PAT iJr' '-----n--n I_. ~j,; 1~ ::~" . . ----.--.. --.-...--.-.. ------- . g~;g~ ~i-~~~~·H~t- ._ _ . ~"'IJ {[v"IHrr 06'.112270168210095 ·1Imr--------------·t)(;5~ttH It 36 ,/ 'Ilo:fjj-IRHjRli 06536 41 06512 01200 INi l 'H-4d, 10,1 0 Not?&lZ ·fjfkhN~·~·r(r------·------·-· -.. ----- --.-----------.. ---.---..-----.--.-- .... (jr,r.4f1 14 IZ03' oooJo Ob~60'46 06680 01200 'HifI-'o ·oo"6:.:tr-·---- ol,ij-1~-lf .~~;¥*~-4~.!.~.~.d.(J.!!.~~.!..rQ~ ..I.~v.~~ __.____ . _________ .____ gt~~~·l~:U{i~~¥.~~--- ~n· _._.1T ..~~,:•. _ "-'--"'-'" fRpUl-4f,,40,Tii, rnr TnR "Nt 1;(,1 17 1161 11 ~nA; ~~~~;:T'-" .ROIP,~I.(.l?:-:! . . _. flI _ _ _ _ _ _ _ _.......:_ HT tNVIH,INPUl-4(, 06608 21 01S44 12026 IHH rKflN"lo,f"tNfll TRANS 06660 n 1\"" NTITJ-vr tNl'lIl 4o.0,\h,IEST t:hit 1i01 W/fttANS 0(,(,20 14 17032 OOOJ) ~ ~~,~ ~~~ ~tig!Ti{n;I""---' ---.- -....-.-.-.- -----.--.-.-.---.- ·g~~l~ ~~·g·tl:~Y1J~~-····H ~n ~liJrz -:~ ~~~~:~i~W-·i!.~·-----··-······- ----.-.-.---- -..-.-.. --.-.---.. ----...... -g~~~~ H~~~~-ug;}- 1 11 I"~; -n-lf11;-----.rr;;;- lJ, , ,I) 1 H P / .. "t1' I"''(,1 ~l ~~~.Pl~··~-'~·'S!.~T_,_,~_S_! ... !.I~M .l~~!_P_' __ ()!' .. _P_A_T- .. -__ .• - - - - - - - - •• - - - - - . -•. il x•.", 1'1 . . . . . . . 1 . ~7 H18 11 Hlil ritiRl;' .:·4-··-----·-·----·--··-·--· .. -----. - ..- .. - .. -.- ... --.--.------ -.... -+'l~ ~::~ 7 I\HO_".!~..f.!1 f iml ::~r /J1l~ 11 . 11 ._t.7I . lNPlIT-40,11,10,HST r'ON ROI W/O tRANS liN[ NflT n 17 ~U~: 1~~n'~;:11'" --"--"- '-' ......--- . -.--.-.-.-.----.-.-.--..-. II NO I? 0(11((;' .;:i,··· ---"'---"-"- ..... 1-411(. . ~! ..!!l1H'1 1l~2 Nul Hijr 11 2'7 tNtt l. L1 gt::g· 4 ObU~O J4 00000 00\02 I" 110 -. t~n~ . iI~jFNiJHH~ijf:lA.;i;·A·,~ I'lnCl,l'tNll"tEST FOIl I.IIST SUHPATTERN IRyr-.ii . ~~ ;-t~~g;~~:~V6!"~'~~; ~gl~~~iifOR lIfF7 .~_[t~ U Ob" ilb2 rUT Filii' ;!iju GEN PAr ... .... "-'- -- ... C nA . . -'-- -·g~~U-·;·;· y~gu·tnU----·-------·----'-..----. RCty orllHj --4 ---g"flo t ?;~H~H···U~;·~·H~*··-:--·-· - ~ORli .~~~ J U l 0; 71'; ANYMi) Of, 13(, 14 12032 OOOJl 0(,148 4 06680 Olioo o,;Hf-" - -- ...... -.- .... -.--- --Tf" nr-·-----rnr (l1J f PU f - 50, ao.t (} ~H ~!t ~:~p~~~r~-~~1::4!!.~E.~~!!IALJlE .p)i.A.Sr2_ 1" O&7Q4 ·iif~·4n(rlJl~·----24 l14t~ Ut81 ijl~H7j 06728 ... 49 06128 '0;,1j"i)' - .... --...00000 -----.-.. ----- 06192 14 12808 OpOJl 06004 41 06524 dI2Do [&[2 .-- . ~~t . . ~n~;Hl~~=zT H~::~ U~~~ 000-0 -00 06164 't9 06668 00000 -.-.- - -".- ... -. --.--..... -.-- .. --. lRII"Il-31tll,lo,Trsr FOf( ?NIl GEN ROf W/O!RANS CM "0 'J16 ()hf,"2 46 01218 01100 Of.'11& !'.!tT_. .. -"-' ..--. U· gJ}~ gt~t~· g~gil-·--···-·---·---··-·-·····--------··---··--- o PG lIN lAIIH fII' lIP[.R.~'::I.I].~ 1') 1 j4 1;;04 . af -tJ(, ---R·~ 1S H4 1', 714 15 'H 1~, 7 J/, .. ___.. ____. ___ . yl~N ... ~P .. !4_~.~~C T I ON~..:.._-.-- ________ _ ____ .. ____ .__._ ... __ .__ . __ ... 06951t 46 06990 01200 eR' . JlliJ[:';Jq;TO~lffiF1fH-·~TTi1' IlNE NOGO -... --.... --.---.------... -.. -O-(;'i"l.b·n .. -. Oh~ln 0('9QO 12 12195 000-1 SA PUICl,O\.Iu tR OU1PlIT-41,PlOC2.1l FS OU rp.~~.-:-_ ~2.!.~~;!~~ ___________ ..... FS (ICII'01-n,Ft,2Y 01001 31 129t6 1219N '-"'--"-'-'" -070H-n 121rrlJlJO"-=t------- . AM ... 'lIlO1:,; 1;10--··--··-_···_·-·-· . ··-· ..--.. --.-.---.- . -..--.. . . g;g~~ . {: ·gg:~;~·~U·---·-----· ------.----010~O pj lV, --p'··n..-. 10; "~ 49 00402 J14J! 07067 16 00469 -1091 01074 16 00445 J294i . n7olH, 4'j oC)"4-ot""Hli4lJ"-·---·---·--------·-·--·------------- IS lH I', n t• 1'; 144 1') 114 rR ST-10,r';l.x-q . 9 JI·!1l~.8191J!.--1!!~ .._._. ___ .._.. __.• ____ ._.._______ .. ____ .. TrR ·Tft:NT;·,·,·THn·tJlI, INUunm;"" A COm-OKA' srrw'- ···Y·----····'I 1 1V 0 Co t: uuuu 1 IF C(JMr<)·tll~;,.~',~r~l.;.;.IJ~c~7.,--~l"TTrTll"'-r'llT".,.7'"?l;nt.",!)"In'?nr---_--_--.~0ir7nlr2~:>:_;:7.7./,:___ii1i:14;.m11tr.:;2ri".,;9 ':-_____,--_,--_______ 1'j 7i", 1 c; 1?4 lJi1R'l~'i-4"'- 2,'0 !! . i 011]1, [6 01>206 -1154 .. 'TS ·lJ1i·-----.,.,,-n::;TI-I.';If,-t.Zu, ,""fItAn oN A cum hR stnr 1,} 1"l4 IS n rXFC ~;r .._. __ .. __ .___ ..... 27111~.1: 4'1 01j8'4~._l!.~~~~________ ...... _ ..... _.. _. ___ ~ _ _ ._____ ._ ..-.-.-.-..-.... " ,n 071~4 1 R 1 Gf .- .. -. -- --.-- -' - 07Hj;' "RAR 37 13261 00000 ~:' 'lHj1;{r,OlJlnllr~'---"--"-:' -.. "-- .----.. -- - ..-- "-"--' -. 1~4 ~r 15 734 A~ PlOC?,I,lO 1~ 'J~ 1') 114 A JRYGO-12 i-4 ---- ..-............--., .-....... -.-----.- ...... . 01202 .... 49 -... 06930 00000 . . 1)"7<'M - -.--.--------.-.----... -.--....-.. -.-.---. -"'-'-'-'-"'-" PlOC2.1,L0 OP(I{FT'l'lfin; ll,Ift[Oi\Tf 21m &£N oUHHS PUNO, SIJRF2£l 077.10 L2 12795 oou-t llj···"Hlt-----· IlNR • P,f'LOt7,lI,tNO[X Oiil{l~' 1'> 1.il. NOI;O Stl 1'> 140 TF 1'; 13a Ti~ mo 0711/1 11 121'15 000-. 1'; 1S '44 "ljO 1<' AMI 1') i'J(H) Of I" I l~ ~Olj -..- •. TJ;" II~I\ 07217 o12lhJ l PFI.ACI.II".TtIl>or 2Nl)j~n~ PAT' 1 DTRAN",Rfsrf lNn Gl~ CONTROL SWITCHES SF 1~ qO~ 1" r hilt l1\SI :,OtrAI It PN --1~, '1311 nil 1" (1);' AH 1<> 'n'> fit 1S 'Hl 'n" rr 01367 \4 01514 OfF ., rUff. l.l.l!)~ IlmFX JftI1\~PS flOTNT(R' rIlJ"'O,SlIflF H.2 1114 I'LnC3,I,tO 1\ flllli!. POII:-lrl.1?,.,Oll1l'Ur .. OI3Uf, 46 Q1466 01100 ..-- '-'-U7J1J1l 11 l]R4't'"ooO~l"-' 01410 31 12~'O 1264M "01422 01~34 "!Ti( _-I, I~(xr I'AIHRN 131~9 - - - - - - - - . . . , oITT 7:T"n"I,r·rr I:;>ll;;;; 12849 :·(J"[I!3..,I,I'lOChll,REI.OAIl 31lt) GFN 1JI'f:RAIIIlN hliFFtMS '-rr,-'rr, r---'- .TfTtr"·':".!:.-::"[~;,';,I~'L~':";Jt:":'.1r-•...,[.,.l-.·l.,,-;~";-.l)rl. ""'mr.•.-.'"'C.,.t-'::lr lj5'l "OTIU",TjfMll>"t 16 ""';1 11> II 11) TJ270 0000'0'-- 07290 32 09120 00000 04 CI' 12001 Unlr'" 0175R 45 01Zlfb 1219N ----.-... ---. '07270 1i.9 if40l"20lfOOU---···----···----· .... -.-:- ..- -.- ....-.:-.....-.--. II PDIIMI't.12' , ,'OUTI'UT THIS' PJ. T r,r n 072)4 26 12890 12a06 OI)Hf, .-4 .,rrrM 1<; 1.11222 -----------'"'Ol'"'lt,..,2;'J:4T(,"""'T"(r 1271)', 000-1 '_.... _. 011/1,'; ···It 13269 00000 ·-"·--···'··--·-·Orn'4!R083)rom096trOoO~00.O0°0 .,,, .. r O·',-- . . . 01'::;2['-'-- -.. ;...-.---.-----.- -..• - - - - - - - - - - - - ' ..... _... _. _ .." () ",2Z ofj~i4" 44 015H 13269 ii9 Ofi9lf"oofiOir-"-- . - -.......- - H 1.10 . 'J'II:t,I-,' . . 'Ol'1H -.-1"'7'100 - - - - -.... ·.. ·-:-ii-· --1ii1/jiVvIF:it't~sTI'{ojijNssOO'ijffi'itit(iilJirrl1mINiii[r----------------~.;..:.:~------------....----------11 JO~ INVIH ttl III1H'IH-46,lltlO,TE:5' FUR 1 COORO 01544 H 12917 OOOJI .. .. 1 Iii)] ~lIOt· 1111, ' 11 ll? . ... -.,. ,- -. -- ---. -----..... 0IS·56-7t;;'"ofl.Jir·oUmr-·--·--···-·--------.--· . ....... _ ..... _.. _ ...._. 01568 11 01S1t3 00012 ._., ··-·oJiiiilflt,·o16filH100··.. -· ..·.. , ...--- ... - - - . - - - - - - - II r ,. fI iMifi.l7. . . ..... -...-.-.-...---"'''---''--''-'--'-'- . (,.,: ....,.,.. FI.l~yI{P-·Y!·-·~,·.H-!..~~).. Wi,. l'fIlMX,IlIJTPIH-J6"TEsr rflR JNVlRr RICHr UK HFT . --,.r3,,-,--'··-:-rr·- ,.--.--.-.--............ -.--,- _._. III,H'ill-16".Rf!Cms TO jl(ils 1" HZ ,III .. : ... iI.J24 n'lv,i; --lo H', "'OP1X . ,m' sr " "1'~~U.'-.~f,.! •..!PtU!i..!_'.I.!.4_....N.'J.~ .. , .. ...- .....11, 1tli. t 1HO "!J«li '-10 ~---- ann .......... 016111:12 12'111 001100 ti U,'\o i,2oool)(iij(j'(jM ---·rr-~rrT~y~.~k~N~F-~r"r70"wHwY~,T.unuTTnpITIJTI--~;(L,-,-,1~[~·:~.r~i~lnlR~i~Nr.V"E~Rnf~umr~I)~M 11 hI) 11 1',: . 01"44 33 12931 00000 ',,'11 ... tm;ij6 4;(0000r; ooi)oo" [)!lI\G '-10 "jnMY --'IT ',1)". 01b';1 'iilffpi,r~i6 Sf" "14 othi';ii '32 fi I SIJIlMllllfiNf. 17 ",'..I;' ,((j'1l f1 ','11. it 01to'll III'WN'" ··--·-.. ----(~-0165lJl2'137 CF. mHrUI.-.7.~ 11 \H l1 i H. 11 \\.'1 II '.16 ii iiTo'" 11 11 izisK ooojl' o 116l 01100 12187. 000-3 07800 00000 07762 12 12162 001)-) d~f14 47. oob(W 001/00 01H'> tofJll() I)NlY ·-··-··--·-·-··------:'iIJ:-;,-7'-7'6-;--1~4...129TToo~o;;:-J.,.,1;---.,.--------------Ollaa 46 08180 012110 .. 0;;;6616 0046'f ':"781ij 11 4t4' 11 414 015?7. 44 07618 l2921 0'''04 33 lzQ27 00000 07b16 42 00000 00000 01812 26 01260 12'121 iiltiiit 49 iHi62 J34i9 ,r IlHHN,'1'J.. x.(:osnIfEJlln OI/"""-16"1l:;1~ FIf TO DIIHI.R 1216~ 00099 (10469 -1U8·~J;----------------016bO 26 01260 12931 (70)" 2(. 01048 41~ 4i~ 16 bt~ti 4~ diit.2 jj4~~' .... ----'.IL-~j. --1'1 477. 17 422 11 42'2 1742', t r'.~6· - 61no ll> 00"'''9 -195 ', 07Q1Z ~6 01260 12937 . "·I'11'11oi." 4<) urn, til u1Jf.pOf-26, fJltIIs .... - ... -...................--.-...--.----......... TF ~'M .. IcIUf'FO,'J9" Jnl~·--'--·- ~mJl~UT::Jr.;TfrsTN .. · . -...,-.-.. ... ·oN1.0·1/,· iiO"6'J ':'l)olH ..{{- ~.t;--------------------------:--------------O;;o~~~"')~~~~~;,;~:r-;~. t~,·~ . g :~= 17 4lR 11 410 F~ ...... ~'l!~.~~.!.~~_ .....____...... __._........ . , H2 "ff' ff .. ···I,.·iijlt~---·-· .. W v.C(IS~flfr:rAD&X~HNtHI[TAD d(HPUI-46,l(..Iti.Hsr fOR A SlOpe n~f H'M '.'H';TJr.rTrUr";lb';·Ul'~nmnCAlTY ·SIJM"ARGl.H 11 440 11 440 FA . ... OllT~~'.'::I.~!.~I~.r.'!-'?:~_ .... , ..............." '_'" ..... _....• : ,.. 11 4'.4 l'11 1.46 44» 11 ""0 . 17 452 '&62 B .-14 .. .:. ji),'OUl'l'UF'l> ." .-14 11 ',04 ItPJR$' SM 8~~ .-12,P1CNT,1IiIN~EX P01NrfR BACKWAROS PTCNT,26,bIO,fe5T FOR'" GOlO O~ ,,'STOP 17 '!iW) ,8I't .&l,l,OllTl'lH-',O 11 '.tIl'J 11 "Ii~ 11 Oq R . &20 11 'HO H' ., 'ill. (ill 17 'HZ (;115 f CM "'Oj~ .-'--,---- 1 t 'jD Ilf 1'1 'il'. 11 ',l', "" tfnRtV .. · · · · · · · v v ................. 08196 45 08184 1218K olil0if II 12r1l2 000':1'''''' .. , ......... . 'Otl2,l0 14 1218KO i-0T-0;:Kx-·6;r.~_ _ _ _ _ _• _ _ _ _ __ '--"---~'2JT4T1'fI:'i7i'P v nUJI""-'il ,prc;;", ll; SHIH 4 lOC!lTlONS lfH i.lrr"lI.,\;Nil·"rrSr ~iJIl r--hl~f (;r. .. vAl ()IHPII'-4o. 11,,0, '[Sf FI,n l C(JORO 1)1: ItU . .. fll/trll'-:;6,'~I-.l --..::.;:.;::::..:;.------- n £. 01 )00 "'~Il~!~ 41'~ IO')~b,l71'1~ 01~!0~~ 'lrJ\«(; *:-(,' SF NORrV·. ... - 17·-ijo'r----·~. .Wifrf:::.·[r-.•'-1(n(..,----.---------'--:--~---'---' _. H.~H.i.-- .. ___F"_c·~I.:.:'\J:..:l~·P~U:..:'_-..:l~('..:,..::S~T_-..:l...:;I 12~1 't lHHlJ1f'" 08150 47 ORIlh OIJOO "(Hill'!!' H un" -;;ftNl;1.,T6--- . . . . . . ... . Sl' -1681 ·· ..·,... "["4 1... · 12·1·...' ""0"'-'1' A'I 17 ')09 004Z~ . 081 J7 oiuA3 .RF.V(R5E ~lIlHt-I)UrrNI: ·"'CN1.I' ro·. ·· ..... ' . .. 0'1 Bl 08100 16 004"" -81J~ .. ..01;'112 16 'M44!j J2O}.,."'