Digital_Computer_Graphics_Volume_1_1967 Digital Computer Graphics Volume 1 1967
Digital_Computer_Graphics_Volume_1_1967 Digital_Computer_Graphics_Volume_1_1967
Digital_Computer_Graphics_Volume_1_1967 The Eye | File Listing
User Manual: Digital_Computer_Graphics_Volume_1_1967
Open the PDF directly: View PDF 
.
Page Count: 404
| Download | |
| Open PDF In Browser | View PDF |
COMPUTER PROGRAMMING SERIES
DIGITAL
COMPUTER GRAPHICS
VOL. I
PUBLISHED BY
AMERICAN DATA PROCESSING, INC.
19830 MACK AVENUE
DETROIT, MICHIGAN
©Copyright by American Data Processing, Inc.
PREPARED UNDER CONTRACT NO. NAS 8-21227
By LOCKHEED MISSILES & SPACE COMPANY,
SUNNYVALE, CALIFORNIA
ABSTRACT
This document represents a study of data collection, application analysis, and user
requirements for the following five computer graphics applications:
• Numerical Control
• Electrical Network Analysis
•
Flight Mechanics
• Structural Analysis
•
Engineering Drawing Retrieval
A discussion of "study of Man-Machine Solution Optimization" and a hardware analysis
report of the MSFC Univac 1108 computer complex using Univac specifications are
also presented.
The results of 'a Computer-Aided Design/Computer Graphics literatu~e search is
compiled into an annotated bibliography as Volume II of this report. Conclusions and
recommendations in respect to computer graphics and the various areas of computer
graphic applications are summarized.
This study was prepared by Lockheed Missiles & Space Company under COJ;1tract
NAS 8-21227, "Digital Computer Graphics, " for the National Aeronautics and Space
Administration, Marshall Space Flight Center.
iii
ACKNOWLEDGMENTS
The following LMSC personnel made major contributions to this document:
P. W. Card
D. L. Drew
W. R. Dunn
S. Elster
S. K. Ferriera
K. R. Gielow
J. F. Golden
F. S. Kessler
M. A. Krop
D. C. Miller
R. L. Moll
N. E. Nielson
W. C. Randels
C. J. Skogh
P. L. Taulbee
D. L. Trask
Significant contributions were also provided by D. A. Biederman of Lockheed
California Company and M. M. West of Lockheed Electronics Company.
iv
CONTENTS
Section
Page
ABSTRACT
ACKNOWLEDGMENTS
iii
iv
ILLUSTRATIONS
ix
TABLES
xiii
INTRODUCTION
1
2
NUMERICAL CONTROL
xv
1-1
1.1 Numerical Control (N/C) and Computer Graphics
Involvement in the" II.ldustrial Process
1-2
1.2 F~asibility of N/C - Computer Graphics at MSFC
1-16
1.3 Numerical Control and N/ C - Computer Graphics
Specifications
1-20
ELECTRICAL NETWORK ANALYSIS .
2-1
2.1 Summary
2-2
2.2 Network Analysis Program Requirements for the
Graphics System
2-2
2.3 Description of Available Network Analysis Programs
2-4
2.4 Program Evaluation
2-11
2.5 Program Selection
2-13
2.6 Program Modifications
2-13
2.7 User Graphic Procedures
2-15
2.8 Network Analysis Graphics Applications at Lockheed
2-21
2.9 Network Analysis Graphics Applications at Other
Companies
2-22
v
Section
3
4
5
Page
2.10 Programming Requirements
2-23
FLIGHT MECHANICS
3-1
3.1 Mode of Application of Graphics
3-1
3.2 Graphical-Computer System for Orbital Transfer
Studies
3-4
3.3 Requirements Specification for Orbital Transfer
Studies
3-15
3.4 Graphical-Computer System for Ascent Boost
Optimization.
3-29
3.5 Requirements Specification for Ascent Boost
Optimization
3-50 .
3.6 Conclus ions and Recommendations
3-55
3.7 References
3-56
STRUCTURAL DESIGN ANALYSIS
4-1
4.1 Background
4-1
4.2 Introduction to Specifications
4-17
4.3 Program Requirement Specifications
4-19
4.4 Hardware/Software Requirements
4-32
4.5 Program Specifications
4-32
4.6 PEANUTS: An Example of Graphic Program Requirements and Specifications
4-70
4.7 References
4-87
ENGINEERING DOCUMENT RETRIEVAL
5-1
5.1 Introduction
5-1
5.2 System Requirements
5-2
5.3 Storage and Retrieval Systems
5-20
5.4 Recommendation
5-37
5.5 Hardware State-of-the-Art
5-78
5.6 References
5-83
vi
Section
6
7
Appendix
Page
STUDY OF MAN/MACHINE SOLUTION OPTIMIZATION
6-1
6.1 Introduction
6-1
6.2 Description of Program
6-2
6.3 Experiment Data and Operations
6-4
6.4 Instructions
6-22
6.5 Suggestions for More Efficient Division of Labor
Between Man and Computer
6-23
6.6 Simulation
6-24
HARDWARE ANALYSIS
7-1
7.1 Introduction
7-1
7.2 Hardware System Configuration
7-1
7.3 Software System
7-3
7.4 Display Operations
7-4
7.5 Conversion Considerations
7-6
GLOSSARY
A-I
vii
ILLUSTRATIONS
Figure
Page
1-1
Symbols Defined
1-38
1-2
Subroutine START
1-39
1-3
Subroutine POINT
1-40
1-4
Subroutine LINE--POINT-TO-POINT
1-41
1-5
Subroutine LINE-POINT. DIRECTION
-1-42
1-6
Subroutine CIRCLE
1-43
1-7
Subroutine PARALLEL
1-44
1-8
Subroutine FILLE T
1-45
1-9
Subroutine RETOUCH
1-46
1-10
Subroutine LINE TYPE
1-47
1-11
Subroutine TRANS FORMA TION
1-48
1-12
Subroutine NUMERICAL CONTROL
1-49
1-13
Subroutine DIMENSIONING
1-51
1-14
Subroutine MACRO
1-15
Local Area Layout of Display-Oriented Support Equipment
1-82
2-1
Typical Component Templates
2-17
2-2
Completed Schematic
2-18
2-3
Proposed Control Surface Layout
2-24
2-4
Proposed Keyboard Layout
2-26
2-5
Sample Component Group Structure
3-1
Orbital Transfer Display Scope
3-9
3-2
Ascent Boost Optimization REVIEW Display
3-39
3-3
Ascent Boost Optimization MONITOR Display
3-45
4-1
,PEANUTS - Structural Design
4-2
Data Input Display
- 1-52
Evolut~on-
·2-32
4-12
4-23
ix
Page
Figure
Communicat~on
and Data Display Layout
4-28
4-3
Rectangular
4-4
Square Communication and Data Display Layout
4-28
4-5
Shifted Square Communication and Data Display Layout
4-29
4-6
Communication Message Area Layout
4-29
4-7
Frame Types
4-31
4-8
Application Data Base Structure
4-34
4-9
Data Base Access Program
4-35
4-10
Resident Name Table (RNT)
4-37
4-11
Program ID Table (PIDT)
4-12
Case ID Table (CIDT)
4-38
4-39
4-13
Data Requirement Table (DRT)
4-40
4-14
Program Data Reference Table (PDRT)
4-42
4-15
Sample Data Set Format
4-43·
4-16
Library Program
4-45
4-17
Sign-on Program
4-46
4-18
Stop ProgIam
4-47
4-19
Program Selection Program
4-49
4-20
Program Next Case
4-50
4-21
Input-Output Flow for Card, Magnetic Tape and Data Base
4-51
4-22
Input Data Flow for Cathode-Ray Tube
4-52
4-23
Input Program
4-53
4-24
Card Input Program
4-55
4-25
Tape Input Program
4-56
4-26
Data Base Input Program
4-57
4-27
CRT Input Program
4-58
4-28
Compute Program
4-61
4-29
Standard Output Program
4-63
4-30
General Output Program
4-64
4-31
Void Output Data Program
4-65
4-32
Cathode-Ray Tube Layout
4-68
x
Figure
Page
4-33
Functional Block Diagram of an Ideal Structural Program
4-71
4-34
Functional Block Diagram of PEANUTS
4-72
4-35
Function Input
4-81
4-36·
Single Item Input
4-82
4-37
Standard Output Display
4-86
5-1
Documentation Repository Input
5-7
5-2
Engineering Documentation Released by MSFC Release Center
and Forwarded to Central Repository
5-8
5-3
Documentation Repository Output
5-14
5-4
Program Distribution of Manpower at MSFC
5-17
5-5
AAP Documentation Processed by MSFC Engineering Release
Center Between June and December 1967
5-19
5-6
Documentation Flow
5-21
5-7
Sample of MSFC Form. 2896 - Document Input Record
5-27
5-8
Sample of MSFC Form 433 - Request for Documentation
5-28
5-9
Sample of MSFC Form 2598 - Distribution Data Acquisition List
5-29
5-10
Release and Accounting Procedure Flow
5-31
5-11
Saturn SA-5 Main Assemblies
5-43
5-12
Breakdown of Saturn SA-5 Payload and Stage I Assemblies
5-45
5-13
Logical Flow Chart Segments SL1, SL2
5-14
Logical Flow Chart, Segment S
5-58
5-15
Logical Flow Chart, Segment D
5-60
5-16
Logical Flow Chart, Segment E
5-61
5-17
Logical Flow Chart, Segment U
5-63
5-18
Logical Flow Chart, Segment I
5-64
5-19
Logical Flow Chart, Segment A
5-66
5-20
Logical Flow Chart, Segment DL
5-68
5-21
Logical Flow Chart, Segment R
5-70
5-22
Logical .Flow Chart, Segment IN
5-72
5-23
Typical Display - Interface Structure
5-73
5-24
Typical Display - Digitized Drawing
5-74
xi
·5-56
Page
Figure
5-25
Typical Display - Structure Breakdown
5-75
6-1
Flow Diagram Used for the 144 Experiment
6-3
6-2
Complete Data Display
6-6
6-3
Display of Data Before Acceptance Into Data Set
6-7
6-4
Display of Functions of 10 Best Curves
6-8
6-5
Display of Best Curve
6-9
6-6
Display for User to Select a Function, Transform and Degree
6-12
6-7
Display of Chosen Curve From Box I, Fig. 6-1
6-13
6-8
Run 1, 144 Curve Fit Program, Preplot and Curve
6-16
6-9
Run 1, Computer Output
6-17
6-10
Run 2, 144 Curve Fit Program
6-18
6-11
Run 2, Computer Output
6-19
6-12
Run 3, 144 Curve Fit Program" Preplot and Curve
6-20
6-13
Run 3, Computer Output
6-21
7-1
Hardware Configuration
7-2
7-2
Display Console With Permanent and Programmed Controls
Shown
7-5
xii
TABLES
Page
Table
1-1
Console Procedures for Start (Keyboard Oriented)
1-21
1-2
Console Procedures for Numerical Control
1-22
1-3
Console Procedures for Erase
1-24
1-4
Console Procedures for Line Type
1-24
1-5
Console Procedures for Fillet
1-25
1-6
Console Procedures for Retouch
1-25
1-7
Console Procedures for Parallel
1-26
1-8
Console Procedures for Circle
1-27
1-9
Console Procedure s for Point
1-28
1-10
Console" Procedures for Line Through Point at Direction
1-28
1-11
Console Procedures for Line Point to Point
1-30
1-12
Console Procedures for Dimensioning Information
1-31
1-13
Console Procedures for Geometry Macro
1-34
1-14
Console Procedures for Transformation
"1-36
2-1
Network Analysis
2-2
Network Analysis Program Evaluation for Graphics
2-14
3-1
Sample Output of Orbital Transfer Program
3-19
3-2
Timing and Storage Requirements for Orbital Transfer System
3-23
3-3
Light Button Actions for Orbital Transfer System
3-27
3-4
Storage and Timing Requirements for Ascent Boost Optimization
3-53
4-1
Representative Listing of Structural Analysis Computer Programs
4-2
4-2
List of Input Requirements
4-22
4-3
Command Table
4-66
4-4
Register Table
4-69
4-5
Description of Input Quantities
4-73
Program~
2-5
xiii
Page
Table
4-6
List of Possible Output From PEANUTS
4-79
5-1
Summary of R&D and 10 Usage
5-13
5-2
Six Most Active Requestors - October 1967
5-15
5-3
Machine Readable Documents
5-77
6-1
Time Elements
6-11
7-1
Hardware Comparison - CDC 273 vs. Univac Mariner Display
7-7
xiv
INTRODUCTION
This country's scientific and technical environment increases in proportion to the design and complexity of our space program hardware. The resulting data which represent research, development, and design effort have become less accessible because of
the sheer mass of informati9n that must be stored, manipulated and retrieved. As a
result, the data originator is literally separated from the information he needs and the
cleavage increases as the data collection grows.
With the advent of on-line techniques and direct access devices the problem is diminished, but only in a limited manner in the area of alphanumerics, i. e., the advisability
of using on-line devices to examine mathematical equations -is questionable when the
changed data output must be transferred to hard copy and compared against graphic
representations. This is a time-consuming
i~erative
process. Graphics computer in-
put devices that allow this iterative process on-line in an interactive', 'real-time mode
are the keys to meaningful use of the computer as a tool to support the originators of
information. Data capture at its initiating source rather than at
th~
first paper
~is
closure as is presently inferred by the term "source data" is a prime goal in computer
graphics. Some of the same benefits 'and advantages which are only promised in nongraphic, computer-aided design can actually be realized with the addition of graphics.
These benefits include:
•
Ability to observe the results in graphic form without having to resort to
hard copy output
•
Immediate program interaction capability
•
Ability to revise data directly without a manual data reduction step by allowing users to work with problems in the graphics environment
•
Ability to drive other output devices directly and produce hard copy or program tapes to operate such units as numerically controlled machine tools
xv
Computer graphics at Lockheed has been conducted primarily on Lockheed independent
development funds. While Lockheed interest in computer graphics dates back to 1958,
the
p~esent
operation, which began in 1962 at LMSC, is an outgrowth of an attempt to
improve the design effort through computer graphics utilization.
Most of the early effort was concentrated on system evaluation and use of a printed
circuit design program as the first application of software. Subsequent efforts have
included such applications as reentry vehicle deSign, logic design, structural dynamics,
and trajectory analysis.
Lockheed-Georgia Company and Lockheed-California Company have been actively engaged in this area on both independent and joint cooperative ventures within LAC. Lockheed Electronics Company (LEC) is examining applications presently relative to entry
into graphics. Lockheed-Georgia Company
~as
research-oriented applications in pro-
gress, both in computer-aided design and in computer graphics.
The computer graphics
programs include structural analysis, electrical circuit layout, and numerical control
center path development for the operation of numerically controlled machine tools. At
present, parts are being fabricated for the C5A program from input.using a computer
graphic terminal. Applications have been developed for both aircraft and shipbuilding
operations. Lockheed-California Company has four major areas of computer graphics
applications: design, design analysis, flight test, and numerical control of machine tools.
This report is segregated into eight major areas:
•
Five computer graphic applications (see below), discussion of data collection
and the application analysis, and a statement of user requirements ( a requirements specification) for each application:
1.
Numerical Control
2.
Electrical Network Analysis
3.
Flight Mechanics
4.
Structures Analysis
5.
Engineering Document Retrieval
xvi
o A "discussion of "Study of Man.,..Machine Solution Optimization"
•
A hardware analysis report of the MSFC Univac 1108 computer complex in
MSFC Computation Laboratory taken from the Univac specifications supplied
to LMSC especially for this study
e An LMSC Computer-Aided Design/Computer Graphics literature search bound
separately as Volume II of this report
Each of the five computer graphic applications contains terms which are common to all
along with special terms peculiar to the application. These are grouped together as a
glossary for the total report in the appendix.
When applicable there is a summary of conclusions and recommendations, in each of
the applications described herein.
The computer graphic hardware referred to in this report (and upon which all program
applications are based) need not be of any specific manufacture or type, but generally
consists of, as a minimum:
•
A basic computing machinery (mainframe)
•
A cathode-ray tube with a light-sensitive probe (light pen or pencil) with which
the user points at the CRT and manipulates graphic lines o;r alphanumeric data
plus certain auxiliary input devices such as a "button box" or keyboard (both
of which mayor may not be capable of input of alphanumerics or special
graphic symbols and commands)
•
Peripheral hardware for input and output such as a card punch, card readers,
high speed printers, collators, sorters, tape drivers, etc.
The main consideration in all of the following is that the least complex hardware array
allows the user/operator direct interaction with his program which is available as a
CRT display.
xvii
Section 1
NUMERICAL CONTROL
This section covers a portion of NASA-MSFC study contract NAS 8-21227. With deletions specified during negotiations, it is responsive as described in the LMSC proposal
No. 699516, dated February, 1967.
The firms from which information for this study was gathered are:
o General Electric Industrial Process Control Division
G
Gerber Scientific Instrument Company
o International Business Machines Corporation
o
Lockheed Georgia Company
o
Lockheed California Company
o Sundstrand ·Machine Tool Division of Sundstrand Corporation
o Bendix Industrial Controls Division
o Cincinnati Milling Machine Company
o Control Data Corporation
o Pratt & Whitney Machine Tool Division
•
McDonnell - Douglas Aircraft Corporation
'" Bunker Ramo Woolridge Company
•
TRW, Incorporated
•
Boeing Aircraft Company
•
Kearney and Trecker·
•
Universal Drafting Machine Company
Society affiliations employed in this effort include the following:
•
Aerospace Industrial Association
•
Society of Automotive Engineers
•
American Ordnance Association
•
American Institute of Aeronautics and Astronautics
1-1
Another source of information was the illinois Institute of Technology - Research.
Institute.
This document is intended to be comprehensive enough to facilitate successful implementation of Computer Graphics - Computer Aided Design - Numerical Control as one
system when combined with an appropriate operating community and compatible hardware and
software~
It should be recognized that some of the detailed information contained herein is in a
state of change or expansion. The authors have attempted to provide enough information to enable creation of a working system, but they could not hope to foresee all
future refinements.
1.1 NUMERICAL CONTROL (N/C) AND COMPUTER GRAPHICS INVOLVEMENT IN
THE INDUSTRIAL PROCESSES
1.1.1 General Discussion
Numerically controlled machining, processing, and testing devices have become so
sophisticated that many users now enjoy a high degree of precision. Also, the interfacing of computing equipment and computer graphic presentations to the N/ C function allows man to develop a concept, review it, analyze it, modify it, and construct
the manufacturing tool in a completely dynamic environment. It is also significant
that, in addition to X-Y plotting eqUipment, another aid to man-machine
co~munica
tion is computer-interfacing, precision line drawing equipment, which enables a
comfortable documentation transition period and completes the total industrial proces s.
The techniques of the use of computer graphics by the engineering community is the
subject of a narrower field of study because of the many varying needs for data manipulation.
However, the greatest realization of the potentials offered by computer
graphics and numerical control will come from an industrial process that essentially
eliminates dependency upon man for any interpretive roles beyond design completion
and choice of production method. Realization of the full computer graphic/engineering
1-2
potential depends on the possession by the design engineer/operator of a complete
understanding of the computer graphic environment. Only a few design engineers are
now taking advantage of the existing new capabilities offered by data automation and
computer graphic systems. It is therefore important that manufacturing and engineering personnel receive complete orientation in the latest disciplines.
A unique capability· becomes available with use of the CRT and light pen for design
engineering communication with the computer, and use of the same CRT and light
pen for N/C parts programming, coupled with non-interpretive use of N/C graphics
equipment, alphanumeric printers, punched tape, film, magnetics, electrostatics,
etc. , for documentation of any or all results.
One of the objectives of this study is to arrive. at ~pecifications which enable establishment of a working system to accomplish these, ends. Existing technology pr.ovides the
capability for implementing all the aforementioned technologies into one integrated
system. It is improbable, under the present Engineering and Manufacturing organizational structure in the industry, that the cost-effectivenes s of numerical control can
be fully
demon~trated.
Consequently, a
c~nsiderable
amount of the computer graphics-
engineering-N/C portion of this total NASA-MSFCstudy is devoted to the involvement
in the industrial processes and to answering such questions as the following:
o What are the optimum techniques for training personnel in the use of these
computer-aided capabilities? .
•
What organizational structures offer optimum realization of the potentials of
these new technologies?
•
What are the most advantageous approaches to linking together the computer,
the N/ C control system, and the N/ C tool?
•
What techniques for introduction of the computer-aided capabilities into the
engineering/manufacturing/operational communities offer the greatest
potential for early acceptance and use?
•
What communication techniques between the users and the computer systems
offer the greatest potential?
1-3
•
What growth potential is necessary?
•
What new computer-aided design techniques will be developed in the next
three to five years?
•
What is the value/need of software interchangeability between various
facilities?
•
What is the opportunity to take advantage of the economies offered by use
of existing software?
1. 1. 2 Answers to Questions Previously Posed
Each of the previously listed questions will be dealt with in the following. Information
has been gathered through visits and conferences with members of the industry, through
searching literature, and through personal research.
1.1. 2.1
~at
are the Optimum Techniques for Training Personnel in the Use of These
Computer-Aided Capabilities? Initially, those who need training must be identified;
and the extent to which they will be trained must be agreed upon.
Th~n
the varieties
of techniques can be evaluated to determine which offers the highest potential for
success.
The degree of success of employment of a CG-CAD-N/C (Computer Graphics - Compll:ter
Aided Design-Numerical Control) system will be in direct proportion to the degree of
accurate knowledge of the system held by: (a) manufacturing, (b) engineering, and
(c) management. The weakest of the three will be the limiting factor for full realization of the potentials of the CG-CAD-N/C system. The technique of training each
group will be different, and yet the overlap of understanding must" be broad.
a.
The N/C manufacturing community (even more than the conventional) could
much more easily and accurately program the creation of a part if the
designers limited themselves to geometrics employing the locus of straight
lines, circle sections, other conic sections, and oft-repeated special curves.
1-4
b.
Engineering personnel generally do not realize the creativity offered by an
N/ C manufacturing facility.
c.
Management must realize the necessity of the timely, dedicated support
necessary to gain maximum potential from CG-CAD-N/C as a system.
Thus it is that Manufacturing, Engineering, and Management must each understand
the role not only they will have, but in a large sense, the role the others will have
also. This is the general training task that must not be overlooked as definition of
skill-developing training commences.
It has been demonstrated that young men fresh out of school can most quickly become
proficient with N/C-CG parts programming as well as with CG as a geometry defining
tool.
