R-577-sec1-rev1
Virtual AGC Links Page
--
1foiza0eer)))}
Ba
9) v-_y
v1es)
eaet
GUIDANCE, NAVIGATION
=
--E_e
AND CONTROL
wn
ys
Approved: (LEKa wt
Date: PaJeq
A. LAATS, DIRECTOR, SYSTEM TESTS
APOLLO GUIDANCE AND NAVIGATION PROGRAM
2=~]
Approved yaa. a. Nratnin
Date: Jisfeq
F.H. MARTIN, COLOSSUS PROJECT MANAGER
APOLLO GUIDANCE AND NAVIGATION PROGRAM
ee
4
= <=
Approved:
KA "SY Tem
Date: *Qo 64
R.H. BATTIN, DIRECTOR, MISSION DEVELOPMENT
APOLLO Cpe
AeNDSNAVIGATION PROGRAM
U 4
ApprD.ovGe.d:HOAG,
ELE
Dates')G an Sf
APOLLO GUWANCE
NAVIGATION PROGRAM
rn
Y4e)
P 3
ApprIRo.NRvS,eTd:RRAULGMhAENaN,T4AdDTEI,POUNT3Y.LADLBiIlORaREA¢4TTa.OO..RRY Date: 3 ae oF
R-577
GUIDANCE SYSTEM OPERATIONS PLAN FOR MANNED CM EARTH ORBITAL AND
LUNAR MISSIONS USING PROGRAM COLOSSUS | (REV. 237) AND PROGRAM COLOSSUS IA (REV. 249)
SECTION1 PRE-LAUNCH (Rev. 1)
DECEMBER 1968
WEE
INSTRUMENTATION
LABORATORY CAMBRIDGE 39, MASSACHUSETTS
ACKNOWLEDGEMENT This report was prepared under DSR Project 55-23870, sponsored by the Manned Spacecraft Center of the National Aeronautics and Space Administration through Contract NAS 9-4065 with the Instrumentation Laboratory, Massachusetts Institute of Technology, Cambridge, Mass.
ii
R-577
GUIDANCE SYSTEM OPERATIONS PLAN FOR MANNED CM EARTH ORBITAL OR
LUNAR MISSIONS USING PROGRAM COLOSSUS 1 (Rev. 237)
SECTION 1 PRELAUNCH
Signatures appearing on this page designate approval of this document by NASA/MSC.
f
Approved:
) are. a DAAf ,
Thomas F. Gibson Asst, Chief, Flight Software Branch
C 4 Manned Spacecraft Center, NASA 7
Approved: _ "=
C
<
James C `Stokes, Jr.
Chief,
ht Software Branch
a
Manned Spacecraft Center, NAGET
Approved: Leh Kaarth Lynwoed ¢ « Dunseith Chief, Flight Support Division Manned Spacecraft Center, NASA
Date: 77-3 ~*oe ~
Date: rfp
Date: %, Ges
ili
(iv is blank)
Date: October 1968
REVISION INDEX COVER SHEET GUIDANCE SYSTEM OPERATION PLAN
GSOP # R-577 Section # 1
Title: Title:
For Manned CM Earth Orbital or Lunar Missions Using
Program COLOSSUS 1 (Rev. 237)
Prelaunch (Rev. 1)
Date Oct. 1968
Revision PCR 596*
PCR 575*
Description Filter gain in gyrocompassing Fig. 1.3.4-1, page 1-18 Changes per NASA review, pages affected: 1-25, 1-26, 1-27, 1-28, 1-29, 1-31, 1-39 1-40, 1-43, 1-51,
*Indicates an MIT Program Change Notice (PCN).
Vv
(vi is blank)
Table of Contents
Page
1.
Introduction ,
1.2
CMC Self Test
Bee
1-3
.2.1 Options Available in SELF-CHECK.
1-3
ee ee
.2,.2 Procedure to Start SELF-CHECK
1-3
.2.3 SHOW-BANKSUM Operating Procedures
1-3
.2.4 Control of SELF-CHECK Options (Figure 1.2.4- 1).
1-4
.2.5 Explanation of SHOW-BANKSUM (Figure 1.2.5-1) .
1-7
Bee
.2.6 ERASCHK (Figure 1,2.6-1) .
1-7
-2,7 Check of Rope Memory (Figure 1.2.7~1)
1-13
1.3
Prelaunch Alignment Program Computations
1-15
1.3.1 Initialization Data
1-15
1.3.2 Coarse Alignment (P01)
:
1-15
1.3.3 Vertical Erection Computation (P02).
1-20
1.3.4 Gyrocompassing Computations (P02).
1-20
1.3.5 Azimuth Change Computations (P02).
1-20
1.3.6 Optical Verification of Alignment (P03)
1-20
1.4
Prelaunch Alignment Functional Description .
1.5
Performance Test Computations.
1-25
1.5.1 Gyro Drift Measurement
1-25
1,5,2 Accelerometer Error Measurement
1-29
1.5.3 Gyro Torquing Scale Factor Measurement
1-29
1.6
Functional Description of Performance Tests.
oo
1-37
1.6.1 Gyro Drift and Accelerometer Error Test Description .
1-37
1,6.2 IRIG Scale Factor Test Description
1-37
1.7
Performance Test Data Analysis
1-47
1.7.1 IRIG Scale Factor Data .
1-47
1.7.2 Gyro Drift Data
1-47
1.7.3 Accelerometer Scale Factor Error and Bias Error Data
1-50
vii
(viii is blank)
1, PRELAUNCH 1.1 Introduction
The Guidance System Operations Planis publishedas six separate volumes (sections) as listed below:
Section 1 Section 2 Section 3 Section 4 Section 5 Section 6
Pre-Launch Data Links Digital Auto-Pilots Operational Modes Guidance Equations Control Data
The purpose of this section is to present the program requirements and descriptions for the prelaunch calibration and test operations for manned CM Earth Orbital or Lunar Landing Missions using program Colossus 1 (Rev. 237),
These routines utilize the uplink capability (described in Section 2) of the CMC to load either variables or instructions for utilization or execution during the running of the tests,
The results of the gyro drift and accelerometer parameter tests are used to determine or confirm the IMU compensation parameters used for the mission. The compensation parameters will be loaded prior to launch and used during the mission to reduce the IMU alignment and specific force measurement errors.
This volume constitutes a control document to govern the test methods and data analysis equations to be used for prelaunch calibration and test.
Revisions to this plan which reflect changes in the above control items require NASA approval.
1-1 (1-2 is blank)
1.2 CMC Self Test
Theversionof AGC Block Il SELF-CHECK found inthe program COLOSSUS has been reduced to include only the erasable memory check, the fixed memory
check and the SHOW-BANKSUM job.
1.2.1 Options Available in SELF-CHECK The different options of SELF-CHECK are controlled by loading the
appropriate numbers into the SMODE register. Placing a +0 into the SMODE register forces the computer to go into the backup idle loop where it continuously looks foranew job. Loading SMODE with +11 octal or greater causes SMODE to be changed to +0 and puts the computer into the backup idle loop. Loading SMODE with any other number less than +11 octal starts up one of the active SELF-CHECK options. These option numbers are as follows:
£4 octal checks erasable memory £5 octal checks fixed memory +10 octal performs both previous options +1, +2, +3, +6, +7 same as +10 option -0 same as +10 until an error is detected,
The SMODE register is set to +0 by any FRESH START.
1.2.2 Procedure to Start SELF-CHECK Noun 27 is assigned tothe SMODE
use the DSKY as follows:
register, so to activate SELF-CHECK.
V21N27E (option number) E
This loads the desired option number into SMODE, and starts that option,
1.2.3 `"SHOW-BANKSUM Operating Procedures
The SHOW-BANKSUM routine shows the sum DSKY (equal to plus or minus the bank number), the
DSKY, and the "bugger" word in R3 of the DSKY.
consists of three steps: it is important to perform particular job.
of the bank in R1 of the bank number in R2 of the
The operating procedure
the last step to end this
1-3
Procedure to START SHOW-BANKSUM This routine has its own Verb (91) so it is very east to start, The
information for bank 00 appears in R1,R2, and R3 of the DSKY immediately after starting SHOW-BANKSUM,. Verb 05 Noun 01 is used to display this information. Starting SHOW-BANKSUM puts +0 in the SMODE register. This forces SELF-CHECK to go into the backup idle loop.