- 00174 49 08100 00000 11 '105 11 ')1)6 11 ~01 n····~oii-·· o/HH\l\"(l. 013111 J29~1 . "oi\lI,2 l6 iHillJ J2 u <; " " uv u .~8i]'~~~.)'''~? . 08288 . O~()O.~ ... u l.. OUl16 ]2 08689 00000 OniaH Ii 1118T..;;o:.;;o..;;o,.:-~I----~------08.100 31 12912 1218K "OIlH2 HOM"" 11515' 06.l2'. 14 12911 OilOJI . (jaB;' 46 084611 01200 _ _ _ _ _ _ _ __:__----,------..---.-----~~~:g ~-=~_J{~-.. -~-- . --.----- 1111312 4'1 0047.2. . J2.Q(l() iillj144 '1;' O('i46'1 -fl4l't 5' (lJ 3. Oil : te&*trid==* ,_. ___.. ___ . ______ .____. ____ .. ____ .-.!.Qr;~_.. Q!'.J_'~~'_'W~!.;o..:ION~S_ __ 1)1' o ------------------- OIJTPt/l-';O ,1') nr 1; 08444 II 00048 U9iz Ptc.NT ,;~ 1~: INOH rOINn~· ·-·-·-..--------··---------·lJiJ1t"JblI··nm:·ocf6lrII~- .-- -..---.--.-.-----.- .. -.--.-.-- ... -- ..... - .--.. ·------------·--------------~~~~··U·-!~~-~:irgii..-l5 O·IIH, .-to 0841n o;r. 1 57.0 . 1 ~'-I . 1 11 --rt 527 1)2] N,. -tjM~m~i::~l,··!-~t.'JF..!-/I.~nC.I\.~lfI.N.~-.RJI!~tT . . .------ s!'1 NO)'} .!: ~~2 p rcNI'[ .10 . ~~~~ . ~~:-~~~~~:" . ~Q-~~-~.-. . Ii ;.6<; NORtV . ' (,011 --1T6IT ~ t. 1 H 1 ' ~IIR~. 1, .. -. ~=.=~-._-. ---.. oUfI>UT-4(.,n,fo;rm FlJlt z·cooiHl·--·---·- ----- .. --- ---.-... -- fiE PU,N1 OUtPUT-3b, ··0668s·-U-THlro"oJr----·--.----.-.-----------08100 46 0878lt01Zpo c;t-u OiJ712 16 00469 -un . . ----..----------.. -- . -.----.---.-.--.-.--. .---.---.-.----.-----.. ---.----.- ~~nt__»-·~g~·~~-¥zVOtr-·------·--···------·------ . F.~ _._.~ll.!.~_u!.:_:?~_tH:__~ . .. ~._. ____ . ___ ._. ___ .__.__ ... _..... _._. 08148 16 00469 -8181· ------.--.-----.. - - ····off;t;o-H;ooit~-~-J29Jf--------- 00172 49 00422 JZ9 ~ 011 8 ,. 0 804 ,. 11 618 R~F 11 12182 000-1 ---.-. -...---.----.-. ------·----·--i!!~ -t~-~U~tHgg~~ -~-._.--.---.--.---.----- CM IA b 08402 1i ~iij(,. -ii-II It 11 ~MI..:..... ~f~TII.-NP.~T:-50.t!.H!.!.!.ES-!.-.f~.R .. ~OJCJ .o~ . STllP 081404 --4608884 lit U913 000-0 " " -. . - - .-.- .... --...... -....... --- - - -.. -. --·---QijR"i"6 6UOO--·----·-~------- ~.-----.,-.--- . .-.. -.-----. ---.--.-.- ---.-.---- _· __ ~:~~t-~9--~8!11U~~.~-.-:---- .._---.-------- ..... -..- -.--... - --~11. ((,.2,....~~.• ------jAt~iil:-~ti.0'j'4"ilP_:')~tit(,f;.'0Fi4iTi'Hlr0::..--_ _ _ _ _ _ _ _ _ _ _ _ _ _ _,---_ _ _ _ _--*O~8irfl~28hlH1Hl~!~It:lJ~'t-*O~OO~-r.,I_--------------- 1" o l\:tF It" 08Hlt0 08860 08689 01\"')2 It9 .. - .- ..-. ------.. --.. -... --..--... -.-....-.. - -------·oii"i"j!j·,-·· .. -..08884 -- - -....00000 -.-----------...... -.. -....... -.---- .--....-- .. -- -.- - .&20,Ntll!:.Vt.I n •.•.IRflN4 ij"nRI :." ... -.--....- .-.----.--.....- -... ---.-- -.-..--.- 11 (,75 h<:~ il f _ _-_..__.._-_._-_.- ..... -... ... .-.- ....... -.' ----- --.-.... -.-_.-- _.. -- ..---- P"_LJ~'_ .~_fl!~!! .. _~_I' _._ .._~~.~~~!:?. ___________ ._____ ... ----~.-.--- ..--.---. --.. --- .....-.---- ...-.----~-- .--:-..- ..-....: .........__ . -------_. -.- ...--. ._ ...__ ... ___.___ ..____ .____._. ________...!:-OtN~ __!!fSTRUt f 10._",.;..oS'---_______ 11 (.2f, TR 48,otITI'UI-51 08860 11 OOOltS 129U 11 ~'2 fi -- -- ·-n·---1flTffUT:.-n;-tf1f;";1fH;JTrDlrr1tTOF.iTTOWS-··-- -·--···---·-----------1fiiOff3rl·lijlb"1):xoiro4ii8~----------------,--17 !~J~ TRA'I4 AM cn~.P9&b,I,10 fl888't 11 11414 000-1 --rr61to 'Hm --12,COMP9t.6.11, INoF.)( FlASt tEN "OINTER ilea,,/) 45 OB8'J4 illilM 11 941 An tOHP9t6,1.10 08908 11 I1lt11t 000-1 11 f,)0 Tr-M·· 11 632 CF ·lmrpm:.:~(r;tTJ;j{ntr~jJrr.TlL-Sgnr;rw:"S·---··-· ----.- .. -.-----.----.. ----~Zlrll>·mtr"'OmJ:.·o- NO~(V 08932 33 ~8688 00000 11 6J4 ... - . . cr·-· N"11rrt:'V"f.r-·---·-·--···-·······-·--···--------- ····--·----··---·---·-·-·-··---·-----089lt1j-:-Jr-tm1>ii1J·"lrOmJX""O---~~~~~~----~:~;~m~Ar.G~~.~-'[~or-----------------------------________________ .______-ig~:~~~;6yl-4~2~O~0~0~0~O~0~00~0~O~______________________--------11 6Itil 1 t· ('4'j '~1\3 A~j'- lRA8./'rLAGI.l"fr:5T FOR "lRO GEN I'ATIERN 089$11 44 09098 13269 .. -. -il'lr· ···f!{I,'f;....i naNi~ ; A·iH~- ·l~ij'j-t;!: ~ -·jI·AlTtifrfS [mrn:ll·ToG-n~---· ·0 Ii 916 . 4~-09iffif·O·~"Il·------··--------·-------- -------- OU91l2 44 09002 09349 t 7b SO-·"ij ---··,IUb·-·-· ----.-.--.--.-- ...."-.- ---. -.-.- --.----- - - ..... --... -.-.-.-- -. ---·-·------·---M It 11 6~1 CM TRAN2-Jl,12,10,TFST FOR N~W 1ST PT OF THIS 1ST GEN PAT 09022 14 1280~ OOOJ2 . flUm .---...... .-... -------- . ------..---- .------ -----··------···-----------1ffi)"l7f·-n··08"5Blnrr:nW-·----·-;----.-----.------. TF ST-l1.TRIIN~-21"SAVf FIRST POINT 09046 26 12900 ~l018 1\ fRANI 09070 49 08588 00000 . 091r',·--·-·-··--------···---------·---·-·--·--.-- .. £I[ .----n-?t~!>:..;ilr------.;..rF;..----isMf~-~i:..!,~,..;.t:.;A~fJ~2.::._r[r[!.!..C~:..:..::-.:....!..::~.--:...:::..:..;.:..:..------------------~o~Cj?ioii)i-8-.:l;;6[--11r;2rc9i11~o~1~2i-82;.;8r--.......,-----..:..---'-------l ' "<;9 17 (.·~d·- . .. 11 TRill t 1 o~1 ~(,). . .... 11 ~b4 -·--rr~rRi\8 . ilTit ST-20,FGJX-(J, ,SAVE Pf-.. OF 3RD G(N PAT NANl··-·--.--..... - -FIRST ... -......... --.. -.--------.-... --........ ------.. DOKG .-4 ilNF fRANI ,PFLJlCf.U" rEsf FOR 2ND tEN PAr 09018 31 12891 134.72 ------·---···~qi'i9(r4q·lli15"88·000CRr----·-··---·'-·-·-------- .--.--.- - - ..- ..---- OQ097 0'1098 44 08588 13210 09110 49 08982 00000 .-..- ····--·------·------ll·jlltr -. -.--.... --.. -.----.. --------.----.---.-..------.----.- ..-.- 17 h66 11 661 » TRIO' nOk(; ·.·.;;2· ... -- 11 qOO 17 904 !lrkAt~ 11 (105 • OTMAN SUR~CUTINF COMPUr[S OIFF ~RANSFORMATION WHICH LINKS 1ST GEN 0 , IlNF'·1j l~ (J liH.N'i"i '-1ST"fl ReS-wITW .. -.... - ... - - .. ---.-... - .. -- -----·-·-·---·-091Zif'.f4··{f9:J1+S·-69T21f-------------------·· -.----. IIlif OUT I,Ul&1 09132 44 09290 09Jo\9 ---triiw.- o ·DilR~--i::.-.{·--·-·----·--··--~······-----··--·--------·-.-.- ..--.-.. ---- .. --...... --.-.--...-.----- ..- .. -. .. -.-- .-. -. --. ·-:t;.:;f~·~~O~I-!-t-!-l~~~----------------------------iOii;9~ln4n4njri3~oi9i34~9ii--ioiio~oi;iomor----------------- t~ ~.g~ ~~L ri_~~~~;~1-'..~3-~!O,.!.E~!.-~O'~.-~.~~-~-~!-~~~....~.~~-·-~~~·-~-~--····-----------~:H~·~~~gg-~~~--·-----··~----.----.---.---.~.-I ~ ~ g ·Ttw 1·· ~~r- ~~~~:~;. t!~~~;, n~T HIlt .jlUj.. r;arrAT~·--··· ..... ---.- .···--·-·-------g;H~~}· .~:;:}{~;~}.-__.--.--6----..,. ..-----_--:---- --1IT1~.. ~1~ "l7 91 ~ . 11 'JIZ g ;g . 11 'J12 - __;Clr~I.:~.... r:.---,~IN~P~l~J1~-~4~O~,-1~:!..O.!..,·!.I~O!-,T~f~.:S~T~F~O~R~P.:A.!..T_O~P~[~·R~A:;:T:.!I~O~N~______~_______~0r.l?~2T°i4-i14~·,lm2~Oii3jlZri°iTO~O~J0n-----_...,.--------~'WL 01216 47 0929601200 tHJl1 PT FOP. ENTIRE P A r . 0927.8 14 128"'7 OOOJZ -----------.fir· .fRA,n-Jl,12, .smn--·· ...... -...-~J ..O,NO -.- -- Nl:W - - 1ST . .-_ ..-.-.... --... _-- .. --.-----..- --.. ------------.-- ·-·1JiJ'l7.-o--itr01J2Tlr,jTJtJO"·--·-·--...------- UI ~~.-. ;·~~~:·~:~~tH .- .- .. --.-- . -.-.--.--..-·----------·-------··---···g~~~~-·~~·-H:;}-{_~:;~-·--· --.--.--.. ---.. --------.-..-.- 5 t) 4.. 0'1216 42 00000 00000 1'8 11 'Hz Il(\Rt. t-[\) 11 912 S~tlr1 n FG3X-·J,FG2X-<.J, ,RESET 1ST PT FOR PAl 11 ·.,fj·i1u·rr--· ·;.I~ ...... -.. --. -.. -- ... -----.--.- .-.--...--....-. -....------.-.---... - ... -.----- _ -.-.-~-----~---------.------.---.----- • • - - 09277 3 1 _13 "72 ~131t30 -.-----.----..- .-.. -.----.-.---..-.-.--.. - .. -. ---~~o 4·Z Guo(Jo" "OO"150li .09278 - - - _ . _ _ • • • - •••• _ ••••• - ___ A _. ___ • ___ • _ _ _ _ . . . . . . _ _ • • • • • • _ _ -~.-- •• - - . ""··· ..tllf .. ·Ht·.,.",;" ...,...' .. u ,~ "I I 111'1.,(/\1111', 1 I 9\ I, it f) j<, 11 9\/, --Tr'rr~-'-- .-.- ... 11 'llR it 'jj'i 11 H i f~i')(, 'ilAiL';-lli ,id'SI I T~r Pi" ir·!,: ··-~.·!..I·;'fl!.' !,·.·.~r{·~I;~·.·~·:-,~-l·---·--------···-·--·-·--· .. _.__ ... _ ........ '" • ;jn 11 'I;") 1 j ij?9 26 13449 \2828 26 12900 12818 OH?8 2" 17.910 12828 O'JJ04 . ·--···-····--·rFi H6 0'J31,i) -I, • "'JI' (.1 Ihll;,ii ,r)\. Bt IlMF. :',wffcil '\~H' lit r~; 0'1291 lRhl Jjlll:., fOJ~ Nf,W Nun iiX(;,'X~lr;Sr:.::20;·;~AVr-·Tsr·pf IST-I'nINf fiF'-NtxTB-f O'n60 33 09J4a 00000 lZ806 000J2 09364 46 09580 01100 i~nl ,il\1 .. "(j"939·61Il345I '12891 09408 16 00469 -9443 '-"09420' H,"oo4i15 Jliii;o ...y;-,rr7-r;'""l-.:Y,.--------- --IT-\j){f·-··-~·-f';--'fx1.2 n l' 'H(i ~~~ FAil 1793'J g ~!1~ l' _0l!!.P'y.!.__}(' ...":I~(i?~. 16 00445 J34/0 6i~~>~1 6ijY6~ . -.. . . _- bl266 0~~04 42 00000 OOUOO 0~5U6 16 00469 -9541 ·--O·~--·-··-·-·------···-------------- " .•.• _._ •.. __ .. _. ij~5l~ "'1 l~ 00445 JiQ21 09530 49 00472 J}~60 (f'is42 Il; 004f,ij -9':; it 9',1) 9',') 09554 16 00445 J29J1 --'IT'ii;,i 6-1 1\&1> 4'1 00427. 17 ')/, I Il'l .n· ',J't,' rr It 'j'.f, H= 11 91t~ FI NXG?X,rRANi'-2!ttSAVr: ~I:W TltANSrltRM "NX\;2V,rk.Aliii-iT 69~~~ ~6 1~41ril~u2~ FIlf)-12 --r; . ·946---·----00RG .'::T·" ()/tl\ i 1 ·j<;o DfR.\NZ IIIJ;' 11 951 CF it '152 U, Tfifro---···-------------- 09518 '.2 00000 000011 O'l'i ';;j 09580 26 13460 12818 Of)RL' .'':'''10 11 '1IoJ N[::wr 11 0~456 00402 J3439 09400 14 12911 OOOJI II/I ".- ..(l~f):r;l(vG--=.--T{~O----------------- F" ~9 . ()'ji.66 it? 0(40). ,i3449 ;i~r' ~~I:"1-1,(".~1,10 •. rE~J.. Fl!I~.Lc~~'?-'~() --'---rr-~-'--" 0~417 . .- -.....- ..... --..--....- - - . - - . - 094f.f, Tt,-1f0469 -,..94·'<'----··--·· 11 9:10 11 oOijoo ij~j'~ '.'1 . . . . .-.----------.=0 ------~iff;f·14 .NX.';?,I( ,.FGI.~ ._ ....... _._. ___ .. 11 'll'l 11 q Oiji9) RU'ijij4ij6 09 ;'.1 6j)~~ III cn····TiUiN?--=Jr.I7, 10, HsT 1 I 'I?" J9~_~_. n! .. -'.,,:!~.!!.~UC TI ~~~ .. _... _ ... (Fii9Z·Z6· nUi[28fs-' :>1-1,'-"1\'1-;'·\1 I! I; no 11 ',,7i ni 11 'J72 -----rT ' 49 {}04o~ J34'H' .- .... - .-......-.- ...... -._-.. . 0989R 31 09153 00000 -·elf ·-···utJnUr:;..'4{,~T[-.Ti);Tnl ·rnR-·-CtlmAO··.... . "(i<;i)l(j H 1~ UT-""Ji);N~G'fX-·--~·-·-····-·.. --· 11 '1'11 FA ( f 'J~9~1--- OUII'1I1-_.I_I"._'J_l(,'..;W 11 999 ()(j() An' UURG .-10 18 005 18 OU6 18 007 18 00/0 '/lli 18 010 --liflf~ fin 111 04) 18 ()U 18 041 19 041 18 ()43 til 04\ ltiUlo 14 nllOA oOOJ2. ~;,,;\X. fRAtJL-ll -------i~:.,,~;"'g""~..,..~-;::.~~~~~~HT1T.~------· 10046 26 loofjA49' 10065 ... '1006&':31' 10018 49 NXG1Y, fltANl-ll t\ 07 . .OORG .