Therefore, it is desirable to concentrate on the young for training in these new
technologies; however, consideration of the following elements causes the "educate
only the youngsters" approach to be very suspect: 1) the environments into which they
will be placed (controlled by the older generation), 2) the support that is needed to make
them effective, 3) the indispensable need for acceptance by the established communities,
4) the wealth of. practical experience held by the established communities. 5) the investments in each area. In looking around the country at several of N/C installations, it
also becomes apparent that as the young men become proficient in
th~s
new technology ,
they are moving on to loftier goals and higher rewards, necessitating a continuous
training program and continuing relianc'e upon recent arrivals.
Thus it becomes important to ensure that training in these new technologies is directed
toward the established community as well as the young men.
The answer to the question posed, then, must factor in the foregoing. There must be
at least three specifically tailored, comprehensive training and education efforts in
each of three areas that obviously interlock with each other. Manufacturing, Engineering, and Management must be trained in:
a.
An understanding of the other's problems, needs, and capabilities
b.
Skill in use of these capabilities for older personnel
1-5
This training and education effort must be tailored to complement the recipient's background, study habits, capabilities, and motivations to realize a high degree of effectiveness. A modicum of proficiency with the hardware involved must be realized at the
same time as the capabilities and known techniques of use are portrayed to those involved
in the training and education effort. The most effective training techniques employ
several of the presently used tools such as manuals, lecture-demonstrations, and a
wide variety of proprietary and vendor-supplied audio-visual aids. At least one system also employs the tutorial dialogue in which a statement is displayed upon the face·
of the cathode-ray tube describing the next piece of information needed by the system
as the logical progression from the last information or command given.
1. 1. 2. 2 Which Organizational Structures Offer Optimum Realization of the Potentials
of These New Technologies? At the time any industrial facility involved in manufacturing, engineering, and computing initiates a capital investment in CG-CAD-N/C they
should also endeavor to develop a fresh reevaluation of their approach to the total
industrial process. The organizational structure that would be optimum 10 years from
now is unlikely to be detailed today. It is also unlikely that many firms would undertake to restructure in one large step. Summing up these factors leadE? to the conclusion
that continued success will require frequent reevaluation and faster response to new
needs than many organizations have either given or experienced •.
In studying the methods that effect organizational structure changes, it becomes axiomatic that the larger the organization the less likely it is to accept a radical change
over a very short span of time. It is important, also, that existing product flow not
be impaired or interrupted to any significant extent. It has already been demonstrated
by the N/C users alone that those who make a strong, fast move to the highest possible
(actual) use of their N/ C tools realize a faster profit on their investment.
An example of the misconceptions existing in some parts of the N/C community is the
idea that "N/C is unprofitable in small-quantity machining." The facts show that it is
most profitable for short runs (less than 10 items) in small shops; and in large shops
it can be shown to be at least as profitable if all costs of conventional manufacture are
1-6
equated obje"ctively with costs of N/C manufacture. Support organizations must be
correctly oriented in their roles in the CG-CAD-N/C implementation.
The computer sciences, engineering, and manufacturing disciplines must be encouraged
to work together to realize maximum benefit from technical knowledge and economy of
operation.
A vast difference exists between: (1) the current employment of a ''batch mode" wherein
time intervals between input and output of minutes. hours, and even days are tolerated
(because nothing better was available), and (2) necessary insistence upon response in
seconds or fractions of seconds by both Engineering and Manufacturing. A relatively
small portion of the computer world has actually experienced the rigors and pressures
of involvement in the real-time environment; and most of the systems that have experienced these rigors are special-purpose,
speci~ically
designed systems, owned and
operated by independent organizations.
Of equal significance is the impact on engineering methods, including accustomed time
frames and
tec~niques
of differentiation between design engineering, engineering sup-
port, administration, and such functions as manufaoturing, engineering, engineering
standards, specifications, preferred parts, and procurement.
It is possible for engineering personnel ,to train their own parts programmers and thus
deliver to Manufacturing the tapes for the N/C tools, tool and fixture instructions, and
material specifications. It is also possibie for Manufacturing to take an 'engineering
design of a piece part and redesign that part with superior functional weight, strength,
cost, and schedule'results using manufacturing techniques about which Eilgine'ering has
no prior knowledge. Elimination of, or a significant reduction in~ such 'l~ck of ~om - .
munication between manufacturing and engineering is an undeniable effect of a properly
structured CG-CAD-N/C industrial process. As in Manufacturing, Engineering and
support organizations must be reexamined with respect'to their roles, in new industrial
processes.
1-7
1.1. 2.3 What are the Most Advantageous Approaches to Linking Together the Computer, the N/C Control System, and the N/C Tool? Computer graphics is progressing
toward a capability for enabling a design engineer's rapid communication with a computer for design concept evaluation and a resultant improvement in optimization before
design release. Use of computer graphics for numerical control parts programming
in a profiling operation (where depth is anything but curvilinear) has in 1967 become
operational. The outputs of the CG-N/C parts programming are either basic tool
movement geometry definitions or completely processed for particular N/C tools and
ready-to-cut parts (in the case of N/C machining centers). This information can be
in any of the several computer output forms such as punched tape, magnetic device,
or alphanumeric printout. The control systems for N/C tools are generally of various·
levels of sophistication and accept input data from devices such as magnetic tape,
punched tape, and, under recent development, direct computer-processor interface.
There are a number of efforts in progress to identify a computer for the N/C operation.
Certain installations (Bendix, G. E., Bunker-Ramo-Woolridge, and Boeing) have been
visited and were found to be in some phase of study or implementation of on -line
computer/processor rUDning the N/C tools. They fall into two distinct categories:
• A
spe~ial
control system that relies upon a central computer to do the inter-
polating and that runs the machine tool from that point
•
forw~rd
A much simpler device that interfaces with a complete, conventional N/C
tool control system where a tape reader would normally interface and that
uses the local control system's interPolating capabilities
There are several advantages of the dedicated, on -line computer to manufacturing.
They fall into these categories:
•
•
Scheduling
Elimination of routine use of punched tape
• Calibration and fault isolation
.• status reporting
•
Local computation for simpler part programming
•
Rapid turnaround
1-8
In most installations, no one of the above advantages will pay for a computer/processor
capable of running 10 or more three-axis N/C tools simultaneously. A control system
can be estimated at $10,000 per axis, and a properly sized processor with necessary
peripherals cannot be purchased for less than $350,000. However, a system is established which can by-pass the punched tape reader, do calibration and fault isolation
and some of the part programming work now being paid for elsewhere; improvements
in the N/C tool utilization can be realized which can approximate the costs of computer
lease and operation. By adding to this the worth of accurate, flexible scheduling and
near-instantaneous statusing, it can be shown to be profitable to have a dedicated computer system for an N/C manufacturing facility employing 10 or more three-axis tools.
The initial configuration of the system to be profitably employed would operate on a
three- or four-level priority interrupt basis - a software-hardware based technique
interfacing with as many complete, local N/C ,tool control systems as there are tools.
The interface would be at the output of the tape readers (which should be left switchable
on the N/ C tool for backup). The interrupt to the computer would be generated in the
same way as the punched tape advance is initiated when buffer storage is unloaded.
The computer would draw its increments of data from a combination of long- and
short-term storage media such as magnetic tape and core. When not honoring interrupt commands, fault isolation routines might be going on. When neither N/ C toolgenerated interrupts nor fault isolation is going on, next-job orders might be prepared
for moving tools, fixtures, materials,
~nd
instructions to arrive with a setup crew at
an appropriate time before the previous job is completed. If these direct support
functions of the N/ C shop do not need to be done at any particular moment, the computer might honor a request for status or. a parts program request and, if no other
operational requirements are levied, scheduling information can be factored in to
accumulate experience-updated commitments on every N/C tool. This technique of
loading the computer has been demonstrated in other applications, and hardware for
implementation has been in development for some time. The basic processor/computer
systems are exemplified by the mM 1800, the CDC 1700, and the GE 4020.
The initial tie-in of such a dedicated, on-line N/C computer tp a computer used for
computer graphics is obviously through hard-line disc, magnetic tape, or punched tape
data transfer. Subsequent developments in the linking of the computer graphics to the
1-9
N/C tool are in various stages of development and experimentation. Generally quiet
activities are going on with joint participation between computer-processor manufacturers, control system manufacturers, and N/C tool users.
1. 1. 2. 4 What Techniques for Introduction of the Computer-Aided Capabilities Into
the Engineering/Manufacturing/Operational Communities Offer the Greatest Potential
for Early Acceptance and Use? In the preceding answers, part of the answer to this
question has been given. Actual experience has shown a slow-rising exponential curve
when linearly plotting elapsed time against effective use of the computer-aided
, capabilitie s .
The introduction of these new capabilities into the established community will best be
done by establishing the strongest possible ties with familiar entities. The computer
graphics hardware used by the various engineering and manufacturing personnel should
be in or adjacent to the areas in which they already work.
The introduction must also be preceded by careful indoctrination and education of the
management community.
The answers to the question of introduction techniques to be employed grows out of the
need for a professionally administered, deliberate effort at pre-selling the capabilities
at all levels, followed immediately by successful demonstrations by persons already
lmown and respected by the audience in environments in which they feel secure. All
of this is necessary to overcome the almost universal resistance to
change~
1. 1. 2. 5 What Communication Techniques Between the Users and the Computer Systems
Offer the Greatest Potential? Experience with computer systems in operational situations has clearly demonstrated the value of communication with computer systems in
a way that uses established languages and fast response. In terms of encouraging
wide -spread use of computer systems, development of user-oriented languages has
been most beneficial. The next most advantageous characteristic is that of the user's
getting instantaneous responses to his requests of and instructions to the computer.
1-10
The employment of computer graphics as a communication medium with a computer
system has already been demonstrated to be a very powerful technique. This is not
surprising since man's most effective communication techniques involve some form
of graphic symbolism.
Those who have researched the problems connected with communication between man
and man, and between man and computer, conclude that man has significant advantages
over a computer as a recipient of information. This advantage lies in the ability of
man to (1) compensate for a faulty input by drawing upon partially related information
and (2) bring more than one type of sensor to bear upon a phenomenon. Man also has
an ability to question the sender when reception is not clear. The more nearly the
man - computer commWlication approaches a man - man communication, the more
reliance men will place upon the computer to do those things best done by computing
equipment.
Awareness of the advantages of this approach has led to the development of "questioning"
and trouble-shooting software routines (as accessories to basic software packages)
which are automatically called up when the computer cannot continue with a process.
Another technique of improving the communication calls for in -depth training of persons
who will use a computer to learn a language the computer accepts.
In the use _of computer graphiCS, two baSically different philosophies were early employed for user communication with the c,omputer. One used an elaborate accessory
keyboard and the other used a simple keyboard and a listing (menu) flashed on the face
of the cathode-ray tube from which the next desired information was manipulated by
means of the light pen. Experience with these two and addition of some special features
has resulted in a movement by each group toward a mix of menu and keyboard.
To use these systems, instructional'reference manuals are -being created. Early
,
samples are enclosed with this study -report.
,
-
Another technique that simulates man 'spreconditioning for the next sensation is the
use of a "tutorial" system which displays on the CRT the next information needed by
the computer to act upon the previously communicated information.
There is no doubt that a combination graphic-symbolic communication medium that
preconditions both the man and the computer offers the greatest potential. Many
isolated, and a few integrated, techniques for accomplishing this are being employed
by groups realizing significant improvements in communication accuracy and efficiency.
In the case of N/C part programming, communication with the computer through CRT
and light pen has proven effective from the viewpoints of cost, time, and accuracy.
It's obvious superiorities over manuscript-card -tape communication techniques are
today limited to applications involving two-dimensional curvilinear and third-dimensional
fixed depth and slope activities. It is
expecte~,
however, that third axis curvilinear
definition can be acco?lplished in 1968.
1. 1.2. 6 What Growth Potential is Necessary? !crom a management position, the
answer to this question is important for capital needs and budgetary apportionment.
It is obvious that once the capability of the new techniques 'of CG-CAD-N/C is known,
there will be an increase in demand for system time. It is also obvious that it would
be extremely costly to provide every man with a CRT and light pen at his desk (at least
at present). Thus, a firm's most creative talent should be given initial access to the
system so that their creativity can be expanded at the earliest possible moment.
Several factors have not yet been accurately determined or are still far from reaching
a plateau of development in their area. These factors include the following:
•
The amount of time a man remains productive when paced by a system that
can consume and process data faster than man can conceive it - this raises
the question of number of men per console.
•
The number of CRTs (with light pens) that can practically be tied into one
system.
1-12
•
The continuing improvements in hardware and software enabling efficiencies
that expand effective power without change in main frame hardware.
•
Innovations such as data compres sion that expand the effective power of the
system.
1. 1.2. 7 What New Computer-Aided Design Techniques Will be Developed in the Next
Three to Five Years? In numerical control, it is not possible to deny the value of
direct processor control of many machine tools simultaneously. With that advent will
come fault isolation, machine tool loading, scheduling, and accounting as well as
statusing, preventive maintenance direction and other functions better handled by
machinery than men. Another area that will interface with, and be aided by, the
dedication of a computer to the N/C operation will be adaptive control.
Because part of this study was devoted to ascertaining and understanding the highly
probable forthcoming developments " it was found necessary to consider adaptive control; at least three
~qmpanies
are already deeply involved in its development. In
brief, adaptive control involves sensing the cutting edge -work piece position relationships to enable. near instantaneous modifications to motion - for the purpose of safeguards and more precision in manufacture. Sensors in this field are not yet capable
of ideally accurate definition of position; but all the approaches being talten make a
very intelligent approximation based upon sensing of a variety of reactions within the
system, and then selecting an appropriate, preconditioned, immediate response from
the machine and its control system. The potential for expansion of the sophistication
of the adaptive control's response by employing a processor or computer with significantly more capability than the normal N/ C control system computer's capability is
apparent. The ability to achieve a significant improvement in precision in positioning
and movement of an N/C tool (over today's state of the art) is a direct dependent upon
more accurate, timely definition of true position relationship between tool and work
piece. Thus, a manufacturing-dedicated computer/processor suggests a path that
can lead to the inevitable demands for greater accuracies.
1-13
Tying an N/ C manufacturing-dedicated computer to a computer graphics computer
will enable greater accuracies and economies through reduced handling of data. The
ties will initially be of the telephone grade transmission line category.
1.1.2. 8 What is the Value/Need of Software Interchangeability Between Various
Facilities? In many computer-employing installations, a lack of communication between developers of software results in duplications that are three-fold wasteful:
(1) repetitious effort, (2) waste of creativity, and (3) restrictions on growth and improvement. Often continuance of this lack of communication or support for this
continued duplication can be traced to hardware and software differences. Most
experienced users of computers now realize the time and dollars of software conversion and thus make an attempt at hardware standardization; however, when this is done,
management often still hears that employment of one group Is software package by
another group as an accessory to existing
so~ware
(even though hardware is identical)
can It be done without conversion costs. Searching out the reasons for this statement
reveals that basic operating software used in the two or more installations is not
identical even though objectives of data handling were stated to be the same. All of
this has been, is being, and will continue to be experienced by computer users until
a learned, well-defined, well-disciplined management of software/hardware systems
is instituted.
The advantages of software interchangeability are many and include (1) effective extension of creativity, (2) conservation of computer resources, (3) reduction in programming costs, (4) better understanding by users of computer capability and needs because
of need for tighter specifications to enable conformance to interface requirements,
(5) back-up by multiple installations, thereby counteracting down time and work load
peaks.
Much of this applies directly to N/C tools. It is desirable to be able to quickly define
alternate tools upon which work can be done when the preferred tool is down or overloaded. There is no well established reason why post processing for more than one
machine tool cannot be an automatic function in those designs that lend themselves
1-14
to computer graphic definition. It would be beneficial to have backup to parts programs
to allow use of N/C tools located in another facility to which you can gain access.
Computer graphics -software should be capable of interfacing with material displayed
or produced in other facilities to enable efficient communication and eventual centralized
call-up of most programs or drawings.
1.1.2.9 What are the Opportunities to Take Advantage of the Economics Offered by
Use of Existing Software? There are so many considerations that must be satisfied
to enable an overall working system of people, machines, and software such as
CG-CAD-N/C that a considerable advantage can be gained by evaluating existing working systems before moving into new areas. Well documented, fully modularized software will reduce implementation time'1
The value of early demonstration of an expandable, but minimally satisfying, overall
system is difficult to ascertain. There are many advantages to employing selected
existing software
packag~s,
at least at the outset, even though eventual modification
or replacement, of some of those packages
~s
1-15
ini.tially understood.
1.2
FEASffiILITY OF N/C-COMPUTER GRAPHICS AT MSFC
1.2.1
General
Consideration of N/C-Computer Graphics at MSFC falls into the following three
categories:
•
Economic feasibility
•
Technical feasibility
•
Functional feasibility
Economics will be discussed in the following subsection. Technical and functional
feasibility considerations are discussed elsewhere in this document.
1.2.2
Economics
Numerical control parts programming by use of scope and light pen by itself allows
a small percentage of improvement over conventional techniques; however, it can
show a considerable time advantage over parts programming when used in conjunction
with APT or some similar computer-employing, batch-process technique. The time
taken for a man to copy a print by manually exercising a light pen is, of course,
eliminated when the original design is developed l:lsing a light pen. In addition, it
will be feasible to eliminate the, many interpretive errors and dimensioning omissions
now experienced in entering data from design sketch through engineering drawings
to scope and light pen re-draw of original design.
1.2.3
Facets of CG - CAD Applicable to N/C Operations
Other sections of this document deal with the potentials of the applications of computer graphics to a variety of analyses and design efforts. In most cases, the end
product of the analysis programs result in hardware of some type. The very fact
~
that these hardware configurations are digitized will aid in the eventual conversion
of the manufacturing processes to support this type of hardwa~e to N/C. Three major
areas are identifiable now that lend themselves today to a total CG-CAD-N/C system.
They are described in succeeding paragraphs.
1-16
1.2.3.1
Electronics. Electronics will have considerable CG-CAD-N/C potential in
several areas in which N/C manufacture can be employed. ·Those areas include:
o
Discrete component/solid state
(I)
Integrated circuit/microelectronic
o
Interconnections
Suggestions of the N/C applications in each of those specific areas are:
a.
b.
Discrete component/solid state:
1.
Total etched board manufacturing processes
2.
Component selection and insertion
3.
Soldering/welding operations
4.
5.
Subassembly integration
6.
Enclosing, packaging, etc.
7.
Storing, retrieving, etc.
Testing
Integrated circuit/microelectronics:
1.
0
Total photo-optic and mechanical operations in the manufacture of
discrete components
2.
Integration of discrete components into logic configuration
3.
Integration of logic configurations into functional configuration
4.
Welding and soldering control by many precision techniques such as
conventional micro point, unconventional electron beam, aI1d laser
beam
5.
Testing at many steps through functional system build up and
final acceptance
6.
c.
Enclosing, packaging , etc.
Interconnections:'
1.
Wire/cable sizing, selection, cutting, trimming, and tinning
2.
Connector selection, mating, welding/soldering, etc.
1-17
1.2.3.2
3.
Continuity and insulation testing
4.
Labeling
5.
Large cable build up, forming, lacing, etc.
Enclosures.
a.
Sheet metal forming, punching, bending, cutting, fastening, welding, etc.
b.
Solid material machining by cutting, electro-discharge, chem mill, etc.
Co
Plastic molding, curing, cutting, etc.
1. 2. 3. 3
Structural Members.·
a.
All machining operations
b
All welding operations
0
co
Many assembly operations
d.
Many testing operations
e.
Raw material storage, retrieval, and sizing
f.
Finished part packaging, storage, and retrieval
1.2.4 MSFC Computer Involvement in N/C
It will be a long time before computer graphics at MSFC is capable· of N/C parts
programming and another lengthy time period will follow that before all design and
N/C parts programming is done with computer graphics. During this interval, the
NASA-MSFC computer support of N/C operations can significantly
affec~
the success-
ful employment of N/C tooling. Card punch, printing, tape punch and computing
operations are, by necessity, sequential operations. It is most desirable for NASAMSFC Computer Operations to enable the shortest practical time spans between parts
programmers' attention to each job. Both off-line and on-line activities are involved
and special attention should be given to their needs.
If we hypothesize a situation in which a remoje terminal with reasonable main frame
interrupt capability and appropriate peripheral equipment is located in the N/C parts
1-18
programming area, some comparisons with conventional techniques of computeremploying parts programming can be shown. Assume that the remote terminal is
a Univac 1004 remote to an 1108 which in turn has full APT and post processor capability. Let us further suppose that the 1004 has as its own, local peripherals: card
read and punch, magnetic tape and/or magnetic disc read and write, and punched tape
(both paper and mylar) read and write. In addition, a key punch would be set up in
the same area as the parts programmer and the 1004 and its peripherals which,in
turn, are located adjacent to the N/C tool area. With an interrupt system into the
1108 which essentially handles complete batches on a first in/first out basis, and with
sufficient storage space in the 1004, jobs needing 1108 power ,can be loaded into the
1004, processed a few Ininutes later by the 1108 and output back into the 1004 for the
format desired. Tape punching, card punching; alphanumeric printing, magnetic tJlpe or disk-pack-to-punched-tape conversions, and verifications can all be done as needed
by the 1004 without waiting for 1108 availability.
With this hypothetical facility, it becomes possible for parts programmers to reduce
the number of parfs programs they have in work from a high of 6 to 15 month
to an average of three to four. The reasons f.or this lie in the ability to get key punch,
computer runs, printouts, paper tapes, and all computer or computer peripheraldependent products returned, completed in a few minutes or few hours as a routine
instead of the present (without special priorities) minimum of one day for each .
activity that now exists in many installations. This kind of support will greatly
enhance the posture of the Computer Services organization. It will demonstrate the
problem to be expected in servicing dedicated, on line operations. It will. ~emonstrate
the worth of quick response to requests for off-line work. It will enable manufacturing
to effect significant economies in the N/C organizations by cutting programming and
tool try costs as well as lead times to finished programs.
1-19
1.3 NUl\1ERICAL CONTROL AND N/C-COMPUTER GRAPHICS SPECIFICATIONS
1.3. 1 Console Procedures and Flow Diagrams for Keyboard-Oriented Systems.
The console procedures (Tables 1-1 through 1-14) and flow diagrams (Figs. 1-1
through 1-14) are mutually complementary. A good basic understanding of the graphics
system user's viewpoint and function can be obtained from these data. A glossary of
terms used in the tables appears after Table 1-14.
The console procedures and flow diagrams herein form a solid basis for developing
a final graphics system configuration. The configuration shown is a minimum basic
package that can be used for engineering production de sign and drafting and numerical
control programming applications.