STARTING PROCEDURE V91E
(The computer must be in program 00 or a V36E should precede V91E)
Procedure to Display Next Bank
The" proceed" verb is utilized to display the sum of the rest of the banks. Each time the proceed verb is entered from the DSKY, the information for the next higher bank appears in Ri, R2, and R3 of the DSKY. If another "proceed
verb enter" is performed after the last bank in a particular rope has been
observed, the information for bank 00 will be displayed again. Continued proceed verb entries will allow you to observe all the banks a second time.
CONTINUE PROCEDURE V33E or PRO
Procedure to Stop BANK -SHOWSUM
The operator must punch in the "terminate" verb when heis through with
SHOW-BANKSUM.
This terminates the SHOW-BANKSUM routine in the
EXECUTIVE,
TERMINATE PROCEDURE V34E
1.2.4 Control of SELF-CHECK options (Figure 1.2.4-1) The program starts at the entry point SELFCHK after whichit stores the
address of the ERASCHK routine in register SKEEP1, A check for anew job is made and if no job is waiting, proceed to test register SMODE. If the contents of SMODE is +0, idle by looping through the check for a new job or, if greater than +10 octal, change SMODE to+0 and idle, For any other contents of SMODE increment the SCOUNT register and test SMODE again, following with either A, B, or C below.
A. Ifthe contents of SMODE is +4 perform ERASCHK, the check of erasable memory diagramed in Figure 1.2.6-1, CNTRCHK, the check of all counters and other special erasable registers (Figure 1.2.6~-2), and CYCLSHFT, the check of the cycle and shift registers in Figure 1.2.6-3. Then increment SCOUNT+1 register, store the address of the ROPECHK routine in register SKEEP1 and check for a new job starting the erasable memory test option
1-4
G-T
Load by DSKY Option Number intoSMReOgDiEster
OVp2t1NN2o7.E E | c(sKEEP)) + ADRS(ROPECHK) |
SELFCHK
C(SKEEP1) + ADRS(ERASCHK) |
ne!
"I
Cheek for new job
-Non Zero
c(SMO IDE) + +0 } +Non Zero
Increment C(SCOUNT)
SMODE
' TC SFAIL
£#-1(001,2o,c3ta,l6, 7)
ERASCHK
Bad.
Check Erasable Memory
Good
| CNTRCHK
Bad ROPECHK Check Fixed Memory
[Good
CYCLSHFT
Bad
Check Cycle and Shift
Registers
Good
Increment C(SCOUNT +1)
Fig. 1.2.4-1 Control of SELF-CHECK Options
C(ALMCADR+*1) + BBCON SELFCHK
C(FAILREG) = Octal 01102
Program Alarm light turned on
Increment ERCOUNT C(ALMCADR) + C(Q)
ERRORS C(SFAIL) + C(Q)
Restore Erasable Memory
again. Normaily the program continues to cycle as above until the content of SMODF is changed by DSKY or until an error is detected. If the contents of SMODF is + 5 perform ROPECHK, the check of fixed memory in Figure 1.2.7-1. The program then cycles back through the starting point SELFCHK and continues tocycle in a manner similar to that of option +4, as described in the preceding paragraph. If the contents of SMODE is -0, +10 octal, £1, +2, +3, +6, or +7 branch to the routine indicated by the address inregister SKEEP1, For the first pass this would be the address of ERASCHK. Complete the HRASCHK, CNTRCHK, CYCLSHFT loop. At the start of the second pass, the content of SKEEP1 has been changed to the address of ROPECHK. Therefore, after the second test inthe loop of SMODE, the branch (TC SKEEP1) is to ROPECHK. At the end of ROPECHK the program loops through SELFCHK changing SKEEP1 to the address of ERASCHK for the third pass. This alternate cycling of ERASCHK and ROPECHK continues indefinitely until the content of SMODE is changed by DSKY or an error is detected. In the event that an error is detected, the program stores in register SFAIL the address of the location following the location in SELF-CHECK that detected theerror, This address is also stored in the register ALMCADR for the ALARM routine. If ERASCHK is running, the program will also restore the contents of the erasable registers under test, The register ERCOUNT (set to+0 by DSKY FRESH START) is incremented and the ALARM routine is called. The ALARM routine turns on the program alarm light and loads into register FAILREG the alarm code for SELF-CHECK (octal 01102). The BBCON of SELF-CHECK is loaded into register ALMCADR +1 and returns control to the SELF-CHECK program. The contents of SMODE is then tested, followed by D, E or F, If SMODE is + Non-Zero, change the contents to+0 which puts the computer into the backup idle loop. If SMODE is - Non-Zero, start the option again from the beginning (at entry point SELFCHK). If SMODE is -0, branch on the contents of SFAIL to the location in SELFCHECK immediately following the location where the error was detected and proceed with the option from that point. Alarm Display: A SELF-CHECK error initiates program alarm by calling subroutine ALARM2 with C (A) = C(Q) = C (ALMCADR) = C GFAIL) and ERCOUNT incremented by one, The alarm code for self check error is 01102...
1-6
H. In the event that the check for a new job finds one waiting, the job will be executed and at the conclusion will return control to SELF-CHECK,. Since SELF-CHECK is runas part of the backup idle loop it cannot runas long as there are any active jobs.
1.2.5 Explanation of SHOW-BANKSUM (Figure 1.2.5-1) SHOW-BANKSUM consists of a routine called SHOWSUM. This routine
essentially does the same thing that the routine ROPECHK does; that is, add up the sum of separate banks inthe rope. After this the similarity ends, ROPECHK makes sure the sum of the bank is plus or minus its own bank number while SHOWSUM displays the sum of the bank in R1 of the DSKY irrespective of what the sum may be. SHOWSUM also displays the bank number and the bugger word in R2 and R3 of the DSKY at the same time. The sum of the bank and bank number in R1 and R2 are shownas the least significant bit instead of bits 11-15 (the actual bank bits in the computer),
Undoubtedly the greatest use of this routine will be in restoring the confidence of personnel in the computer and in verifying that the correct rope modules fora particular mission are actually the ones in the computer package. Following is a short description of the SHOWSUM subroutine.
Each bank in the rope is summed separately; from the lowest address to the highest address used in that bank. The contents of a higher address are added to the sum of the previous addresses, If this creates an overflow condition, a +1 is added to the new sum; a -1 is added to the new sum if an underflow condition is created, The sum of each bank should be plus or minus its own bank number. The sum of the bank is displayed in Rl of the DSKY. The bank number (actual bank number used to sum the bank cycled 5 places left) is displayed in R2 and the bugger word is displayed in R3. Entering a proceed verb (33) from the DSKY will display the same information for the next higher bank. Entering a terminate verb (34) from the DSKY will end the SHOWSUM routine.
1.2.6 ERASCHK (Figure 1.2.6-1)
This part of SELF-CHECK makes sure that it is possible to read a "1" and a "0" into and out of each bit position of erasable memory.
1-7
Load by DSKY V9IE
PY
SHOWSUM Turn off SELF- CHECK
----------
STSHOSUM (=ROPECH+K2)
initialize for first bank
Y
Sum one bank
Display by DSKY
R1 = sum of bank
R2
actual bunk number
R3 "bugger' `word
Go to next bank
Operator DSKY Action
V34E
Terminate Job
V33E Proceed
Yes Start SHOWSUM again
Fig. 1,2,5-1 Control of SHOW- BANKSUM
put +1 in SKEEP2 intialize to check EBANK zero
L
i
t
INHINT
put C(EBANK) in SKEEP 4
store C(X) and C(X + 1) in
SKEEP 5 and SKEEP 6
put C(SKEEP
i7) in ERESTORE
put addresses X and X+1 into registers X and X+1+
!
check that the sum of C(X)
NO
and the complement of C(X+1)
is -1
YES
put complement of addresses X and X+1 into registers X and X+}
check that the sum of C(X+1) and the complement of C(X)
NO
is -1
yes
restore original C(X) and C(X + 1)
put +0 in ERESTORE RELINT
check for new job
put C(SKEEP4) in EBANK
NO
Increment SKEEP7
YES
Check if thru checking bank or
unswitched erasable
ALARM | ERRORS
put +0 in ERESTORE
restore original C(X) and C(X+1)
put +0 in SMODE and idle go to the start of SELFCHK
erasable bank or unswitched erasable
checked last
bank checked last
unswitched erasable checked last
initialize unswitched
erasable memory
J
put +0 in SKEEP2
3 J +
.