-4 Tit NX(;lx':;"i.ST-2.d "07 Tr OORG .Lt, • III 0'.0 18 10009 bbCH·nAN2:':Jr,-12.10· l~·g~l----- ~~ SICO SUIIRUlITINE; - COMPUIES SINE AND CnSINE OF I{WlIIv;··SICO-L·-· HCO .. ! 09984 16'lm4~'J29JI 09996 49 00422 J3'H2 "TMO{\ 42 O()!)M liOOOO 17 991 18 , 09935 . "09916 '16 M4bi) -<)lltt 09948 16 00445 J2Q27 . oqqb() 49 bo4~2 J 3!)ol 09912 16 0046'1 JOOOl .... --........-..-.--.--- .-..... _--.-.. 17 9'}i\ ; .. ROT ANGLE .... t· i " 13511 12828 0982(, ooo()() -O-j-_I B493 126t}1 09826 00000 10095 I ' -- .. --.-t 11)096 16' 004(,9 Jdl31 "IU120"41 TF ?,9?,.ANGlf IN ~AOI"~S ij12~1 JOa~S 10132 26 00009 00099 ;..F.;;sTI....N-.::.l:..H:;::.::.I::N.!...;..9=!:.!:.......!..:.:.....:.:..::!::.!..:~:-----...:--------------ilouil7;4r4 T6T4TI(IJOTrcr-· ..- · - - - - - - - - lu15h 16 142~R J3419 111168 4?·UV.n -Mi19 ! .. _.. J 10108 26 012"0 1340? .~ "1 ·t).. ·C" a.~ wrtt t t intt t$ un'· o 12911 OOOJ~ njitli15rm---·--.---------.--------~ 12182 000-1 . ooOdjj-·oo:jjfio----------:-----------·------4_ 10924 (6484 ~~--u:n Ugi}--·-·--~--r--·..c-------- o Of' 16 n(, . _~"I~~_~~!lj __ ._______ ...________ .. ____ ._ . __ . __________ ._!£I_~_~___ .(~~ _ H3 flNIl ~~~T'~U~.!J_~~L til 2 r l l l r k -- ;,,-FrotJii;o04-0Z------------------CM MlllR-l,32,hlo,.rST rOR INVfRT RI ';H T ----.---------.;l:..;;~:-;:~:;~.;:g---:--!.,ri-r.t~;<-:~r.:<-;~:....;;g~~~g-=~i-~-. -&/4 16 280 H4 -----ni~----I\!=jf 18 784 1 Ii I ii6 hfAh' III till> 111 1M, 16 261 rFM r: A- -----T"-iij'-"--111 lll9 111 7. ul --Lu --------rlf --- -1\1 ij::-ij"";<-,7Tl,rTU.,."m,,"'Jjr'_'4r-"lr ,-,"sT;/i"1-r.-I'TI-rItTT~Tn('rl'/lrTui'lx"'IITIuT.'rr.-F·I,CIli'I"H.6 '\ aC.'>') 1:,11:,_ • • _i. 46 1069~ !t'l 106~~ ~,,-~o_,,-"_ 01200 --Tilf191 10892 14 l023N OOOLl, 10')04 46 106 1j'2J ---UI-'ll ;::-----omrr;-.=r.-----t0924 18 ]10 Mb SF AORS",SET AKC OR SlOrf SWITCH ----- 10936 11l 370 I'll 41n-I,.~ , IllJfr l tl-47,-,-NOn PT to AUx ",jH 10<)48 Iii j.'? . fR OlHrl.ll-2!>,MJX IN, ,SeT CIltITR!.Il REC MARK ------lO')hO III J~'i, rrtl 'OIJH'lll-!t6;12dO,srl t(lilRO triot 1li971. 111 12;, I; ·H7.f.;>4 1t1 ijrj 10812 16 10116 -0402 1\ MFAO ()tlilt:; ".:""4 ---- --- ' ..- .. "-'~-' '\!.l(, 1 H lOH 16 16 16 49 14 10799 10800 [4 12942 NZ900 10812 47 106~0 01100 If)>>24 11, 12942' N)lllO- - ----------------------------------------------- I 10A36 46-10892 01100 ; I08~0 l:~-~gf----~....;~,;.;~;..;:~,:.;g~J,.;,.l""Or,r.onll--r',"b"i'l.----------. It1 __________-'-_____ 10131 J3J89 oo4'6 i J -linn - --- -------------------------- - - - - 00445 J2941 004Z-" "lHbr-------------------·---13013 OOOJ4 10144 41 10100 0[300 10156 14 1]013 000K6 'iiji6s 41 I04SiI oUo(f ---- .-----------------.--------------------10180 ~3 10484 00000 MFADtt~,0210 iR ____________.__________ 10636 41 10800 01100 --------------------10648 [6 1072& ·;:;o4cfl"--------------·------------ OUTPUT !:lUFF 32 1048" 00000 31 13012 12916 31 12936 12104 rt,' 1'2911 --OooJ'Z 49 10450 00000 10900 44 IIOIH ll)\i92 -]11.?916 11004 33 10980 1101642 0_0060 11017 [1618 11030 1104-2 11054 j[ 13012 26 11 45 - UOf.·(, 11 00009 0000('; 11047 0000'1 10980 }j(H2 00000 OOO(){) 129(h 11414 000;"-1 OOOOR 000-1 --------rlr.t~~~~g t~ ~f~1~~~~.~~~~~~~~~:~-------------------------------111~2 49 lllSa oouOO ----lIIO'} - - .--5-u 6 "(, I It! I.Aii' I. 1)(1 to :41) 'it U _ ~"r~~ ItUII'IIl··,""AIII··lO,,\IIIFT 0[1'''' ' " I~,I fit ji,() iR.i!I.. I\ln,,;·o.I\I/XII,,;·;(ff.iitjj;H·,r iHTj1'iiHK 1\11l-/,I,c". )(r.",,;~(;[ (:(11'111) r.O[ll:S ,IH' '" III .... I I I HZ ··-----fili,--··iif/fl',j"j·::IjT,-;-,·--·.. ···- --.--:-.-.-...:.-~--- . - - -.. --1u :JIt l ~," /II.' . fij ji.4 18 l/.'j " ( 6 Illj lil(if\ll, 16 1 H. M[~sr, \1134 15 1l01l 00002 29 00003 . ·-·------.rT/.6 ~ ;'1\ 1I1~ r.1(~';I.II\;"i()ll,ij· II i;~ i i' rHilo' 01'400 ......- ...- - - - -..,...--.----.-......- - . - - . - "-\(. 3,,' 1I11S lrE 11111 ,'( I;, til. I (lilt· 4'0 'ooooi)" iHioiHf-' ........ - ...----.~.----.--- •.-.• -:- .. ~-- (11./fPIll-47,INPU'-'.1 11226 3112916 12025 11114 i)i>ili;"ij(HNiTTiilS"R:[c' "~'NI'.III ,1\\JxIN ~ 11q 19 olli 11<1" . "1I?!ii n 1l4fi4 .:.0000 .. · ....... - .... --.- .. -._-..---...---._- 11250 JJ 12012 12101t .(1~62 31 N[~T tiijlt(; irr.;>· fi{' rTAI',i.:t,',[T;!,frmCTi-il!;·;ftWHti' fNhiHERN IM,tJ, 020 AM P'A",l,JO t3 g~~: ~n 19 0',0 IJIo:, t q 0"') ,iii tit Dr- !'I\L~TIII(A(,r . (l1)R G • - I 0 l~f;~::!g(ijtt, ~~~~12 19 "t, .. rAfI;l( !4!~ ~! !~1l: tg~~ ~r:Z~" 1l1 ,11 1t5 113(,2 Il~RI' ' lllt.O 42 00000 oOl.ino . I I 16 1 i TIT\-T\Tt. 2-riT'i'"TIT'12..."X"1.........0"'o""0::-_""1..-----------:------- w&I /,,"P-1tlJ ............. . ·-"~~~T;Tif;l·NI)r.x UN(;Hf flF IHC "oTNTT:if"'" ..·..---·-..~- --W-fl\i'. - 117.9ii· 3j I J26- hilil;' 1'1'100 11 13260 000 ... 1 · ...... -----·---rmT'lrn26;,-O~ 10 ~l;t';1·J')'1(J.1, ,~s~_ ~o~ ~NI) UNR 'Pi .. 'Il~~' ·SI)flIWUIINI.' '-'I'·-SfOl{CS VARIAl\LE R[t IN PA' ,/lI.ll' --f\f':li25'-"-"----'"7I'r-Pf('"if;r. J, [79~3-rJ0~rl[r--------~------------------- 11274 49 01100 00000 'q 1)lc)Sj') ot', . 11201 39 11185 00100 ~/.J 111 . --·n . .·r'Hi--..~--· fiI----'i'iJ I ;p.r.iJ~I~,A:;';':;';.JX:";,;"I- - - - t~ J 11·1 ')OJ 1 I iffiIZi'1lno:TIj------....--------- :1·-··.. '::·r-;ii",( I III H~~SG " , . , i;ii5HsrA~rfWWNfHiiJ(" H. 17') t: ;:! n Ag~·gg ~gggg .. --.....--.. .-.-.--.-"-~-.--.--~. --- 1-10 WAry . --. 16 t 11• OP .J~.~.T.~~JCn~I.~§.. __ ... __ .... _ .. _.. __________ ._ ... _ ...•..• 11111,·, -----n;-nn'l{:;tITi· . " 16 11) .. ..- HH~HH~~ ·Bgn-.. ·-..----··--~--------·. -·· I'rd f..{s,m.l J·)I).ll'I\LU}W.'1I1J.1~ rAf .. :>TORAGr: ULrwr;1I HUi 49 II M~ P qtJj) P3P.~ ..OO~~.0._. OP10,!. OfmG • r,? 'U3'J'; 10 '1.00 • S\llr'lOIHF~l' P.\Tlfl(.-:iC4"'1S I'AI r.\ill;. III I.'J!./I,j· PAl rn HI'. !.lUI 1'111 --'''2 (j{1 ffi' i;-"1'[i1C"1f-Tjjnr·'}I~(,·;I;NfiT;"'i7,' , , fl. r Sf" lJi'f'TIiiRT'ii'i'.r, Ti.iiiT -....... ---... --. - - - ' , i TiT.lT.ITit P;-J4 ')09 . 2() 0 /,11 (I1M!''I C ooOnO,I'I\TlIH:-t,.rrsr r-ll't PII' Nl!' It'IIlR l4 OOUOO 11395 2U ()~;il iiN,." ii, .. .. ili.lO 4 t i143 i j(l1ioo ZO OhO Uli 11412 '.Z 00000 00000' 20 Of,? 11111',,1-10 iitin ... 1'I un 22"O~I "OIl.-,11~"·" . _..... ~.~ _I\A.~~~I'H'J ,I ,(.IJI""'J.~ _ _ _ _ _ _ _ _ _ ..... _ ... ___ ·__ ._· __ .•__•__ · ___· ___ !·!.!~~·rr26 '-H~~-L!.!.~!'i-It~----------.-:.----...;,. " f)1}'Jld ,11'" II) cm11''J1.6,IJO(lO'.l"MflVl flOINHI< HI 1151 N!:Xr Pili I\r. 7(01)011111)') 70 100 71 017 A II COIIP'·j . ,InIH; .(," . . . . IN I'AI fllllt.( ..... .' . . 11 /,"(, 1 146'1 OOIJ"'~, 11/.~)R 21 114l't 000011 ii'tto 49 1l408'()OOOO . . lt4U3 wSliP,R(:.tlflN, ,,"iii-':"-iiiJti'iH~, tARi)::, II~ 1",'1 'til 111< wdl IYPF.O tl:;'lh'G ,'I O/,OOPl,r I\r:;! "f\lClIlJ .... rr~;l Hill C/lltO fl,II1'\H 11',1}4 ') I j','j't;--"---wTJrT IIIJTPII''''Id., ,"lIl'fR f/lPI: IIu'''iir-·-·-·--7''''·-· .. ---..------·--.. ·-----fil.:16 21 011 BU, -"20 l1'iOH 21 oin "fl~~ln li~2ri ;>t 011 - - - - - - - - - - - - - - - -..- -.. 4(, )8 4b 49 115tH, 00200 lNf(;-OOlOO 115211 00400 il~54 00000 G.·- -----~---- , 4PG LIN tAI\[l .L!JCN .. ~!. !~~.!.~.I!~ !.~I?~s.. _______... _. .(.I!'[~.~~.~~.~... rll' 21 01'1 Olll/,(; .-'. 21 ,j2o CK')il eM "'UiJt'POF4b;':H;ro;tTsf'riilf"fNii fJf j'iUIGj(Ai~ 21 O~I ~F FIN. ---- 2I-o~lTwllur-·- nn 21 0('.) ZI O'~6 nOIlG .-10 wrltv mjfr>i.Jt-Tr;·~~·nl.ilrirr-T1SnNG· ilN lillY 21 028 i\ 21 03'} nOlte; --4 21 CH4 IINII Mf PNc.fO,ltSlzr,Il,r.Oll~r LrNGTlI OF 'RSIIr~(jf~lij-" ...... - . . 71 (no 21 041) SM RSIlr.,nUIP(j, ... 41 0\ 'UiHUlc;lisHf'" (11,(, S UUIlOC,ltS'H 21 0'./\ Ii 2t (ll" I'ltdf,i ~: .g~~ 71 21 :J~" )1 O~,~ l t 1"'.7 ll'j'j.l n~;~>/i')R 12911) 00100' Il':>(,(, 34 OOOll() 0010R fVflF.WIHHIIt 11 'inc; ·lT5lf6'ITl3l4Y·ifiH.i-1 11·.. ·1/145 I Ei8{, lJ2 ' ,N T~'.IS RECllltP Ht.lo 11 FUlt FUll l.Iou':"1 . o~!~~_e_lJ_r!:£!~ .... --.---.-.------H*~~~-}#.:~ ~#o;<.lo...O-.:........------'------.-----. lIbl0 II 130'; 132 /,5 ltbi)~ 41 1l<)1I. OOtJlih 11(,R" F.lll· 1 ((.'il) 21 U 132.HI 00000 ---- ..-----------rrnr"lTTTIj" 000-[ t 14 13235 JJ211' 11 13;'j 1,7 Ill020Doo"' 1I/:'1l 11 1-'2'>0 OOU-l Hit,? 26 Uti!; 132~O . OUTlOt.ltSI zr . 01. 13<,i,$ 11622 12 13245 J291" Ili,3it 21 l:iHr, BillS fmltG • -It 21 J~" IHr.·; ' " .... . 21' "IV;'--'-"--~if 71 ()'>II t~, "'0---------------- 11~40 46 12006 01200 -+llr.,;"'j'j~7~.-,4;..;2;-..;00:;0nO;.;0;;OMOriOriOriO lli;lil 49 lir..'ii 110000' .. ___ ....... ~~:_._~~(ARII~I-'>" ,en "I f)';? ".),::( I ·lb2814.. 12ijllMojr9··---·---·-..... ~..-- .. ~-.------ C"r<'.j'j .. -.. lr-tI)Tf;NtITIf'-7iM-- ksllE,Ol,lo,GIlfW oulpur 7.1 lI:'~r; __... _._. 11~21 :JUIl()(:,OC.,)l)o,6,I~rSl:1 .u:r4AINor:R or- CAitO HI ll'tO fj'jfl.t1C,ol, 10 .-. 1126 ourLOC,(,ll{np-', OIGo" ....... -.... .. AM SEUN(!,OI,tO II' o.,\(jijisl:"JNiI .. WNf.P C'I,tP"-1 '. sn stlJll.:~ltt·NU 21 (1M, fN~: eARO ,_ .. __ ... _._._._ .. __ -!:lr7,.,~I-( ..,-~-lul!.n,3.{~-!Af,:~~ ... __·An 1·99·7·Cl......- - . _ - - . _ -.• '21'(;(,1 SVirVY"lU;"' ·RHi.:ii;.~ffsi.. nri{ ..lT~~·-- .. · ..·..·-·.. ·---,-..-.. -...... l"O 0 " ' / U""UfI ~ CI 21 (/1\ I'M~II\ ,I:IIF RHIl."'UflC,J,l unR 44'11818 lJ2421 nl/ W'lrv I'ltlnt."~;'l'ilHl'ut ,,,IS ~,.c. ON tv~·tWfllttrli tLlilO 3/1 1324- 00100 21 (117 ZI IIll 11 01'. 7. i ,17'. '} 1 I) '(. II 1111 21 ,lr,1 IU:rl' 7.1 I)Ill .:U·1I.I 21 I)H~ IIHv /lr~ j1l41( iF; l1An 34 00000 001U8 ,,It!.iJr.,(H,Io .-I7.,I'~L(lC,t1 .. p·iHTil....;oT".'Til..-_ ..... lIIUt, II Bi',o·OoO.:.l · .--.-.---...... ....... .:'1 '1: • i'l :) "I '~'I: ,'I ."1'. 7. .I I 1.1 ··...---·--.. TfT:·;;;r-rrfT,J7i'('..mm=r------- ,. l'I{"Hll 1111 10 I) Irk I.• 'I''' • 1'"11 na I'RUlt,CAf!I.l/l·-I'I"nrS!"f PnJNrrRS OIHi.!lC,CIII{flfl-1'-J _1, .. I(rru, 'RI"rl.1 7. t II II i I rn .' -n '-fii liT ;Tr-;FiTfi1rrr=n;T.;TiTTS"'iIT'C"-TlfIii iTi'iIT'lilifT nr.··-· 71 fl,II. 'I IlllTlI'L.'t\Jl.I "llmrx fllIlPIH nllrr IU :;I .. lrrt 71 fI',', unix 'rJ~ It!; 1 H.IIlJli'ill-', , 71 tilL. CI 'I,Jf"IH-',lo,J'/,II'tfrs, Fir, I ;~Il Of "!ltl(;ICHI . ;,;\' 111\4(. 45 118J1t 1324- JIm ::':'Ir t1. .•. 'will'" f"1·'JI.IR!ljf"l:I. .. CT .. ·-T:Hi·'.T::T------....--.. - - · - - · - · Ill'.' ",I'IINCII .f.\:,r CARll f ': tl~ \. • - r, ..._.. .- ... __ .... ',"I II1'H'. OOUOO l1f118 16 13;40 J3136 (}. I11J'IO It. l1Z)'; J1B(,' 11'1(17. I.'. 