The console hardware configuration'for this system requires, in addition to the
cathode-ray tube, three input devices: a function keyboard, a typewriter, and a light
pen.
Console procedures and flow diagrams should be completed and agreed upon between
the graphics user and the graphics software programmer prior to any coding. These
will eliminate much misunderstanding by the user, serve as a basis for instructing
users, and eliminate much costly reprogramming of software.
The most successful graphics systems have been derived by user-programmer teams
developing and checking the applications programs jointly.
1-20
Table 1-1
CONSOLE PROCEDURES FOR START
(KEYBOARD ORIENTED)
Step
User Action
Machine Reaction
MENU DISPLAYED
1.
FKSTART
CALL
-
Retrieve file (drawing) from
storage device
FILE
ID
'CKPT
-
File drawing on storage device
Key in ID for starting new dwg
(Checkpoint) file model on
storage device for recall in
case of bomb off
RECALL -
Recall model from storage
device after bomb off
2.
LPD CALL
CALL under lined
. Message
3.
Key or pick ID
-
KEYID
Retrieves drawing from storage device
and displays drawing
2.
LPD FILE
--
FILE underlined
Message
3.
Key or pick ID
-
KEYID
Files dwg on storage device
un~er
in
2.
LPD CKPT
--
CKPT underlined
Outputs model on storage device
2.
LPD RECALL
RECALL underlined
Retrieves and displays last model
checkpointed
1-21
ID keyed
Table 1-2
CONSOLE PROCEDURES FOR NUMERICAL CONTROL
Step
1.
User Action
Machine Reaction
FK NUMERICAL CONTROL
MENU DISPLAYED
SETPT
- Key in or LPD cutter starting
"-
position
ZDIM
- Key Z axis, value of bottom of
cutter
FEDRAT
- Key or pick feed rate (velocity
in inches per minute)
POST (Postprocess)
Retrieve CL data from bottom of
model and write APT CLTAPE.
ABORT
- Abort present cutter path and
exit from N/C mode
2.
LPD SETPT
SETPT underlined
Message - SETPTX=
3.
LPD point from display
Message - SETPT X = XXXXX
SETPT Y = XXXX
SETPT Z =
4.
Pick or key SETPT Z DIM
Message - SETPT Z = XXXXX
CUTTER =
5.
Pick or key CUTTER DIA
Message - CUTTER = XXXXX
2.
LPD SETPT
SETPT underlined
Message - SETPT X =
3.
Message - SETPT X = XXXXX
Key SETPT X DIM
SETPT Y=
4.
Key SETPT Y DIM
Message - SETPT Z =
5.
Key SETPT Z DIM
Message - SETPT Z = XXXX
CUTTER =
6.
Key or pick CUTTER DIA
Message - CUTTER = XXX
-
*SELECT DRIVE SURFACE
*MESSAGE - Select PT/ELEM
1-22
I
Table'1-2 (Cont.)
Step
2.
User Action
Machine Reaction
LPD Z DIM
Z DIM underlined
Message
3.
Pick or key Z DIM
Message
-
Z DIM =
Z DIM = XXXXX
* SELECT DRIVE SURFACE
2.
3.
LPD FEDRAT
FEDRA T underlined
Pick or key FEDRA T
Message
-
FEDRAT =
Message
-
FEDRAT = XX
*SELECT DRIVE SURFACE
2.
LPD POST
(APT INTERFACE)
--
POST underlined
Message
3.
Pick or key MACHIN
Message
-
MACHIN/
MACHIN /XXXXX
STATEMENT
3A.
FKACCEPT
4.
pick or key SCALE
SCALE =
Message
-
SCALE = XXJCX
Writes CLTAPE to input to APT system
2.
LPD ABORT
ABORT underlined
All N/ C output deleted from the model exit from N/ C mode
2.
LPD PT
Cutter displayed at PT selected
-
CL placed in model
, *SELECT DRIVE SURFACE
2.
LPD ELEM
3.
LPD CKSURF
Message
"
-
SELECT - CKSURF
Cutter displayed at intersection of elements
Message
*MESSAGE -\Select PT /ELEM
1-23
-
SELECT- CKSURF
Table 1-3
CONSOLE PROCEDURES FOR ERASE
Step
User Action
Machine Reaction
1.
FK ERASE
Message - LPD ELEMENT
2.
LPD ELEMENT
Removes element from display and model
Table 1-4
CONSOLE PROCEDURES FOR LINE TYPE
Step
User Action
Machine Reaction
------'-
1.
FK LINE TYPE
MENU DISPLAYED
SOLID
HIDDEN
I
CENTERLINE
PHANTOM
-
2.
LPD SOLID
Message
3.
LPD ELEMENT
Displays element as solid line
2.
LPD HIDDEN
Message
3.
LPD ELEMENT
Displays element as hidden line
2.
LPD CENTERLINE
Message
3.
LPD ELEMENT
Displays element as centerline
2.
3.
LPD PHANTOM
Message
LPD ELEMENT
Displays element as a phantom line
1-24
-
-
LPD ELEMENT
LPD ELEMENT
LPD ELEMENT
LPD ELEMENT
I
Table 1-5
CONSOLE PROCEDURES FOR FILLET
Step
User Action
Machine Reaction
1.
FK FILLET
Message - KEY RADIUS
2.
Key or pick RADIUS
Message - LPD ELEMENT 1
3.
LPD 1ST ELEMENT
Message
4.
LPD 2ND ELEMENT
Message - LPD LOCATION OF FILLET
- LPD ELEMENT 2
CENTER
5.
LPD SIDE OF FILLET
Displays fillet tangent to elements selected
CENTER
Table 1-6 .
CONSOLE PROCEDURES FOR RETOUCH
Step
1.
User Action
Machine Reaction
FKRETOUCH
MENU DISPLAYED
. EXPAND
CONTRACT
BREAK
2.
LPD EXPAND
Message - LPD ELEMENT
3.
LPD ELEMENT
Message - LPD EXPANDED DISTANCE
LPD ELEMENT TO EXPAND TO
4.
LPD ELEMENT OR
Expands element
DISTANCE
2.
LPD CONTRACT
Message - LPD ELEMENT
3.
LPD ELEMENT
Message - LPD END TO BE CONTRACTED
4.
LPD END
Message - LPD LOCATION TO CONTRACT
--
TO
LPD ELEMENT TO CONTRACT
TO
-5.
LPD LOCATION OR
Contracts element
ELEMENT
1-25
Table 1-6 (Cont.)
Step
User Action
Machine Reaction
----
2.
30
LPD BREAK
Message
-
LPD ELEMENT
LPD ELEMENT
Message
-
LPD LOCATION OF BREAK
4.
LPD LOCATION
Breaks element at point normal to LPD
location. Element is now stored as two
separate entities in the model.
Table 1-7
CONSOLE PROCEDURES FOR PARALLEL
(STORES NEW ELEMENT PARALLEL TO OLD ELEMENT)
Step
User Action
Machine Reaction
1.
FK PARALLEL
Message - LPD ELEMENT
2.
LPD ELEMENT
Message - LPD SIDE NEW ELEMENT
TO BE
-
3.
LPD SIDE
Message
4.
Key DISTANCE
Displays parallel element
--
1-26
KEY NORMAL DISTANCE
T·able 1-8
CONSOLE PROCEDURES FOR CmCLE
Step
1.
User Action
FK CmCLE
Machine Reaction
Message - KEY X DIMENSION
Message
-
LPD POINT
2.
Key or pick X DIMENSION
Message - KEY Y DIMENSION
3.
Key or pick Y DIMENSION
Message - KEY RADIUS
4.
Key or pick RADIUS
Displays circle
2.
LPD POINT
Message - KEY RADIUS
LPD POINT 2
3.
Key or pick RADIUS
Displays circle
2.
LPD POINT
Me,s sage - KEY RADIUS
LPD POINT 2
3.
LPD POINT 2
Message
-
LPD CENTER DffiECTION
LPD POINT 3
4.
LPD CENTER DffiECTION
Message - KEY RADIUS
5.
Key or pick RADIUS
Displays circle
2.
LPD POINT
Message - KEY RADIUS
LPD POINT 2
3.
LPD POINT 2
Message
-
LPD CENTER DffiECTION
LPD POINT 3
4.
LPD POINT 3
----
Displays circle
1-27
Table 1-9
CONSOLE PROCEDURES FOR POINT
Step
1.
User Action
Machine Reaction
FK POINT
Message - KEY X DIMENSION
Message - LPD PHYSICAL LOCATION
Message
-
LPD 1ST ELEMENT OF
INTERSECTION
-
2.
Key or pick X DIMENSION
Message
3.
Key or pick Y DIMENSION
Displays point
2.
LPD LOCATION OF POINT
Displays point
2.
LPD 1ST ELEMENT
Message - LPD 2ND ELEMENT OF
KEY Y DIMENSION
INTERSECTION
3.
LPD 2ND ELEMENT
*Displays point
*If one or both elements are circles or curves an additional LPD is required to
indicate which of multiple intersections the point is to be stored.
Table 1-10
CONSOLE PROCEDURES FOR LINE THROUGH POINT AT DmECTION
Step
1.
User Action
Machine Reaction
FK LINE THRU POINT AT
DffiECTION
MENUDISPLAYED
HORIZONTAL
VERTICAL
PARALLEL
PERPENDICULAR
AT ANGLE
1-28
Table 1-10 (Cont.)
Step
2.
User Action
Machine Reaction
LPD HORIZONTAL
Message - KEY X DIM:ENSION
Message - LPD POINT
Message - LPD TANGENT CIRCLE
3.
Key or pick X DIMENSION
Message - KEY Y DIMENSION
4.
Key or pick Y DIMENSION
*Displays Line
3.
LPD POINT
Displays Line
3.
LPD TANGENT CIRCLE
Message - LPD SIDE OF CmCLE THAT
LINE IS TANGENT
4.
LPD TANGENT SIDE OF
*Displays line
CmCLE
2.
LPD VERTICAL
Procedure same as HORIZONTAL
2.
LPD PARALLEL
Message - KEY X DIM:ENSION
Message. - LPD POINT
Message - LPD TANGENT CmCLE
3.
Key or pick X DIM:ENSION
Message - KEY Y DIM:ENSION
4.
Key or pick Y DIMENSION
Message - LPD PARALLEL LINE
5.
LPD PARALLEL LINE
*Displays line
3.
LPD POINT
Message - LPD PARALLEL LINE
4.
LPD PARALLEL LINE
3.
LPD TANGENT CmCLE
I
*Displays line
Message - LPD SIDE OF CmCLE THAT
LINE IS TANGENT
4.
LPD TANGENT SIDE OF
Message
-
LPD PARALLEL LINE
CmCLE
5.
LPD PARALLEL LINE
*Displays line
*Program will set value of line length to some constant. Retouch function will be used
to trim line to desired configuration.
1-29
Table 1-11
CONSOLE PROCEDURES FOR LINE POINT TO POINT
Step
1.
User Action
FK LINE POINT TO POINT
Machine Reaction
Message - KEY X DIMENSION
Message - LPD POINT
Message - LPD TANGENT CIRCLE
2.
Key using typewriter or pick
Message - Key Y DIMENSION
with light pen X DIMENSION
3.
Key or pick Y DIMENSION
Message - Key X DIMENSION
Message - LPD POINT
Message - LPD TANGENT
Message - CIRCLE
..
'
4.
Repeat steps 2 and 3
Displays line
4.
LPD PREVIOUSLY STORED
Displays line
POINT
4.
LPD CIRCLE
Message
-
LPD SIDE OF cmCLE THAT
LINE IS TANGENT
5.
LPD SIDE OF CIRCLE LINE
Displays line
IS TANGENT TO
2.
LPD POINT
Message - KEY X DIMENSION
Message - LPD POINT
Message - LPD TANGENT CmCLE
3.
LPD 2ND POINT
Displays line
3.
Key or pick X DIMENSION
Message - KEY Y DIMENSION
4.
Key or pick Y DIMENSION
Displays line
3.
LPD TANGENT CmCLE
Message - LPD SIDE OF CmCLE THAT
LINE IS TANGENT
4.
LPD TANGENT SIDE OF
Displays line
CmCLE
1-30
T~le 1-11 (Cont.)
Step
2.
User Action
LPD TANGENT CIRCLE
Machine Reaction
Message - LPD SIDE OF cmCLE THAT
LINE IS TANGENT
3.
LPD TANGENT SIDE OF
Message - KEY X DIMENSION
CIRCLE
Message - LPD POINT
Message - LPD TANGENT cmCLE
4.
LPD TANGENT CmCLE
Message - LPD SIDE OF CmCLE THAT
LINE 'IS TANGENT
5.
LPD TANGENT SIDE OF
Displays line
CmCLE
4.
LPD POINT
Displays line
4.
Key or pick X DIMENSION
Message - KEY Y DIMENSION
5.
Key or pick Y DIMENSION
Displays line
Table 1-12
CONSOLE PROCEDURES FOR DIMENSIONING INFORMATION
Step
1.
User Action
Machine Reaction
FK DIMENSION
MENU DISPLAYED
HORIZONTAL
VERTICAL
PARALLEL
OFFSET
ANGLE
RADIUS
D:rAMETER
CURVE
1-31
Table 1-12 (Cont.)
Step
2.
User Action
LPD HORIZONTAL
Machine Reaction
HORIZONTAL underlined
Message - SELECT POINT 1
3.
LPD 1ST POINT
Message - SELECT POINT 2
4.
LPD 2ND POINT
Message
5.
LPD LOCATION OF
Displays dimensions, dimension lines and
DIMENSION
arrowheads (they are stored in the model to
-
LPD LOCATION OF DIMENSION
be used for hard copy if required)
2.
LPD VERTICAL
Same as HORIZONTAL except dimension is vertical
between points selected
2.
LPD PARALLEL
PARALLEL underlined
Message
3.
LPD 1ST POINT
4.
LPD 2ND POINT
.
-
SELECT POINT 1
Message - SELECT POINT 2
Message - SELECT ELEMENT THAT
DIMENSION IS· PARALLEL TO
5.
LPD ELEMENT THAT'
Message
-
DIMENSION
DIM IS PARALLEL TO
6.
2.
LPD LOCATION OF
LPD LOCATION OF
Displays dimension, dimension lines and
DIMENSION
arrowheads
LPD OFFSET
OFFSET underlined
Message - SELECT 1ST ELEMENT
3.
LPD 1ST ELEMENT'
Message - SELECT 2ND ELEMENT
4.
LPD 2ND ELEMENT
Message - LPD LOCATION OF DIMENSION
5.
LPD LOCATION OF
Displays dimension, dimension lines and
DIMENSION
arrowheads
1-32
Table 1-12 (Cont.)
Step
2.
User Action
LPD ANGLE
Machine Reaction
ANG LE underlined
Message - SELECT 1ST ELEMENT
3.
LPD 1ST ELEMENT
Message - SELECT 2ND ELEMENT
4.
LPD 2ND ELEMENT
Message - LPD LOCATION OF DIMENSION
5.
LPD LOCATION OF
Displays dimension in degrees, minutes,
DIMENSION
and seconds, dimension lines and arrowheads
LPD RADIUS
RADIUS underlined'
2.
Message
-
SELECT CIRCLE OR ARC
3.
LPD CIRCLE OR ARC
Message
4.
LPD LOCATION OF
Displays dimension with R, RAD, or RADIUS
DIMENSION
fol~owing,
LPD DIAMETER
DIAMETER underlined
2.
LPD LOCATION OF DIMENSION
dimension line and arrowhead
Message - SELECT CmCLE
3.
LPD CmCLE·
Message - LPD LOCATION OF DIMENSION
4.
LPD LOCATION OF
Displays dimension with D' or DIA following,
DIMENSION
dimension line, and arrowheads
LPD CURVE
CURVE underliried
2.
Message . - SELECT CURVE
3.
LPD CURVE
Message - LPDLOCATIONOF DIMENSION
4.
LPD LOCATION OF
Displays dimension, dimension .lines and
DIM:ENSION
arrowheads, parallel to curve
1-33
Table 1-13
CONSOLE PROCEDURES FOR GEOMETRY MACRO
Step
1.
User Action
Machine Reaction
FKMACRO
MENU DISPLAYED
START
ABORT
RETURN TO DRAWING
DISPLAY MACRO
CALL
STORE
START underlined
2.
LPD START
Message - SELECT ELEMENT
Message
-
SELECT FK (This instruction
is to be followed only if the
console operator is constructing
the macro from previously
stored geometry)
3.
LPD ALL GEOMETRIC
Message
-
SELECT ELElY1ENT
ELEMENTS TO BE USED
IN MACRO
4.
LPD DISPLAY MACRO
DISPLAY MACRO underlined; computer displays only those elements
select~d.
Operator
can then see if he has completed the macro
Message - SELECT PIVOT POINT
50
LPD RETURN TO DRAWING
RETURN TO DRAWING ·underlined
Message - SELECT ELEMENT
6.
LPD additional elements
Message - SELECT ELEMENT
& repeat step 4 until macro
is complete
1-34
Table 1-13 (Cont.)
User Action
Step
LPD DISPLAY MACRO
7
Machine Reaction
DISPLAY MACRO underlined
(same as step 4)
8
LPD a point (becomes pivot
Message
point)
point and pivot line will be used to locate
- SELECT PIVOT LINE (pivot
geometry on face of drawing-location
and rotation)
LPD a line (becomes pivot
9
Message
- SELECT STORE OR ABORT
line)
10
LPD STORE
STORE underlined; stores macro
11
LPD RETURN TO DRAWING·
Returns original drawing to display
3
Select any function key
Display goes blank
4
Construct geometry
Geometry stored in model as macro
5
Repeat step 1
6
LPD DISPLAY MACRO
DISPLAY MACRO underlined
Message· - SELECT PIVOT POINT RETURN TO DRAWING
7
LPD a point
Message - SELECT PIVOT LINE
8
LPD a line
Message - SELECT STORE OR ABORT
9
LPD STORE
STORE underlined; stores macro
2
LPD DISPLAY MACRO
Displays 1st macro stored
3
LPD DISPLAY MACRO
Displays 2nd macro
4
LPD DISPLAY MACRO
Displays Nth macro store d
5
LPD CALL
CALL underlined (th Nth macro which was
--
displayed when call was selected is the
macro being called). Original drawing is
displayed
Message - SELECT PIVOT
1-35
Table 1-13 (Cont.)
Step
User Action
Machine Reaction
6,
LPD a point
Message - SELECT PIVOT LINE
7.
LPD a line
-
Macro geometry is displayed in conjunction
with original drawing (macro geometry is
stored in model as part of original geometry)
Any LPD ABORT
ABORT underlined
Any + 1 Sele ct any function key to
get out of macro sub routing
Table 1-14
CONSOLE PROCEDURES FOR TRANSFORMATION
Step
User Action
Machine Reaction
1.
FK TRANSFORMATION
MENU DISPLAYED
MOVE
ROTATE
SCALE
2.
LPD MOVE
MOVE underlined
Message - KEY X
3.
Key or pick X TRANSLATION
Message - KEY Y
4.
Key or pick Y TRANSLATION
Display is moved
. X& Y dimensions. keyed will be new
position of the window center
2.
LPD ROTATE
ROTATE underlined
Message - KEY ANGLE
3.
Key or pick ANGLE"
Display rotates to angle keyed in
2.
LPD SCALE
SCALE underlined
Message - KEY SCALE
3.
Key SCALE
Display is scaled by amount keyed in
1-36
GLOSSARY OF TERMS
FK
Function key
LPD
Light pen detect
Pick
Select item from display with light pen
Window
CRT is considered a window through which
we view the drawing
Key
Key in data with typewriter keyboard
Drive surface
Surface the cutter is to drive along
Check surface
Surface cutter stops against
Perp
Perpendicular
1-37
c. ._______)
[
]
=
USER ACTION
=
MACHINE DISPLAY
=
MACHINE ACTION
=
FUNCTION - FUNCTION SHOWN IN
BOX REMAINS IN THIS UPPER L. H.
CORNER OF DISPLAY UNTIL ANOTHER
FUNCTION KEY IS DEPRESSED
= MESSAGE
Fig. 1-1 Symbols Defined
1-38
OR MENU
FK START
(START] \ -_ _ _ _
~l\: . :cl: : .E.!. :N. : :U_ _ _ _ _
___1
CALL - FILE - ID - Ch."PT - RECALL 1 - - - - - - - - - - - - - - 1_ _....,..._ _
~
KEYID
KEY ID
KEY ID
OUTPUTS l\IODE L
ON STORAGE DEVICE
RETRIEVES &: DISPLAYS LAST
DATA OUTPl'T OX STORAGE DEnCE
SEARCHES TAPE
DISPLA YS DRAWING
LEGEND:
FUNCTION REMAINS ON DISPLAY
UNTIL ANOTHER FUNCTION IS
SELECTED.
OPERATOR CONSTRUCTS DRAWING USING
FK'S AS REQUIRED.
MACHINE DOES AUTOMATIC CKPT EVERY
TEN USER ACTIONS.
ID DISPLAYED l\IESSAGE
Fig. 1-2 Subroutine START
SELECT
FK
_....::F:.....:K~P..::::O.::IN~T=--~ [
M_E;;;..s;;..;s~A;;..;G-E-----_I
pornT ] t -_ _ _ _ _
KEY X DIM
LPD LOCATION
SELECT ELE!'rlENT 1
LPD ELEM
KEY Y DIM I----=:M;.:.::E::.:S:.::S::.:A:.::G:.:E:....-_)~-----_{ KEY X DIM
KEY Y DIM
DISPLAYS
POINT
1'---------------1 LPD LOCATION
t-------------------- NO _ _ _ _-(
SELECT
ELEMENT Z
ARE EITHER OR
BOTH ELE11ENTS
CIRCLE OR CURVE
?
LPD
LPDINTERSECTIOX r---------------------~INTERSECTION
Fig. 1-3 Subroutine POINT
FK LINE - POINT TO POINT
....!...!!....:=~-:...=.:...:......:.,:;:...:....::=.:.....--f
tINE - POINT
MESSAGE
TO POINT 1 1 - - - - - . . : . : = = = - - - - - - 1
-
KEY Y DIM
--i(
J-_ _ _ _l\_lE_S_S_A_G_E_ _ _ _
KEY X DIM
SELECT POINT
SELECT CIRCLE
KEY X DIM
1 - - - - - - 1 LPD CIRCL0
LPD
TANGENT
SIDE
(LPD POINT
KEY YDIM
KEY 2ND
YDIM
LPD SIDE
....._ _ _.......:.;M:..:E:..:S;,;;;S::.,;;A..;;,G;;;;,E_ _ _ _- - i KEY X DIM
)-
KEY 2ND X DI:\!