- lin SKEEP
2 |
initialize EBANK for next higher bank
YES Check if EBANK2 is
next bank to be checked.
NO
initialize to check EBANK2
_ |
Check if EBANK7 has been checked
ves!
NO
go to CNTRCHK
initialize to check EBANK 1,3 4.5.6.7 |]
Fig. 1.2.6-1 ERASCHK
The RESTART program tests the contents of ERESTORE (the ERASCHK
activity indicator) before proceeding with RESTART.
The contents of
ERESTORE (set to +0 by any FRESH START) should be equal to the contents of
SKEEP7 (address of the first of the pair of registers under check by ERASCHK)
or equal to positive zero if no pair of registers are being checked, If the test determines that the contents of ERESTORE isnot "a positive number less than
2000 octal and equal to the contents of SKEEP7", the program switches to
DOFSTART (programmed FRESH START). The reason for the DOFSTART is
that the improper contents of register ERESTORE causes one to doubt the
contents of erasable memory, (The exception occurs when ERESTORE itself is
being tested,)
If the contents of ERESTORE are positive zero, do not restore erasable, proceed with RESTART. If the contents of ERESTORE are positive, less than 2000 octal, and equal to the contents of SKEEP?, then the original contents of the pair of registers under check are restored to those registers, ERESTORE is set to positive zero and the program proceeds with the RESTART.
The non-special erasable registers are checked for correct addressing and content by placing their own address in two successive registers and making sure thereisa difference of -1 when the contents of the lower address register is added to the complement of the higher address register; if it is not, this subroutine branches to the PRERRORS subroutine.
The previous contents of the erasable registers had been preserved and are restored to the two registers by PRERRORS which then performs a TC to the ERRORS subroutine.
If the difference is -1, the contents of the two registers are complemented and the complement of the lower register added to the contents of the higher register; the result is checked for-1. If the result isnot -1, TC to PRERRORS as noted above. If the result is -1, restore the previous contents to the two registers, and proceed tothenext iteration. The higher address register of the past iteration becomes the lower address register of the next iteration. The erasable memory banks are checked from zero through seven with common erasable (60-1373) being checked after each erasable bank.
CNTRCHK (Figure 1,2.6-2) The CS instructionis performed onall erasable registers from octal 60
through octal 10. These include all counters and other special erasable registers. It is not feasible to put their own address in these registers and
check their contents because of their special use.
CNTRCHK
put 00050 in SKEEP2 and A register
Decrement
\
SKEEP2 =]
add 00010 to c(A)
y
CS erasable addresses
60 through 10 octal
+ NON-ZERO
+0 go to CYCLSHFT Fig. 1,2,6-2 CNTRCHK
CYCLSHFT
put 25252 in CYR, CYL, SR, EDOP registers
a
add c(CYR), e(CYL),
c(SR), c(EDOP), anda
NO
constant and check that
result is -1
ERRORS
|
add c(CYR), c(CYL),
c(SR), c(EDOP), and
NO
+1 and check that result
is -1
increment SCOUNT +1
go to SMODECHK (put address of ROPECHK in register SKEEP1, check for new job and check register SMODE for SELF -CHECK option,
Fig. 1,2.6-3 CYCLSHFT
CYCLSHFT (Figure 1,2.6-3)
The octal number 25252 is placed in the two cycle registers, the shift right register, and the EDOP register. The contents of these registers are then twice checked for correct contents.
1.2.7 Check of Rope Memory (Figure 1.2,7-1)
The routine for checking the correct contents of a rope is called ROPECHK. Its purpose is twofold. First, it is a check on the computer. It makes sure all current drivers, sense amplifiers, and associated circuitry used in connection with the fixed memory are operating properly. Secondly, it is a check on the rope itself. It makes sure none of the sense or inhibit lines have become shorted or opened (essentially guarantees content of rope is correct and can be read correctly by the computer),
The sum of each bank should be the same as its bank number in the low order bits of the computer. A special word, which is called a "bugger" word, is added to the normal sum of the bank as the last word to be added, This "bugger" word forces the sum of the bank to be plus or minus the Bank Number, As an example, the sum of bank 33 octal may be 00033 or 77744, Two TC SELF words indicate the end of the summing process for each bank unless the bank is full, The "bugger" word immediately follows the second TC SELF word. If the bank is full, the "bugger" word is in the last address, and the two TC SELF words arenot necessary to indicate the end of the summing process for that bank. Of course, all addresses in a bank up to and including the "bugger" word have to contain words of good parity. Following is a short description of the ROPECHK routine.
Each bank in the rope is summed separately; from the lowest address to the highest address used in that bank. The content of a higher address is added to the sum of the previous addresses, If this creates an overflow condition, a +1 is added tothe new sum; a-1 is added tothe new sum if an underflow condition is created. The sumof each bank should be plus or minus its own bank number. If the sum of the bank is its bank number, the subroutine proceeds on to checking the next bank. If the sum of the bank is not its bank number, SELF-CHECK goes to the error routine, The banks are checked in ascending order.
ROPECHK
rs [oY
STIOW SEPM
put +] in SKEEP 6 put 10 in SMODKE, initialize SELFRET of SELRCHK
to address
i
set flag to check banks 00 and OL
common
fixed
|
initialization required
a common fixed bank
1
to check
!
add SUM of bank (check for new job between additions)
isfsum} of bank the same as bank number
NO YES
ERRORS
display (1) SUM of bank, (2) actual bank number and (3) bugger word in RI, R2, and R3 of the DSKY
wait for PRO
put +0 in SMODE and Idle
go to start of SELFCHK
has last bank YES
----
f
NO
common
what kind of bank is to be
checked next?
fixed J
fixed fixed
NO
is bank 02 next banks
to be checked
YES
is bank 04 next bank to be checked
YES
NO
set flag to check fixed fixed banks 02 and 03
1
initialization required to check banks 02
set flag to check rest of common fixed banks
initialize to check banks 03
CCS SKEEP 6
go to start of SELFCHK
start SHOWSUM again
Fig. 1.2,7-1 Check of Rope Memory
1.3 Prelaunch Alignment Program Computations
The stable member will be aligned at launch with the stable member Z axis vertical, down; the stable member X axis in the direction of the desired launch azimuth in the horizontal plane; the Y axis completes the right-handed triad.
The prelaunch alignment operation is depicted in Figure 1.3-1.
1.3.1 Initialization Data The initialization data required for the pre-launch alignment programis
the following: 1, Load vehicle azimuth (VAZ), scaled in revolutions, measured clockwise
from north to the Z axis of the spacecraft. Load latitude of test pad scaled in revolutions. 3. Loaddesired launchazimuth (AZ), scaled in revolutions measured clockwise from north. 4. Load azimuth of optical targets, scaled in 1/2 revolutions measured clockwise from north to target. 5. Load elevation of optical targets, scaled in 1/4 revolutions, measured from horizontal plane passing through the SXT focal point. Load performance parameters,
n
1.3.2 Coarse Alignment (P01)
The computations for coarse aligning the stable member as part of the
initialization of the prelaunch alignment are as follows:
1. Compute the desired stable member orientation matrix in a vertical, south
and east IMU-centered earth reference coordinate system,
0 -1/2(cos AZ) 1/2 (sin AZ)
[sua =
0 1/2 (sin AZ) 1/2 (cos AZ)
-1/2
0
0
2, Zero CDUs and wait 10 seconds.
Compute the navigation base orientation matrix ina vertical, south and east
IMU-centered earth reference coordinate system.