11?31l 13l611' . " . ··--·-·-..·-·--·TPiT;;..'1T"'T17m.. TNrr.----·-.---.--------1I9J.(, 21 1)23'. 1)11t5 11'DII '6 132'.'; J,?'Ht. 11'1';0 14 12<117 0001('1 11'1(·7 41 ll~<.l 01100 . . -- .._ .. _ .._ .._·11~1',~I~.-.-!..,...·.4.lr7.J:~0(-,(,:I~·,!)0320(o:'·o~-." - .....-.....- .. - - .....- .. - - - - - - . - - - - - ,." .>.> , 11"'/11 1t'1 1110,: OOl)on •• d V"PU'WMYWNN'errIl'M'p'yg" t • ,~ _if 'T • lll~. o ~I"ty "j\ 21 O'}I, rtNI 7. I 100' 21 IIJI .- "'''fOTif'lFJPiTTI15-' 1 lII'; otlJ' I HESSI" ,um IJI'P1I1I5E2 iN f r;~· ;r i.fIT(A[TU·· fiiil'"NFiTToir" --~ 12001> 3'1 13339 00100 PC iN AU)( h~ '18 1,ii),INPUf . ·.ii······ ..•.... - nc 1"~'''IJl(IN h:;'" -il" ..... l1 01' i;Z') fiifV 17.025 [ 20 12 12012 12121 2 -0 i2 j 12134". 1 OZ'} Rtypr no j 1 O',() rllLNII 0'> 1 0'.2 TIll T" Il~ fh 1 0 ' - " " - ' ' ' - ' - ' ' .......... -... - .....--- ..-. -.- 1 !lit', IlliHn lPi 10 l 1 --'---1' 0 1.1 O',} It -·'('7.H·ij-·---·1O ·fflio06oOOij- .. -·--·---·.. --------·-·---·--·r-UUIt . to . i.Jt'·· [o,'"e;-oociMo'i) -.....-.--.. - ... -----.. -. ---"'-'" fwn Biil'fN 1)4'i;-fjj(Mfjfl (IS oc 10 b(' t ~~~ ~:~~f;l g~" ~~~. t 1I'i" OC l.l1 .. ·--·l1iT.ij--·----'-· ur. ~;tiooilu"'-' O e ? , 19 t ~~ ·tn~~ ~~~l ~. ::~~ ~~: ~:~3 1 110 tu~r7 1l';11 -&1 1 I H. 1)( 7..3;> 7., lit l • .1(, Ii4 Of. 7 ." () IJC lif. ·.,.OiHlt)1l 1)(. i'Tiii .. ----~ I rit I', 13'1 jVtll1. It. ..,11 rlIlC.~ If, 141 f'(IIL' I (. I',.! .1,)!·!~L IIC tt, I'll 16 1',(, JI~IIN-I. (IS if. 1',,) "lI1C~ 1',f, I'ClII' I)C flC 1" \lC' 16 1"'1 If. 1'0'1 fRAiH IlS ·-i :g H.l1-·",?,,' 'i2~ ~l.22-~~~""'" - - . - - - - - - - - ...- . - - - - - . - - - - 5~iJ()i'i()i.i .:li~8~qOO'i. ~,- :'01°0,,00.°0,'--... c 2,00 Z.uu 2 -0 2 -0 1180(' .5.~O()'!O o;,O(lOllO n .... i;,>1i3i1 17.0"" 'i -0000 ~ -0000 'i2l\49 -------..-.----.----.. . -.. ---.- ~~:!~ ~-:...,~:------------------- Il8~t;' i281lR 2,On H .l:II~~. ue "'1)0. I::> 2\ II riJ"g;j ......... . .. . ec OS 1 (, 1".7 I\IJXI1 110 I'd 1" 1<" 1\111 1'(' .. 2,tH. . . , ....... g~n· 11'113 . ~!'~'(llnpiil lZ'J63 46 l"i AII1 r"i; '~~:;{I: H. I (.b H: 11',1 1(. 1.-.11 11- I{.:' PU!t H. 1(,1· I'll V I" 1/01: ((, , (.,j 48 1~~11. 110lt 13054 . ~Ol' 4J 2 -0 :;~: ;: ~~~ HZl1 l -0 ----.--.--.......... -·HN~ 131q3 •• H .... __ . . . .•• ... - -. . - - . .- -. . . --.. ·-----}Jn. InB I\Ht. PItA',I.' r "1'\ ItT,"': 1n, ".\ I' to .J ii,i1dtl!1(,,~ '0, "'.d ,,, I" V' 1 •. '. ~ l/U'!I"llJt:'~ i'. ;·ft-'!iri~;'·'·;) I ~H !) J291li ') ,-II(I,lOO h 171\ III 1I\1I III tU~', 'J2191 (" (I{)()OIJ'1 1\1 16'-/ 110 , 0; 1.( 1<0') 111 I,'t.' .---~ ilL lIC If> \1>'1 jf, 11,'1 . 'flj' 80. tJJOJ IJJl1 \'\.';1 i " fl',,, .., ';'0060 5 H)OO .~f-~'--7~~~;~~'--'-- ------.--------..---.-.--,----.---..-fJ'i'b'-'------ ~.~()f)- II II iTi':A I'" .. --.....- .. ---.-.-. -----..---. 1~7Z0;j~~;t \f, \f.') 16 11.,9 Ie.. 1(," 16 ,(,') li. If, 1 f, If, 1(. 11, II, . 2f·:--~-·-:--:-· 2 -0 -~l~-...;Or-.-----------~------,~ Uq(,j H, l'j(, V.I'" I :j~ ~g 1321'> 11> 1','.' U';t. (,IIRlll·-1'l Ii> 1'.,{. t (, 1'i1 IAIH'I I)~" tBU:"',t;I, U22!> } Ih AI'III ',Il';A.__ ."ILf-';!jI,------__ --.-- '-"-"'-'"'' .-. --.-----.-....... --.- .. _.-- --- ~ !B~ \i. t I,'j ',HllL'liC:' 11'~ t~~ uilto 16 I II ~' I'Rlt!\. 11';/1 CMlIl-l} 13245 16 1 II~,I/! tJ\JtNll-1t1 llZS0 1(' If.... Sf I."JII IlC ~,Ol)()()() t325ij 1f> I f.7 ~,Vl\.' 1'1. '~.OOMU U.'i,O ,.... _-----------,.--- .~-----1_h_.--------.--.---.---------------.--.---.-----. -........;~~~~~~~;.:;:~---.::8'~~------------------- Ii, Fd 0 2 -0 ' 3.3 UP INS'RU~.TI~.~~. ......-.. ...!..£~-)-----------.~--------.---.-.-----------"IT'iTS- If, I ~() t f, 1',1 IlIJTrur ~~ .._.J _---;l:--..-.-:0r-._________________--' 12AOl 12004 . 3) fir Phl"If) ........ _ .... _... _._ ..... ____ .. ". "_"' __ '.'_" J ~.OOI)1l0 1(' 1 /.'» ",;\ . -----..-..... --.---.. .---.-. ·--..-~ . ·---1 at; PI, I.lr-I Lllill'l. lb \',/, o;r . - ........-_.-.-- ·-·------4 _ _ _ _ _ _ _ _-ilr.2;.,1;.:8~1:-----~Z;---::l'71t:-...---------------------__t 121113 2 l6 I ..... --~ '--'-" - ... ~ -11221~··98 -05 ~ "-O'?O.'O-O' ... 16 ~ L ... -'~H .-.------.-...... gn~ :~ :Z'i~ ~:~-:i~+---::~- ~:~g IIC Z Jl 2 38 ... ~ ~9r'- -.-..- ....-- ..... -..... -.-.....-.--...----...-.. -.-... -..'-"'-1 12166 2 1(0 ----------,tr.Z'l-:1;-:f,;'-;8:-----ii2, . ; ; k C ; ; 9 ; - - - - - - - - - - - - - - - - -......- ......Qf... __..l.!!!~ __._ _.__. ______.___.________ H;'ji,'i '>IJIH') 5~~·Jn~~-----· .. ---:-·--· -.-....~ ... --.-.--..------~ 1215" ~~. I Ii) .. 1,)'·:..:0000 .. -·· ..-·- .. •·· ..- ..... -------... -.,--.----.•,---.. ~, ". . IHU· .-.. . . -....... 2, lH 1 II )'} ~C; 2,111 '1 ·ii;j'i ..·.. · .--- .. l'iE.. ~.-.-------. o ·IH81 } ~~:~'.: nt j 5 -OOUO 12182 1 01\) "HO 10 12l71· 5,~()OUn iZ7'th .1[.22-"1'6~00' I -------ro .--.-.-----------..------.. --.--.---.--.- , -.... -12164.. 12110\ j I 04h PTtNr l oltl, RtcNI ltd 1 ---..-.--.-.- ....-~ ..-.,.--....--.... -... - ... - ·-··----··-··-·---·---···· .. -····1-2104"" ····-·-·~j2.... -·-·-··--·----·--:-------·--~------i 12\0" ....·---- -21""'--'" 1 · , ! -.- .... -.-- ...... --.... --.-.. ---..----.--.....- -.---.... --.-----. ---· .. -· ..-·.. --IHill· --'-' -------.-----~. UC 2,1)'0 ..--- -r;no Tvrr--lE"'-" ~ . ..----.. --. -----.... - ·----·-riH 1'8 ··\"9s685·ji·..aifeei-----49-19522···---·-----·---· ....-.. -.. llnKli .-It . ----.. ----.. --.------- 7 -0 3-0 11 1l,ENIl P"A~f2iJ to N~qOI)OOI)Ob • 10 NllunOIlOOO __ . ___ ..__ . ____.. H_. ______._.__ - .. -Ti1hir--·---·-To"NTnmm 1U i o · li~? OOOOOO l.F,O" ll',I'} lu H·Il.".S3l<)) 10 l3',;> III I 11.1'1 10 l\ii7,'j' I I', '.11 , I',(,() , ---"-T(r-I 10 ~, " Ai ...... ,-~~~-----------------~",","""~ ;~" ... .~II "I i p(; .LIN '-ABet 18 ')1' nf'I'{I\lM~ (!(. I "~ Ii'. III . . -.. ~-··:I·~:4.~~1;;r.~:~~~~~~~:r~!!:~==--=--==~-=-~ 11')/\ NXr.,7.Y 18 8'19 18 ')0(; F.; Jl( '--'rr-?OTTi~lY- 18 '10<' lu .. i)< <)02 NX(; IX ·18 'ItH NXGIV . . t R ijo', 18 900; (iN" '--'-"600'- pr ~ -..--..·------0--..- -.. ----. ······10·--·-----··---1,:1 ,', io I'~ PI, 10 f. '4 . tHaI 13.491 13492 ._ .. -.--- ...-. ····-·--····n50~ 13512 to 10 I ·-·16-----·---·----··--------··----··-- to - ....•....- - .. rj~lj--·--···-··--·l··--·· . - -.. - -.- ....... -....... . ... -.. -.-.-.---.. - - - - - . - - - - -. ·-·*hJi.·~;!:.;~:-:d~:------------'-......;------------------lAMln;r:;~:i~:--_--:l~-..:O'---_-_------------_ .--_...- ..... _--_._--- ... - ...--"-.--. ---I , ·.. ·--ri -.. -~ ----------{ji I .+ o "IIP"'E• •t' o Cd IT' : - MODIFICATION TO PHASE 2 OF AUTOSPOT TO RUN FROM DISK PATCH PROGRAM 19466 26 13225 12033 19478 11 13225 00010 19490 26 13225 12055 19502 11 13225 00004 19514 49 11226 0 19522 34 19560 00701 19534 49 19976 06850 19546 10540020000000 19560 10535001001900 19574 bbbbbbbbbb 0 Program patch that was in the way moved to here End of this phase - seek for next phase Disk control word this section Disk control word next section : MODIFICATION TO ORIGINAL PROGRAM 19584 36 00000 00500 Read in last cards of modification 19596 36 19680 00500 19608 26 01762 19712 Modify "read a card", instruction 19620 26 03219 19705 Modify error routine 19632 41 00000 00000 No Op 19644 26 11780 19720 Modify "write a card" instruction 19656 26 12024 19727 Modify end to call next program 19668 34 19546 00701 19680 38 19546 00702 Write these modifications all on the disk 19692 48 19694 46 19466 01200 19706 49 197420 19714 49 19912 4919522 READ DISK DATA 19728 10000000119600 19742 36 19728 00702 Read a card from disk & seek if necessary then go back and read again 19786 _ 03600 19754 47 19766 34 19728 00701 19778 49 19742 0 19786 11 19733 000Q1 Increase sector address by one 0 TRANSFER FIELD JUST 19798 25 13056 19600 19810 14 19809 19679 19822 46 19866 01200 19834 11 19809 00001 19846 11 19804 00001 19858 49 19798 0 READ TO EVEN LOCATION Transfer data to area beginning with an even address Check for last transfer Increase disc location by one Repeat 5j H MODIFICATIONS TO PHASE 2 of AUTOSPOT o TO RUN FROM DISK TRANSFER FIELD JUST READ TO EVEN LOCATION 19866 16 19809 19600 Housekeep because all data is transferred now 19878 16 19804 13056 19890 49 01768 0 Return to program WRITE DATA ON DISK 19898 10200000113136 19912 38 19898 00702 19924 47 19956 03600 19936 34 19898 00701 19948 49 19912 0 19956 11 19903 00001 19968 49 11786 0 19976 19988 36 49 19560 02318 Write data on disk If addrE!B check - seek first then write data Increase sector address by one Return to program 00702 End of this phase - read in control program for next phase 00000 Load program to load in modifications into core 00000 00012 00024 00036 00048 00060 00072 36 19466 00500 Load modifications Into core and branch to the first modiflcatiO 36 19546 00500 36 19626 00500 36 19760 00500 36 . 19840 00500 36 19920 00500 49 19584 0 o ..•. _ - .•....... _ ... _ ..... -.