SELECT POINT 2
I~LPD
(
CIRCLE
SELECT CIRCLE 2
(LPD POINT 2
( KEY Y DIM ' J - - - - - - - - - - - - - - - - - - - - - - I .
DISPLAYS
LINE
Fig. 1-4 Subroutine LINE-POINT-TO-POINT
LPD
TANGENT
SIDE
LPD SIDE
n;
L1~.;"
- PIll:>:T DIIlFCTlo:\
\IE:\C
110nI/' - \TllTlC'AL' I'AH.\LLEL - I'EllP - .\TA:\GL
~IESSAGE
:\0
:>:0
YES
Fig. 1-5 Subroutine LINE-POINT DIRECTION
FKCmCLE
[
CmCLE
]
DISPLAYS
- CmCLE
0_._._ .• 0.
MESSAGE
J~
KEY X DIM.
SELECT POINT
KEY X DIM.
MESSAGE
LPD POINT
KEY Y DIM.
KEY Y DIM.
~
C)
<
rJ.l
rJ.l
~
KEY
RADIUS
KEY RADIUS
~~
MESSAGE
J
KEY RADIUS
SELECT POINT
SELECT POINT
/ ' LPD POINT 2
MESSAGE-
2
KEY
RADIUS
MESSAGE
3
LPD CENTERDmECTION
LPD DIRECTION
LPD POINT 3)
Fig. 1-6 Subroutine cmCLE
F_K_P_A_R_A_L_L_E_L--.. [PARALLE L
1
MESSAGE
SELECT
ELEMENT
LPDELEM
LPD SIDE OF
OLD ELEMENT
NEW ELEMENT TO
BE DISPLAYED
LPD SIDE
DISP'LA YS NEW
ELEMENT
. Fig. 1-7 Subroutine PARALLEL
KEY DIST.
KEY NORMAL
DISTANCE
FK FILLET
--------------~
[
FILLET
]
MESSAGE
~--------~
KEY FILLET RADIUS
t-----4
KEY RADIUS
RADIUS = XXXX
SELECT
ELEMENT 1
DISPLAYS FILLET TANGENT
TO ELEMENTS SELECTED
LPD 1ST ELEM
( LPD APPROX. LOCATION OF FILLET CENTER
SELECT
ELEMENT 2
LPD LOCATION OF FILLET CENTER
Fig. 1-8 Subroutine FILLET
MESSAGE
LPD 2ND ELEM
__FK_R_E_T_O_U_C_H--I [ RETOUCH]
~
r-C
LPD ELEM
r
SELECT
ELEMENT
t--M_E_N_U~
MESSAGE
I--_~_-tr
I
LPD EXPAND
(LPD CONTRACT
Co:)
~
<
00.
LPD ELEM} -
Co:)
<
00.
00.
~
00.
~
~
:;;
LPD EXPANDED DISTANCE
SELECT ELEMENT TO
EXPAND TO
LPD ELEM. OR DISTANCE
LPD BREAK
EXPAND - CONTRACT - BREAK
LPD ELEM ) - -
SELECT
ELEMENT
1-----,
MESSAGE
SELECT
ELEMENT
LPD
LOCATION
OF BREAK
DISPLA YS CONTRACTED
ELEMENT
DISPLA YS EXPANDED
ELEMENT
I
I
LPD END ) - - -
l\1ESSAGE
LPDEND TO
BE CONTRACTED
LPD ELEM. OR LOCATION)
LPD LOCATION TO
" CONTRACT TO
SELECT ELEMENT TO
CONTRACT
( LPD LOCATION)
DISPLA YS ELEMENT WITH
BREAK AT POINT XORl\IAL
TO LPD LOCATION
Fig. 1-9 Subroutine RETOUCH
FK LINE TYPE [ LINE TYPE]
MENU
I - - - -_ _~
MESSAGE
SOLID -- HIDDEN - CENTERLINE - PHANTOM
LPD SOLID
SELECT
ELEMENT
HIDDEN
UNDERLINED
SELECT
ELEMENT
PHANTOM
UNDERLINED
CENTERLINE
UNDERLINED
ELEMENT DISPLAYED
AS SOLID LINE
SELECT
ELEMENT
MESSAGE
SELECT
ELEMENT
ELEMENT DISPLAYED
AS HIDDEN LINE
ELEMENT DISPLAYED
AS PHANTOM LINE
ELEMENT DISPLAYED
AS CENTER LINE
Fig. 1-10 Subroutine LINE TYPE
FK TRANSFORMATION
[
~
MENU
------------------~ TRANSFORMATIONJr-----------------------------------i
MOVE - ROTATE - SCALE
~---------T----------~
ROTATE
UNDERLINED
SCALE
UNDERLINED
MESSAGE
KEY Y DIM
~
0
~
0
~
fIl
<
fIl
::is
::is
~
fIl
~
I-'
I
~
00
MESSAGE
KEY X DIM
KEY ANGL
Fig. 1-11 Subroutine TRANSFORMATION
KEY SCALE
FK NUMERICAL
CONTROL
[
~l_E_N_L_-1
NUMERICAL] t -__
CONTROL
SETPT - ZDIl\i - FEDRAT - POST - ABORT
L-~---------.------r-----.-----------~--~
CK;:~~I1ED ~-------~~LPDSETP~
11ESSAGE
(LPD ABORT
LPD POST)
( LPD ZDl:lI
(KEY SETPT)---
SETPTX =
H
l\IESSAGE
LPD POIKT
ZDIM
UKDERLIKED
ABORT
CKDERLIKED
CCTTER PATH DELETED
...£.QIT
CKDEHLI~ED
KEY ZDl11
SETPT
FEDRAT
l 'N""DER'LiNE D
= XX. XXXX
( KEY ZDIM
SETPTY=
KEY :ll.-\ClIl~:
~D1"OX.=X
KEYSETPTY
( h"EY l\IACHIN/xx.x..X)
KEY FEDRAT
r--
SETPT X = XX.XXXX
SETPT Y = XX.XXXX
!\!ACIlH-;lXXXX
r-(-EDRAT' XXX. XX
ACCEPT KEY
SETPTZ =
SCALE
(
= XXXXX
\
KEY SCALE'J----t
:/
SCALE =
MESSAGE
KEYSETPTZ)
SETPT X = XX.XXXX
SETPT Y = XX.XXXX
SETPT Z = XX. XXXX
OUTPCTS
CL TAPE
CUTTER=
CCTTER DIA = X. XX-XX \-___.......:;:I.;.;;IE;.;s~S.;.;;A..;;.G.;;;E_ __t
SELECT
DRI\T SCRFACE
t--------( LPD DRIVE SURF
\
:/
~IESSAGE
SELECT
CHECK
SCRFACE
1-----1, LPD CIlECKSLR~
CCTTER DISPL\YED AT
I~TERSECTIO~,
CCT YECTOR STORED FOR
\\'R1TI);G CL TAPE
Fig. 1-12 Subroutine NUMERICAL CONTROL
MENU
FK DIMENSIONING
HORIZ - YERTICAL - PARALLEL - CuRVE - A:\GLE - OFFSET - RADIuS - DIA)\
CALCt:LATES & DISPLAYS
DIMENSION. DI;lIE:\SIO:\
LINES AND ARROW HEADS
Fig. 1-13 Subroutine DIMENSIONING
FK GEOMETRY MACRO
raEOMETRYJ~_______M_E_N_U__________~
------------------------,------;L
MACRO
START - DISPLA Y - RETURN - STORE - ABORT - CALL
~~--------~--------------------T_--~
MESSAGE
ADDS NTH MACRO
GEOMETRY TO
DRAWING
SELECT
PIVOT
LINE
Fig. 1-14 Subroutine MACRO
MESSAGE
1.3. 2 Console Procedure for Display..;.Oriented Systems
1.3. 2. 1 Points. Methods of definition of points are as follows:
1.
Using the Computer Console Typewriter:
a. Push the "typewriter" button.
b. Type P v X..; Y (now space several times) •
c. Push FINISH on the computer console.
2.
Using the Alphanumeric Pick Table:
a. Push the PICK button.
b. Pick P"; XJ Y
c. Push the NO MORE button.
3.
Positioning the Tracking Cross:
a. Push the POINT button
or
pick the POINT function
b. Position the tracking cross where the point is desired.
c. Pus.h the ACCEPT TC button
or
pick the ON function.
4.
Intersections of Surfaces:
a. Pick the INTERSE CT function.
b. Position the tracking cross on the first surface.
c. Push the THIS SURF ACE button
or
Pick the TO or PAST function.
d. Position the tracking cross on the second surface.
e. Push the THIS SURF ACE button
or
Pick the TO or PAST function.
1-53
NOTE:
Where multiple intersections exist, the point nearest the
first tracking cross location will be defined.
THIS POINT WILL
BE DEFINED
1.3.2.2 Lines. Methods of definition of lines are as follows:
1.
USing the Computer Console Typewriter:
a.
Push the TYPEWRITER button.
b.
Type
c.
Push FINISH on the computer console.
HL J K1 (now several spaces)
or
VL I K2 (now several spaces)
NOTE:
This format defines horizontal or vertical lines extending to the boundarie s of the display area.
2.
3.
Using the Computer Console Typewriter:
a.
Push the TYPEWRITER button.
b.
Type - L J Xl J Y1
c.
Push FINISH on the computer console.
v X 2 v Y2
(now several spaces)
Using the Alphanumeric Pick Table:
a.
Push the PICK button.
b.
Pick
HL I K1
or
VL J K2
1-54
c.
4.
5.
6.
7•
Push the NOMORE button.
Using the Alphanumeric Pick Table:
a.
Push the PICK button.
b.
Pick L J Xl
c.
Push the NOMORE button.
...j Y 1 ...j X 2 ...j Y 2
Two Tracking Cross Locations:
a.
Push the LINE button.
or
Pick the LINE function.
b.
Position the tracking cross.
c.
Push the ACCEPT TC button.
or
Pick the ON function.
d.
Reposition the tracking cross.
e.
Push the ACCEPT TC button.
or
Pick the ON function.
Through a Point and Tangent to an Arc:
a.
Push the LINE button.
or
Pick the LINE function.
b.
Position the tracking cross for the point condition.
c.
Pick the THRU function.
d.
Reposition the tracking cross near the desired tangency point.
e.
Pick the TANGENT function.
Tangent to Two Arcs:
a.
Push the LINE button.
or
Pick the LINE function.
b.
Position the tracking cross on the arc near the desired point of tangency.
c.
Pick the TANGENT function.
d.
Position the tracking cross on the second arc near the desired point
of tangency.
e.
Pick the TANGENT function.
1-55
8.
Through a Point at Some Angle to a Line:
a.
Push the LINE button.
or
Pick the LINE function.
b.
Position the tracking cross for the point condition.
c.
Pick the THRU function.
d.
Position the tracking cross on the line.
e.
Pick the AT ANGLE function.
f.
Pick the angle value in decimal degrees.
g.
Push the NOMORE button.
NOTE:
If the point falls on the line a degenerate line display
will be given. The display is from the point to the line.
9.
Parallel to a Line by a Given Normal Distance:
a.
Pick the LINE function.
b.
Position the tracking cross on one side of a line.
c.
Pick the PARALLEL function.
d.
Pick the. normal distance.
e.
Push the NOMORE button.
NOTE:
Connecting the end points of the two parallel lines will
form a parallelogram. The parallel lines are the same
length.
1.3.2.3 Arcs. Methods of definition of arcs are as follows:
1.
Using the Computer Console Typewriter:
a.
Push the TYPEWRITER button.
b.
Type A / Xc ..; Yc ..; X ..; Y (several spaces)
or
A Xc V Yc ..; ~ ..; Yb / Xe ..; Ye (several spaces)
center
point
beginning
point
1-56
end
point
c.
Push FINISH on the computer console.
NOTE:
The arc will be gene rated clockwise" froni' beginning
point to end point.
2.
U sing the Alphanumeric Pick Table:
a.
Push the PICK button.
b.
Pick
c.
3.
4.
A ..; Xc ..; Y c ..; r
or
A,"; Xc ";Yc ";Xb ";Y ";Xe JY
e
b
Push the NOMORE button.
By Two Tracking Cross Locations:
a.
Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross.
c.
Push the ACCEPT TC button
. or
Pick the ON function.
d.
Reposition the tracking cross.
e.
Push the ACCEPT TC button.
or
Pick the ON function.
By a Center Point and a. Radius:
.
a.
Push the' ARC button
or
Pick the ARC function.
b.
Position the t~a,cking cross'to establish the center ,?oint.
c.
Push the ACCEPT TC button .
or.,.,
Pick the ON function.
d.
Push the PICK button.
e.
Pick R Jr.
f.
,Push the NOMORE· button.
1-57
5.
6.
By a Center Point and a Line to Which it is Tangent:
a. Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross to establish the center point.
c.
Push the ACCEPT TC button.
or
Pick the ON function.
d.
Position the tracking cross on the line.
e.
PiCk the TANGENT function.
Passing Tangent·to Two Lines With a Given Radius:
a.
Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross on the first line.
c.
PiCk the TANGENT function.
d.
Position the tracking cross on the second line.
e.
Pick the TANGENT function.
f.
Pick the RADIUS value.·
g.
Push the NOMORE button.
'.
'
NOTES:
1.
The lines need not visually intersect.
2.
It is necessary to position 'the tracking crosses
on the appropriate sides of the lines.
3.
The arc will be generated in a clockwise direction
from the-first tangent line to the second.
7•
Passing Tangent to a Line and an Arc With a Given Radius!
a.
Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross
... on the line or arc •
c.
Pick the TANGENT function.
d.
Position the tracking cross on the remaining surface.
e.
Pick the TANGENT function.
f.
Pick the RADIUS value.
1-58
g.
Push the NOMORE button.
NOTES:
The arc will be generated clockwise from the first
condition to the second.
Case I: Where the line does not interest the arc, positioning the
tracking cross inside the arc will indicate one of the
possibilities illustrated below.
-,
1
~,~
"\ \
/
J
,./
--'"
,Positioning the tracking crqss o'n the line determines which of these
two possibilities is selected.
Case II: Where the line and arc intersect, positioning the tracking .
cross inside the arc indicates an internal tangency and
positioning outsIde indicates an external.tangency.
1-59
8.
9.
By a Center Point and an Arc· to Which it is Tangent:
a.
Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross to establish the center point.
c.
Push the ACCEPT TC Dutton
or
Pick the ON function.
d.
Position the tracking cross on the arc near the desired point of tangency.
e.
Pick the TANGENT function.
Passing through a Point and Tangent to a Line With a Given Radius:
a
0
Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross to establish the THRU point.
c.
Pick the THR U function.
d.
Position the tracking cross on the line.
e.
Pick the TANGENT function.
f.
Pick the radius value.
g.
Push the NOMORE button.
NOTE:
Due to the clockwise convention of arc generation, steps b •
. through d. must be reversed for certain possibilities (see
illustrations below) •
.""..,.-
/
,/
I
......
"
\
THROUGH \
1 I
,
\
-1iPOINT
\
"- ""--
I
l-qO
10. Passing Tangent to Two Arcs·With a Given Radius:'
a.
Push the ARC button
or
Pick the ARC function.
b.
Position the tracking cross on the first arc.
c.
Pick the TANGENT function.
d.
Position the tracking cross on the second arc •.
e.
Pick the TANGENT function.
f.
Pick the radius value.
g.
Push the NOMORE button.
NOTE:
Always position the tracking crosses near the desired points
of tangency. Refer to the illustrations below for internal and
external tangency conditions.
1-61
1.3.2.4 Motion Commands. Motion commands are described in the following
paragraphs.
One-Surface startup.
a.
Push PT-TO-PT MODE button.
b.
Pick the CKSURFfunction.
c.
Position the tracking cross near the surface.
d.
Pick TO or PAST or ON function.
NOTE:
Step a. is modal until changed.
TO OR PAST
FINAL POSITION
INITIAL POSITION
FINAL
POSITION
1-62
Two-Surface startup.
a.
Push PT-TO-PT MODE button (modal until changed).
b.
Pick DffiECTION function
c.
Positionthe tracking cross on first surface.
d.
Pick TO or PAST or ON function.
e.
Position the tracking cross on second surface.
f.
Pick TO or PAST or ON function.
NOTES:
1.
The second surface indicated is a check surface-and must be
the next drive surface if peripheral mode is contemplated.
2.
In the case of arcs, TO, ON or
~AST
conditions will limit
the two surface startup to two possibilities. The first tracking cross location must be nearer the desired location.
TO OR PAST
TO OR PAST
TC-to-TC Mode.
a.
Push TC-TO-TC MODE button (modal until changed).
b.
Pick CKSURF function.
c.
Position the tracking cross as desired.
1-63
d.
Pick the TO or PAST function.
NOTE:
This mode operates independently of the "gravity field" ~ The
cutter path is from one tracking cross location to another.
PT-to- PT Mode
a
0
Push PT-TO-PT MODE button (modal until changed).
b.
Pick the CKSURF function.
c.
Position the tracking cross as desired.
d.
Pick the TO or PAST function.
NOTES:
1.
Major use will be for drilling, boring, and reaming operations. However, it can be used to offset surfaces in the
same way a one surface startup is used. Reference the
one surface and two surface startup.
2.
This mode is modal within the system.
Peripheral Mode
1.
Step Method:
a.
Push PERIPHERAL MODE button (modal until changed).
b.
Pick the CKSURF function.
c.
Position the tracking cross near a surface.
d.
Pick TO or PAST or ON function.
NOTES:
1.
This method will generate one cutter path at a time. It
essentially steps from one surface to the next, as indicated.
2.
A startup is required prior to using the peripheral mode.
1-64
3 TO OR PAST
TO· OR PAST
2.
Automatic Method:
a.
Push PERIPHERAL MODE button (modal until changed).
b.
Pick DIRECTION function.
c.
Position the tracking cross near the next surface.
d.
Pick TO or PAST or ON function.
e.
Position the tracking cross near the final surface.
f.
Pick TO or PAST or ON function.
NOTES:
1.
All surfaces must be connected by end points to form
a continuous string of
2.
line~
and arcs.
Step c. must be the next line or arc.
1-65
NOTES (Cont.)
3.
There must be no ambiguous connections; that is, there
cannot be more than one connected path to the final
surface.
4.
This routine should be avoided because it has not been
. debugged.
TO OR PAST
~--.~
Go To Point.
a.
Push the PICK button.
b.
Pick G
c.
Push the NOMORE button.
vX
..j Y ..j Z.
Go Forward.
a.
Pick GOFWD function.
b.
Position the tracking cross.
c.
Pick the TO or PAST or ON function.
1-66
WILL GO FROM 1 TO 8
WITHOUT HELP
·NOTE:
This feature overides the convention which states that the last
CKSURF will be the next drive surface. The last drive surface will be retained and motion made to another CKSURF.
Direction. Direction is used in three ways:
a.
As the first surface in a two surface startup.
b.
To indicate the direction of movement in automatic peripheral mode.
c.
To establish direction of motion around arc
dri~e
surfaces. By conven-
tion, direction of motion is established by measuring the angle of sweep
between the cutter center and the tracking cross position for the CKSURF. Movement is in the direction of the small sweep angle.
This convention may be overridden by
e~tab1ishing
a small sweep angle with judi-
cious positioning of the DIREC.TION tracking cross.
AUTOCKSURF. When operating in tracking cross and peripheral mode, repetitive picks
0
== the CKSURF button are required.
AUTOCKSURF. eliminates this
repetitive pick by reactivating the CKSURF routine after each movement.
Reject Cutter Path. The last, but only the la.st, item of motion data may be rejected. This item might be a PARTNO, Z-VALUE, cutter path, mode button,
etc.
Number of Cutter Paths •. The number of cutter positions which are displayed is
modally set at four. This may be altered at any time by:
a.
Pushing the PICK button
b.
Picking NCP
co
Pushing the NOMORE button
~
K
Where K = the number of cutter positions to be retained on display.
Check Points. CL-DATA may be segmented by entering check points from time
to time. This permits faster access to portions of CL-DATA deep within a part
1-67
program. Thus, when creating a program, the more check points which are
entered the more directly accessible positions within CL-DATA are available
when replaying.
a.
Push the PICK button.
b.
Pick CKPNT.
c.
Push the NOMORE button.
1.3.2.5 Machine Functions. Machine functions are described in the following
paragraphs.
Part No.
a.
Push the PICK button.
b.
Pick PARTNOXXXXXXXXXXX.
c.
Push the NOMORE button.
NOTES:
1.
A PARTNO should not exceed 68 characters in length.
2.
A PART-NO is required to dump to magnetic tape.
Cutter.
a.
Push the PICK button.
b.
Pick CU ..; Kl ..; K2 •
Push the NOMORE button.
c.
NOTE:
Kl
= Dia, K2 = Corner Radius
Z-Value.
a.
Push the PICK button.
b.
Pick Z ..; K.
c.
Push the NOMORE button.
1-68
· NOTE:
K is a positive or negative number.
Feed Rate.
a.
Push the PICK button.
b.
Pick FR. v' Kl v' K2 •
Push the NOMORE button.
c.
NOTE:
Kl = Desired feedrate in inches per· minute.
K2 = 88 for auto acceleration and deceleration.
K2 = 92 for step acceleration and deceleration.
~
K2 is not given, the modal value in postprocesso.r
will be use d.
Tolerance.
a.
Push the PICK button.
b.
Pick TOL J K.
c.
Push the NOMORE button.
NOTES:
1.
For external profiling around arcs the K value represents
the maximum tangential deviation from circular.
1-69
NOTES, (Cont.)
2.
For internal profiling around arcs the K value represents
the maximum cordal deviation from circular.
Machine Tolerance.
a.
Push the PICK button.
b.
Pick MCHTOL
c.
Push the NOMORE button.
v K.
STOP.
a.
Push the PICK button.
b.
Pick STOP.
c.
Push the NOMORE button.
Reference Point.
a.
Push the PICK button.
b.
Pick REFPT V K.
c.
Push the NOMORE button.
Coolant Control.
a.
Push the PICK button.
b.
Pick ¢>NKUL
c.
Push the NOMORE button.
or
pick ¢>FFKUL.
1-70
Spindle Speed.
a.
Push the PICK button.
b.
Pick ¢>NSPIN v K
or
pick ¢>FSPIN.
c.
Push the NOMORE button.
NOTE:
K = spindle speed in rpm.
Positive values of K mean clockwise rotation.
Negative values of K mean counter'clockwise rotation.
REWIND.
a.
Push the PICK button.
b.
Pick REWIND.
c.
Push the NOMORE button.
a.
Push the FINI button.