Ix wp|
1/2
0
0
0
1/2 (sin VAZ) 1/2 (cos VAZ)
QO -1/2 (cos VAZ) 1/2 (sin VAZ)
4. Given the|X.7/and|X, [orientation matrices in the vertical, south and east coordinate system, compute the CDU angles required to bring the stable
member into the desired orientation. (Ref. 5.6.3.2.2)
Load initialization data and initiate program
(1.3.1)
|
Coarse align stable member to:
Zou. vertical, down
Xy Level
along launch azimuth (1, 3. 2)
1
platform using
vertical erection prog.
(1.3. 3)
Is
Yes
liftoff or
backup liftoff discrete
present?
Have n* seconds
elapsed ?
On astronaut command
Align Xonq to the
do optical verification
azimuth using gyro-
computations
compassing program
|
{1. 3. 6)
(1.3. 4)
Transfer to Pll
Yes
Is liftoff or
backup liftoff discrete
present?
n* = 640 for initial vertical erection = 320 for vertical erection after azimuth change
astronaut commanded anew azimuth?
Torque to new azimuth (1.3.5)
Lo
Fig. 1.3-1 Prelaunch Alignment Program Operation
1-16
Ay
piY p
a ESAVMEPRLYE,5 I & SEZCERO
[xsm]
--_
MATRIX
"EAST" PIP
FaVN;
SE
| I
| ' ! t ' !
(om
| !
MATRIX
! { 1 \ ! I
ESVAEMRPYLE 5 &SE| CZERO O¢, Os
PALNAGTLFEORMTOREQRUREOR VECTOR
TGOYRRQOUSE
EARTH RATE COMPENSATION
LI-T
Fig. 1.3, 3-1 Vertical Erection Loop for Pre-launch Alignment
"SOUTH" Pip
[os
--
MATRIX
SEVAEMRPYL,ES &SEZCERO
"EAST" PIP
95
PLATFORM ERROR
by
(os
MATRIX
ANGLE TORQUE VECTOR
TORQUE GYROS
85 ta
ESVA-EMaPYLE 5 &S|EZCERO 6¢, 85
EARTH RATE COMPENSATION
j
Fig. 1.3,4-1 Azimuth Alignment Loop for Pre-launch Alignment
Transfer to Pll
Astronaut load new azimuth using extended
verb.
Rotate Xonug to the new desir ed azimuth,
Add azimuth change angle to the vertical gyro torque command,
Rotate Platform.
Change alignment program to vertical
erection mode.
Do vertical erection.
Is Liftoff discrete present ?
Have 320 seconds elapsed ?
Continue gyrocompassing
computations
Fig, 1,3.5-1 Azimuth Change Computation
5. Command IMU gimbals to the CDU angles using coarse align mode.
1.3.3 Vertical Erection Computation (P02) The vertical erection computations are depicted on Figure 1.3.3-1.
1.3.4
Gyrocompassing Computations (P02) The gyrocompass computations are depicted on Figure 1.3.4-1.
1.3.5
Azimuth Change Computations (P02) The azimuth change computations are depicted on Figure 1,.3,.5-1,
1.3.6 Optical Verification of Alignment (P03) The computation for calculating the stable member alignment error using
optical line-of-sight information should be according to the following steps: 1. Display pre-loaded azimuth and elevation data for Target 1 and Target 2. 2. Convert azimuth and elevation of Target 1 and Target 2 into two 1/2-unit
line-of-sight vectors L,. and Ly in the launch pad vertical, south and east coordinate system. 3. Compute the Target 1 and Target 2 line-of-sight vectors in the desired stable member coordinate system,
|Xsa| x Ly = T, referenced to Xsu desired
|Ssa x Ly = Ts, referenced to Xsm desired
4, Compute shaft and trunnion angles for Target 1 utilizing Ly and [sv] (Ref.
1.3.2).
5. Drive to the computed shaft and trunnion angles when optics mode is trans-
ferred to CMC control. 6. Wait for "mark" on Target 1.
From "mark" data compute T, line-of-sight vector in actual stable member
coordinates,
8. Compute shaft and trunnion angle for and IXnp| (calculated in 1.3.1).
Target
2. utilizing Ly
8. Drive to the computed shaft and trunnion angles when optics mode is trans-
ferred to CMC control,
10. 11.
Wait for "mark" on From "mark" data
Target 2. compute
--To/;
target
2 line-of-sight vector in actual
stable member coordinates,
1-20
12. Given Ty and Ts referenced to desired IXou | and Tana T, referenced to
actual IXsnal compute rotations required to
and |X._,| actual, (Ref, 5,6.3.2.4 and 5,6.3.2.3)
co-align [Xsaa| desired
13. Display the three rotations as gyro torquing angles since gyro torquing
sequence is Y, Z, X.
1-21
(1-22 is blank)
1.4 Prelaunch Alignment Functional Description Prelaunch alignment functional description is in Section 4, under P01, P02,
and P03.
1-23
(1-24 is blank)
1.5 Performance Test Computations
1.5.1 Gyro Drift Computation
The physical hasis for gyro drift measurement during prelaunch operations is the detection of the vector rotation of the gravity reaction acceleration, The IMU accelerometers provide thenecessary data. The data is corrupted by accelerations due to launch vehicle swaying motion and by quantization in the Pulsed Integrating Pendulous Accelerometer,
The effect of gyro drift on the vector rotation of gravity is small, therefore an optimum data processing method is required.
The datais processed byasimplified optimum filter which includes in its state vector estimates of the launch vehicle disturbances, The 13-dimensional state vector is described in Table I.
The simplified filter design recognizes that the gains for the optimum filter may be precomputed, since the measurement times will be the same for all trials and the a priori assumptions for the statistics of the initial state vector will not change.
The filter gains are precomputed by operating on a digital simulation of the system with a complete linear optimum filter. The gain functions are reconstructed piecewise in the CMC during the operation of the filter process using data loaded into the CMCerasable memory. The operation of the simplified optimum filter is depicted in Figure 1.5.1-1.
Figure 1.5.1.1 is a block diagram representing the following computations:
A. Measurement
The accelerometers are sampled every second. The sampled
accelerometer outputs are transformed to the vertical, north and
east
reference
AV. x
T
coordinate T
system.
AV v
T
2] av |xsm| =
AV
y
s
AV,
Av,
Where | xs | is the transformation matrix from vertical, south, east earth reference to stable member coordinates. The sign of the AV;is changed by AVs=-AV5
The measurements are used to update east velocity. It is corrected for disturbance.
estimates of south and the effects of wind
1-25
po, ==
Pow ==
AM, =
SM, =
w"~ e
. CAV, 9 (2°)
Po
.
CAV, (25°9)
4(Cov,) - PO,
4(Cyv,) - Po,
Cy= . 9.763768n3a3s
Cy = -0. 52223476
Filter gains The filter gains are pre-determined in the design process of the simplified filter, The gains are updated every second. The following gains are used.
1 . K, multiplies the total pulse counts from the accelerometers (po).
wn AR TaARpw OA NAe
2
multiplies the estimated east axis leveling angle (7).
3
multiplies the estimated, azimuth axis angle (a).
4,
multiplies the estimated vertical gyro drift (dx),
5.
multiplies the estimated north-south gyro drift (dy).
6
Zero.
7
wind induced sway velocity gain.
8
8 wind induced sway accelerometer gain.
For the first 30 seconds K, and K, have the form Ae thy
(see K,fig=ure0.9315,5,1~°--20)9.12t
Ky = 0.262e ~0.208t
The gains are modified at each sample as follows:
K,a, = K, [k,(0) . 93505870] Kya, = Ky [K,(0) . 26266423]
Ky, K,, K,, Ke are zero initially, then modified as follows:
Ky ta, = K,
Kytag, = Ky
K, + ag = Ky
Kg tag = Kg
The values of a, 7 a, are applicable over specified intervals. The values of a, 7 as and their applicable intervals are tabulated in Table 2.a is zero.
K 7 = 0.17329931
Kg =-0, 00835370
State vector update
The state vector variables are updated as follows: a+AM, (K,) za
B + AM, (K,) = 6 y + AMy (Ky) = ¥
+ AM, (K,) = pO,
po,
+ OM, (K,) = po,
Vv
+ 4M, (2K,) FV,
v
+ AM, (2K,) = v,
P
{ These parameters are updated during
launch vehicle parameter extrapolation
a,
+ 4M, (2Kg) za,
.
a,
+ AM, (2K,) = a,
4,
+ AM, (K,) = ay
4d,
+ AM, (K,) = 4,
D.