-- .. '-~ .. wm _ em .rt o o o t":,? a j - II I o G o 513 ,," ,n" r·t ±m". o GOOOfiE4R GOODYEAR AEROSPACE CORPORATION ARIZONA DIVISION LITCHFIELD PARK. ARIZONA GENERAL RAY TRACE PROGRAM o Presented to the 1620 Users Group (Western Region) Meeting at Tempe, Arizona, December 12, 1963. Presented by D. H. O'Herren AAP-18375 o December 12, 1963 o INDEX !!.I!. Description of Program - - - - - - 1 Program Input .3 - - - - - - - - - - Sample Problem - - - - - - - -. ~ - a. Sketch of System - - - - - - 8 b. Discussion of Input - - - - - 9 c. Listing of Sample In,Plt Data d. L1st1n~ 12 of Sample Output Data 13 0 fj 15 1 [I''''!7 trtr ri t i# '. - 1 - o GENERAL RAY TRACE PROGRAM This program is based upon a paper by Gordon H. Spencer titled itA General Ray Tracing Procedure", IBM Research Paper RC-549o Spencer's paper applies, with a few extensions I to the problem of tracing a light ray thru surfaces which may be rotationally symmetric, cylindrical, or conic and may have arbitrary orientations with respect to a reference systemo included in the program. Certain adjustments and additions have been This paper does not attempt to delve deeplY into the mathematics behind the program. It is felt that a brief description of some of its advantageous points along with a sample ray trace problem would be of more interest. The sample will attemp-t to illustrate the value of t he program as an optics system evaluationotool. In general, surfaces are described in a local coordinate system. This local system is then positioned in the optical system thru translation and rotation of the local system in relation to the basic co.ordinate systemo It is therelSy possible to position surfaces off the optical axis and "tilted" to the desired degreeo One feature the program possesses which is not common to some other ray trace programs is the capability to trace rays thru cylinders and prisms. These types of lenses are fairly common in modern optics systems making this an important featureo The basic surface equation suggested by Spencer which is used in the program is: F(X, I, Z) = AX2 + 8I2 + CZ 2 + Z • 00 This represents the surface obtained by revolving a conic section with vertex at the origin about the Z axis o o points. In general, a ray intersects such surfaces at two By using vertex equations set up so that the firat iteration point lies on the plane Z =0 thru the vertex, the iteration starts closer to the desired ;; j 6 tJi:ad'~ - 2 - C: intersection point than to the extraneous pointo For c.ylinders with conic F = AX2 cross~~ection in the XZ plane, the surface equation becomes: + CZ 2 + Z =- 0 The rulings of this cylinder are parallel to the Y axiso A similar equation will describe a cylinder with rulings parallel to the X axia o The equation: + Z2 (by)2 = 1 describes a cone with apex at ~ the origin with the Y axis as the principal axis o Cross-sections parallel to the XZ plane are ellipseso This equation may be rewritten in the form: F Z + D -JEX2 ::: + y2 = 0 To iterate with this equation, the following rule holds: Z :> 0 for D <: Q Z -=: 0 for D :> a Therefore it is possible to represent three types of surfaces with the single equation: .. Z .. D = 0 The constants A, B, C, D, and E are program input data o o mil b#ttt - 3 - o GENERAL RAY TRACE INPUT A header card and 3 cards for each surface are required inputo Ray information may be entered in 2 ways, either one ray per card or as a fan of rays. To enter single rays, switch No o 1 must be ono CODE REFR (1) HEADER CARD: Cols. 1 - 11 Initial index of refraction (may appear aqywhere in Cols o 1 - 11). • Ex •• If the initial medium is air, the number 1.0 can be punched in Cols. 1 - 3. Col. 12 NOSUR Blank Cols o 13 - 14 Number of surfaces. If the number of surfaces is 9 or less, puneh it in Col. 14 and leave Colo 13 blank. o TOL Cols. 15 - 29 Iteration tolerance in form + O.lOOOOOOOE.YI. This number i. used to establish a criterion for convergence of the iteration processo Convergence is assumed when the increment magnitude il leIS than the tolerance. It is recommended that +O olOOOOOOOE-06 (01 X 10-6 • 10.7 ) be used, as onlY 8 places are carried in computationo It may so_times be necessary to relax the tolerance to + O. lOOOOOOOE-05 , either to obtain convergence or to .peed up the program. Cols. )0 - 80 Blank ·0 518 -4- o SURFACE CARDS: Card A: IO(I) Col. 1 - 11 x ... coordinate of origin of local system" punched in 8a.. ~ as the initial index of refraction on the "adar card. YO(I) Cols. 12 - 22 Y - coordinate ot origin at local system, punched in aame tashion as X - coord1na'te ZO(I) Cole. 2.3-" z- coordinate of origiJl of local system" punched in a. .a tashion as X .. coerdina'te. Alpba(I) Cola. 34 .. lab Y &xia euler a~l. (C)() in decimal degrees, pUl'lchad in aaM tashion as X - coordinate. Beta(I) Cola. 4S -SS X axia euler angle (/3> in decillal degr.es" punched iJt c alUll8 ruhiGll as I - coordiRa'te. Cela. 56 .. 66 zan. euler angle (7' in decimal d~gree., punched ill aame ta.hion as X • coordina'e. Cola. 61 - 1T Blank. Cola. 78 ... 80 It i. 8Uggea'ed, but Rot required, that the surface carda be punched Oll .. OlB .. OlC, W, 02B, etc., in Cola. 18 - 80 \0 eneure that t hey are kept in the correct order. Card B: 'l'he tirat tift tielda gi... the ooettie1enta in the aUrface equatiOil r .-AXI + BIt • CZ! + Z • D u.2 + yt • 0, a. tollows: .1(1) Cola. 1 .. ~ A (punched in _tubioa a. X .. coordinate on Card A) B(I) Cola. 12 .. -2f,B (punched in __ ta.hion as A). C(I) Cola. 23 .. 3lP (punched in a... tasbion as A). D(I) Cola. lb ....w..p (punched in a_ tashion as B(l) Cob. 16 - .A). 5~ (pancbecl ill a_tashion as A). 519 ----------- o HNUrnM' y- •• : sr. tst ft" " T· - S• o Refr(I) Cols o S6 - 66 Index of refraction of medium following the surface, except in the case of reflection, when the negati~ of the index of refraction for the previous surface is us.do (Field is punched in same fashion as A)o Cols o 67 - 77 Blank Cols o 78 - 80 May be punched 8S suggested for Card Ao Card C: APl(I) Cols o 1 - 11 X - coordinate of center of circular-~nnular aperture or coordinate of center of hyperbolic aperture or X lower bound of base of rectangular-trapezoidal aperture o AP2(I) Cols e 12 - 22 Y - coordinate of center of circular-annular aperture or coordinate of center of qyperbolic aperture or X upper bound o of base of rectangular-trapezoidal apertureo AP3(I} Cols. 23 - 33 Inner radius of circular-annular aperture or length of semi-major axis of hyperbolic aperture or Y lower bound for rectangular-trapezoidal apertureo AP4(I) Cols o 34 - 44 Outer radius of circular-annular aperture or length of semi-minor axis of hyperbolic aperture or Y upper bound for :rectangular-trapezoidal apertureo APS(I) Cols o 4S - SS Y lower bound for hyperbolic aperture or reciprocal slope of left hand side of aperture 0 o recta~lar-trapezoidal Enter 0. for c ircular-anJilular apertureo 52U - 6 - Co18 0 56 - 66 o Y upper bound for hyperbolic aperture or reciprocal slope 0f right hand side of reetanular-trapezoidal ape,rture o Enter Oofor circular...annular apertureo NAP (I) NOUT(I) Cols o 61 Blank Colo 68 Ape rture code circular - annular 1: rectangular-trapezoidal 2: hyperbolic Colo 69 Blank Colo 70 Output code Cols RAY INPUT Blank: 0 71 - 77 Blank: no output a t surface 1: output at surface Blank C·\ 1 (Single ray per card, switch Ion) One record, either typed or punched, is used for each ray. with a bar over them are system coordinateso NOTE: Coordinate. Those without a bar are local coordinates 0 x- coordinate of 1st point on rayo XA Cols o 1 - 11 YA Cols o 12 - 22 I . coordinate of lat point on rayo ZA Cols o 23 - .3.3 Z "" coordinate of 1st point on ray. RPAR 1 Cols o 34 - 44 X- coordinate of 2nd point 0n ray. or! direction cosine of ray at 1st point. RPAR 2 Cols o 45 - 55 Y - coordinate of 2nd point on ray or RPAR .3 Y direction cosine of ray at 1st pointo Cola 0 56 - 66 Z - coordinate of 2nd point on ray or ! direction cosine of ray at 1st pointo Colo 67 .---------~-.--~---~-- Blank 521 '±z btt"j - 7- o Input code Colo 68 NIN NAXIN Blank: 2 points 1: 1 point and direction cosine Colo 69 Blank Colo 10 Optical axis intersection computation code Blank: b,ypas8 computation of intersection of image ray with optical axis o 1: Cols o 71 - 13 IRAY Perform aboTe computation Ray identification number (right justified) Cols o 14 - 80 Blank. (Fans of rays on 2 cards) RAY INPUT Card 1 Cols o 1 - 11 FXUl, the maximum Cols o 12 - 22 FZ1, the Cols o 23 0[010 X coordinate at t he 1st pointo Z coordinate at the 1st pointo 3) XGAP1,the X spacing between fans at 1st pointo Xcoordinate Cols o 34 - 44 FXP2, Cols o 45 ... 55 FZ2, the! coordinate at t he 2nd point. Cols. S6 - 66 XGAP2, the X spacing between rays at the 2nd pOint. the maximum of each fan at the 2nd pointo Card 2 Co1s. 1,·'11 FYU2, the maximum Y coordinate at the 2nd pointo Cola. 12 - 22 YGAP2, theY spacing between rays at .he 2nd point. All fans are assumed 'to originate ~t 1 :II: 0 at the 1st pointo The following sample ray trace problem will better illustrate the useage of the Horn pro gram 0 o The surface s.Yste~ to be traced is as follows: 522 Ii' OBJECT 0.0000 SURFACE III 3.0000 SURFACE 3.5000 c - - - - - R SURFACE 112 3.8525 113 = 1.15 ' - R = .85 - -- 4.3525 SURFACE 4.8525 4.9525 ------ - -- 114 -- --- ---- ---- e .500 l+l /ELLIPTICAL LENS 6.0000 6.5000 6.7500 1.000 IMAGE NOT 1'0 -..f 8.0000 - x SCALE y z z FIGURE • SAMPLE RAY TRACE SYSTEM .. ------ ...._---_.. -_.--- .... _._-- --'--'- I 01 523 r iW $1 #rim • - 9 - o DATA INPUT TO GENERAL RAY TRACE PROGRAM FOR SAMPLE PROBLEM SURFACE DATA SURFACES 1 & 2 These are both spherical surfaceso F = AX2 =L.. 2 + By2 X2 ... CZ2 + Z 1 + -y2 R a: .. L.. 2 R 2 The surface equation becomes: R 0 Z2 + Z == = - 0435 x2 - 0435y2 - 0435 z2 .. Z == 0 0 (Surface #1) = 0588 x2 ... 0588 y2 + 0588 z2 ... Z = 0 (Surface #2) These are not coneso o Therefore D =E = 0 The vertex planes for these surfaces are perpendicular to the Z (optical) axis 0 Therefore Ct. , J3, and r are zeroo The vertex coordinates are: Surface #1 (0, 0, 3 00) 5\lrface #2 (0, 0, 3050) 0 The index of refraction following Surface #1 is 10523, that following Surface #2 is 1 0 00 0 The apertature on both surfaces is a circle of 005 inches diametero o - 10 - o Surfaces .3 & 4 Both surfaces are planes arranged to constitute a prism o The first, No o .3, is arranged so that rays are deflected in the negative X direction o The second surface, No o 41 is perpendicular to the optic axis o The surface equation for both surfaces is: F = Z = 0 A=B=C= D=E=O Surface No o .3 is rotated 45 degrees when positioned in the system o To accomplish this rotation, it is necessary to specify the angle alpha equal to 45_degreeso The rotation angles, alpha, beta, and gamma are defined: o ~~ yt = yw Zt xt ____ ~yl Jlro----~ X" y xt = Xl' o "Ell m tr $ r it - 11 - o The specified apertature is a square 1" xl" 0 Therefore the minimum and maximum allowable x and y values are (0 0 5, 0 0 5) and (-005 9 0 0 5) respective1yo These After rotation of surface No o 3, values apply to the surfaces before rotation o the actual minimum and maximum x apertature values will be corrected by the program to (~005/ 0 0 5/ -{2, -{2). AP5 and AP6 are the inverse slopes of the left and right edgee of the apertature as viewed from the object point (positive Y up)o In the sample, these slopes are reciprocal infinity or zeroo equal to 1 to specify a rectangular apertature o output data at the surfaceso NAP is set NOUT is set equal to 1 to obtain The index of refraction of the prism is 1 0 60 Surface 5 & 6 Surface 5 is an elliptical cylindero Surface 6 is a simple plane with input similar to surface No o y2 0 a 2 + The equa tion of the ellipse in the YZ plane is: 40 Z2 2 == 1 b a == b 1 == 005 The surface equation, adjusted so that the origin is at the point nearest the object point, is: F == -0 0 25 y Therefore A = D 2 = E = 0, B = -0 0 25, C = -1 0 00 0 Note that surface No o 5 is a cylinder with rulings in the X direction o If it had been desired to translate this surface off the optic axis in the Y direction, then YO would be specified according1yo This pro~ram was originally written by William Webb of Goodyear Aerospace Corporation, Akron, Ohio and to that gentlemen goes the credit for t his signifi- o cant contribution to the lens desiGner's kit of toolso The program ha s been altered slightly to suit current needs and is being maintained b:y the writero ----------------------------------------------------- ~~--' 1:1 - 12 - SAMPLE PROBLEM DATA INPUT SAMPlF PROBLEM. PROGR.M NO. 14~'A-6~ • 1~/(1f'-/63 C' 7+0.1000~000f-~6 1 •."