FINI.
A message will appear in the upper left-hand corner above the frame stating:
"Verify readiness of proper dump tape."
b.
Communicate via the intercom with the computer operator and verify that
the paper tape is mounted on Channel 0, Unit K, where K is the console
number of the requesting station.
c.
Push the FINI button again.
All data relative to the job on Console K will be written on magnetic tape.
This is the master record of replay tape to be labeled.
CL Tape.
a.
With the FIN! procedure complete, push CL-TO-PP button.
A message will appear stating: "Verify readiness of proper postprocessor
tape. "
1-71
b.
Communicate via the intercom with the computer operator and verify that
the proper tape is mounted on Channell, Unit 2.
c.
Push the CL-TO- PP button again (Ref. Operations Section for Tape Control
Procedures) •
1.3.2.6 Special Features. Special features are described in the following paragraphs.
Arc Centers. Picking the Arc Centers button will cause displayed center points to
disappear and undisplayed center points to appear.
CANON. CANON is used to obtain canonical forms of elements.
a.
Pick the CANON function.
b.
Position the tracking cross on the desired element.
c.
Push the TillS SURFACE button
or
pick the TO or PAST function.
d.
See Note 5 below.
NOTES:
1.
The canonical form is displayed above the upper left hand
edge of the frame.
X I Y (coordinates).
2.
The canonical form of a point is POINT
3.
The canonical form of a line has the end points.
Line
4.
Xl I Y 1 I X2 I Y2·
The canonical form of an arc is the center point and radius.
ARC
5.
=
=
=
X IY IR.
After observing the canonical form of the first element, other
elements may be interrogated by repeating: steps b. and c.\
above. The feature is terminated by picking the CANON function again.
1-72
Delete Geometry. This feature is used to delete from display and memory all record
of the geometric element being removed.
a.
Pick the DE LETE GEOM function.
b.
Position the tracking cross on the desired element.
c.
Push the THIS SURFACE button
or
pick the TO or PAST function.
The tracking cross will go away and reappear. If it is desired to delete
additional elements repeat steps b. and c. above.
d.
Pick DELETE GEOM function again. This terminates delete and removes
the display.
NOTE:
Do not delete an element used as a DIRECTION surface or aCHE CK
surface of motion command.
They are required for REPLAY.
DUMP. This feature is used to periodically store on magnetic tapethe information
relative to a specific console.
a.
Push the DUMP button.
A message will appear above the left hand edge of the frame stating,
"Verify readiness of proper dump tape. "
b.
Communicate via the intercom with the computer operator and verify that
the proper tape is mounted on Channel 1, Unit K, where K is the console
number of the requesting station.
c.
Push the DUMP button again (Ref. Operations Section Tape Control).
ERASE. This feature is used to abbreviate the display of lines and arcs to provide
clarity and/or connectivity.
a.
Pick the ERASE function.
b.
Position the tracking cross on the surface to be shortened.
c.
Push the THIS SURFACE button
or
pick the TO or PAST function.
1-73
d.
Position the tracking cross in space, near an end point or on a surface.
e.
If an end point or tracking cross position:
Push the ACCEPT TC button
or
pick the ON function.
If a surface:
Push the TIDS SURFACE button
or
pick the TO or PAST function.
f.
Repeat Steps d. and e. for a final condition.
NOTES:
1.
For arcs, ERASE functions in a clockwise direction from the
first surface or point to the second.
2.
ERASE should not be used in place of DELETE.
EXTEND. This feature is used to lengthen the display of lines or arcs for clarity
and/or connectivity.
a.
Pick the EXTEND function.
b.
Position the tracking cross on the surface to be lengthened.
c.
Push the THIS SURFACE button
or
pick the TO or PAST function.
d.
Position the tracking cross near the end point to be affected.
e.
Push the ACCEPT TC button
or
pick the ON function.
f.
Position the tracking cross in space or near a surface.
g.
If in space:
Push the ACCEPT TC button
or
pick the ON function.
If near a surface:
Push the TillS SURFACE button
or
pick the TO or PAST function.
1-74
· NOTE:
EXTEND functions from one end point of the surface to a
tracking cross location or to another surface.
INPUT TAPE. Any feature which may be input via the alphanumeric pick table may be
read in from magnetic tape. Punched cards, beginning in Column 1 and using the
identical formats which would be used if entered via the alphanumerics, are put on
magnetic tape. The last card must be double punched 7 and 8 in Column 1. Several
groups of input cards may be put on the same magnetic tape if separated by the double
punched 7,8 card. The order and number of groups must be known by the part programmer.
a.
Push the INPUT TAPE button.
b.
Verify, via intercom, that the proper tape is mounted on Channel 1, Unit O.
c.
Push the INPUT TAPE button again ~
Invisible Geometry•.. This feature is used to temporarily remove from display geometric surfaces in .order to provide clarity or .ease in "attaching".
a.
Pick the INVIS GEOM function.
b.
Position the tracking cross on the surface desired.
c.
Push the THIS SURFACE button
or
pick the TO or PAST function.
d.
Repeat Steps b. and c. as many times as desired.
e.
Pick the INVIS GEOM function again.
JUMP TO CKPNT. This feature is used to replay through CL-DATA without performing all calculations and displays. You may jump to check points and begin stepping
through all operations. This feature shortens the required time to make corrections
or check certain areas for errors if you are replaying and in tracking cross mode.
a.
Push the JUMP TO CKPNT button.
b.
Pick the CKPNT number.
c.
Push the NOMORE button.
1-75
NOTES:
10
If the last CKPNT number is not known, a very large number
(i.e., 1000) will jump to FINI.
2.
The last cutter display will appear and the last Z-VALUE, feedrate
value, mode, etc. will be activated.
3.
After stepping beyond the PARTNO, the JUMP TO CKPNT feature
must be used only when in tracking cross mode.
Light Pen. The light pen may be activated by depressing the switch on the pen or by
depressing the light pen button on the button box. This button further serves to indicate an active pen by remaining lighted when switch is depressed.
LOAD. This feature recalls from magnetic t~pe storage only geometry data. Thus,
if a REPLAY TAPE is read in using.the LOAD feature, no CL-DATA data will be
entered.
a.
Push the LOAD button.
b
Verify readiness of proper LOAD tape.
0
c.
Push the' LOAD button again.
MASTER REPLAY. This feature is used to select and load a REPLAY TAPE from a
magnetic file containing numerous REPLAY tapes.
a.
Push the MASTER REPLAY button.
b.
Verify readiness of proper LOAD tape.
c.
Push the MASTER REPLAY button again.
d.
Pick the desired file number.
e.
Push the NOMORE button.
NOTES:
1.
See Operations for further information on master files.
2.
Do not use MASTER REPLAY unless the file number is
small or requirement is imperative.
1-76
OBTAIN REFSYS. This feature is used to reenter a closed reference system in order
to REFSYS additional geometric inputs.
a.
Pick the OBTAIN REFSYS function.
b.
Position the tracking cross on the REFSYS axis indicator.
c.
Push the THIS SURFACE button
or
Pick the. TO or PAST function.
Origin. This feature moves the coordinate system about the face of the scope in order
to view different portions of the display.
a.
Push the PICK button.
b.
Pick (> J X J Y.
c.
Push the NOMORE button.
NOTE:
X and Yare values from the center of the scope to the
coordinate system.
Origin and Scale. This feature combines the ORIGIN feature with a scaling feature.
The ORIGIN shift is performed first then the scalillg is performed about the center of
the scope.
a.
Push the PICK button.
b.
Pick(>JXJYv'S.
c.
Push the NOMORE button.
NOTE:
S is the desired 'scale value.
Reference System. This feature allows the temporary establishment of a second (or
more) coordinate system.
a.
Push the PICK function.
1-77
b.
Pick RFS VX VY V a
or
RFS J X J Y
or
RFSva.
c.
Push the NOMORE button.
NOTES:
1.
A half arrow is displayed from the origin of the reference
system along the + X axis to visually denote its orientation.
2.
If a second
ref~rence
system is. input with the first one active
this second REFSYS will be with respect to the first REFSYS.
The first will become dormant and the second active.
3.
No geometric formats which require tracking cross assistance (i. e., a line parallel to another line) may be used with.
a REFSYS active.
4.
Reinitialize.
To deactivate a REFSYS:
•
Pick RFS •
•
Push the NOMORE button.
This feature restores the console to its initial state removing any
record of previous work done. All modal values are reinstated and the console set up
as if at the sign-on stage.
a.
Push the REINIT button.
NOTE:
Do not use REINIT until all information is safely stored on
magnetic tape.
See All Invisibles. This feature recalls the display of all surfaces previously made
invisible.
a.
Pick the SEE ALL INVIS button.
1-78
Scale. This feature is used to reduce or expand the size of the display. The change
of scale is accomplished about the center of the scope.
a.
Push the PICK button.
b.
Pick S y K, where K is the desired scale value.
c.
Push the NOMORE button.
System Dump. This feature is used to store on magnetic tape all information necessary to reestablish the present condition of the system. If a computer stop occurs,
the last System Dump may be ,read into the computer and work may continue from that
point. This feature is an expansion of the DUMP feature in that it dumps all data for
all console s •
Every 30 minutes the computer operator will interrupt console operation to take a
system dump. No action of the part programmer is required. After the system dump
has been taken the computer operato'r will notify the part programmers that they may
continue.
Tracking Cross. If for some reason the system fails to return a tracking cross when
required,one may be obtained by pushing the Tracking Cross button.
REPLAY. The REPLAY feature is used to:
•
Re-enter an incomplete part 'program in order to complete it
•
Check a part program
•
Make corrections to a part program
•
Make additional "dash-numbered" parts from existing part programs
a.
Push the REPLAY button.
b.
Verify readiness of proper load tape.
c.
Push the REPLAY button again.
The geometry will appear in the configuration existing, at the time of dumping or FINI'ing.
d.
Push the STEP button.
The PARTNO will appear to further verify that the proper tape was loaded.
1-79
•
STEP:
If you desire to check the entire cutting sequence, push the STEP
button. If at any time you wish to alter CL-DATA you may pick REJECT C/p
which will reject the last STEP performed. If at any time you wish to add
motions or CL-DATA, you may do so by normal procedures. Additions must
eventually connect logically with the function of the next STEP s1
~
1(0+)
IN
XXXX
2N
XXXX
TYPE
FREQUENCY
DUTY CYCLE
PEAK AMPlJTUDE
FREQUENCY SWEEP
~ c[
BODY
E(O+)
LEADS
}
Fig. 2-1 Typical Component Templates
2-17
USER DEFINED
CONDUCTOR LINES
6000
1
20
COMPONENT
~LEAD
5
3
~----
5
1
4
2N335A
10K
1000
__- -..... 5
500
Fig. 2-2 Completed Schematic
2-18
o Point/Point Connect: By picking two points on the working surface and
activating the connect function, .an automatic connection will occur. The
picked points may be located on conductor lines, component leads, void
areas. A point pick on any component lead will be reset to the end of the
lead opposite the component body. Additionally, picks on conductor lines
other than endpoints will cause the conductor lines to be segmented (and
introduce a connectivity circle at the intersection point on the wire).
o Offset Connection: Only horizontal and vertical conductor lines are generated. Therefore, if the point picks are offset, a vertical line is drawn
from the first picked point to the horizontal plane of the second picked point
and extended horizontally to the second point.
o Point/Line Connect: The connection of a point and a picked position on a
line is accomplished by a normal being drawn from the point to the line.
The intersection of the nor:r:nal and tlie line will be segmented.
o Line/Line: The picking of two lines which are normal to each other will
cause both to be segmented and a connectivity circle will be created at the
intersection.
Erase. Components, conductor lines, and guide lines can be selectively erased by
the light pen picking a point on the desired entity and activating the erase function.
The templates are protected from the
~rase
function.
Zooming. This function provides for the magnification of schematic presentation and
is activated by requesting the ZOOM function. This magnification can be increased
or decreased in integral increments of powers of two.
Duplicate Function. This function allows the operator to produce a duplicate copy of
a collection of conductor lines and components at another position on the work area.
The duplicate copy has the same orientation as the original. The operator may make
as many copie s as needed.
2-19
Annotation. The circuit schematic must be annotated. Annotation includes:
•
Parameter data
•
Command data
•
Node numbers
•
Wave forms
Node numbers are assigned automatically, a node number appearing by each component lead. Parameter data and command data are entered by the user employing
general purpose light registers in conjunction with a label function and displayable
alphanumeric font. Parameter data for a resistor would be a numeric value denoting
its resistance, whereas parameter data for a generator would include the type, frequency, duty cycle,. etc. Command data includes:
•
Transient time interval
•
Time step selection
•
Frequency values
Wave forms are also input by the user. A minimum and a maximum value are input,
both for the
x- and Y-axes. The independent and dependent variable(s) are also
entered. Grids for these are then generated and displayed. The user picks points on
the grid(s) which are then connected by a best curve fit.
List. When the user is satisfied with a schematic, he calls on the LIST function. This
function verifies that parameter data exists for each component and
inform~
the user
of unconnected component. leads , dangling conductor lines, etc. If the annotation is
complete, the nodes are numbered automatically. If the user desires certain node
numbers to remain fixed, a manual override capability exists. The user picks those
nodes to remain fixed and the other nodes are renumbered accordingly. Input for the
analysis routines is then extracted from the schematic and stored for later use. A
completed schematic is shown in Fig. 2 -2.
2-20
2. 7. 3 Analysis/ Graphic Displays of Output
Analysis. After making any necessary corrections to the schematic, the user calls
on an analysis routine by picking a light button for the appropriate program. The
user then selects from the types of analyses; AC, DC, or transient. A log of activities
is output on the printer, as well as the topology of the schematic. Printer output of
final results is optional.
Graphic Displays. Graphic displays of output are provided, both in table form and in
plot form. Table displays are given by selecting the table function and the type of
data desired, such as node voltages, transfer functions, plot data, and branch currents. Linear or semi -logarithmic plots are also available. Prior to performing the
analysis, plot variables are selected from a list of variables. The schematic and the
selected plots are displayed at the same time. Plots are generated in real time and
plot points appear" in a random order as the analysis proceeds. The user thus sees
the trend and may terminate analysis before completion if results appear unsatisfactory.
Plots may also be selected after analysis is completed in the same manner. In this
case, plots are generated immediately in their totality.
Hard Copy Plots. The same displays mentioned previously, including the circuit
schematic, may also be obtained as hard copy output from the SC 4"020 or Gerber
plotters.
Review. The user interprets the results at hand.
He may then return to any point
above and continue in the design phase. For example, he could redesign part of the
schematic, modify the input data, or call a different analysis routine.
2.8 NETWORK ANALYSIS GRAPHIC8.APPLICATIONS AT LOCKHEED
Graphics status at two Lockheed facilities is described in the following paragraphs.
2-21
2.8.1 status of Graphic Electrical Network Analysis at Calac
The graphic electrical network analysis program at Calac is being run on an mM 360
computer with two 2250 display consoles. It has been demonstrated to several potential
users. A new version of this program is being developed and is scheduled for release
by November 2, 1967. The approximate cost per scope is $37.00 per
h~ur.
2.8.2 status of Graphic Electrical Network Analysis at Gelac
Gelac does not have a Graphic Electrical Network Analysis program. Their future
plans are to participate with CAD/LMSC for program application and development.
2.9 NETWORK ANALYSIS GRAPHICS APPLICATIONS AT OTHER COMPANIES
MIT is using an on-line capability to implement AEDNET for circuit design modification. Circuit diagrams are fed into the computer with a light pen and oscilloscope,
and circuit parameter data entered with a typewriter. The operator views the oscilloscope and, using the light pen, locates and moves circuit elements. Component
o
parameter values are assigned using the
typewr~ter.
•
The output displayed is selected
by the operator, and eliminates all unwanted information.
Bell Telephone Laboratories is using a console system for problems requiring "dynamic
scratchpad" capabilities. Included in such problems have been:
•
Printed circuit component and wiring placement
•
Schematic circuit design
•
Block and flow diagram design
•
Text composing and editing
•
Placement of cards on a chassis to achieve connecting wire length
IBM uses a small digital computer connected to a large-screen buffered display
equipped with a light pen to aid in the design of mask artwork for hybrid integrated
circuit modules. The operator uses the light pen to construct a circuit schematic
2-22
on the display and the schematic is stored in the computer memory. The operator
then uses the light pen to generate the artwork for fabrication of the circuit mask.
While the artwork is being generated, it is automatically checked against the stored
schematic. When the artwork is complete, the computer is used to drive a plotter
that generates the final mask artwork.
Norden Division of United Aircraft Corporation utilizes CADIC (Computer Aided Design
of Integrated Circuits) to produce mask artwork. Circuit definition is entered by the
use of punched cards. These input data are then translated into size and shape of every
component, recalled from computer memory. From this input a layout is displayed on
the CRT. The operator then uses the light pen to manipulate the components and make
desired changes immediately. CADIC displays the initial layout, but the detailed mask
geometries can be called up by the operator by means of the keyboard. Hard copy of
mask geometries or detailed layout can be obtained, but a direct interface with maskplotting equipment is not presently utilized.
Future work will include this capability.
2.10 PROGRAMMING REQUIREMENTS
2.10.1 System Ground Rules
System ground rules are as follows:
a.
At completion of each program (function) a message indicating completion
appears in the message (l\ffi) register located on the control surface.
Requests to the operator during execution of functions are displayed in the
l\ffi or text (TX) registers (see Fig. 2-3).
b.
The operator will be locked out (cannot interrupt function execution) until a
mes sage appears in the l\ffi or TX registers with one exception - during
analysis function the operator has the option of stopping execution.
2.10.2 System Programs
Initialize. This program is called by picking the primary light button (LB) UC and its
secondary LB AP.
First, the program searches the computer graphic list for the
2-23 .
MR:
PICK:
TX:
RES
CAP
. IND
ANAL 1
ANAL 2
ANAL 3
DIO
~TRA
:GEN
~ BOX
EN
UC
PLOT
LIST.
WAVE
HALT
'X:
Y:
VALU:
FREQ:
TYPE:
DUTY CYL:,
PEAK AMP:
NOTE: WORKING SURFACE IS 11 BY 17 IN. ON 22-IN. CRT
Fig. 2-3 Proposed Control Surface Layout
2-24
J
frame entity. If this is not found the computer graphic list (CGL) is assumed to be
empty. The frame and control surface (CS) controls are created, the keyboard keys
are defined, the template library is entered in the CGL from card images on disc, and
the model library is input on cards and written on the disc. If the frame was found,
initialization was performed previously.
MR: ENA SYSTEM READY
Component Select. The user picks the LB acronyn for the desired component on the
right-hand side of the CS (see Fig. 2-3). The program examines the pick table for
destination picks and rotation picks. (If rotation -is desired, the rotation pick must
directly follow the destination pick.) Translations are made using the top right-hand
lead of the template as it appears in the library as the origin point. As translation
occurs, the category fields are changed to differentiate between template and component
and the component is entered in the CGL.
MR: END COMPONENT SELECT
Connect. This program is called by pushing keyboard key 5 (see Fig. 2-4). The pick
table is exam,ined and picks are taken in pairs for connection purpos,es. Depending on
the types of picks in each pair, one of three connections is made: point-to-point,
point-to-line, or line-to-line. The connection lines are entered in the CGL.
MR: END CONNECT
Erase. The user makes picks on those items to be erased and then depresses the
erase keyboard key. The entities selected for erasure are removed from the CGL.
If a pick is on any part of a component, the entire component is erased. Only tem-
plates and the application frame are protected from erasure.
MR: END OF ERASE
Zoom. The primary LB FM (frame manipulate) is picked. Three secondaries appear:
MF (move frame), DP (double viewed area),
~nd
HP (half viewed area). The second-
aries can be picked singly or in combinations. The operator pushes the accept key
after picking secondaries. If the MF is picked, the program requires a point pick in
t!J,e pick table and the frame is centered about that pick.
2-25
.1
IlEneT
--- I~~a
L ABEL
-
ERASE
Iiu
...
C
--
P OINT
-Fig. 2 -4 Proposed Keyboard Layout
i~NDi
POINT
Duplicate. This function is called by pressing the DUPL key. The program interogates
the pick table for items picked. These items are the ones to be duplicated. The user
is then requested to make an origin pick on a component lead (this pick will be trans1ated to that terminal pad) and then will ask for a destination point for that terminal.
After the copy is made (entered in the CGL), the user terminates the function by pressing REJECT, or makes another copy by pushing ACCEPT.
MR: END OF DUPLICATE
Label. This program is called by keyboard action. It examines the pick table for a
component pick and register picks. The value is extracted from the picked register(s)
and the component pick is converted to that component's annotation group. The value
is entered in this group and displayed beside the component, the position of the annotation being determined by the orientation of the circuit component.
Wave forms are input by selecting the wave LB. The user is requested to enter the
variable names in some register and the minimum/maximum values for the X and Y
axes in the X and Y registers. This information is extracted by the wave program
and a corresponding grid is constructed and displayed. The user is now requested
to pick points on the resulting grid and to push ACCEPT. The program connects the
points with a best curve fit and extracts sufficient data points on the curve for future
use.
MR: LABELING COMPLETE
List. The primary LB LIST and its secondary LIST are picked to call the LIST function. This program searches the CGL for -components and checks these for proper and
necessary annotation. The user is informed of any problems via the MR or TX registers or printer output. The connectivity of components is then traced and the operator
is informed, as above, of dangling wires, unconnected leads, etc. Any missing annotation or connectivity problem results in termination of this function and control is
returned to the operator for corrections. If no problems arise, node numbering can
begin. The operator is requested to pick those nodes he wishes to remain fixed and
2-27
push ACCEPT. Picks, if any, are examined and the node numbering is accomplished
accordingly. The topology of the schematic is stored for use by analysis routines.
The connectivity algorithm follows.
MR: END LIST
Connectivity Algorithm. Definition: A pad is the end of a lead opposite the component
body. The following steps are performed by the list function:
1.
Search the CGL for circuit components. Check each for necessary anIlotation.
If a component is properly annotated, continue the search. If not annotated
properly, assign a unique name to that component; for example, HI for the
first resistor encountered which is not properly annotated. This name is
displayed to the component and the operator is told that the component of
that name needs annotation.
At the finish of the CGL search execute Step 2 if all components have the
necessary annotation. If not, return control to the operator so that he may
add the required annotation to the schematic. The operator makes changes
or additions to the schematic and selects the LIST function again. On subsequent passes through step 1 names entered during previous searches are
removed from the display.
2.