Extrapolation of launch vehicle parameters.
The launch vehicle parameters are extrapolated for the next
measurement using the following equations:
p(tn+1) =[C,p(tn) + Cyv(tn) + C,altn)] 2
v(tnt+1) =(C gpttn) + C,v(tn) + C,attn)] 2 a(tn+1) =(C,p(tn) + Cgvitn) +Cya(tn)] 2
Where the coefficients C of the transition matrix are: om = 0, 47408845
Cy = 0,23125894
C3 = 0. 14561689
Cy, =70, 06360691 Cy = -0,16806746 C, = 0,15582939
C, = -0.06806784 Cy = -0,75079894
fy = -0, 24878704
1-27 [x]Revised
E.
Calculation of the sines and cosines of alignment angles for ex-
trapolation of platform variables.
This simply involves computation of the sine and cosine of the various angles using the interpretive trigonometric routines in the CMC program,
The following functions are evaluated:
sina , sinp , siny ,
cos @ cos B cos ¥
F,
Extrapolation of stable member variables,
The Euler angles for aligning the stable member to the reference coordinates are computed as follows:
dx
sm" dy az
+ Yom
Ww, (Won is the angular velocity of the stable member)
cosB
0 -sing
cosy
siny
0
1
0
0
|¥snq| =
0
1
0
-siny cosy
0
0
cose sina
sing
0 cosf
0
0
1
0
-sine cosa
For vertical drift measurement Wow = Won + Ww,
y
¥y
y
a
cos
0 sin
cos ¥
cos ¥
é
= {sin y cos B
1 sin y cos 6
sm
cos ¥
cos y¥
y °
-si.n B
0 cosBp
BYa}
=
¥JaB
+
&4& | aT
ee (radians)
[x]Revised
G,.
Computation of estimates of velocity to be measured,
This computation adds to the previous value of south and east velocity the component of velocity expected due to the rotation with respect to gravity.
caPnO, . oPOO, + si. ny g p>, = p"Ao, + si.nB cos yg
1.5.2 Accelerometer Error Measurement The accelerometer scale factor and bias errors are determined by
comparing measured output with local gravity reaction acceleration, The accelerometer is aligned with gravity at the start of the measurement using the estimates of leveling error angles generated by the simplified optimum filter (1.5.1), Pulse rate from the vertical accelerometer is measured. (Figure 1.5.2-1). The pulse rate is converted to em/ sec? and displayed.
1.5.3 Gyro Torquing Scale Factor Measurement The computation of the gyro scale factor is performed by comparing the
number of gyro pulses required todrivea CDU through 22.5°to thenumber for the ideal scale factor. The result is then scaled for display in units of parts /million (ppm). The effect of CDU quantization (40 arc sec) is eliminated by starting the gyro pulse count at the receipt of a CDU bit and stopping at the receipt of the last bit. A gyro pulse corresponds to only approximately 0.62 arc secso this quantizationisnot important. Figure 1.5.3-1 shows the flow of these computations.
.
1-29 [xeyJRevisea
TABLE 1
Prelaunch Calibration State Vector Components
Azimuth Alignment Angle (a) South Axis Leveling Angle (8) East Axis Leveling Angle (y) South PIPA Velocity Increment (po,)
East PIPA Velocity Increment (po,)
Launch Vehicle Velocity; North-South wv.)
i
Launch Vehicle Velocity; East-West Ww.) Launch Vehicle Displacement; North-South (p,)
Launch Vehicle Displacement; East~West (p,)
Launch Vehicle Acceleration; North-South (a,)
11.
Launch Vehicle Acceleration; East-West fa.)
12, South Gyro Drift (dy)
13. Vertical Gyro Drift (dx)
1-30
pastaay| x]
roe
¥GLS
TIME (seconds)
0-30 31-90 91-100 101-200 201-450 451-790 791-1200 1201-1700 1701-2100 2101-2700 2701-3400 3401-4000
a 1 (Time Constant
PIPA Counts) 0, 91230833 0,99122133 0.99971021 0, 99550063 1 0,99673264 0.99924362 0,99963845 0,99934865 0.99947099 0.9995 7801 0.99966814 0,99972716
I
a. 2 (Time Constant Leveling Angles) 0,81193187 0,98940595 0.99852047 0,98992124 0,99365467 0,99888274 0,99913162 0,99868793 0,99894799 0,99916095 0,99933952 0,99945654
I
a 3 (Slope Azimuth Angle})| -0,.00035882 ~0,.00079010 0,00042697 0,00043452 0,00003767 0,00000064 0,00000090 0,00000055 0,00000018 0,00000007 0,00000002 0,00000001
i
a 4
a 5
(Slope
(Slope
Vertical Drift) | North-South
Drift)
-0,00000029
0,00013262
-~0,00000265
0,00043154
-0,00000213
0,00011864
-0,00000401
-9,00021980
-0,00002317
~0,00003305
-0,00004012
-0.00000195
-0.00002927
~0,00000026
0.00001183
-0,00000005
0,00000300
-0,00000001
0,00000096
0
0,00000028
0
0,00000010
0
I
l
Table 2
Time Constants and Slopes
Té-T
BUGEb 490
Ea
EE
EE
Cel
as
EE
Compute estimate of velocity to be measured
Z >
Extrapolate stable member variables according to platform dynamics for next measurement
Extrapolate launch vehicle parameters for next measurement
Calculate sines and cosines of alignment angles for extrapolation of platform variables
i
>
ce-T
Subtract extrapolated launch vehicle para-
meters
Incorporate current PIPA measurements in state vector
Measurement Sampled velocity incre-~
ments from PIPAs transformed into vertical south and east reference coordinate system.
Z r
Pre~load all filter gain calculation data
Reconstruct gains for
current
-time
measurement
Fig. 1.5, 1-1 Operation of the Simplified Optimum Filter
CURVE
A
8
v.28
0.24
0.20 1.0
0.16 0.8
GAIN 0.12 0.6
0.08 0.4
0.04 0.2
0.262e "0-208: kK (Curve A)
é0.935e79-0M%2t
2K) (Curve B)
|
!
4
8
12
16
20
TIME (SECONDS)
Figure 1,5,1-2
Gain Variation with Time,
1-33
60
40
20
K 5 (NORTH-SOUTH DRIFT)
--_--
GAIN (*10-3) ° -20 -40 -60 3)
--
K4 (VERTICAL DRIFT)
K3(AZIMUTH ANGLE)
50{ 0
7,0I00.
1,50| 0
2,00| 0.
2,5l00 3,0|00 3,500
TIME (SECONDS)
Figure 1,5,1-3 Gain Versus Tiine
1-34
Store the contents of the Scalar at time of occurrence of a AV from selected PIPA as Tl. Store PIPA counter contents as Pl.
Compute Earth Rate
correction for elapsed
time and correct plat-
form alignment
ein lat
-cos lat
2 9 0 47)
mz,,|
E) T |X, am
Tk
* [Exe| > [Exel *
Stcre the contents of Scalar at time of occurrence of a AV from selected PIPA as T2 store PIPA counter as P2
ge-T
Compute AP and AT P2 - Pi* AP T2 - Tl? AT
Compute g measured
*K = AoTP =g 7
--_--_--___--_TM DISPLAY
*K = Ideal Scale Factor x 3200 (Scaling Constant) #m = , 24339048 gyro pulses/10ms JE xe = gyro torquing error vector
|* `sm |
=
coordinate matrix
transformation
Fig, 1,5.2-1 Accelerometer
Error
Measurement
Computation
Initialize LGC giving gyro to be
tested, direction, N.B. orientation
and other data (See sec. 1,4, 2)
Coarse align then Fine align SM so test gyro IA is EAST (West)
Calculate gyro torquing required to compensate earth-rate,
E,,| sin lat
- cos lat m* = |
2°
0
wT) | E
+/|Erel
||
--
x sm| Vx
| |rE,e|
*
Start
Gyro Torquing
Save gyro pulse
counter
Exit. Disi play Alarm
1670
Zero CDU coun-
ter corresponding to gyro
under test.