-' 0.0 0.P 3.0 0.'" 0.0 0.0 lA -.435 -.435 -.435 9.0 0.0 1.5~' 18 0.0 0.0 0.0 18.5 0.0 0.0 0. 121. 3.5 0.0 0.0 0.0 2A 0.588 0.~e8 0.588 0.pj 91.0 1.0 28 0.0 0.9 0.0 0.5 0.0 0.0 0.0 "'.0 4.352'5 -45.~ 0.0 0.0 3A 0.e 0.0 0.0 0.0 0.0 1.6 3B -0.5 121.5 -0.5 9.5 0.0 0.0 0.0 4.9525 0.0 0.0 e.0 e.e 0.0 0.0 0.0 13.0 0.0 1.Jlf -0.5 e.5 -0.5 0.5 0.0 tIJ.0 0. 0.0 6.0 0.0 "'.-tIJ 0.0 5A 0.0 -0.25 -1.(21 tIJ.0 121.0 1.5 58 -2.0 . 2." -1.0 1.Ci'J 0.8 0.0 0.'" 0.13 6.75 0.0 0.0 0.0 6A 0.0 0.0 fJ.fJ 0.0 0.0 1.0 68 -2.13 2.0 -l.et 1.0 ttJ.ttJ 0.0 13.0 tIJ.0 8.0 0.0 0.13 0.fJ 7A 0.0 0.0 0.0 0.0 0.13 1.0 78 0.0 0.0 13.0 121." 0.fJ 0.8 0.1 0.0 0.1 0.3 3.0 0.! 0.1 0.~ , o o lC 1 2C 1 1 1 3C 4A 4B 4C 1 1 1 1 5C 1 1 e C' 6C 1 7C o 521 Oh 0 t rtrr rirbH rirlt - 13 .;. SAMPLE PROBLEM DATA OUTPtrl' SAMPLE PROBLEM, PROGRAM NO. 143A-63. OBJ PT RAY SURF 1 1 1 1 1 2 1 2 4 5 NO INCIDENCE ~ 0 2 3 2 4 2 .3"'568 -.~55S9 .30fl'CfI'~ .30e~2 -."'35e9 ZOR M .0~PJ0'" ~.08521 .99781 XOR K X OR K .100091 .~00~0 y~ L Z OR M • "''''C'J00 .309J~ .28848 .29772 3.39209 -."4675 -.01308 4.30574 -.42369 -.15807 -.32813 Y OR L Z OR M .10000 .""000 • "0QJ00 .10000 .0001'30 .0000O 3.0t'WJ&2'0 .30"'0" .OO~00 3.00"'P0 4.9525~ X OR K .10"0 .00000 .00000 .10000 .00000 .0M00 .10"'00 .00000 .00000 .10000 .QJ000QJ .80000 • 30D)00 3.00fJ00 -.00136 .30409 3.04095 .rlJ~00 -.01~C'J6 .3Bf1fJ0 3.00C'J00 .00000 .3009."0 .29198 !.44-823 -.03302 .06923 4.31947 -.26404 I( 2 y OR L l OR X OR y OR Z OR X OR Y OR Z OR X OR 5 M K. L M K ,..l K Y OR l Z OR M XOR K YOR l 6 7 NO IMT W/OPT AX 2 '3 3 2 3 '3 '3 ,. '3 0 1 3 5 6 • 000t2J0 • 10.fJ9.JfJ .00000 .0fJ0011 .10000 .3e0~0 3.0~000 .,0fJ0fJ .30900 3.00~t"" 3.~012J .0000fJ • ~l'J0f'0 3.1JfJN0 ~0"'fl'00 .3~" ,.0QJ00QJ .0PJfI01J .fJ0tHJf8 .30000 X OR IC. Y OR L .QJ0000 .1000fJ • 00P100 Z OR .00fl'0~ S.9fJOO0 .0rm.!J00 .30"''''0 lOR fit 2 DIR COS .300"0 Z OR M )( OR K 2 tNT PT .100~0 Y OR L 2 PT/DC .0000121 X OR 3 2D X OR K Y OR L Y OR L lOR M 1 12/02/63 M X OR K y OR l Z OR M X OR K Y OR L Z O~ M X ~ K y OR l Z OR M X OR K y OR L Z OR "4 X OR K y OR l l 09' M X OR K .10000 .80f/J0flJ .89000 .10000 • 00000 .00000 .10C"091 .00000 .00080 .1emJ0 .00000 .00900 .1f2'000 3.~00 -.30000 .311000 3 .80f/J0flJ -''''~633 4.95250 -.99486 -.;7026 6.03553 -1.272!9 -.43702 6.75000 -2.08!77 -.63295 8.0Pft09J -.300fJ0 -.31159 .30869 3.08697 -.30660 .3000'0 .l'?55 ~.00000 3.38486 -."'1229 -.02183 4. 34192 {'I -.14898 -.14338 4.95250 -.58127 -.5278" -.'M00 .''''000 !.m00021 -.3""'" • 30fJCJJ0 3.00000 -.30000 .3fJ~ .0fM'0fJ .0t'K'J0('J 3.0(?J000 ,10000 -.391fJ00 6.07531 -.74N0 -.3QJ4~3 .894~9 -.494!5 -.19014 .84810 - • ..,9091 -.3042' .53fJ8S -."'211~ - •.0.2."1 .99933 -.92552 -.24762 .96851 -.'1870 -.15416 .928~7 -.'4192 -.24"62 .8e'311 -.36128 -.08675 .92140 -.54192 -.1301~ .83028 -.54192 -.13181' .83tt28 .01673 -.03736 .99916 .28045 -.3",.,• .91~'34 -.21391} -.19fJ21 .95815 -.~4225 --.3fJ434 .!8895 -.22816 -.fJ8569 .96984 -.3.125 528 ;"iI' - 14 .. y :Z :3 • It 4 ,. 4 5 5 5 5 ~ 5 X C!JR·K 7 Y OR( NO tNT W/Op·T 3 OA l .DR M Z M~ X ~ ~X 1 ~ y OR L. Z OR M X OR Y OR 2 , l em X OR "fOR z OR X OR Y OR Z OR 4 5 APERTVRE STOP 1 6 6 6 6 6 1 2 3 4 5 , •.,.e .J.,.. .s.,.. J.00fJefJ· .3~fJ .J~ '~fJ86"8 -'t'~t~~ - •• "!~ .,~ft'·' - ....1"2 , .29166· ,.• Z'166 ~.1"'12· '.51'49 ..... "21126 ·~ •••·1" -.82,,26 •• "!2.3 -.'6'll _,"2"1" ·-.t." ."~,.·1 .0eff00 .~ .00080 .000fJ0 .00080 .,~ • 304fJ9 3.~fJ !.QJ4.'~ OR K ~0QJ0fJ0 .~ • f)~ • Nfl• YOR l Z OR M X OPt K "f OR L OR M X OR f( .1¥'0f&fJ .3fJ1J08 .292'.,1 -.,24751· l M. f( OR L "fOR L Z OR M X OR K y OR L Z OR M X OR K y ~ L Z OR ~ X OR K y OR L z OR M X O~ K y O~ L Z M fill X OR K y OR l ·.efIeH _000e" •~"e .000021 .fJfJNfJ .~""'"' .3.." 3."" • 0f1ee0 .,.s. . . 3.000~fJ .QJ0fJfJfJ .,0MfJ .~ .~ 3_~fJ _~8 .seeee, _0.00 -".,. .. .MJ08" .000ee .00000 3.0e0ClJfJ • 00t.?JfJ" .0fff19QJ 3.fJ00ee -.15."" .. '.44t2' .... ",vtt,a2 -.38412 ·.568.5 .".,.,. -.,",2964 ~9'9S6 .. SZ,.,.. 9'.8. -.-.1,. •• ." -.ZI,.,,, ' .912'" -.52287 &.m'.·S8 -.888l1 .91111 .e61fJl 4.,5!58 ..... "3853 4.952'0 -.98414 -.36698 -.2.,.,1 -.8156' -1.11t28 -.S2Z87 .3~ -.4'459 6.75000 -1 •••145 -.6J101 .... 13231 .fJ3614 -.,,3614 .99869 .000fJfJ .3~N0 .~00 3.00000 a.0e000 ~·fJ00fJ0 -.!~0fJ .00"'00 .~" .!fJ0fJfJ -.3986m .30860 .e42.1 -.~l287 -.132'~ .,,.Ze7 3.00fJf1Jt!J ~.086fJ8 .0~ -.300(1!1e .000t!f0 .0~ .300MJ 3.00flJ00 .fJ000fJ -.!0000 . ,.0MJ00 -.100'-'0 -.29166 .29166 3.38849 .2t4"85 -.iJ4085 ~.""5 .3fJ412 -.316.12 .911218 -.28566 .... C2J18CJ3 .95998 -.!2ge6 .'0~" -.151S6 .... 3.412 • 0e9JH _00MJ0 .. ~ .00~~ •. 0(:11000 ,~00 .0"000 .~IJI00" '.0MC'Jt'J -.!0G'J0r1 .!000t'J 4.9'25" -,49297 -_51421 C; w.3~"8 ~ .00000 ~. '\J -.1"54 ·.. 952se ~ ~ X OR K INT W/OPT AX .00~C1J ..,~ .'~.71 '.80fJfJ~ X M NO .~..., •.C800fJ" Z OR M 5 ."."" .'''86. .3886. .,. •• 161~ .!12JIJe8 K L Z~ M ·X OR K y OR L Z OR M 7 ••~e77· ·~.la8M ,.~ y 5 8.00"*8 ,011J9!JfI" ,0000fJ YOR L , ~.HfJIIIJ .80088 z 5 ~" .•.m.,. ....!sa?,.1 .0CJJ000 .. e0000 .00fJ0fJ ~00000 )( 4 6.15etee -1.1996. ... 76011.5 ~ .~ X OR K L Z OR M , .PJC1JfiIJ8e 3:. -J'''''' .,.,.. .eeeN ~ YM 2 .1_ ."...., "'~l'el81 .89~99 ... 21911 -eIl13.! 529 0 - 15 z 0 6 6 6 .., 7 ., '7 7 0 K • flJ0f1JflH6 ~.0~e00 -.!~ .0fH!M8fJ ~.00000 6.1~f!NHj .9~e7 -.32986 l M .!000~ -.76"46 .00000 3.0000" a.~ 1 X Oft K Y OR l Z OR M -.100"0 .3fJeflfJ .,1159 -.e16T~ .,~ .'''869 ~."'8697 -.037'6 .99916 ? X ~ K YOR L -.lM00 3;,:t14t6 -.!flJlt, • .91133-' I( lOR M )( OR K YM l 3 Z~ M )( OP K 4 5 3 APERTURE STOP 9 -.le~tt .,." -.llfC'YfJ~ .~fJ .IJf.JfJIJQJ0 .!~0 ."650 • !0fJfJIIJ -.~2S11 !.~0 .!~ .... 31611 .!fINI10 3."",," 4.S!'00 ....4.95250 15''''''' -.2M4~ -.46886 -.19121 ,16254 -.75ell -.'1',.34 ,!mrN -1.918"5 .5tT"2. -.5.12 • fJ09.J0e .eeelJfJ 3,0fJHe 6.'06"2 .85."., -.11MJl'J(1J .~'-'0 .3~" 3.lJ&WJfJfJ • •1'6 .~0~ .~ _"4.9 3.114"'95 .ell" -."'971 . . . .'3 -.lMmJ .NNP .t919.8 -.14?2 • ""tl X OR K y OA L Z OR ~ )( 0" K V~ Z~ l ... XQR Ie Y OR L .~0 •.~N0fJ -.18eee • fNJ8ee Z OR M .~fJ XOR K -.1MfJ", X~ VOR lOt K l M )( OR K Z OR • X Oft t( y ~ l Z Ott M 7 .e~ .lJtMJfJfJ -.1tJ0fJfJ .0fWHJ0 .fJIIJfIJtlJ0 -.1""''' .0fJee9 .tM'Jt'J0" -.10(MJ0 .~ .~ NO tNT W/OPT AX , .'''66'' .!91'5 M 6 2 ,!MtfM lOR 5 1 .'0f1~0 K )( OR K Y OR t. Z ORM X Oft Ie y O~ L Z OR M X Oft Ie ¥ OR L Z OR M 0 .......,e • 0fMJ"~ .000e0 '.0ftflfI'" ,.0",," yOtlt L 4 8 9 .~0tJ" .fJ0N)fJ xOR V OR L 9 .93581 -,1~f9fJ YOR t. ZOft M 8 -.12590 NO tNT W/OPT AX ·2 S -" ••,9 .... ~f2!" -.!0M0 e 8 .30000 .P0fJ~e .0QJ~ 6 1 8 .~fI X ~ y OR L Z OR M e 8 .912182 -. J29fJ6 -.12599 .~. YOR L Z OR M 7 6.17732 M -1.08431 I> 7 OR X OR V OR Z OR -.leeN .8fJQJfJfJ ."" -.1"-""''' .009Jf1f.J .0~" -.1~ .fJfJllJee .~ .,."e .,-- -.SfJ89fJ _"'1""6 3.~" '! •••e23 .1"''' .e!lt. .. 118fJZ .fJ25~2 -.'1"'• .1JfJ0fJIJ •• 3481 3.0~ .~9J .1~e ~.~PJ 4.18..,'1 -.15".., • ~IJ -.a1469 -.365!8 -.!~, ,.efJlN '.83457 .~ -1.e1115 .93'1'1 -.58.1' .sfJfJC.IJe .1fJfM0 3.~ .~ff .3fJ9Jf1fJ ..... -."41~1 4.95~50 -•• ,,54 6.15,"," -1,8"976 ... 62984 3.~ '.'~fH! -.s..- -.'''S68 .1"852 -.3,""-",, 3,""'" -.2884' .2.,?7l !.'92'" .1.94 .StM!JfMJ -te6!'J! !."0~ 4.45'44- 3.~~ -.,~ .3~e 1,08~21 -,15.1• .'H~6 .. ,5fJ4t~ -.24162 •• .8~"'~ -.e89'J2 -.1'3" .8"t-' -.5fJ41! -.1!~ .",t? ••55" -.8'5e, ,9,..,81 .12811 ... ! •• !t .I~H -_191'61 -.19814 ,961'~ 530 - 16 9 9 6 9 ., 10 10 1" Ie 10 OR Ie M l OR M -.1~ OR -.10fJ00 f( M L ~ ,.. OR K y OR l Z O~ M X ~ K y OR L Z M M )( NO I"T W/OPT AX 9 10 y l X y Z 5 9 10 )( 4 ~ X OR K y OR l l ~ M )( OR K YCR L lOR flit X~ K YOR L Z~ M X OR I( YOR L O~ M )( OR It 6· Y OR l Z OR M x OR K 1 2 3 4 z Y OR l Z OR fit .." X OR It. , OR l l 11 Oft M OPT AX INTERSECTION Z • 1m 1 )COR K YCR L ZOtit M 11 11 11 2 ! 4 XOR K y~ l Z~ M XOR « YOR L ZOR M X OR t( Y OR l lOR N 11 5 11 6 11 7 X OR K y ~ l I OR M X ~ It y OPt L lOftM X ~ I Y OR l .~.,'" .~fJ .fJfJfJfJ" .~ . -:= -.211}81 5.". -.t'8S.. - • .,8126 6.fJle42 -'~4'.4 -.:)16tl ~.~fJ -.i.MG1J0 -.1!1ge -.!_~0 6.7S0fJfJ ..... 96!~.., -.~1621 .. ,1219fJ .5_~0 -.1611,. "QJfJ(JJpJfJ 8.~ .3fJIJfJfJ ..,.,.,'" .~e 3.~ ~.".fJ51 • 'fJIIIJIIJfI • 2_206 .~"'~ .8fJIfJ0~ • fJefJfJff ,leeR '.~0 •""em .JI21" 3._0 • SfJ11JPJ0 .eeeee 3.4'185 ."2fJ1. .~ .941882 .9.",2 .....'.12 .fJIIIJ" ,99,'1. -.2129• ."~ • 96l"• -.4,e16 .fJ8lfHIe .-.•.1.,.,., 1.-..... .-.1_ ..,., ..--. ..,. .-,,.._0 ... .1_ ...... .-- .-•.,.,N .1~ .JM0e .".,., '.fJ000S .1",.., !.0fJC8~ .t~ .00000fJ .fJee0fJ .'"""'''' .!1JIJe8 .8181&8 3 • .".,.,1' .1~ .~flJ0C!J0 • GJIJIBfJft .0f""~0 !.~0 5."5485 • lIN. .tJf1C!J00 .eeeee ,.I.f.JIJ .1fJ00e .fJfJfMfe • .JCMIJH .CHm0e !.1I91JG!'0 .~fJ .1M1N1 .fJlJfJ0fJ 3.8IJfJee .8.",0 .1. ." . . .00 .fJe000 .~ !.~e .1~ .~""" .1lf1lefI12J .fM008 " .1JfJ8fJe .1..", ."IftMIJe -.2"1'89 .~e 4.9525fJ -1.4N84 .~ 6.lJIJNe .", .efJ0J0~ 4.37t2" -1.11695 ,1JfJfJ0e 6.75.., -!.1741~ • ~0QJ 8."""~ •"M.,e .fJ."