Search the CGL for component leads. For each lead encountered the following
data are added to a table:
• X, Y coordinates for the lead's pad
• A count value
• A group number
where the count is, with one exception, 0 and the group number is, with
one exception, incremented by 1, beginning with 1, for each lead found in
the CGL. If the X, Y coordinates of a pad match the X, Y coordinates of a
pad already in the table, then the count of both is set to 1 and the new pad
entry is given the same group number as its match.
When the entire CG L has been searched, continue to Step 3.
2-28
3.
Search the CG L for conductor. lines. Let the endpoints of a conductor line
be (Xo' Yo) (Xl' Y1)·
Is the pair (X , Y ) in the table described above? If the answer is yes,
o
0
retain the group number, N, of the match, set the count of the match equal
to 1, and determine if (Xo' Y 1) is in the table. If (Xl' Y1) is not in the
table, add the coordinate Xl' Y1 to the table with count 0, and group
number N. If (Xl' y 1) also matches some entry in the table, retain its
group number M, set all group numbers equal to M or N equal to the smaller
of M and N, and set the counts of those entries equal to 1.
If (Xo' Yo) is not in the table, check if (Xl' Y1) has a match in the table.
If (Xl' Y 1) is also not in the table, enter (Xo' Yo) and (Xl' Y 1) in the table
with counts 0 and group numbers J +'1, where J was the last assigned group
number. If (Xl' Y1) is in the table, retain the group number of its match,
N, set the count of its match equal to 1, and enter (X , Y ) in the table with
o
0
group number N.
When all conductor lines have been found and interpreted the table is complete and its contents are examined .. If any pad entry has a .count equal to: 0
it is an unconnected lead; if any conductor line end point has a count equal to
0, the conductor line is dangling and, in either case, the operator is notified
of the problem. The group number of each pad with non-zero count, with
one exception, will be identical with the group number of one or more additional entries in the table; these pads (leads) being connected. A lead may
be connected to a dangling conductor line, and therefore have a non -zero
count, but no pad with matching group number in the table. The analysis
and the technology of the network is node to node. Using the above table,
nodes are now assigned sequentially, with the ground always node zero.
Leads with the same group number are of the same node and are numbered
as such. If leads A and B are of the same group number and the A pad
coordinates and the B pad
coordi~ates
are identical, then only one node
number appears on the schematic; whereas, if the pad coordinates are not
equal, the node number appears alongside each lead.
2-29
Analysis Driver. Select an analysis LB on the left side of the control surface and
then pick one of the secondaries which appear, denoting the types of analysis available
(AC, DC, transient, etc.). This calls the analysis driver program. The analysis
options selected and the schematic topology are output on the printer. The selected
analysis routine is then called and execution continues. The operator may halt analysis at
~ny
time by picking the HALT LB.
Final output of analysis is
op~onal.
Analy-
sis results are written on disc.
MR: ANALYSIS COMPLETE
Graphic Displays. To call a table display, select the PLOT LB and the TABLE secondary LB. The program displays a list of available tables on the working surface and
requests the user to pick the table(s) desired. The data are retrieved from the disc
and displayed in table form.
MR: TABLE COMPLETE
Plots generated during analysis are in real time. Before calling the analysis routine
the operator selects plot variables from a list of those available. The analysis driver
constructs grids for plots if real time plot option was indicated. Points are extracted
as analysis proceeds and appear as X's on the re.spective grids.
Hard Copy. Hard copy plots are available by picking the PLOT LB and either secondary 4020 (SC 4020) or Gerber. In either case the operator is requested to hang magnetic tape and put the identification card in the read hopper and push ACCEPT. The
program searches the CGL for all displayed items, performs and outputs plot data
only for graphic entities currently displayed on the 11- by 17-in. display area. Tape
is then submitted to the appropriate plotter to obtain the hard copy.
MR: END HARD COpy
2.10.3 Component Group structure
The component group structure contains specific subgroups: the body subgroup, the
annotation subgroup, and one or more lead
s~bgroups.
2-30
Each subgroup is made up of
one or more graphic or alphanumeric entities grouped together by a linkage entity.
For
~stance,
a lead subgroup consists of a line entity and a dot entity whereas an
annotati~ subgroup
consists of one or more alphanumeric entities.
A capability of identifying the type of component represented has been established.
The reference word of a· component or template top group contains two 12";'bit binary
fields: (23-13), not defined and (12-0), unique element type no., which is the same
for all interchangeable symbols.
The reference word of the top group of a lead contains two 12-bit binary fields also:
(23-13), number of leads on the component and (12-0), lead series number.
The component group structure is depicted in FIg. 2-5 as both a schematic structure
and a group structure. The schematic symbol is shown with its value designation.
The group structure categories are given in octal (RW, reference word; CAT,
category).
2-31
TRANSISTOR
2N1302
GROUP STRUCTURE
- - - ---------
",--/
(
----.......1
I
~N1302
~-:---.....,
((Q ~
\
"
__ ANNOTA TION
GROUP
CAT = 54
CAT 4Y,
~-~,
BODY GROUP
CAT = 44
~
/------...,
I
I
\
\ CAT = 56 )
~~~-50 ,
CAT = 51
(
.
'\
)
\
\
'"
TOP GROt (
...........
LEAD GROUP RW = '3/1
CAT = 50
)
RW = X/T* "
/
CAT = 40.........
./
LEAD ,GROUP RW = 3/3
I
CAT = 50
(
I
\"
I
I
,
~
I
I
-" ,
~--
I
I
I
\
""""'-
-- -
"
LEAD GROUP RW = 3/2
CAT = 50
------".
~-- - - - _ . . - /
*WHERE T IS THE CODE FOR THE TRANSISTOR
Fig. 2-5 Sample Component Group Structure
2-32
/'
Section 3
FLIGHT MECHANICS
3.1 MODE OF APPLICATION OF GRAPHICS
Computer-graphics can be an extremely helpful computing tool and adjunct.
To get the
most from it the user must capitalize on its pictorial and interactive aspects.
He must
incorporate these special aspects into his problem-solving plans at an early stage in'
order to integrate the graphics capabilities and controls into his computational solution
and to bring graphics to bear as effectively as possible on this solution.
The following
paragraphs discuss ways to most profitably employ graphics, in solving analytical
flight mechanics problems, and cautions against certain pitfalls based on this task
group's experience with graphics programs.
3.1.1 Recommended Modes
First the operating mode should be kept interactive, that is, it should provide for communication between the computer and the user during the run; he shoulq attempt improvements and note their effects. Furthermore, the user should depend upon graphical indicators from the console scope to show the progress of the calculations, to point
out error sources, and to indicate reactions to changes made during the run. If the
system is not pictorial and interactive~ preparation of a graphics program' is probably
not required or justified; results could just as well be obtained with a QUICKTRAN or
TYMSHARE terminal or a batch-processed CRT output device, possibly at less expense.
Full use should be made of the user's sensory powers, deductive powers, and memory.
When the results of a serie s of events are properly presented to him, he can interpret
and effectively monitor their progress and direction. His abilities to draw conclusions
and act in a nonlinear manner should be heavily relied upon; linearized computations
and decisions should be left to the computer.
3-1
The heart of every problem should be a good working picture. It should show the
principal working variables, or at least it should clearly present the results of changes.
From the information and control options on the console display, the user should be
able to control all the significant variables. Just what these displays are, or their
best choice, may have to be learned from experience with the graphics
~ystem
itself,
with the system being improved in later modifications. Conceivably, the displays may
have to be changed to suit a new and different application.
All displays, messages, and controls put before the user should appear in his language,
not that of the computer or the program. General computer or console software functions should not be his concern. All indicators and commands should be phrased in
physical and computational language peculiar to his problem so that he can quickly and
clear ly understand and react to them.
All commands the analyst gives should require only a single action for implementation.
His action should set off a series of automatically sequenced, logical and computational
steps within the computer-graphics system. His instructions are, in essence, "macroinstructions" because they trigger a large number of other
machine-~riented
instruc-
tions. , Single-instruction commands keep him free from routine, distracting decisions.
For each action he takes, the system should display a positive indication that it recognizes and accepts it. For example, the response to selecting a command instruction
or to requesting a change in a displayed variable might be an immediate underlining of
that display. The response to typing in a new value of a variable might be· the display
of the numerical data 'in place of the old.
It is certainly desirable, as suggested in Ref. 3-1, to give the user control over all
input variables and system parameters. In this way he is 'able to perturb them to compute the sensitivities of his solUtions, to changes in them, and to possible tradeoffs
among them.
The displays should come fast enough to insure that the user can remember them and
sense changes. It is even wise to keep the displays driving the user and forcing him
3-2
to make comparisons and decisions in order to stay abreast of the progress of the
calculations. Pauses for evaluation should be allowed only as absolutely required by
the complexity of the final product. Rapid operations will require
car~ful
use of data
storage devices and careful planning of the transmittal of large blocks of data, and may
ultimately call for increased hardware capability for storage and data transmission
devices.
Many of the misuses of graphics can be easily inferred by taking the opposite approach
from the preceding recommendations. One of the commonest classes of pitfalls is relying too heavily on generalized computer programs· and display· software. The results,
are generally exceedingly slow-acting programs and overly cumbersome graphics systems.
For example, in developing the ·PRESTO trajectory optimization program (to be
described in Para. 3.4), the Flight Mechanics Task Group saved an order of magnitude
in computing time by discarding the general-I:>urpose, variable-step-size integration
subroutine commonly in use and developing one specifically tailored to powered-flight
problems. The hypersonic aerodynamic formula (FORMAC) computer-graphics system
developed earlier by LMSC pointed out the problems associated with multiple-command,
sequential, general-purpose, computer-orient-ed display software.
.
'1;he commands were
"
in some cases so cumbersome that, at best, the user was considerably delayed in working out a problem and, at worst, he could lose his train of thought and purpose in conducting a particular investigation. Many routine data inputs and display changes'
required him to make a repetitive string of choices and decisions and, consequently,
bogged him down in unnecessary detail.
In summary, the features of a good computer-graphics system are that it is particu-
lar ly keyed to graphical display and interaction, that it is fully controllable by the user,
and that it is fast enough and its displays are effective enough to utilize his deductive
abilities to a significant degree.
3-3
3.2 GRAPHICAL-COMPUTER SYSTEM FOR ORBITAL TRANSFER STUDIES
Two specific applications of interactive graphics techniques have been studied in depth
during the current study. The first deals with performance calculations for orbital
transfers and is discussed in Para's. 3.2 and 3.3; the second is concerned with ascent
boost trajectory optimization and is covered in Para's. 3.4 and 3.5. The orbital transfer graphical-computer system itself will be described in Para. 3.2 and a requirements
specification, written from the viewpoint not of the programmer but of the user-analyst,
will be presented in Para. 3. 3.
3.2. 1 Objectives
The orbital transfer system could be applied to both interplanetary and planetocentric
intersatellite transfers. A multitude
of direct. and flyby interplanetary
transfers have
.
'
already been computed, tabulated, and graphically presented in various Parts of the
Planetary Flight Handbook prepared by LMSC for trips to and from Mars and Venus
and to Jupiter and the asteroids Ceres and Vesta (Refs. 3-2 and 3-3). Part 6, prepared
by the Douglas Aircraft Company (currently being printed), includes Venus swingby
missions to and from Mars (Ref. 3-4). All these interplanetary compilations can be
kept to a manageable size because the orbital elements of the terminal bodies are definitely known and finite times of interest can be postulated a priori. . Extending these
handbooks to other bodies of the solar system greatly enlarges the scope of the work;
the bounds on the quantity of material could grow "astronomically." Furthermore, preparing similar reports on general interorbital transfers would result in an unmanageable volume of material, most of which would likely turn out to be useless or of value
to only a limited number of analysts.
What does appear to be most useful would be a means of readily extending the interplanetary data or the interorbital transfer data to bodies and time periods as needed by
mission and system analysts. In order to. develop those means, the following objectives
were drawn up for the orbital transfer graphics system:
•
Develop a system to generate mission contour plots 3:nd data as needed, that
is, create a "working handbook. "
3-4
o Store and recall to view all the contours generated during a computer run.
o Permanently record the contours for publication and future use.
Specifically, the first objective must encompass the following:
o Accept orbital elements of any two bodies moving about a specific central
body.
o Enable the user to detect and correct errors in the input data.
o Compute mission velocity requirements and trajectory geometric data for
rendezvous or intercepting transfers between the two, bodies, and calculate
these data in useful coordinate systems.
o Compute and plot contours of specified mission/geometric data.
•
Adjust the ranges of the independent variables and the contour levels to show
the most interesting regions.
3.2.2 Computer Program Description
At this point it is appropriate to .describe" a computer program already in existence that
can perform several of the above objectives in a batch processing, non-interactive mode.
This program (MACON) was derived from the Medium Accuracy Orbital Transfer _
(MAOT) program already ~n wide usein NASA (Ref. 3-2) and includes contour plotting
features
execu~ble
on the SC-4020 cathode-ray tube plotter. Although versions of
MAOT have been developed which treat planetocentric transfers, and a similar graphics system could, in a straightforward way, be developed for them, the current discussion will, without loss in generality, be restricted to interplanetary travel.
The inter-
planetary program could be useful in studying missions among the planets, the asteroids,
comets, and hypothetical solar system spacecraft and space stations.
Inputs to MACON consist of planetary constants (such as orbital elements, obliquity,
and location of the vernal equinox), designations of departure and arrival planets, departure and arrival date ranges and incremental step sizes, contour specifications
(such as line legends, names of variables to be contoured, number of contour levels,
increments between them, minimum levels, or a list of their values) and plotting
3-5
specifications (such as x- and y-axis labels, minimum trip time, and grid-squaring
option indicator). For each pair of departure and arrival dates within the specified
ranges, the heliocentric conic trajectory is computed which connects the two terminal
planets on these dates. The method used to find this trajectory employs a numerical
iteration of Lambert's theorem.
culate~
The characteristics of this trajectory which are cal-
include the semi-major axis and eccentricity, inclinations relative to the ter-
minal planets' orbital planes, the true anomalies of the terminal points, and their
heliocentric speeds and directions; of primary importance to the mission analyst, the
magnitudes and planetocentric directions of the hyperbolic excess velocity of the departure and arrival hyperbolas at the two terminal planets are also figured.
The subset of all these computed values which the analyst wishes to contour are stored
as functions of the two terminal dates. At the option of the user up to four variables
may be selected for contouring and up to fifty steps may be taken along each data axis;
these maxima are set by the amoun~ of core storage available in the computer.
Thus,
the analyst can control the spacing between contours and the fineness of the date grid
as well as the number of variables to be plotted and the number to appear on each plot.
The contour subroutines examine one stored variable at a time and linearly interpolate
in each grid squar~ of the departure date-arriv3;l date plane for the contour line segments. Each contouring level is marked with a unique alphanumeric code, and a table
of the contour level associated with each code is printed alongside the contour plot. H
more than one variable is plotted on the same date plane, a different type of line is
assigned to each one and a different table of contour level codes is printed. Finally,
all the contour plots and decoding tables are transmitted to the S(}-4020 plotter and permanently recorded on·microfilm or full-size photographic prints.
3. 2~ 3 Interactive Aspects
To improve the turnaround time of a MACON-aided study, the objectives listed in Para.
3.2. 1 for a graphics system were drawn up.
To accomplish these objectives, graphics
interactive features must be added to MACON. The following cycle of operations has
been devised to de scribe a typical graphics .study:
3-6
1.
Input data are prepared. The analyst must estimate the likely range of
interesting dates and limiting values of the contours. He may wish to specify
limiting constraints on velocities, radii, declination angles, etc., the number of such constraints being dependent on the capacity and coding features
of the particular graphics-computer system being used. He should be prepared to check the computer run for input or execution errors and to jot
down frame numbers for recall during the run.
2.
The computer program is read in and initialized. The graphics scope background is initialized and displayed; this background material consists of data
descriptors, control words, graph axes, and a list 'of the variables which .
are available for contouring (see Fig. 3-1).
3.
The data for the first case are read 'in. Certain quantities are displayed on
the scope for monitoring by the analyst.
He may correct errors by means
of his light pen.
4.
The mission data and contour lines are computed.
5.
The contour lines and all the data specifying the number and spacing of them
are, displayed.
6.
There is a pause in the computer's operation while the analyst monitors the
display; he may then choose to:
a.
Correct an error and recycle the operation to 4.
b.
Change the axis limits or contour specifications and recycle to 4.
c.
Store the present display for future recall and recycle to 6. The store
operation saves the present display in a large capacity memory device
and causes a permanent record to be made of it on a peripheral CRT
plotter.
7.
The next case is read in and the cycle returns to 3.
8.
The analyst may recall one or more of the previously stored displays in sequence for a rapid overview, evaluation, and comparison of the entire run.
This step is like flipping the pages of a handbook to obtain a rapid comparison of mission opportunities; detailed study of the plots should be postponed
3-7
until the permanent copies are available. However, this step may indicate
that more detailed or extended calculations in certain areas are needed during the current run.
What the analyst sees to guide and control the run is the scope display shown in Fig. 3-1.
This figure is a typical illustration of the layout and contents of the scope. The heavy
border encloses a· ''working area" that is 11- x 17-in. in size. Around the working area
is the "control surface area." The printed alphanumeric information composes the unchanging background display; fields for the variable data which will be displayed during
operation of the system are indicated by rows· of
space for a + or a - sign.
XIS
preceded, as appropriate, by a
The date plane for the contour plot occupies the left side of
the working area; the contour, axis, grid and conBtraint specifications
ar~
on the right.
Phase indicators, control commands, and a numerical keyboard appear in the control
surface area below the working area.
face that are sensitive to light pen
and circles on this figure.
The "light buttons" (that is, regions of the scope
~ignals)
are marked by individual dashed rectangles
A typical sequence of displays occurring during a cycle of operations might be as follows, the numbere.d steps being those previously defined:
1.
No· display.
2.
The background material is displayed.
3.
The DATA indicator, which appears in the control surface area, is underlined.
The axis labels arid the limiting departure and arrival dates are
printed along the axes of the contour plot. The title is printed at the top of
the working area. Under Contour Specifications will appear the line legends
(LEGD) and scaling factors (SCALE) for the variables (PARAM) to be contoured. This information will show
selected.
w~ether
the proper variables have been
Parameters recommended for use with MACON are indicated in
this specification table; the symbols in this table refer to the following
variables:
VHl
Departure hyperbolic excess speed
.VH2
Arrival hyperbolic excess speed
3-S
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX TITLE
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXX.X
CONTOUR
/
I
/
I'
:.<
/
/
:><
/
~
/
/
xxxxx.x
~
~
I
/
~
ft1
I
-:t:
H
I II
en
(--:1
r------, r:------,
r---- --,
r------,
r------,
I
r-----'
L
_____ -1 IP::::V=-=J
[x=-~~~]
r-----'
L-±3_x_x_~!J
L!=_x_xy_x_:J
r------.,
L
_____ .l tP_~l_"t.2..J
r-----'
LX_~~~.:J
r------,
r------,
rn-E-c-L"fl
L _____ .J L::
_ _ _ _ :..J
~~~~.:J
r:----,
,-;------,
~...:c3_xY_:.J
1.!2'_X_X_x_o J
r.----.,
r------.,
r------,
/
/
I
X
~
x
~
(_ / /
x
~
~
/
r------,
c:::3_·...1
0_x_x_~:J
CHANGE
FIRST
CHANGE
LAST
rX-X'
C _:.I
C..:c..:c_x_x_ •...J
r------,
rX-x-'
1
__ :.J
r:-----,
r xx-'
L±_x_x_~~~
L±_~~~~:J
r£X1
~-!.j
~_:J
fX-Xl
1.:_:.1
S PEC IF ICAT IONS
ARRIVE
AX IS INC REMEN T
NO.
xx
xx
OF INCRM'rS
/
/
/
/
/
/
/
/i
/
/
/
/
/
/
/
/
CONS TRA INTS
/
/
/
GRID SPACING
/
/
x
x
tI
/
/
V
/
/
xxxxx.X
xxxxxxxxxxxx( X-AXIS
VHl
LE.
~~~3::::X~X3=-~~.J
VH2
LE.
r:=-.::-~~~~~~~J
DVl
• LE.
DV2
/
.xxxxx.x
xxxxx.x
L±_!.!~Y:_·...I
L±Y:_~~~_~
/
I
~
0_~~21_!J
L~2<
rT------,
/
/
~
r------3
/
/
~
NC
Z
/
I
MAX
r-----'
-E"cL21
L
_____ ...J rL::D____
/
I
/
H
/
/
/
/
/
/
/
~
MI N
r-----;'l
/
~
~
/
/
')
~
//
/
I
I
SCALE
L______ ...J
,
/
/
/
5
~
PARAM
r:==--=-_
/
/
/
~
LEGD
./
./
/
I
~
/
~
/
I
~
SPEC IF ICAT IONS?
/2"
/
I
I
/
./
I
/
I
/
I
/
/
~
~
REMAIN XXX
FRAME XXX
xxxxx.x
xxxxx.x
~~~~~~~~~J
r,- -------.,
LE.
~ ~ ~ x_xy 3 ~ ~.:J
DVl+2
• LE.
~~X~~~~~~J
DEC Ll
• GT.
~~~~~~ . LT.
DEC L2
· GT.
L±_x_x_,y_x..J
r-----l
• LT.
r;-----:-1
C~~:....x_x...J
[±=-x.=x=·=x.=xJ
LABEl,) xxxxxxxxxx
DATA
COMPUTE
CONTOUR
r,;:----,
I~~](_'!J
?
~R-E-D-Q1
L ___
,,1+)
, ....
rS;YO-REI
/"'
I - ,
~
•_______1
iST-Op1
L. _ _ _ _ _ •
1- - - - - - - ,
LR_~C2.~~!:.J xxx
\
" -'
I
..........
\ ...
,
~I
Fig. 3-1
",-"
r 0 )
,_/
'-1'\)
1
'-
(2\
,_ ..... 1
(;,
''4'\
a I
,...-
\
I
'-7"
\
... _,
'-
'-8....\
... ..... I
\
''''3''\,
Orbital Transfer
Display Scope
1
,~
/-,
~ 6 J
--'
;'9.... '
',_I
3-9
DVI
Impulsive velocity increment required to transfer from a
specified circular orbit about the departure planet to a coplanar tangential hyperbola having an excess speed VHl
DV2
Impulsive velocity increment required to decelerate from a
hyperbola approaching the arrival planet and having excess
speed VH2 to a specified speed (e. g., orbital or entry) at a
specified altitude in a tangential maneuver
DVI + 2
Sum of DVI and DV2
DECLI
Declination of the departure hyperbolic asymptote
DE CL2
Declination of the arrival hyperbolic asymptote
In Fig. 3-1, the line legends appearing alongside the parameters VHl and
VH2 indicate that these two parameters have been selected for contouring.