Alarmt duirf i CDU
coun
ur:
RELINT
18
Alarm code 1660
Monitor CDU
Corresponding to
gyro under test for pulse, Interrupt inhibited.
Pulse
Load 22.5° into
CDU counter
Release i«nshyibsit every 160ms to prevent alarm
Torque gyro
2,8° even passes
Monitor for
2048 pulses
22. 5° in CDU
counter
odd passes
sth odd ° Pp: 238
Torque for earth rate comp.
CDU moved 22. 5°
Stop Test Alarm Alarm Code 1670.
Read gyro pulse count at end & compute number of pulses equal to 22.5°
Unreasonable result |
SFE >20 x 10° ppm
Compute SF Error (PPM) =
(Ideal no. of pulses) -(no. actually used))+ K
_ 22.5 ° x 3600 sec/id eB 1978 ~ ,617 sec/pulse
Display result in R1 in PPM.
R3 display position.
Repeat each
portion of the
test three times and operator average results
m* = 0, 24339048 gyro pulses/10 msec [Xml = coordinate transformation matrix
|E.o| = gyro torquing error vector
Fig. 1.5.3-1 Gyro Torquing Scale Factor Error Computation
1-36
1.6 Functional Description of Performance Tests 1.6.1 Gyro Drift and Accelerometer Error Test Description The gyroand accelerometer calibration program requires initialization of 166 erasable memory addresses prior to starting the test. The complete determination of the performance parameters requires repeat of the test 13 times, Eachrepeat test will reorient the platform with respect tothe following reference coordinate system: Xaxis - in the direction of local gravity Yaxis - south Zaxis - east The initialization data include constants for determination of filter gains (1.3.1), desired stable member orientation, and spacecraft latitude and azimuth. The initialization data must be pre-loaded for each of the 13 repeat tests. Each test is terminated witha FRESH START (V36) and assumes a FRESH START has been executed prior to its initialization. The following flow diagram provides a detailed description of the operation. (Fig. 1.6.1-1) 1.6.2 IRIG Scale Factor Test Description The stable member is positioned separately for each of six portions of the test. The CMC then positions the platform, torques the gyros, and computes the results without further operator action. The following flow diagram describes the CMC and ground/operator actions required. (Figure 1.6.2-1)
1-37
CMC OPERATIONS
OPERATOR OPERATION
Load K-Start tape with Initialization data
COMMENTS
Start program with VERB 92 ENTER
-- ________|
Initialize program. Set mode 07. Display latitude and azimuth, VERB 06 N41 Azimuth XXX,XX DEG Latitude + KX,XXX DEG
Ot Is Azimuth and Latitude correct ?
=
Calculate coarse align angles to position stable member to preloaded
orientation
NO
YES
VERB 33 ENTER
Load correct azimuth and latitude VERB 24 NOUN 41 ENTER Azimuth + XXX. XX ENTER Juatitude + XX, XXX ENTER
{
Fig.
1,6.1-1 Gyro Drift and Accelerometer Error Test Description
(continued on next page)
1-38
CMC
Operations
t
Coarse
align
t
gimbals
Do calculated gimbal angles
result in gimbal lock?
YET S
tNO
Change IMU mode to inertial
Wait ~ 225 seconds
Operator Operation
Observe NOATT light on DSKY
Comments
Presence of IMU or
CDU fail signal at this
,
time will result in auto-
matic test termination
07 will be blanked in
mode lights
Alarm code 01601 displayed
[
Sample IMU accelero-
meters every 1 second and estimate southerly gyro drift
!
Check for computation
overflow
No Overflow
Overflow Occurred
Hsaescon8d9s6 elapsed?
NO YES
v
(Continued on next page)
T01u6r0n0 on alarm Terminate test
Ovserve alarm determine , cause for system failure Terminate test
with V36E,
Possible causes of overflow are large initial alignment errors, (>5°) errors in initialization load or degraded accelerometers
Fig. 1.6.1-1 (Continued)
[x ]Revised
1-39
Oct, 196
c o
CMC
i Operations
Display south gyro drift
VERB 06 NOUN 98
Operator Operation
REIN R2: XXXXX R3: XXXXX
ERU Position code
Comments
Do I] wish to proceed to
accelerometer
ment
error
measure~
YES PROCEED
NO VERB 36
The normal test flow will proceed if conducting test positions 2,4,11,12
! Align platform to local
vertical using estimates of leveling errors computed by previous test section. Correct for earth rate errors
{ENTER
Test terminated
Alarm code 01601 will be displayed at this time if IMU or CDU fails are present at end of platform alignment
y
Determine rate of vertical acceler-
ometer pulses, Coarse align to O°, O°, O° after rate determination
y
Determine rate of vertical accelerometer
pulses
¥
Display measured
gravity
VERB 06 NOUN 98
R1:itXXXXX. R2: XXXXX
cm/sec ae
R3; Position Code
Load estimates of previously measured east-
' west drift.
(continued on next page)
Do I wish to proceed to vertical drift measurement
YES
NO
PROCEED |
VERB 36 ENTER Test terminated,
The normal test sequence will proceed if conducting test positions 2 and 4. Vertical drift measurement in positions 2 and 4. must be preceded by south gyro drift measurements in Positions 1 and 3.
Fig. 1.6. 1-1 (continued)
1-40
i CMC Operations
Torque platform
to move
Accelerometers out of
deadzone region (~0.36°)
Operator Operation
Comments
Sample IMU accelerometers every | second and estimate vertical gyro drift
Check for computation overflow
No Overflow
oul etton Occured
Has 3987] {Turn on
Observe alarm
seconds | jalarm 1600
L pee Determine cause for system
elapsed? Terminate test
I
failure. Terminate test with
VERB 36 Enter,
NO YES
Mc
earth
rate
caused
misalign-
ment in
south axis
only
Display VERB 06
NOUN 98
Rl:tXXXKX } ERU
R2: XXXXX
R3: XXXXX
SM POSITION CODE
TERMINATE TEST WITH VERB 36 ENTER
Fig. 1.6. 1-1 (continued)
1-41
Flow of IRIG SF Test CMC
Accept UPLINK data
S
Aap
Ground/ACE
Load KSTART Tape to initialize test. The following data is loaded:
L. Set flag to provide branch for required delay after set gyro torque enable relay.
2, Set flag to provide small increment of torquing (640ms) before start test.
3. Set count of earth rate torque passes co
zero,
Set index for CDU to be read. . Set flag to show direction to torque gyro.
Set indicator for gyro to be torqued. Initialize register to show no CDU pulse yet. Initialize so it will compensate for earth rate odd number times through. 9. Initialize a matrix which determines desired SM position. 10. Partially load the matrix for the Nav. Base position(remainder filled in by program based on N.B. azimuth and latitude. ) ll. Partially initialize matrix used wi calculation. 12. Constant for scale factor error calculation.
Enter V25N26E 04001E XXXXXE YYYYYE. (Where XXXXX = Starting Address and YYYYY = Contents of B Bank)
V30E
'
{continued on the next page)
Fig. 1.6.2-1 CMC and Ground/ Operator Actions
1-42
Flash V06 N41 with R1 = Azimuth
R2 = Latitude
Monitor Display
val
v22
V33E |V21E
V22E
Change azimuth
Clahtaitnugcele
Calculate sin and cos az. Store in matrix giving N.B. position. Calculate gimbal angles to align to desired position Zero ICDU's Coarse align
Command 360° about OA of gyro under test.
Change if desired V21E change azimuth V22¥E . change lati: tude V33E Proceed
Observe No ATT light on DSKY
Presence of IMU or CDU fail signal at this time will result in automatic test termination, Alarm Code 1650 will be displayed.
Fine align mode
(continued on next page) [ X]Revised
Fig. 1.6. 2-14 (continued) 1-43
Fine align to desired angles Calculate earth rate vector in sm coordinate. Set gyro torque enable Wait 20 ms Start gyro torque with POSMAX in gyro torque counter Wait 640 ms Zero CDU Counter
(continued on next page) Pig. 1,6,2-1 (continucd)
Check for CDU pulse
1 puise
more than | pulse
no pulse
Alarm Kxit Alarm Code
1660
Load 22.5° into CDU counter
160mrs withont interrupt
No Yos Check for higher priority gob
Monitor Alarm
Save contents of gyro torque counter
Lr
Sid
Torque gyro for 2. 8° and monitor for CDU counter = 0
CDU = 0
Even # exit CDU 7 0 odd # exit CDU 7 0
Compensate for.carth rate
oth odd pasLs ___]
EXIT Alarm
Fig.