~ ! • .,.,~ .. ..,1093 .~0 '.4"92 -."3274 •""eN') .fJMmt .68el' ".4'eT., .MJNJe ,872.4 - • ..,!!.~ -. ..,~,.! .~,.. •• 'eI6 -.1121" .efJfJN .99916 -.82658 i . . .NJ .99.64 -.!'4!8 .QJIIJIJefJ .91t2~ .~0 .fJfJfJl3e 4·.9525" .e449! """"" 6.""''''' ••"fJ -1.2e657 6'.'"'' •.,.,.,0 -1.,9189 '.~00 .~~ C _' •• 16 ".51488 !.~fJ .0800. ." ••6 -.1""~ 4.'1915 -.2'118 -.'2fJ31 C'I .91414 -.5.914 .~2J • fJIJIlIJ0fJ -.'''4~' .• e98~8 -.~~3ee .50000 ..",,, ..1'" .,.., •,1,..,.,. ...,ee -.~1621 4.95250 .~0 • .,M00 .fJIJ00e -.16~~ t.~"''' -.!M00 -.1ftWJ~ .!~ -.l~fJ -.HI62 -.3~6'9 ."..ftIJ .,3426 -.5'••8 .1-"'2 -.!S4" rJ 3 1 0 o o o ...... ....... W N ...... ... ... .\Ja \JI .... .... \at ~ 0' ... W ~ .... ~ .. .... ....w .... N.... ~ fill .... .... ... W '" .... -." CI' \JI \Jo) ....~ ~ .. ....N .... N '" N ... .... .... .... N 0 " ,.. ..... ,.. -4 )( )( > )( Z ....Z f'11 Ttl -.... "' "'..... .... ~ (It ::cJ Cft "' "' n - n iiijiiii~iiijiiii'iiji i-i Ns~~zr~~~~~r~s~~~~~Z~~ N~~~Xr~s~~~r~.~~x~~~r~ HS • • • ... ('\ "''''.t!''~s''" .... "0 ~ (A j'iiiiiii N 0 .... ....Z -CXN-C)cN-()( N N-C.N<)CN~)CN~~N~~N~)(N~)( iiiiiij~i~ijiiiiijijii ......... N~~N~)(N<)CN-CMN-C)CN-C)CN-C)( .. ~ ' ....N llilltl~ ~ljll'ijjlliltlljjl'I' 11)~lli'li)~~tl~tt}l}~ ~ )tlllit' i~lli~lillitllii)lill Iliiliiliiliitilliili I • N., . """, ,. '\It .. • ·1·li~·I~~I·I&S·li=·I~··li'&i I~I! . ~~ §! !=I~t I \ :i ~ . Ii !! ii iii; I liiiiiii '·I·~~·I·~·I·I~;I·~~·I~ii~'·I'~ ~ ", .: iiiiijil ~. • • ~ , • \It If' f ~i~~I·~~·I}i·I~}·I~'·I~j·l~ N ~N •• ~N ~w ~~. ~. • .... t •• •• ... • t t • • ~ ,_ t ~ w_ _.I~ ~. ~~ M~ ~.·I~~·I~~·I··~~·li~·I~~·I~i·I·! • ~ ~. ~~ W~ ~- a~ .~ W~ • • • ~.s~. ~•• ~~ ~ • . •at ... : I .... ...., I ....~~~~---------------~-----------------............-~ -.,.,..~.--.- 14 14 • 5 14 6 1'4 T Z ~ X OR ~ YM L % ORM X OR, It .., Ott L z OIItM X CM K •_111.,,, • ...ee , Oft .L I OR M X (JIlt It Y OR L Z OItM !4 OPT AX tNT£RS!eTIGRz 1 l! 2 l' S • fM Ie y CIt L ! QIIt,1It X. I It z."K 1~ 1-5 5 • _Oft.: l' ,o.t %." Y'ott L '(M'L ORM i •••• Z xOit K ~M Y"• .fJ__ tIJ k L Z ... ·15 7· 16 1 2 JOR It Y eM L Z Oft ... X GIlt Ie. y OIItL 1. J 16 ,. 16 16 l' ••1" , 1~ 16 .0It k Ott l Z GIlt flit OPT AX tNT!RS£CTtON Z 5 6 1 z." a., X K Y. L 101M J( 0It ~ Y~L lol'- Y. X Oft I( L Z 01 '" )( 0IIt1C VORL 1. 0It M k OR K 4 ••• • •18e •• Z X .. i •••• L YOR L , 15, 6 ..... •••• ._e ..-.,,,... .-,.." ...... .-.'1' ..... ••••• ...... .... ..... ., ••'1., ..... .,..•....., ....•••• •..,,,e, .... .... ..... ._e :=: ;., • 4.'.11. - 18 - . ! ...... •88e"" ~. ,.IJ00IJIJ .1JeIJf10 4.352'" -.2.'.5 .".4,9125" -.1"" 6.MMIJ .9468. .""". ·.51••' 0 .."~ .,,.2. .... ..... -".. .-.•IJ_ .1_., •1M.,.,. ""..",., '.8" •., ••••ell I •••• .t38'• -1.1 ..71 • .. 6.1S" -1 ••' " ' ' ".,,~ -.1_ ..,,..,,,, '.'.1.5 -.3_ .1'•• • • •,fJ , .1II8II!Je ... ....,,... -.,.-,..•• -., I •••• • 8IJ8,1J i • •a" • I •••• ,."''''' ..... ....,.., •• ,88fJtJ • •"IJI J . . . .., -."!2~ .. -.e• -.~148' ."'2~ -.~t41.e 411.".. .~"2' ...,. ...... -..,"" •••• ._ . .... -.1,.." .. ...,.,. .... ! . . .9 -_29tel ,•..-t. .et964 .~ • 9ft;!' .2_751 .968t8 -.• 'l.~ -.~!'11 4.2""'4 -4\2'2." ••eMf -.'54!1' 4.M2!e ,".'1 .".,.,'" .9J.Sl -.61892 -.2H·S2 ,.".NY ••"'''6 0 -.14'..,' ".", 6.1HM -1.15,.,! •.,,,.e ' ' fJIRJ I • .. ,''''31 -.J4~7' -" .9'.'1 -,tJe9." ....,. ...... ...•. , .'''' ...... ....,' .,... -.t- s._"e -.222'" - .."_1, ...,,.,,, .... -.!...... ... .e_ ..,. ..,.,., .J_ .-.1_ .-e ...., .tfJ5~1 -.lfJfJfJtJ . . . .fJ J.'41M .1MJIJfJ ••••iI s ...._ ~ t .... .-.fJ -.1_ ..... .........,.,". -.1_ -.1"" , • .,.fJllI ' ·.J.168 .e.. I . . . .' • 4 ••J.~' -.1?!t19 ._e. !._,ft -.,... I ..... • JIlIN .8fIIIJfJ" J.fJ1Ne0 .~ ..... .~ 4.9'2S8 -1.11361 6._19• -1.~6263 .8NJi11e 6.15"., -2.""44 .MJtJfJ ,91496 .... 4"166 _,e2fJ! -.69fJ66 ,f.Jef.ttM ...£72'17 46,," .Nfl• .8',6' -.6te66 .fJfIJfJM .?2317 -.6"'66 t':C)3 at) 0 0 16 ot>T AX y Ott L Z Ott " JNTE~S!CTtOM Z .NH" • • .M I( 11 .2 ZOR X __ I y _ L- 11 ! 17 ,. y~ ."••"8..,,, -.1-" L .118',., Z (MIt M • OR -,1-' ••'B. Ie. Y OR L 17 17 IT 0 )( OR It Y (Mt L •••••e lOR" 6 -.l~ )( '"' It Y OR L Z OR fill .,.." -,I"" • eM K Y OR L "7 OPT AX 11 18 1 CMt " 5 lORN I"T£"S£CTIo.. Z .., • )( Of' tc 1 Y OR L I ~ fill OR I( 18 2 ~ 11 , Y "" L Z~M X Oft I 18 4 VORL z~, M )( OR K VORL z.tIII 18 11 11 6 1 18 19 0 XOIIt y ~ L 5 OPT AX 1 19 2 1. , zX~· Ott " y ~ L Z ~ ~ N X OR IV~· L lOR" INT!RS!CTIOIII Z • X OR VCM Z ~ X aIt K L M Ie YORl z._~ X ott It !.11'Il _~tIJ Z Ott M )( otIt· K Y _ ., .NlJfJfJ .~fJ K L 1 11 ..-.1_.., ,.•• .._8. ,..... ,1.,. -.1.... . ,. .•.,.,. . ••1." .-. -.1_ .... . ...... .,."., .•....._e. ...... -.1..,. -.,..",., .IJi"" .~0 .."" ..,"019 .8".1 s.e_. .~"_I ..,., '.e.... , " s,..,fJfJ '.I"~" .1'2111 .e~ .f'J2181 .0efI • ....1 • .".., t .... "18" .""t.,. ."".,~ ,.49992 ,I"'." .".. •• l.'!! .e""lf., . .,..,." 1.8".'8 .... -1._"" ... . . . .111 .1. . . S._"" ..... ...... .-." -.1•••• ..... ...... •••• ._M ,.•..,ee -.1_ •••• -.'''.N 4.''7'" ....,,, ...,. -19 - .e.",. 4.95De -.T..'. ' . .~lJlle ." 6."'.. -1.721" ..".,. ."'64 -.,,"....,..,.,,, •.•eM .".,.' ••49'21 ...... ...."'''' • 86e?1 ... ••,te13 ' 4"~" ....-"...,., .MNtII _,•• n - ••• ~2 • I • • •"" .-.2.2,. _." -.Ie!.,,, .MfJtl . ..•...." ."8". .. ,.,. .,.,'" .•..,...". -.,,..........- .... -.1_ -.. 1""" ."",..., ._11 .-'" ........,- J.'.,IJ -.1_ .-- .,.•...... ,. -., . " •••• .1...,. ,J_ ... ,. ._e .1_ -'."2" .1_ ",I." .... ..•. I ..... ,. . . . . .fJ -.1-' .fM8ee 1. . ._ -.~ ...",. ·••11"., • e••• , . . . .8 I ..... -,t"" ,.""fJ .... tfJIJfJIIJ ~ . . .8B ..1!JIIJe '.4,t'5 "., _HI? , .",,,,, .962•• -.2865. 4.!1~." .97842 -.1!IJ~' ... 1'71.1 , .M~?9 4 ••52-5", -."a41 -,22",5 .H.e .9ft~1 6 ..... -,7.'91 -.~!"~~ ••nMfJ ,M'79 -.!.,! ...8.,. .94'7• -1,1456' . .".,., I •• " • • 4. TIl•• ,1fJfJefJ •••• .zta.fJ 1•.,."57 -,J_ _,"'8 I.""" ' . . . .21 -.JIINe -.19"2 -,M5" .8"'9 ."'11 -.'Hl' .%8'8'.' " .194" - .f': 04 at)· ,~ ------------.---... ------.-..------,.~~~~-~---~---~.............. --~~ - 20 - 19 19 20 20 4 5 NO INCIOENCE 1 y OR Z OR X OR Y OR Z OR .00000 L 2~ 20 M .~H 3.~00 )( OR K Y OR l Z OR M .10081' _,,""00 .0H0fJ -,30000 3.0fr0C'J0 • 09t0"" -.1~" ~.M~PJ0 .0_ _ .""'ee -,3011Me 3.fJ0efJ0 .800fJfJ -.:J1JIIJfJC! 1.,,"0fJ 4.9'2~0 • .",,,0 -.99416 -,30"fJ .11"'26 .~09'" .1000'" .. 00000 .10fJ0" 20 7 21 11 1 , 21 ,. 21 : ._01) ~"~ .".., X y tMK )( ORK y ORL Z ORM X y aRK ~ L Z OR JIll )( OR K Y OR L Z OR M 5 X 0Jt K y O~ I.. l M N X OR It I> y 0fIt L Z OR M 21 X OR It 7 Y OR L NO INT W/OPT AX 21 22 22 1 \2 .tIiJfIJe8" .fHIJ00" ~ V Z ~ M OttL Z Ott" 2 21 21 X NO tNT W/OPT AX 20 ..1"'" .,.,,, .'-'~fJ .1~1J .~ X ~K Y ~L 6 Z 20 .~ .1~ Z ORM X ORk y OftL Z MM 5 .94820 -.79091 .30423 -.3f'000 X OR Ie y OR l Z OR M X ORK y OfitL 4- 4.30S74 .e00fJ0 J( X y OR L 3 .191314 3.0000QJ K L Z OR M 20 _"'1308 .~00 .l0~0 M 0" 2 -.30000 Z OR M X OR y OR Z OR )( OR y OR ,108INJ " .1_ •••• !.MCPM8 -.~fJl'6 -,"211~ .02971 .99913 3_"'4Q1t~ .... 131"06 -.29198 3 ....... 23 -.025~2 -_13'02 -.e6923 -.3:187. _.31947 .92807 -.54192 .24162 .SfJ'1.1 6,15N~ -.3"""'" -1.271S9 .4''1eZ .N0fJ8 -2_,,8811 .8H18 -.154192 ,~ .6'2'~ e.0HfJ0 .830'8 -.31159 .016.,5 .031" .99916 .. 28"4-5 • 384!4 .91fJ!4 -.2lSfJe • 1ge21 .8fHPJefJ ,.~ ... ,.~ -.,~ -.297"~ .~e 3.~fJ .~ .~ .1""" ."".,." .N. . . .1'" .. -.i_. -.,.,., -., ..... ,..,., ! .HefJG!J 3.~f!JfJ -.3."" .... 3~ . .0fJfJfJe .,,~ ,000. .00"H .'21t! 4.14fJ28 -.'''225 _38.'4 4.9~25'-' .88'9~ -."121 -.22616 .~8~'9 -.7.0. -.!42l5 .11854 .'.741 6.15~ .. 9'''11 -.3422~ -1.1'K4 .16N5 .lzaS4 e.0(IjffJfJ .9'8'1 .3e86~ !.0fMIJfJe -.30860 3,08688 -.03614 ,03614 .9986. . ..30",,' . 3.,"'" .2"66 -.3.41~ -.29'66 -.3~ 1.NfJ,0e .3"''' -.3"'"'O C; .95815 -.14198 .14"1 ~.JfJft(M -.",.,. ."., 3.J84'6 -_,1229 .969t. .1NN _"fJfJ0e .13"13 , • fJIJlJIIJe -.:1""" 3.80.,., .0fJt2JftfJ .1,e1' .5218£'9 6.07531 .fifJPjfJ" .~e .086'5 .92840 -.'4192 -.30",. ,11J8t1fJ -.36J?8 6.'35~3 .~ .QJS8ft .1~476 ,e36" .8.,."" .10IMHJ ••ee, .24162 .968~1 -.2'4~4 -.39869 3."8691 -.38660 .1eaee .53~85 -.313409 -.308N -.3._ 3.Neee L K L .00~PJfJ .1580'17 4,95250 .etJfJee K M .3fK?'' ' ' -.42~69 C .3e412 r:C)~ a t} ..t 0 t n , _or ,," !I! II - 21 .... 0 22 22 22 23 l ~ M X OR k YOR L Z OR M 3 )( Oft K , Oft l Z Oflt M It 5 APERTURE STOP 1 2 3 23 • 23 O~ K .0';"0;' OA L .~ X OfItK • •11. .1SefJIPe Y OR L ~ M X Oft K Y Oil L Z ~M X ORK V OR L Z Oft fit X Ott Ie YOR L 5 Z OR ... 23 0 23 X OR 6 I( y 0tIt L Z Oft M ., X OR I( Y OR L Z OR .. 23 24 24 NO INT ."/OP' AX 1 X OIl K Y 0tIt t. l OR M X OR K Y OR L 2 Z OR ,.. 24 ! X 4 L Z OR M X OR K y 24 y O~ K ~ ~ l. Z OR M 24 5 X OR K Y ()tit L l OR M 24 ~4 #) )( OR K Y OR l ., Z X <* ~ M K Y Oft L Z OR ,.. 24 0 25 NO 1 ..",. .e~ .~ X Z 2' . .~ y Z ORM 23 .;00N'" .00NtJ .1JII!JefJ0 INT W/OPT AX X aRK T OR L l • fit .-'. ._e8 1.000910 V'. 3Qf.000 -.30000 3.~" .' .3~QIJ" -,380m' • el!Jll8ftJ ."."" .~ •••• ...,., ., .~ .f)fJf'fte ."."". .e..,1 .,.1.1'., .....e .• eiPJl8fJe .fJMN .NfJH -"..,. ..... .".,.., ..... ."..,.m • •111 .e.... .~ .~ -,2'211 .241~1 .~ .96888 -.32619 .15 ... 69 -.,,,,,,,fJ .~ ,.~ .~ -.3~ ..,.". 3.4.,ae ..... QJ61.1 4.'5Z50 .... ll""" _",,"3 4.",2'" -.,.,414 .36691 ,9'23_ -.522.7 .2.7~1 .$l~ •• -."'",~& .e'.~l J • fJt'JCIIiIJe 6.8'.... -1.17128 ,43459 ."'11 -"2~1" ,.~ 6,75Nl' .e42fJ7 -.S2287 .IJ232 .842" -.,.,.,., - .10IPJGJfJ , • •Me -.,-., .... ,.--" -1.... ,4~ .631"1 e.~ .1'2'~l . ."3614 -.HI6" -.~" ~ .03614 .99869 .38412 .38412 .Ha?1 !.ee."8 -4ItfJefJIJ -.2',66 -.J~ ,.N1IHlJIJ 1_,,'.9 -.t_ -.s_ -.,,.8..,. ..... ·.1_ .e4fJ8~ -.~66 • .,4"15 .It.7 -.se. . -.:t".. ,.~ ! • G!IJIIJNJ8 -.UJfNIJe -.~f9I!HI!tf!t .-.~e .~~ .02964 ,99956 .CM!JtM9fJ .••eN 3. _ _ .t.,.. -.,fJIN18 .1..,. .!56f1J'" -.,~ ...... .fM!IINJe .384.12 .0"8 -.,O4" 3.fJ4095 I.e".,. -.s..", .eeIINJ .lDfJIIJ8fJ -.77"~2 -.3." 3.~ .1'''07 .85561 .158"5 4~9525~ .~" .90278 -.4814' -.36.28 !.e. . .....--..... ,..eee_ -.,... ."'" ..,., .".",,,. ...... ......,.,,," • fJfJ!ItIIJ ,f'8.,e,. ,.,88.9 "."2'-'26 .02026 4.332t3 -.1".., 3.""" 3."""" .... 29766 4.1.'15 -.""'