Under the Axis and Grid Specifications, the axis increments, number of
increments, and grid-line spacing for the contour graph will be displayed.
Finally, the constraints selected by the analyst in step (1) may be compared
with those read into the computer.
4.
The underline of DATA will disappear .. Next, the COMPUTE and then the
CONTOUR indicators will be underlined, one at a time. These indicators
inform the analyst of the progress of the nearly one-minute-long calculations
occurring during this step.
5.
The underline of CONTOUR will disappear. The minimum (:MIN) and the
maximum (MAX) values of the scaled contour levels and the number of contour levels (NC) will be displayed for each variable being contoured. The
contours will be plotted and the scaled value of each printed at a regular frequency on each contour line.
6.
The question mark to the right of the NEXT indicator will appear to signal
the analyst that the system awaits his light-pen command.
The discussion
of his actions and subsequent display reactions will be postponed for a few
paragraphs. In general, each command will elicit a positive response and
a display of the changed quantities (if any) from the graphics system. Each
new picture will be assigned a unique sequential frame number for identification.
3-11
7.
The cycle of displays is repeated.
8.
One or a series of displays will be recalled from storage and redisplayed.
Each will include axis labels and values, the contour plot, contour level
labels, title, frame number, and all the data and legends in the Contour
Specification.
Program operation and console displays having been described, the final portion of this
subsection will be devoted to an explanation of the system controls implemented through
the light buttons assigned to the display scope face.
This explanation will dwell on the
meanings and reasons for the controls rather than their software implementation; the
latter aspect will be part of the Requirements Specification (Para. 3.3.3). The controls are triggered by the user touching any light button with his light pen during the
run. At the next pause, the system will indicate a positive response to this triggering
and will wait for the analyst to enter in any cnanges by means of the keyboard in the
control surface area. He may choose to make several changes, each new one signalling
the completion of its predecessor. Repeating a change indicates that values just typed
in were not acceptable •. Changes are completed and the operating cycle is resumed
when the analyst
trigg~rs
the continuation button NEXT or the repeat ·button REDO.
In the Contour Specifications section of the display, touching a line legend button under
LEGD eliminates the associated parameter from the list of those being contoured. This
command allows the user to prepare plots of different contours on succeeding frames
or to reassign line legends.
Touching a light button under PARAM adds that parameter to the list of those being contoured and assigns the next available line legend to it. An order should be assigned to
the various line legends; the following is suggested to insure easily distinguishable
contours:
Order
Line
1st
(heavy solid line)
2nd
3rd
4th
(light solid line)
3-12
Of course, computer storage, scope resolution, and/or data transmission rates may
limit the number of variables that can be simultaneously stored and contoured. This
PARAM command is obviously needed to change to a new set of variables during a run
and may be employed to change the line legend assigned to a particular variable in case
of a data input error. Failure of the user to· supply a scaling factor and contour range
and/or incrementing specifications for the newly chosen variable will cause the question
mark after CONTOUR SPECIFICATIONS to appear when the next control command button is selected and will prevent the resumption of operation.
Touching a light button under SCALE allows the user to change an existing scale fact<;>r
or to add a new one.
Touching a light button under MIN or MAX permits the extreme contour levels to be
modified from the numerical keyboard. Normally, these extremes are input in the
data or computed by one of the contour subroutines from the values of the variable being
stored. If the user dislikes the input data or has a reason for distrusting the computed
values, he may change them with these commands.
Touching a light button under NC leads to a change in the number of contour levels.
The user may wish to do this for a variety of reasons. Perhaps an input error needs
to be corrected or a coarse interval between contour levels needs refinIng. One 'of the
contouring subroutines may have computed too many contours or too fine a step size
between levels and, consequently, one end of the range of contours was drastically
truncated because storage capabilities were exceeded. In any event, after. these light
button signals, a change in the number of contours can be typed in.
Under the Axis and Grid Specifications the first two rows of light buttons allow the user
to shift the portion of the date plane being calculated and contoured left or right, up or
down, or to look at a larger part of it or to zoom in on a smaller area of it. Touching
the CHANGE FIRST light button under LEAVE (or ARRIVE) allows the user to increase
or decrease
t~e
first, that is, the smallest, limiting value of the departure (arrival)
date axis by a numerical keyboard input. Si~ilarly, the CHANGE LAST button permits
the last, or highest, limiting
v~lue
to be altered. Grid limits and plotter scaling
3-13
factors will be adjusted so that the new portion of the date plane fills the 9-in. -square
plotting area on the scope.
Touching the light button to the right of AXIS INCREMENT allows the user to change
the intervals between successive values of the departure and/or arrival dates and
thereby to refine the results produced by the linearized contour line computations.
Furthermore, these intervals should be changed when the date ranges along each axis
are varied. Incidentally, the contours plotted in Fig. 3-1 were obtained with an axis
increment 1/18 of the date ranges. Associated with this light button is the display of
the number of increments (NO. OF INCRMTS)' along each axis.
These two displays are
not light buttons, but they are quite important to the analyst because they show whether
the maximum storage capacity for date and contoured variable values is being exceeded.
Excesses can be corrected by use of the six previously described light buttons.
Touching the light buttons alongside ,GRID SPACING enables the analyst to adjust the
spacing between background grid lines; the data appearing in these fields are the increments between these lines measured along the two data axes.
To eliminate all grid
lines, spacings greater than the range of dates along the axes should
~e
input.
Touching anyone of the CONSTRAINTS light buttons allows the user to type in new
values. This command is needed to correct input errors or to modify initial estimates
of regions of interest.
The control commands appearing in the control surface area have the following effects:
•
NEXT causes the system to proceed to the following normal operation such as
commencing calculations after the input data have been verified, executing a
pre-specified series of recalls of previously stored displays, or reading in
the data cards for the next case.
NEXT is the analyst's signal to the computer
graphics system that all changes have been made, all data are acceptable, and
a normal continuation is in order.
•
REDO is a command to repeat the present case with the changed specifications
and constraints. Depending upon what has been changed, this command will
be interpreted by the system in a variety of ways.
3-14
o STORE saves the present display in a large capacity memory device for subsequent recall and causes a permanent record to be made of it on a peripheral
CRT plotter.
o RECALL enables the analyst to use the numerical keyboard to specify the
frame number of one or a series of previously computed displays for redisplay. Each touch of this light button is a signal that another frame number
is about to be entered. NEXT triggers the actual recall or series of recalls.
The frame number being specified appears to right of RECALL.
C
STOP ends the run.
3.3 REQUffiEMENTS SPECIFICATION FOR ORBITAL TRANSFER STUDIES
3. 3. 1 General Guidelines
General guidelines are outlined in the following paragraphs .
3.3.1. 1 Objectives. The objectives listed in Para 3.2. 1 are the objectives of the
Orbital Transfer Graphical-Computer System.
3.3.1.2 Nature of This Specification. This specification sets forth, requirements on
the performance and capabilities of the system from the viewpoint of the user-analyst.
As far as possible it is independent of any particular computer or display equipment.
3.3.1.3 System Operation. The system shall be' so designed that its operating
displays and instructions shall be expressed in the language of the user in that it shall
use the vocabulary of flight mechanics and engineering plotting procedures.
3.3. 2 Performance Requirements
Performance requirements are outlined in the following paragraphs.
3.3.2.1 Sequence of Operation. This system shall operate in the following sequence
of steps as described in Para 3. 2. 3:
1.
Data preparation by the analyst
2.
Program read-in and initialization (E)*
*Agents for the various steps in the operation of the system are indicated as follows:
(E) main computer executive program, (M) MAOT routines, (C) contouring subroutines., (G) graphic display routines, and (L) light-button-directed control program.
3-15
3.
Data display and monitoring (M, G, L)
4.
Computation of the mission variables and contour data (M, C)
5.
Display of the contour data (G)
6.
Review of the contour data by the analyst
6. 1 Error correction and recycle (L)
6. 2 Specification change and recycle (L)
6. 3 Display storage (L, G)
7.
Continuation to next case
8.
Recall of stored displays (L)
3.3.2.2 Inputs. Inputs are as follows:
1.
Program: The system shall read in from magnetic tape or cards an
object deck of the MACON program suitably modified for graphics
(E).
2.
Data: The system shall read in from magnetic tape or cards the
following input data (M):
2. 1 Number of cases
2. 2 First, . last, and increment in departure dates
2. 3 First, last, and increment in arrival dates
2.4 Departure and arrival planet numbers
2. 5 Mission speeds, which include the following:
• Orbital speed at departure
• Escape speed at departure
• Escape speed at arrival
• Post-propulsion speed at arrival
These are needed to compute the impulsive velocity increments
at departure and arrival which were defined in Para. 3.2.3.
3-16
2.6
Case title, 72 characters.
2.7
Departure and arrival axis labels, 36 characters each
2. 8
Departure and arrival grid line intervals
2.9
Parameter number(s) of no more than four* mission variables to be
contoured, the parameters being represented by the following symbols:
Parameter
No.
Symbol
Mission Variable
1
VH1
Departure hyperbolic excess speed
2
VH2
Arrival hyPerbolic excess speed
3
DV1
Departure velocity increment
4
DV2
Arriva:l velocity increment
5
DV1+2
Sum of DV1 and DV2
6
DECL1
Declination of departure hyperbolic asymptote
7
DECL2
Declination of
arriv~l
hyperbolic asymptote
2.9.1 The order in which a parameter is designated in the input shall
·determine the line
leg~nd
used to draw its contours.
2.10 Contour specifications for each parameter; the following options being
available for each parameter:
Option No.
Specifications
1
Number of contours
Minimum contour level
Increment between contour levels
2
Number of contours
Increment between contour levels**
*The upper limit on the number of variables that may be contoured shall be set by the
storage and data transmission speeds of the computer-graphics hardware.
**Thecontouring subroutines shall determine the minimum level from the computed
values of the particular parameter.
3-17
Option No.
3
Specifications
Number of contours
Resolution of the increments between contour levels *
4
Number of parameter already input whose table of
contour levels shall also be used for the current
variable.
3. 3. 2. 3 Outputs. Outputs are as follows:
1.
Printed output: The system shall compute and tabulate the following mission
data in the same format utilized in Ref. 3-2 (a sample listing is shown in
Table 3-1) (M):
Explanation of Symbols. The symbols used in Table 3-1, along with their
column numbers, are explained in the following:
Column
Symbol
1
DEPART
Julian date of departure, reckoned from JD 240-0000.
2
ARRIVE
Julian date of arrival, reckoned from JD 240-0000 .
3
. SPEED
(VHI) **
Explanation
Magnitude (speed of hyperbolic excess departure
velocity vector, normalized to Earth's mean
orbital speed.
4
RA
Right ascension of hyperbolic excess departure velocity vector (hyperbola asymptote); measured in'
degrees along the local planetary equator eastward
from the "vernal equinox, " i. e. , where the Sun's
path moves north across the departure planet's equator. For Venus, since the orientation of the planetary North Pole is unknown, the equater has been
*The contouring subroutines will first find the minimum and maximum computed values of the current parameter and will then assign an increment between contour
levels which will be the minimum multiple of the resolution that can span the range
between these extremes.
**Parameter names used on the graphics display scope for four of these variables are
enclosed in parentheses.
3-18
Table 3-1
SAMPLE OUTPUT OF ORBITAL TRANSFER PROGRAM
OEPAKTURE PLAi~ET= Ei,RTH
ARkIIiAL' PLAI.!::T= Il:ARUS
OEPAKT ARkIVE
3-99(J0
3-9%1)
3-99UO
3-99uO
3-9900
44-
5
10
SPlEu
~
O~PflRTIJRF
l\PRTVAL
~
.20.)
DiCL
I 1
V 1
5b.t1
.0:.73
.024
7J .t.i
7.4Y
~. 77
4.31
.o€>:'
H7.6
8ei.U
3.(11
• "til)
f..h.b
.Ll30
b9.3
• (J4(l
fl,9.Y
.l)!"4 1. 07R
.124 1.1?7
4-
15
44-
20
25
74.9
.14~
7'J.?
.094 7u.2
.ll:;! 217.1
.lJ3~ 249.0
4~.5
1.h<+
1.p?~)
3-990u 43-9900 I~3-9900 4'3-9900 43-9900 4-
.30
35
40
45
50
.039 249.0
-.~
.057 247.9 -2b.3
."76 24u.4 -3-.l.M
.09~ 24~.4 -40.4
.11j 2u~.1 -51.U
.77
-.2.:)
-1.10
-?07
1.n4b
1.1:67
1.11£13
3-99UO
3-9900
3-99UO
.3-99UO
3-99lJU
~5
.129 23':1.5 -50.u
.144 23u.3 -59.~
.15d ~3,-.7 -6~.~
.17.2 220.5 -6~.1
.18~ 223.7 -67..;)
-~.77
1.117
60
65
70
75
-4.~7
1.1P5
44444-
.3-990~
3-9900 4-
43-99US 4-
80
5
lU
3-99u5 4-
15
.19b 21".3
.201
76.1
.14u 70.0
.U91 7u.4
6j.ii
-oY.2
5u.2
0';'.1
1.PClh
-2.Q.) 1.107
-5.30 1.'3?
-h.lj
-f.HY
1.1~h
1.14~
-7.64 1.147
7.3j
.~6b
5.hl.
.(·1'1
91103
9U .r~
9U.H
9u.Y.
9(J.9
90.Y
90.'1
9U.R
9U.7
90.~
~5.U
.7..) 1. (:4t
-.2.:: 1.(1f-7
3-9905
3-9905
3-99U'.>
3-99U5
3-9905
44444-
~U
.11u 24U.1 -5u.l
55
60
65
70
.12'J 237.3 -54.u
.J.4~ 23<+.0 -51.-.::.
.1'.>.) 2.3u.4 -nlJ.u
.Hlo 220.2 -f,~.3
3-99U5
3-9905
3-9910
3-9910
3-9910
44444-
75
dO
.BY 21th 7
-6~.Y
-7.0~
7u.5
5~.7
6.::.~
5.4~
,·.Jt~
A70l
4.L'~
.059
fl7.Y
.3-9910 43-9(HO 4.3-9910 4-
. ;:'. nu
.094
1.7U .1. ~l2j
.7U 1.n47
-.21 1.nA7
hn.6
3-9910 4-
20
25
.)0
35
.~s.)
179.0
~u.7
• lJ ,33 24 ol. 2
.u39 244.7
47.5
-1.7
.U50 245.P -25.3
!lAS.7
3'JS.J UH7.?
355.7 4Rn.l\
3~n.2 490.4
3~6.B 492.1
.9A7
.gR7
.gR7
.gR7
.YP07
.35~.1
3S7.5 4Y3.F.
l(,n.'~ 267.'1
10:,).6 ?6 0 .<)
1:::'7.5 ?n7.n
12r-.3 ?uO.q
.Q711
.RO?
In?~
103.?
.7nA tnu.o
.71\C,
.7""~
104.<;
.7."Q 105.0
:>7.Q?
.7?~ 10"".3
20.40 1.?O~ 74.7
20.Hl 1.117 AO.l
21.?S 1 .()~?
AU.n
21.71 1.001
AR.4
-:-'01.1
:>4.n
?n.S
?R.U
'0.3
:'oon.<;
?-?~
?gq.p'
?qp..g
?OA.O
?ol',.q
?9S.R
34.?
:">nl.F.
~n.1
-:-'R.n
30.R
41.7
.A117
.770
.7?:>
.n7?• Fo:'o 1
.c,g<;
.<;F.U
.<;37
.""'4
.404
3n1.q
301. 0
.'01.7
301.-:-'
-"00.7
:>-:-..n
?4.A
:>n.h
:>R.4
..,o.?
?4.41
?4.R4
?5.2H
?'i.7?
.A?~ In1.4
.Rnu to?€>
.7R7 lO.~.""
?(,.1.,
.75<) l04.R
30n.n
?Qg.:>
:>oo,.?
:>97.:>
?gF..n
.9 Dl.?
26.SQ
1.7 13?0
27.03
.741 lOS.?
.7?7 lOS.S
1.lg0 74.?
1.11<; 7g.R
?9Q.O
:'oon.s
p. 4 • 4
~nl .~
4n.o .4.QQ
4?S
.u71
In.1 1.49?
17.g 1.?q9
19.h 1.1UR
AR.3
gl.€>
04.4
9F..7
301.R
3n:>.0
:'001.0
-"01.t:;
:>1.4 l.n;:>7
.0?7
?4.9
.R45
:>n.n
.77n
9u.2
90.2
qU • .:>
9Uo!
H9.Y
.?1~
1.?~t',
~~R.F\
4R~.('I
.?~5
1.291
~~A.Q
4H4.<)
4HF..1
A9.R
flY.(,
.2g;> 1.~u6
.jOl 1.415
.~7P,
.77q
.17 0
.A44
.U97
• qop'
166.u ?67.-:-'
IhO.3 :>h4.A
.(J33. hY
.lJ4(1 1.('127
• U Ilf. 1. n r I
ol?6 1.1~?
131.q
27.4
0.4
50!
R9.4
30:>.n
:'>n 1. 0
Ol.h
04.:'1
.P.C}lj
ah.n
.RhA C'!A.<;
.~44 100.1
-2. 7~ 1.10b
-~.51 1.1td
AYd
.q:>:'o
.AO~
1 0 • 4 1. 1 ,~ '1
:>1.:> 1.n?<)
9U.?
f'.Y.lI
?-n 1 .'i
301.Q
17.h
1.U~3
.~:'3
l.~??
~!)9.2
.260.
1.3~n
.?R1
1.~-''"i
359.7 4t\7.7
.3 129. ~
0
04.?
q6.4
.RnF. 9R.:'o
.R4~
oq.q
.k?l 101.?
:">on.~
It:;.5 1.47-:-'
17.3 1.?R7
1 0.1 1.141
:>P.O 1.n:>:'1
??R
.0?7
3 n1 • 1
3nl.h
1.('97
H~.H
RO.4
A4.R
RA.'"
01.€>
?9A.n
;:>00.7
?lon.l
-1.11
-f> • .3~ 1.143
1.147
7.?j
.1'59
7~.;:>
4~.S
-1.9~
-4.?:::I 1.1?h
nFCI C;PFEO
pc; I ?
?9U.<)
?QR.S
BY.1
tiy.h
fl9.Y
9Ll .t
-4.97 1.133
-5.f.J 1.13B
?C,.?S
27.3(\
?4.7;:>
.qqo
.qcn
.9Qh
1.0(,1
?~.9h
.030 241.1 -1.2
.u5u 240.2 -2!J.Y
."7'J 244.6 -~7."
.09.) 24~.5 -4 J .U
7~.j
UR4.~
35R.94tH.7
30
.35
40
45
.U9u
• .30r· 1.409
.272
.7f;L':;
:'05<).0
2~.h4
2U .IH
?:>.1';
?2.b?
1.70 1.0.24
1~
1. 39U
UR3.n
2:>.9 13p..n
S.!) 122. 0
I.? 1?O.f,
:'05 0 .(, 4RO.7
4 0 .u
77.0
7 cJ. 7
1.~f.q
3~,5.<
?~.10
.()4('1 l.('l~h
.UI\:' 1.0H1
.12<', 1.1311
• 1St) l. 1 71.
.1pC} 1.21"
191.~
.i9Y
.1.3d
1.21"h
1.317
1.:'11111
;:>;:>.54
.9R7
.9A7
.9r.7
.gPo7
.Auq
24u.2
~
.233
.;>51
.2hh
.27q
.290
1.~ 1;:>4.?
357. n '.A? 0
:'o~<).g 4M2.1
l.nll('l
.011
.u3~
10
1.17?
.lEn l.?D
• 211 1. ?~)1
?f,0.~
.Oql.
.U3C'
.u5~
.17d 221.7 -64.2
ol':>f:.
154.7
V?
19.7:~ 1.;:>nR
20.2M 1.1?n
2n.H'i t.n~u
;:>1.421.nnl
?1.9Y
.g<;o
.174
2~
.n9b
124.H ?42.4
1R.3 l:'lR.C;
.gt.:lA
.QQ7
1 ?
hu.i.J
£-''1.7
88.6
20
2.~~
• c?1 "
.q71
1.0iJ"
.79:>
.170
• OCH
4-
.3-900~
163.1' ?70.!l
IhO.9 ?71. 0
1.0nl
.R')~
.?6~
~o.?
.3-9905 4.3-99lJ~ 43-99U5 43-99U5 4.3-9Q05 I~-
.:16L.
PFRII-l j\PHFI.
B4.4
7~.U
Rl.~
4.1u
TH~T:>
TIIFT!
PSI 1
PLAtJFT=
F.I'.PTH
PLAt-JFT= TCAQIIC;
It)o.O
?~.r.7
?~.~?
.9RR
.gp,R
.'lRA
.9RR
?6~.r::;
?39.4
?1.0~
.q<;q
.l.J?U
.77n 104.:>
21.31
2 1 • f, ~ 1. II 5 n
1.001
21.'n t.nnn
137.t:;
??3:">
.q5 0
121.El
119.U
2:>.€>Q
.Q?4
.Rq<;
2~.O!J
?q4.R
?o..,.u
.476
l'->.R 1.4R?
1.?'9~
.0?7
.A4f.77R
.7?0
.n71
~-:-'.Cl
.n:>A
.<;q?
'<).7
."'hn
''17.4
.,g.?
.<;33
.'->In
"':>.1
??-.1
Column
Explanation
Symbol
assumed coplanar with the orbit, and the "vernal
equinox" of Venus is taken in the direction of its
orbital perihelion.
5
DECL
Declination of hyperbolic excess departure velocity
(DECLI)
vector, measured in degrees positively northward
and negatively southward from the planet's equator.
6
II
Inclination of transfer orbit to departure planet's
orbit, in degrees; zero is forward, -90 0 is perpendicular southward, +90 0 is perpendicular northward,
and ±I80 0 is backward.
7
VI
Heliocentric speed of departure, normalized to
Earth's mean orbital speed.
8
PSII
Heliocentric angle of departure, in degrees; measured
counter-clockwise from the outward heliocentric
radius vector to the departure velocity vector in the
transfer plane.
ECCEN
Eccentricity of heliocentric transfer orbit.
10
SMA
Semi-major axis of the transfer conic, in AU ..
11
THETI
True anomaly of departure point, measured in the
9
transfer orbit in degrees, reduced to 0
12
THET2
0
.:5
91
:5
360
0
True anomaly of arrival point measured in the transfer orbit, in degrees; THET2 minus THETI equals
the heliocentric transfer angle.
13
PERIH
Perihelion distance of the transfer conic, in AU.