Alarm Code
1670
1.6.2-1
(continued)
( Continued on next page ) 1-45
Save final contents of gyro-torque counter
Unreasonable number
of pulses (Ref, Fig, 1.5.3.1)
Alarm Exit Alarm Code 1670 Compute number of pulses
corresponding to 22. 5°
Compare to ideal number and compute scale factor error
Display VO6N98 Rl = SF error ppm R3 = gyro and torque direction
Record results of test
Terminate this position Resynchronize AG and CDU by FRESH START
Terminate this position Recycle for additional positions
Fig. 1.6. 2-1 (continued)
1-46
1.7 Performance Test Data Analysis
1.7.1 IRIG Scale Factor Data The data for each positionare displayed in Ri at the end of the running of
each position in units of ppm. The gyrounder test and the direction of torquing is displayed in R3 as follows:
+1
X gyro positive scale factor
-1
X gyro negative scale factor
+2
Y gyro positive scale factor
-2
Y¥ gyro negative scale factor
+3
Z gyro positive scale factor
-3
Z gyro negative scale factor
Plus SF error is displayed witha+ signin Rl. The scale factor is defined as0.61798096sec/pulse(1+SFE), The test should be run four times for each gyro in each direction and the results averaged. This is to smooth the effects of occasional 1 pulse irregularities in the CDU pulse rate.
1.7.2 Gyro Drift Data The model equation used for gyro drift is:
Wy ==Dpgt DSF), + Do(S=F)g + DoSF: g + DY (SaFm)yy2 + Dog SaFanGy2 + Doo SFA + Dyg (SF) ASF), + Dio SFSPg + DoglSPG(SFg
where subscripts I, S, and O refer to input, spin and output axes respectively.
Wy
gyro drift rate, defined as positive by the drift rate vector pointing
along gyro input axis.
bias or non-acceleration sensitive drift rate
NBD in Apollo nomenclature
drift rate proportional to specific force along input axis
ADIA in Apollo nomenclature
drift rate proportional to specific force along spin axis
ADSRA in Apollo nomenclature
drift rate proportional to specific force along output axis
ADOA in Apollo nomenclature
TE
drift rate proportional to specific force squared along input axis
1-47
D,
= drift rate proportional to specific force squared along output axis
Dee = drift rate proportional to specific force squared along spin axis
Ds = drift rate proportional to the product of specific force along input
and spin axes
Dio = drift rate proportional to the product of specific force along input
and output axes
Dos = drift rate proportional to the product of specific force along output
and spin axes
The gyro drift performancetest produces dataonthe NBD, ADSRA, ADIA and ADOA terms in the equation. The other terms are expected to contribute very little. The NBD, ADSRA and ADIA terms are the only ones compensated for by the in-flight gyro drift compensation program.
Position
Stable Member Orientation
Drift Equation (DH = Horizontal Drift; DV = Vertical Drift)
1
Xsuy DOWN
DA, = NBDY ~ ADOAY
You SOUTH
23
WEST
2
Xom DOWN
DH, = NBDZ - ADOAZ
You WEST
DV, = - NBDX + ADIAX
Zsnq NORTH
3
Sour SOUTH
DHg =+ NBDX - ADOAX
Your WEST
Zou DOWN
4
Xouy EAST
DH, = NBDY + ADSRAY
Yom SOUTH
DV, = NBDZ + ADIAZ
Zou DOWN
5
Xeon WEST
No drift data for this position
You UP
Zong NORTH
1-48
Position
10 11 12 13
Stable Member Orientation
SOUTH DOWN EAST
NORTH
UP-WEST
Zz
UP-EAST
EAST Y.,, UP-NORTH
UP-SOUTH
UP-EAST
Y.
UP-WEST
Zz
SOUTH
Xsm UP-NORTH
Y SM UP-SOUTH Zour EAST
x SM NORTH Yom WEST
Zou UP
Xom UP Yow SOUTH Zour EAST
Xouy UP Yom EAST Zou NORTH
Drift Equation No drift data for this position
DH. 7 = -NBDX + J2=1 ADSRAXo-fS1> ADOAX
DH, 8 = V2L (-NBDZ - NBDY)
+ 1/2 (ADIAZ - ADIAY)
+ 1/2 (ADSRAY + ADSRAZ)
DHg =
©
S e
ADSRAZ - NBDZ- 1 ADOAZ 2
D #10 DH, Y
L.A (NBDY - NBDX)
+ 1/2 (ADIAY - ADIAX)
+ 1/2 (apsRAx)+ 1 ADOAY 2
NBDX - ADOAX
DHy9 = NBDY + ADOAY
DH, 3 = NBDZ + ADOAZ
1-49
The equations for compensable drift terms in terms of the horizontal and vertical drift measurements are:
NBDX NBDY NBDZ ADSRAX ADSRAY ADSRAZ ADIAX ADIAY
ADIAZ ADOAX ADOAY ADOAZ
1/2 (DH .3 ~ DH,,11) 1/2 (DH, + DH, 4)
1/2 (DH, + DH,3)
/2 (DH, + 1/2 (DH 3 7 DH, "5 (DH,, + DH,)
DH, ~ 1/2 (DH, + DH 19)
V2 [DHg + 1/2 (DH, + DH,,)] + ; (DH,, - DH,)
PV2 * M2 (DH "PA?
2DH,,10-
+ DH, +
fa=2
DV,
DH,-1 JB=2
- V2DH,
DH: 12
- 5 (DH,
DV, - 1/2 (DH, + DH)
- DH 12)
- 1/2 (DH, 4 + DH) (Not compensated)
1/2 (DHy2- DH 1 ) (Not compensated)
1/2 (DHy3 - DH,) (Not compensated)
1.7.3
Accelerometer Scale Factor Error and Bias Error Data The complete accelerometer model equation is:
Specifapisc Force Indi. cated =_ Ag+ AASF= ), + Ap(SF), + AG(SF)g + Ay (SF) 2 + App(SF) (SF), + Ajo (SF) ASP), + Apo (SF) pGF)o
where subscripts I, P, and O refer to input, pendulous and output axes respectively,
Ap,Ag AT Arp
bias coefficient, insensitive to specific forces
scale factor of instrument
cross coupling coefficients
specific force squared coefficient coefficient for the product of specific force along input and pendulous axes
1-50
Alo
=
coefficient for the product of specific force along input
and output axes
Apo
=
coefficient for the product of specific force along pendulous
and output axes
The accelerometer test data are used to determine only the bias and scale
factor coefficients.
The other terms are not separately measured or
compensated,
The simplified equation for the accelerometer model is:
Specific Force Indicated = Bias + Scale Factor ( Specific Force along input axis )
The specific force used in the test is due to the gravity reaction acceleration. The comparison of the indicated magnitude of the gravity reaction acceleration and the known local gravity provides the calibration of the accelerometer. The scale factor error and bias are separated by reversing the direction of the specific force along the input axis,
For the X and Z accelerometers the orientation of the input axis parallel to the direction of local gravity is easily accomplished by use of the data from the other two accelerometers, For the Y accelerometer the gimbal configuration does not allow accurate positioning, therefore data from the other two accelerometers is used in the data analysis to correct for input axis alignment
errors,
Position 12
2
Stable Member Orientation
Xou UP Yon SOUTH Zou EAST
Xgnq DOWN
Your WEST Zour NORTH
Accelerometer Error Equation
Bm1 7 Py t (1 - SFE)
|
= b, + (1 - SFEX-g)
aa
Xong NORTH
Yom WEST
Zou UP
b, + (1 - SFE)g
|
[x]Revised
4
Xgny EAST
Ema = b, + 1 - SFEM-g)
Yom SOUTH
Zou DOWN
5
Xu WEST
Ems = by + (1 - SFE)g
You UP
Zon NORTH
6
Xony SOUTH
m6 = by + (1 - SFE)-g)
You DOWN
Zou EAST
fy = measured gravity reaction acceleration (em/sec") calculated using ideal scale factor of 5.85 cm/sec/pulse
g = local gravity reaction acceleration (cm/sec")
b
= bias of i accelerometer (em/sec") i=x, y, z
SF, = scale factor of i accelerometer in cm/sec/pulse
SFE = scale factor error in parts-per-million defined as positive
when SF > ideal scale factor
For positions 5 and 6 the misalignment angle 8. between the Y accelerometer and the vertical shall be determined from pulse rate data from the other two accelerometers.
a y = fex +
@ = (AV, x - AV_xB_)x SF. x
Z
AT ¢ local
0 =
x
(AV_ z - AV_BZ_z ) 2 SF Zz
AT g local
where AV = number of velocity increments accumulated in AT
Em5,6 will be modified by the misalignment ay as follows:
Bm51 * 8ms 8° 9y5
&m6! ~ me 8&¢ 86 Emst and Emer are used to determine error coefficients.