! .151S6 4.9S~'" -.... . .95998 -.!t9'6 .3"412 .8t!99 -.49291 -.219t' 6,.,.,32 .,. .5.2'., 6,1''''''' ."t8a .,,.21 -1."8411 .e,'.3 -.",. .t2, • ., .9J~'1 •• 329fJ6 • 125ge .76"'46 I.~ .93',S1 .11159 -.,e8,9 3.1886 • ., - ••16" .817" .9"!6 536 :: 'IJ III i'l 25 25 25 25 XOR K 2 YOR L Z~ M X OR K V OR L Z~ N X OR K Y OR L Z OR N X OR K 'If OR L Z OR M 3 4 5 25 6 APERTURE STOP 26 1 26 )( OR K YOtit l lOit M )( 2 y 26 ~ ~ K L Z OR M )( OR K 3 Y OR L 26 26 )( ORK X OR K 7 NO INT W/OPT AX 1 X ~K y ~ t. 2 , X OR K 'If OR L Z OR M X OR K Y OR L Z~ M It X~ K Y OR l Z7 Z OR M ~ )( OR Ie Y Oft L Z OR M 27 6 X~K y ott L 21 7 t OR M X OR K .y Oft L Z OR M ...... -. 'l9fJfJfJ 3.Nee" .S~" ,-., .,~ -.!fttfJQJ ._8 s •..,_ -.JefJIJIJ !.~ -.29,.,' .3fJ6611 ~.3'.8' • 0tJ65tlJ .82811 4.3~9"" ·.31111 .15909 . .,e.,. .91"3" -."', • 19fJ21 -.28e.~ .'043. -.5.'eZ ... ,9525ft -1,91'''5 •• .,890 6.2fJ682 -.-5_12 -.118m2 .8~7'1 _fJ21!3 3,.4"95 •••e... -.S~ ..,lM, -.%9198 ."29"1 3. . . . '.4.823 ••6851 _e3,,81 ..31~.1 .N"88 -.3. . . . l.fJflJfJtllJe -.fJ~l.' .i5.-76 4.3 • .,'1 .9!6~ .... 155'., -.5"'4t3 .24161 ."'" -.1_ ...... -.1_ .lJI8NfJ • •eeII .eg•• -.,..,.. .e.... I • ...., -".-.e" ... -.,-. 3."_.. .~fJe -.3MfJe -.leMJe .",,_1 -.10fMJe .~~ -.!e41J9 .1J41~2 4.9$l~fJ -.11469 .'65~. 6.fJ3~~'" -1.tlf7l1' •• '3~4 .6.75t30e -.leeee .MJfffJfIJ • fJIlMIIMIJ -.10fJfJfJ """"" -.1... .. -.1"".. -.3fJ!J'1 -.,•• t1J2 ,fJ!5.9 3""". 3,"fJfJ -.,~ ,629'. -.28848 _,2813 3.J92e9 .3842' .89429 .lee94 ", ~.1'76! 4.4~J.4 ,96165 -.".162 -.316~1 .162N • 54S8. .38421 .89858 - •.21811 ,08116 -,'4344 .91.14 -.'1621 -.~98~. 3. . . . 6.1.",.2 -.3eefJ8 . -.'''8 • fMlJftfHlI -:-.,""""' -.'' ' fIIJ" 3.~0e ,.0000fJ ."35" .99181 -.Z.lT2 4.952'''' .59974 6.75""" -.963i17 0 .1~3" 3.""21 -.,..,. J.MJfJ. -.3ft'0fJ .8_8 -.~e~l' .8531'· .M0fJe -.ltMtfJ0 .0fJIIJftJ .e89~2 .9,,.~e •• ~ftI)fJfJ -.lfJfJlJe .~ .'~716 -.3360' J.~ 3•• -.1 . . . .2."6~ -1.8897t -.1"'0 .00"" .0eNfJ .e25~2 3.0fJf'fJfJ • •fJtJtJ -.3tM11Je .", .fJNfJtJ ,999!1 .IJ!9! .85'17 -.58_I! .".- -.,.. -,'.,., ... -., ... ...... e. -., ... ...,fI. -.'''' ... -.,"'"'" ....e 0 .862'4 -.·15ft18 .efJl'. .0eMQJ -.1~ 6 .3fJefJfJ -.1-.8"".," -.3_ 1.8.,." ..,,.., 5 l OR M 27 •• ,8. .lIfJNe Y OR l 26 21 .~0 .eeJfJe -il""" Z OR M X Oft K y ~ l l OR fill 27 -.14JIJfJ0 -,1-.~fJ y Oft L Z OR M 27 .~ .~ 4 Y OR L 26 -.1_ Z ~ M X OR k lOR .. 26 -.1-..0'-' ..,. -22- .19"1~ .12198 .94082 -.316~1 .76170 8.80""'" .12198 .94082 537 0 - 23 - o 27 o NO INT W/OPT AX 538 Po c GENERAL RAY TRACE PROGRAM o 21. CARD '/0 DIMENSION XOC~~).YO(35).ZO(~~)tAlPHA(~~)tAETA(35}tGAMMA(3~)tAC~5) DIMENSION RFFR(36).APIC35).AP2(3~)tAP!(~S).AP4(~5).AP~f35),AP6C'5' DIMENStON NAP(35),NOUTC35).CK8A~(35',ClBA~f35J,CM~ARf35JtX~eAR'3~' DIMENSION Y~BAR(35Jtl~BAR(35)tB(35),C~35).D(1~Jt~t3~) DI~ENSION KKK R(35),KSURC35) 1 ~ GO TO (813. 803), KKK I'~. :',•• 803 PRINT 115 115 FORMAT (20HlOAD DATA,PUSH START) PAUSE 813 KKK • 2 o READ TITLE CARD READ ? 2FORMATC49H ,8H PUNCH 2 PUNCH 100 JB = 50 PUNCH 1 ,PUNCH 1.00 PUNCH 10'" PI :s: 3.14159265 CONY • P1/180. C READ HEADER CARD 1'-'12 FORMAT READ CFl1.5,I3,E15~8) 1~2, REFR(l),NOSUR,TOl C REFR • INITIAL INDEX OF RfFR C TOl = MIN. LIMIT ON NOSUR • NOSUR+l ITERAT~ON ,NOSUR. NO. OF SURFACES o JNC~FMFNT 539 P. o 00 203 I c ~ 25 2. NOSUR READ SURFACE DATA 103 FORMAT CFl1.S. FIt.5. FIl.5, FII.5, FIl.5, Fll.5.J2.12,13) REAO 103, XOtIJ,YO(I),lO(I).ALPHACIl.RETA(t).GAMMACI) c XO. YO, AND ZO ARE LOCAL SYSTEM SURFACE VERTEX COORDINATES. c ALPHA, BETA, AND GAMMA ARE EULER ROTATIONAL ANGLES. READ 1?!3.Af I) .~( I) tC(!) ,DC I) .F"( I' .RFF'~( J) a, c. D. AND REFR = INDEX OF A. c c 203 C REAO o c REFRACTION FOLLOWING SURFACf. 1~3'AP1(t,.AP2(I)'AP3(1).AP4(I),AP5(1).AP6(I)'NAP(IJ,NOUTCI) API, AP2, AP3, AP4. AP5, AND AP6 SPECIFY APERTURE DIMENSIONS. SEE C C E ARE SURFACF FQUATTON COEFFICIENTS. INPUT WRITE UP. NAP • 0 FOR CIRCULAR-ANNULAR APERTU~E. • 1 FOP RECT-TRAPElOIDAL APERTURE, =2 FOR HYPERAOLYC APE~TURE. c NOUT =0 FOR NO OUTPUT AT SURFACE, = 1 OTHERWISE. C QFAD QAV nATA C SWITCH 1 ON TO USE SINGLE RAY IF(SENSE SWITCH 1) 2~5 READ 10~t 2~5, CA~D INPUT 201 XA.YAfZA,RPARl,RPAR2.RPAR!.NI~,NAXIN.IRAY C XA, VA. AND ZA ARE RAY COORDINATES AT FIRST POINT C RPARl, RPAR2. RPAR3 ARE RAY COORDINATES AT SECOND POINT C OR RAY DIRECTION COSINES AT FIRST POINT C MIN C NIH-. 1 IF RAYS ARr SPEC. BY 1 POINT AND C NAXIN C =0 c IF PAYS ARE sPtc. ~Y 2 POINTS Dt~. COSJM~S. 1 FOR COMPUTATION OF INTERSECTION OF RAY WITH OPTICAL AXIS. • 0 OTHERWISE I~~Y = RAY NUMBER. GO TO 206 ~ 201 READ 116. FXU1. FZl. XGAP1; FXU2t FZ2. XGAP2 READ 116, FYU2f YGAP2 Po 26 C RAY INPUT DATA, FXUl • MAX.XR. FZl • lR. FXU2 • MAX. X AT POINT 2 C XGAPI • X SPACING AT POINT I, XGAP2 • X SPACING AT POINT 2 C FZ2 - Z C FYU2 COOROINAT~ o AT POINT 2 MAX Y COORDINATE AT 2ND POINT, YGAP2 • Y SPACING AT POINT 2 II 116 FORMATtFll.'.Fl1.5,Fll.5.'11.5,Fll,5.Fll.5) IRAY • 0 GAP3-flJ. RPARZ-'VU2 e12 ~PAR2=~PA.2-GAP3 IF(RPAR2+FYU2)~0,,814.et4 814 GAP'3.YGAP2 GAP! - fJ. XA • FXUl 4 XA - XA"!""GAPl o IFfXA+FXUl)~l2,8~4,804 804 GAPl • XGAPI GAPZ •. fIJ. ~PARI ~ FXU~ • RPA_l • RPARI-GAP2 IF(RPARl+FXU2)_.e0~,8~5 8"5 GAP2 XGAP2 1; YA • 0.0 ZA II R9AR~ FIl • FZ2 NUt • fJ NAXIN • 1 IRAY • IRAY .... 1 206 NIN • NIN+1 GO TO (287.288). NIN o P. 27 0 208 CKf'ARfll • RPARl CLeARC1) • ftPAR2 • RP.'" C~AA~(t) GO TO 2St 2m.., xo • ~ttARI-XA YD • Rf'Aft2-VA ZD • RPA~~·ZA RALE" • SOATeXD*XD+YD*YD+ZD*ZD) CKeA~(l) • XO/~Al~" CleAR{ll • YD/tltALfN CM~Aft (1) • lO/'AlEN 209 .",eARll) • XA Y0eARfl) • YA ZfJ8AR(1) • O DO 15 lA- I • 2. NOSUR IMl • 1-1 SA • SIN(CONV*ALPHAftt, CA • costCONV*ALPHA{t,' se • SJNCCONV*8ETA(1)) CB • COS(CONV*BETAfJt, 56 • S!"tCOMV*GAMMA(I) COSCCONV*GA~MA(I)t CG • Xl • X88AR{IMlt-xoCt) Yl • V.BARf IMl '-VO( r I Zl • l""ARCIMIl-ZOCI) Rl1 • CA*Cr,+SA*SA*S(; Jt'l~ • Rl' 0 -Cl*SG • c.. Sft*SG-SA*CG ft21 • CA*SG-SA*s8*CG 5 ft 2 P. 28 R22 = C8*CG R23 E -CSA*SG+CA*SB*CG) R31 == ~A*CB R32 II: 58 R33 lit CA*CS X"" .:: Xl*Rl1+Yl*~12+Z1*RI3 YCi7 ::I Xl*R21+Yl*R22+Z1*R23 Z0 • Xl*R31+Yl*R~2+Z1*R33 CK a CKBAR(IMl)*Rl1+CLBAR(IM1)*R12+CMBA~(IMl)*R13 CL s CKBARfIMl)*R21+CLBARrl~1)*R22+CMAAR(IMl)*R23 II: CKBARCIM1)*R31+CLBAR(IMl)*R32+CM~ARCIM1~*~~3 eM J 0 t IF(OtI)) 31~ Z2 C II: 31~.311.~02 -ZC!1 C: IF(CM) 6210. 10,60~ 6"0 X0 Y~ 11: X0-2.*CK*Z0/CM • Y0-2.*CL*Z0/CM GO TO 211 311 IF(CMl 210 X0 y~ 210,3~2.21~ 31: )(l'J-CK*Z0/CM • yt?l-Cl"'l~/CM Z2 • 0. GO TO 211 3t2!2 Z2 • Z0 211 S II 0. 5 J :r: J+l X • X0+Ck*S y • Y"'+CL*S Z • Z2+CM*S 543 0 P.29 o IFeDfI)) l12.21~,214 212 IF fl) 7.6,6 7 Z -Z ft IF (CM) 4e0 4~~ 10, 400 = X0+CK*(Z-Z2)/CM V = Y0+CLc.l-Z2)./CM X 21~,216.216 6 IFf V) 715 V = -V IF(CL) 4~1, 1~. 4~1 401 X = X0+CK*tY-V0)/Cl Z = Z2+CM*CY-V0,/CL 216 DO • Del)*otI) F • ffll*OO*X*X+OO*V*V-Z*Z FX • 2.*X*ECl)*OD o ~V = 2.*V 4 DO Fl c -2.*Z GO TO 8 214 IFtl) 6.6,7 213 F • AfIl*X*X+BfI'*V*V+CCI)*Z*Z+Z FX .2.*A(It*X FV • 2.*Sf.'l*V FZ e e DET~T 2.*C(r'*Z+1. = CK*FX+CL*FY+CM*Fl IFfOETMT) 218,217,218 217 IF(F) 10.9.10 10 PUftCH 104, tRAY. IMl 11 CONTINU~ IF (SENSE SWITCH 1)205,3 o 218 OElS = -F/DETMT , ~ 'il p.30 o D~LS*DElS DELS2 • TOL2 • TOl*TOl IF(DEl52-TOl2) 9. 9, 219 219 IF(J-JB) 220,10,10 220 S • S+DfLS GO TO 5 9 t~{D(I)1 320.321.301 320 IF(S-2.*Z0/CMJ 221. 222. 222 '21 IF(CM) !p,e.·~~lt'0f/.f '~1 IFtS) 221t222,~'2 ~00 IF(5-l~/CM' 221 PUNCH 105. 221,222.222 t~AYt IMI GO TO 11 222 lAP • NAPtll+l GO TO 223 RHSO • (Z2't224.22~),KAP (X.APlft,)*CX-APIC!)+fY-AP'CI»)*CV-AP'Cltt IFCRHSQ-AP'fl'*AP3(I») 12. 226, 12 'UNCH 106. t~AY. c ~26 1M! GO TO 11 226 IF(RHSQ-AP4(1'*AP4(I» 13, 12. 12 224 IFfY-AP3(ltt 12,12.227 227 IFtY-AP4(I'J 2l8.12.12 ?2S tFfX-(AP~fr)*tY-AP3(lt)+AP1(t).) 2?9 tFtX-fAP6(1'*fY-AP3ft))+AP2(I,t. 225 IFfY-AP5Cl., 12.1'.229 1~.12tl~ 12t12t23~ 230 IFfY-AP6CItl 231.12.12 ~~1 FFF • fX-APlfl))/AP3fI) FFFl • "F*FFF 666 • fV-AP2CI)t/AP4(It o Po o GGG2 = GGG*GGG IF(FFF'-GG~2-1.) 13 RAT ALe 1~. 12, 1? = ~FFR(TMl)/PFFPfl' = (RAT*DETMT)/(FX*FX+FY*FY+F7*~Zt IF (REFR(I'+REFRCIM1») 232,233.232 23~ GAMUC • 2.*ALC RAT = 1. GO TO 14 232 BLC = fRAT*RAT-l.)/fFX*FX+FY*FY+FZ*FZ) DIS( • ALe-ALC-PLC IF(DISC) 234,235,2~5 234 PUNCH 1?7, I~AYt IMl GO TO 11 o 235 DISC = SORTfOISC) IFfALC) 236.10,231 236 DISC • -DISC = DISC-Ale 237 GAMUC 14 CK • RAT*CK+GAMUC*FX CL • RAT*Cl+GAMUC*FY CfIIl • RAT*C~+GAMlJC"'FZ X0BAR(I) • R11*X+R21*Y+R31*Z+XOCI, Y~8AR(J) • R12*X+R22*V+R32*Z+YO(I) Z0BAR(I) • Rl,*X+R23*V+R33*Z+ZO(" CKBAR(J) = Rl1*CK+RZl*Cl+R31*CM CL8ARtIl • R12*CK+R~2*CL+R32*CM CM8AR(I' • Rl~*CK+R2~*CL+R33*CM KOUT • NOUTfI,+l GO TO o 16 PUNCH (l~,lb).KOUT 108tI~AY.IMl.XA,RP~Rl.X~RAR.l),(KBAR(I) 31 P. 32 PUNCH 109.YA,RPAR2,Y0RA~(1),CLeAR(I) PUNCH 11~.lAtRPAR~.ZmAAR(Il,CMeA~(J) c 15 CONTINUE NAXIN • NAXIN+l GO TO (17.239"HAXIN 239 IF(CKB~R(N05UR)' 240,241.24~ '41 rF(X~~AR(N05UR)) 18.242.1e Ie PUNCH 111, fRAY GO TO 1'1 242 S c -y~eARfNOSU~) IFf CLeAR(NOSuP) , 243.244.243 244 IF(Y?BA~(NOSU~)' 24~ s c 18,245,le 5/CLBAP.(NOSU~) GO TO 245 240 S = -X~8AR(NOSU~'/CKBAR("O~UR) IF(ClSAR(N05UR) 246,244,246 246 IF(5+Y0BARtN05UR)/CLeARCN05UR1) 18,245,le 245 AXIN PUNCH = Z0BA~{NOSU~)+S*C~BAR(NOSUR) 112,I~AV.AXIN 11 PUNCH 1"'0 IFfSFN5F 5wtTCH 1) l05, ~ 1 FORM.Tt!HRAY SUPF,2~X6HOeJ PT,4XSH'D PT/OC.~X6HI"T PT,6X7HOIR COSt leJe FOR"AT flXt 104 FORMAT f I ~ , I ~ t 13H NO INCIDENCE) Ifl15 FORMAT f13,J3.13H VIRTUAL PATH) 106 FORMAT (13,t3,14H APERTURE STOP) 1~7 FORMAT (I3t13,14H NO REFRACTION) 1~8 FORMAT fI3,Il,16X,8HX OR 10.'19 !="OPMAT (22XfeH~ OP l f( ,Fl1.~,lX,Fl1.~91X.F114~.lX,Fl1.~) ,Fl1.5,IX,Fl1.5,lX,Fll.5.1X.Fll.5) o o Po 110 FO~~AT (22X,!HZ O~ M .F11.5.1X,Fl1.~,lX,'11.5.1X.Fl1.5) 111 FORMAT fI3,4X.15HNO tNT W/OPT AX) 112 FORMAT (I3.4X.23HOPT AX £ND c o INTERS~CTION Z =.F11.5) 33 o
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.3 Linearized : No XMP Toolkit : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37 Create Date : 2014:03:09 11:04:57-08:00 Modify Date : 2014:03:09 12:19:34-08:00 Metadata Date : 2014:03:09 12:19:34-08:00 Producer : Adobe Acrobat 9.55 Paper Capture Plug-in Format : application/pdf Document ID : uuid:6403a539-4d0a-da40-9e68-a9239b6bf578 Instance ID : uuid:0b8639a1-899f-224a-853c-2eb5b05f0860 Page Layout : SinglePage Page Mode : UseNone Page Count : 555EXIF Metadata provided by EXIF.tools