Y accelerometer
scale
factor
and bias
The equation for calculating scale factor error for the accelerometer ts:
SFE, es 1- 8mj TZ~ l8omc(ajl+1) x 10° 6 ppm
The equation for determining bias error for the accelerometer is:
by = --mgoj._+ -mg Gt: ) em/sec"
R-577
COLOSSUS Section1
Internal:
P, Adler (2) R. Battin E, Blanchard G, Cherry E. Copps S. Copps G, Cox W. Day G,. Edmonds S, Eliassen A. Engel (2) P, Felleman J. Fleming L. Gediman (15) K. Glick K. Goodwin E. Grace K. Greene M, Hamilton P, Heinemann J. Henize D. Hoag B, Ireland T. Isaacs L, B, Johnson
M, Johnston J. Kernan K. Kido * J, Kingston A. Kosmala W. Kupfer A, Laats L, Larson R,. Larson J, Lawrence D, Lickly W. Marscher F, Martin R. McKern Vv. Megna J.E, Miller J.S. Miller P. Mimno J. Nevins J, O'Connor W. Ostanek J, Parr P, Philliou R, Ragan K, Riebesell
*Letter of transmittal only.
P. Rye J. Sapanaro P, Sarda W. Schmidt C. Schulenberg N. Sears J. Shillingford B. Sokkappa W, Stameris G. Stubbs M, Sullivan J, Suomala J. Sutherland W. Tanner R. Tinkham J, Turnbull K. Vincent J, Vittek R. Weatherbee R. Werner P. White R, White W. Widnall M. Wombie Apollo Library (2) MIT/IL Library (6)
MIT Instrumentation Laboratory c/o North American Rockwell, Inc. Space and Information Division 12214 Lakewood Boulevard Downey, California 90241 Attn: Mr. Thomas A, Hemker
MIT Instrumentation Laboratory G&N Systems Laboratory c/o Grumman Aircraft Engineering LM Project - Plant 25 Bethpage, Long Island, New York Attn: Mr, James A. Hand
Corp.
MIT Instrumentation Laboratory P.O. Box 21025 Kennedy Space Center, Florida Attn: Mr, George Silver
32815
MIT Instrumentation Laboratory
(3)
Code EG/MIT Building 16
NASA Manned Spacecraft Center
Houston, Texas 77058
Attn: Mr, Thomas Lawton
(10)
NASA MSC HW Building M7-409 Kennedy Space Center, Florida Attn: Mr. Frank Hughes
32815
Mr. A. Metzger (NASA/RASPO at MIT/IL)
AC Electronics Division
General Motors Corporation
Milwaukee, Wisconsin
Attn: Mr, J. Stridde, Dept. 32-31
(13)
Attn: Mr. Reino Karell
(2)
Kollsman Instrument Corporation 575 Underhill Boulevard Syosset, Long Island Attn: Mr, F. McCoy
Raytheon Company
(6)
Boston Post Road
Sudbury, Massachusetts 01776
Attn: Mr. J, Shrmack
NASA/RASPO/National Aeronautics and Space Administration
NAR
Resident Apollo Spacecraft Program Office
North American Rockwell, Inc.
Space and Information Systems Division 12214 Lakewood Boulevard
Downey, California
NASA/KSC
National Aeronautics and Space Administration
(3)
John F. Kennedy Space Center
J. F. Kennedy Space Center, Florida 32899
Attn: Technical Document Control Office
NASA/RASPO NASA Daytona Beach Operation
a)
GE
P.O. Box 2500
Daytona Beach, Florida 32015
Attn: Mr. A. 8. Lyman
NASA/HDQ
NASA Headquarters
600 Independence Avenue SW Washington, D.C, 20546
Attn: MAP-2 Attn: Mission Director, Code MA Attn: Robert Aller, Code MAO
(6) (4) (1) (1)
NASA/AMES
National Aeronautics and Space Administration
(2)
Ames Research Center
Moffett Field, California
Attn: Library
NASA/LEWIS National Aeronautics and Space Administration
(2)
Lewis Research Center
Cleveland, Ohio
Attn: Library
NASA/FRC
National Aeronautics and Space Administration
(1)
Flight Research Center
Edwards AFB, California
Attn: Research Library
NASA/LRC
National Aeronautics and Space Administration
(2)
Langley Research Center
Langley AFB, Virginia
Attn: Mr. A. T, Mattson
NASA/GSFC
National Aeronautics and Space Administration
(2)
Goddard Space Flight Center
Greenbelt, Maryland
Attn: Mr, Paul Pashby, Code 813
GAEC:
GBertuhmpamgaen,
Aircraft Engineering Corporation Long Island, New York
(2)
Attn: MMrr.. J,R. MParraitnto
((11))
NAR
North American Rockwell Inc.
Space and Information Systems Division
12214 Lakewood Boulevard
Downey, California 90241
Attn: Apollo Data Requirements
Dept. 096-340
Building 3, CA 99
(1+ 1R)
NGAASECA/RASPO/ NGBRaeerttsiuihomdpnemaangalten,AAepAroLiolorlncnoarguatfStiIpscalsacEnendcg,arinadnfeNteeSrwpiaPncrgeYoogrACrkdoamrmipnoirOsafttfriiacotenrion
NASA/RASPO/ ACED
National Aeronautics and Space Administration Resident Apollo Spacecraft Program Officer Department 32-31 AC Electronics Division of General Motors
Milwaukee, Wisconsin 53201
Attn: Mr. W. Swingle
NASA/WSMR:
National Aeronautics and Space
Post Office Drawer MM
Las Cruces, New Mexico
Attn; RH4 Documentation
Administration
NASA/MSFC:
National Aeronautics and Space Administration
George C, Marshall Space Flight Center
Huntsville, Alabama
Attn: J, Mack R-ASTR-5
a)
H, Ledford R-AERO-P
(2)
L. McNair R-AERO-P
qQ)
T. Telfer R-AERO-DA
(2)
E. Deaton R-AERO-DA
i)
J. Cremin R-AERO-DAP
(1)
L. Stone R-AERO-F
a)
C. Hagood R-AERO-F
Q)
O, Hardage R-AERO-FM
a)
F. Moore R-ASTR-N
qa)
S. Seltzer R-ASTR-NG
(5)
H. Hosenthien R-ASTR-F
Q)
A. McNair I-MO-R
Q)
D, Germany I-I/IB-E
qQ)
R. Barraza I-V-E
a)
W. Chubb R-ASTR/NG
Q)
J. McCullough I-VE/T
(i)
NASA/MSC:
National Aeronautics and Space Administration Manned Spacecraft Center Apolle Document Control Group (PA 2) Houston, Texas 77058 Attn: A. Alber, FS5 (letter of transmittal only)
Q1) (1) (2)
(23)
(280 +2R)
BELLCOMM: Bellcomm, Inc.
(6)
1100 17th Street N. W.
Washington, D,C, 20036
Attn; Info, Analysis Section
LINK:
LINK Group, GPSI SIMCO
(3)
1740 A NASA BLVD
Houston, Texas 77058
Attn: Mr. D, Klingbell
TRW:
Gilbert H. Friedman
(5)
Bldg 82 Rm 2067 TRW System Group
One Space Park
Redondo Beach, Cal, 90278
pikepdf 1.10.3 dfsg