This value is listed only if the vehicle traverses the
perihelion during the transfer.
14
APHEL
•
Aphelion distance of the transfer conic, in AU. This
value is listed only if the vehicle traverses the
aphelion during the transfer.
3-20
Column
15
Symbol
12
Explanation
Inclination of transfer orbit to arrival planet rs orbit,
in degrees; if approach is directly from the rear (i. e. ,
the arrival vector points in the same direction as the
planetrs motion), then 12 = 00, etc.
16
V2
Heliocentric arrival speed, normalized to Earth rs
mean orbital speed.
17
PSI2
Heliocentric angle of arrival, in degrees, measured
clockwise from the inward heliocentric radius vector
to the arrival velocity vector in the transfer plane.
Note that this angle is measured opposite in sense to
PSI!.
18
RA
Analogous to, departure velocity case; note., however,
that this is the right ascension of the position at which
the arrival velocity vector would emerge again from
the planetrs celestial sphere.
reckoni~g
This convention for
asymptote directions is useful when calcu-
lating flyby trips .. When the trajectory is unperturbed
by the target planet, the approach and departure asymptotes have exactly the same right ascensions and declinations. By noting the discrepancies between the two
asymptotes in other cases, the analyst may obtain a
rough but rapid estimate of the bend angle without performing trigonometric calculations.
19
DECL
Declination of the position at which the arrival velocity
(DECL2)
vector would emerge again from the planet's celestial
sphere.
20
SPEED
Magnitude (speed) of hyperbolic excess arrival velocity
(VH2)
vector, normalized to Earth's mean orbital speed.
3-21
2.
Graphics output (G): The system shall display the data, contours, alphanumeric information and control indicators shown in Fig. 3-1 of Para. 3.2.3.
3.
Photographic output (G): The system shall prepare permanent copies of the
graphical displays which the analyst requests to be stored during the operation of the system.
These copies may be prepared directly from the graphi-
cal display by a suitable camera or from a peripheral cathode-ray tube
plotter.
3.3.2.4 Timing and Storage. Timing and storage is described below:
1.
Detailed Requirements: The code storage, data storage, and operating time
requirements for the system have been specified in detail in Table 3-2.
These numbers are estimates and are furnished only as a guide to and standard of desired performance. They are not intended to be indicative
particular computing equipment.
o~
any
Moreover, because of variations in such
equipment, the number of words of storage for data and code could easily
vary by ±20 percent from the values estimated in Table 3-2.
1.1,
The code size includes the analytical program, the contouring subroutines, a typical graphics executive routine, a typical graphics display
package and a typical control program directed by l,ight button signals.
Code size does not include the resident monitor(s) for the computer(s)
in the system. The code sizes were based on the MACON program as
executed on the Univac 1108 computer and on the graphics software employe,d by the CDC 3300 system at LMSC.
1.2
The data words were estimated on the basis of one word to each floating point number and each group of six alphanumeric characters.
3-22
Table 3-2
TIMING AND STORAGE REQUIREMENTS FOR ORBITAL TRANSFER SYSTEM
SEQUENCE
NUMBER
OPERATIO~
STEP
DATA PREPARATION
PROGRAM READ-IN
DATA R \TE
(WORDS/SEC)
2. ~oO
-lOa
100
10,-100
2.1U0
3. I DATA ARE READ IN
500
150
3.2 VARIABLE DISPLAY DATA ARE SELECTED
500
150
3.3 DATA ARE DISPLAYED
10.50U
2.07U
?
GRAPHICS BACKGROUND IS INITIALIZED
D~SPLAYED
3.4 ANALYST MONITORS & CORRECTS ERRORS
12,500
4. I MISSION DATA ARE COMPUTED & STORED
12, OOU
3, Ooo(B) -1. ;;00(C)
4.2 CONTOUR SPECIFICATIONS ARE INTERPRETED AND TABLES
OF CONTOUR LEVELS ARE COMPUTED FOR EACH
PARAMETER
12,100
2. UOO
3,500
4.3 CONTOUR COORDINATES ARE COMPUTED(D)
12,100
1.900
-1,200
:l1-D
1.70U
-1. 000
4.4 CONTOUR C98RDINATES ARE TRANSFERRED TO GRAPHICS
EQUIPMENT( )
,
GHAPHICS DISPLAYS
NOl'E
-1,200
2.3 GRAPHICS BACKGROUND IS
COMPUTATIONS
TI:lIING
(SEC)
D.HA WORDS
ANALYST ESTIMATES & PREPARES INPUTS
2. I PROGRAM IS READ IN AND STORED
?
DATA DISPL.-\Y
CODE SIZE
2,150
M-O
IUU
:11-0
CONTOUR COORDINATES ARE SCALED AND DISPL.-\YED ON
.
GRAPHICS SCOPE (D)
10,000
CONTOUR REVIEW
ANALYST MAKES LIGHT BUTTON CHANGES; ANALYST
TOUCHES "NEXT" OR "REDO" LIGHT BUTTON; CONTROL
PROGRAM ANALYZES CHANGES AND RECYCLES TO
PROPER STEP
12. 000
3,700
1\1-0
IO/ERROH
9(B) 19(C)
IUU
"DATA" INOICATOH IS UNDEHLINED
FHA:lIE TITLE AND NUl\IBEH; AXES L.-\BELS AND
LThIlTS, CONSTHAINTS; LINE LEGENDS AI'D SCALING
FACTOI\S OF PAHAMETEHS TO BE CONTOUHED;
AXIS INCHE~IE!\TS 8; Nl'MllEH: GHID LINE SPACIl':G:
POSITIVE HESPONS£; COHHECTED DATA
"CO:lIPUTE" INDICATOR IS l'l'DCHLINED
"COI'TOl'R" INDICATOH IS l'I'Dl:HLINED
:!uo
2U
6, OUO
2,ljO
STOHAGE IIEADINGS: IIEADINGS FOR SPECIFICATION
TABLES; CONTHOL INDICATOHS, Nl':lIERICAL
KI::YBOAHD
NO CIIAl'GE
IOU
IU
CONTOUR DISPLAY
NOI'E
I'ONE
:lIINClIDI &; ~IAXDll':l1 LEVELS; I'L'l\IBEH OF CONTOL'HS; CONTOLR LINES AI'D VAU'ES: ALL ..\XIS
YALl'ES; TilE "?" BY "NEXT" APPEABS
IU/ERROH
}UO
POSITI\ E BESPONSE, COHHECTED DATA, TilE " ? "
BY "NEXT" IS Tl:H"ED OFF AFTEl\ "HEDO" OH
""EXT" IS TOL'CIlED
6. I RECOMPUTE
RECYCLE TO 4.1
THE
6.2 CHANGE
CONTOURS
IF NEW PAR.UIETERS HAVE BEEN SELECTED, RECYCLE
TO 4.1
DISPL\YS .-\BE HECYCLJ::O
6.3 STORE
ANALYST TOUCHES "STORE" LIGHT BUTTON; GRAPHICS
DISPLAYS ARE SAVED; KeMBER OF FRA;\IES LEFT FOR
STORAGE ARE DECRElIIENTED
"?"
BY ":\EXT" IS TUB:,\ED OFF
IF ONLY CONTOUR LEVELS CHANGED, RECYCLE TO 4. 2
CONTOUR COORDINATES ARE RESCALED FOH CRT EQLIPMENT AND OUTPl:T ON :lIAGNETlC TAPE
NEXT CASE
RECYCLE TO 3. I
RECALL
GRAPHICS DISPLAYS HETl:RNED FHOl\I STORAGE
HE1\IAIN ON DISPLAY UNTIL REPL.-\CED BY l'EXT RECALLED
DISPLAY
12. 000
3.700(B) 6, 000(C)
2.100
2.0GO
4.3UU
12.000
3,70U
6,000
-l.UlJll
9(B)
19(C)
"STOHE" INDIC-\ TOB IS L'NDEH L\:,\ED. FH.nu:s
HE:lIAI:,\I~;(; 8; TilE ...," BY ";';EXT" .\PPL\l;S
~uu
HECYCLE
2. 000
IL
NOTES: (A) 1\I-D: MACH IKE-DEPENDENT (B) 20 " 20 AXIAL GRID INCRE;\IENTS (C) 30 ' 30 AXIAL GRID INCRDIENTS (D) OCCeR SI:lIVLTANEOUSLY
CONTUL1\ LINES &; YALLJ:S: .-\XIS L.-\BE LS 6:
\'ALLES; FltnlE TITLE 8; NDIBEH: 1I1:.-\DIN(;
I¥T.-\ IN CONTOUR SPECIFICATIONS
I
8;
1.3
The data words of core· storage and the timing (where it could be
estimated) are postulated on two assumed contour specifications. One
specification considers the departure date-arrival date plane to be divided into 400 grid rectangle s, that is, 20 steps along each axis; the
other considers it divided into 900 rectangles, that is, 30 steps along
each axis.
Both specifications further assumed that two parameters
are being contoured with six contour levels for each.
1.4
Rapid responses to STORE and RECALL shall be required. High data
transmission rates shall be needed to satisfy the timing requirements
for these two operations.
2.
Overall Requirements:
2. 1
Storage: Exclusive of the resident monitor(s) for the computer(s) used
in the system, the core storage for operation without overlays or chain
links is estimated to be as follows:
2. 2
20 x 20 Grid
30 x 30 Grid
Code words
19,000
19,000
Data words
9,000
12,000
Timing: The time required to read in the data for each error-free
case, compute the mission data and contour lines,
~nd
display
t~e
con-
tour plot is approximately 24 sec for the 20 x 20 grid and 44 sec for the
30 x 30 grid.
These times probably will vary with the particular hard-
ware and software on which the Orbital Transfer System is executed.
The time required to save a particular display and prepare a permanent
copy by means of a CRT plotter is estimated to be 11 (or 21) sec for
the 20 x 20 (or 30 x 30) grid.
The time required to recall a previously stored display should be only
one sec, and only about ten sec more should be allowed for scanning it.
2. 3
Data Transmission Rate: The time specified for recalling a previously
stored display places the most stringent requirements on data transmission rates between hardware components of the system. The rate
of 4000 words per sec specified in Table 3-2 is about one to two orders
3-24
of magnitude faster than that presently being contemplated for telephone
'cable transmissions. If the store-recall feature of this specification
is to be retained, significant improvements must be made in data
storage at the graphics console or in data transmission rates.
3.3.3 Light Button Controls
3.3.3. 1 Guidelines.
1.
Light button control guidelines are outlined below:
Each light button action shall be initiated by only a single action of the analyst.
It shall automatically set off as long a chain of logical decisions and lower
level actions as possible within the graphics -computer system (L).
2.
Each light button action shall elicit a positive response from the graphics
. scope (L, G).
3.
Sufficient separation distance shall be provided between neighboring light
buttons in an attempt to prevent the analyst from accidentally touching the
wrong button; providing a positive response to each light button command will,
of course, warn the analyst of any such error. Spacing between the infrequently used CONSTRAINTS light buttons shall be at least one-third of an;
between all others, one-half (L, G).
3.3.3.2 Control Actions, Responses and Consequences.
Fig. 3-1 of Para. 3.2.3 shall be implemented.
The light buttons indicated in
Their effects, the display responses
and subsequent system effects shall be as specified in Table 3-3. The specified recomputations of mission variables and contour lines shall occur only after an execution command (such as NEXT or REDO) is given (L).
Two explanatory notes follow.
1.
P ARAM Light Button: Contour line legends shall be assigned in the following
order (C):
3-25
Order
Line
Description
1st
heavy solid line
2nd
light dashed line
3rd
dotted line
4th
2.
" light solid line
System Execution Following a Specification Change (L):
2.1· The normal continuation shall be indicated by the REDO command.
However, after the input data have been reviewed and possibly changed,
the normal continuation shall be either REDO or NEXT.
2. 2
The light button control program shall make a logical analysis of the
series of commands given since the previous execution. It shall then
cause the system to recycle only through those steps needed to change
the contour plot and displays ,according to that series of commands.
3-26
Table 3-3
LIGHT BUTTON ACTIONS FOR ORBITAL TRANSFER SYSTEM
LIGHT
BUTTON
POSITIVE
m:SI'ONSE
,\NALYST'S
INPUT
POSITIVE
HESPONSE
COMMAND
COMPLETION
SYSTEM ACTION
EHROH SIGNAL
NEXT EXPECTED DISPLAY HESPONSES
ANALYST'S HEACTION
TO ElUWIl
CONTOUR SPECIFICATIONS - LIGHT BUTTONS
LEC'END DISi\PPEAHS
NONI': I\E GllID LINES ON THE CONTC)lJH PLOT ARE llESPACED. AXIS
VALUES All\<: IlESPACED. l\1ISSION VALUES & CONTOUH LINES
ARE NOT AFFECTED.
"CONTOUR" llNDEHLINED. NEW GRID LINES AND
AXES VALUE APPEAR. "NEXT?" APPEAHS.
SAI\II': AS "SCALE"
SAl\lE ;\S "SCALE"
TilE ".,,, BY "NEXT"
:; :."; \ 1'1'1·:AIlS.
NONE RI·:qulln:n NONE
NONE REQUIHED SAME AS FIRST "REDO."
AS APPIlOPRIATE
NONE
WAIT FOil ":--IEXT?"
NlJIIH:llICAL
VALUE
VALUE APPEAItS IN
TIH: DISPLAY AREA.
NEXT COMl\IAND TIlE INPUT VALUE IS STOHED AS TIlE CONSTHAINT LIMIT.
MISSION CALCULATIONS WILL BE REPEATED BUT TRIPS
WHOSE Vf<:LOCITIES EXCEED ANY OF THE CONSTRAINTS WILL
NOT BE STORED FOR CONTOURS; THEY WILL APPEAR IN TIlE
PIUNTED OUTPUT.
SA:\1I-': AS "CHANGE FmST" EXCEPT AXIS LThIITS.
VALUES AND INCitEMENTS WILL NOT BE
CHANGED.
SAME AS "SCALE"
SAI\II·: AS "SCALE"
nVI .:l
TilE TOUCHED DISPLAY AREA [S
UNDlmLINED.
DECLI
DECL2
TilE TOUCIIED DISPLAY AIlI':A IS
UNIJ[-:HLlNE[),
NUl\n:IUCAL
VALUE APPEAIlS IN
TilE DISPLAY Aln:,\.
VALUE OF
LOWI':R LIM. IT OH
UPPER LM IT.
NEXT COMl\IAND SAME AS VIII EXCEPT NO TRAJECTORIF.S WILL BE OMITTED
FROM CONTOUR STOHAGE. INSn:AD, TIm U:TTER D WILL
m; PHINTED ON THE CONToun Fon EACH TRIP IN WHICH ONE
OF TIlE DECLINATION CONSTilAINTS IS EXCEEDED.
SAME AS "VIll"
SAME AS "SCALE"
SAME AS "SCALE"
IU:IlO
TilE '''''' BY "NEXT"
IllS A P PE ARS.
NONE m.:qul1lED NONE
NONE RE(lUIHED MISSION & CONTOUR CALCULATIONS ArtE NOW REPEATED
WITH NEW CONSTRAINT VALUES.
AS APPROPHIATE
NONE
WAIT FOIl "NEXT?"
NEXT
TilE" '1"
D1SAI'PEAHS.
NON\': m':QUIIlED KONE
NONE HEQUIHED PROCEED TO THE FOLLOWING NORMAL OPERATION.
AS APPROPRIATE
NONE
WAIT FOR THE "'1"
HEDO
TIlE "'!" BY "NEXT"
DISAPPEAHS.
NONB HEQUIIlED NONE
NONE HEQUIHED I';XECUTE WITH CHANGES.
AS APPHOPIUATE
IF NO CHANGES. THE "'?" BY "NEXT" APPEAHS
IMMEDIATE LY.
MAKE CHANGES OR TOUCH
"NEXT"
STOHE
"STORF." IS
{TNm:RLINED.
NONE REQUIHED NONE
NONE REQUIHED LAHGE-CAPACITY STORAGE DEVICE SAVES THE CONTOUR
LINES & VALUES, AXIS LABELS & VALUES, FRAME TITLE &
NUMBER AND HEADINGS & DATA OF THE "CONTOUH SPECIFICATIONS." THE GHAPHICS DISPLAY IS THANSMITTED TO
A CRT PLOTTER.
"STOHE" UNDEHLTNE DISAPPF.ARS. THE "?" BY
"NEXT" APPEAHS.
NONI':
NONE IlEQUIIlED
RECALL
"HECALL" IS
UNDEHLINED.
FHAME
NUMBER(S)
"Hf<:CALL" TO
ENTER FHAME
NO. OR "NF.XT."
nECALLED CONTOUR PLOTS AND SPECIF[CATIONS
APPEAH. THE "?" BY "NEXT" APPEAHS.
IF THE FHAME NUMBER HAS NOT BEEN STOllED,
IT IS IGNORED.
NONE IlEQllIIlED
STOP
"STOP" IS
lJNDEHLINED.
Nom: HEQUIHED NONE
NONE
"STOP" WAS UNDImLINED.
IlEST AHT 'I'll I'; JOB.
CIIANC;I'; FIIlST
CHANca: LAST
IlEDO
CONSTRAINTS - LIGHT BUTTONS
VHI TllllOlIc;l1
CONTROL COMMANDS
INPUT NUl\1I3EH
APPEAHS TO THE
HIBHT OF "HECALL.'
A TABLE OF FHAME NUMBERS FOH HECALLING IS FORMED.
Tim SPECIFIED DISPLAYS WILL BE RECALLED FHOM
STOllAGE.
NONE HEQunU;D JOB IS TEHMINATED.
3-27
3.4
GRAPHICAL-COMPUTER SYSTEM FOR ASCENT BOOST OPTIMIZATION
3.4.1
3.4.1.1
Objectives
Introduction to the Problem. The ability to rapidly and accurately evaluate
the performance capability of multistage rocket booster systems is a primary requirement for mission planning and preliminary design studies. Achieving maximum
payload capability from a given booster system requires utilization of an optimal
"tilt" or thrust vector orientation program during powered flight. The sensitivity of
performance to changes in hardware parameters 'cannot be properly evaluated unless
the effect of tilt program changes is eliminated by optimizing the tilt program for
each set of parameters.
In addition to TILT program complications,
~he
mission generally requires the
satisfaction of a number of constraints on the trajectory flown, such as terminal
altitude or velocity. Finally, a number of parameters defining such quantities as the
initial launch vectors, the coast periods between powered stages, and the upper stage
thrust cycles may be treated as variables for- certain boost hardware combinations.
.
.
These "adjustable parameters" should then be optimized along with the tilt programs
~nd
the terminal constraint combination to produce maximum perforlnance.
The digital computer program for
Ra~id
Earth-to-Space Trajectory Optimization
(PRESTO), Ref. 3-6, has been developed to solve the performance problem outlined above for orbital, lunar, and interplanetary missions. Optimization of pitch
and yaw tilt programs and a variety of adjustable parameters and the satisfaction of
constraint combinations is accomplished simultaneously with a closed loop steepest
descent optimization routine that maximizes payload.
Unfortunately, difficulties and delays often arise in evaluating booster system's that
are radically different from those with which experience has been gained and data
accumulated. Similarly, synthesis of new missions or trajectory profiles that require unusual pitch and/or yaw tilt programs can present difficulties. The problems
3-29
fall into three general categories: (1) errors in data input, (2) poor initial estimates
for the control programs and launch angle, and (3) slow optimization convergence because of the relative importance assigned to each of the adjustable parameters and
the sensitivity to tolerances on the constraints.
When using the batch processing mode, the analyst is forced to solve these problems
by reviewing the data, making the changes deemed appropriate, and resubmitting the
computer program. When this data
return~,
further examination shows whether
those changes were sufficient or even appropriate. Thus, solution of the ascent boost
optimization process may require repetitive passes at the computer with the associated calendar time delays and increased costs.
3.4.1.2
Objective of the System. The objective of the graphical-computer system
for ascent-boost optimization is to furnish the analyst with a on-line monitoring and
redirection capability.
The graphical display of trajectory and booster system parameters will enable the
analyst to monitor each trajectory profile as the computation
procee~s,
detect
irregularities and, delays in convergence towards the optimum solution, and redefine
those parameters necessary to speed up or improve the convergence procedure such
that the booster system payload capabilities may be quoted with high confidence·
levels. Specifically, the graphical display system will provide the mechanism to do
the following:
e
Review and correct the most error-prone portions of input
~ata:· from
both
system parameters and control histories.
•
Evaluate the desirability of the resultant initial trajectory for the particular mission objectives prior to initiating the optimization cycle of the
\
program.
e
During the optimization cycle, monitor the trajectory profile and key
optimization indicators to evaluate the success of the convergence scheme
as it proceeds towards the optimum solution.
3-30
•
Interrupt the optimization cycle to change tolerances on the terminal and
intermediate constraints, redefine the weighting factors associated with
the adjustable parameters, and/or reshape the control program.
"
Perform ,a rapid evaluation of payload sensitivity to system parameter and
initial condition changes by allowing the analyst to redefine these parameters
of interest and reinitiate the optimization cycle.
3.4.2
Computer' Program Description *
The existing batch process digital computer program that solves the payload optimization problem will be described in this section.
3.4.2.1
PRESTO. The digital computer program for Rapid Earth-to-Space Trajec-
tory Optimization (PRESTO) uses a closed-loop, steepest-descent optimization
procedure to derive flight trajecto1:'ies that produce maximum booster payloads for a
variety of space missions. Trajectories can be computed in three degrees of freedom
about a spherical rotating earth. Four powered stages and three upper-stage thrust
cycles are accommadated. Coast periods are permitted between
ea~h
stage. Aero-
dynamic lift and drag forces are included in the computations. Univac 1108 computing
times under one minute are currently being realized for complete three-stage boost
trajectory optimizations from Earth surface to Earth orbit.
The optimization routine simultaneously considers the launch direction and time,
interstage coast durations, upper-stage thrust sequencing, complete pitch and yaw
attitude histories, and the terminal constraints. Intermediate constraints may be
introduced on angle of attack, coast orbit perigee altitude or on the product of angle
of attack and dynamic pressure. The closed-loop procedure greatly facilitates the
satisfaction of terminal constraints and reduces the number of interations required to
achieve convergence.
*This subsection has been extracted from Ref. 3-6 and is presented here for clarity
in understanding the graphic-analytic program interfaces.
3-31
Four basic missions are accommodated by PRESTO - Earth launch to Earth orbit,
Earth launch to lunar transfer, lunar landing from lunar orbit, and Earth launch to
interplanetary transfer. The lunar and interplanetary transfer missions can also be
initiated from Earth orbit. From one to six terminal constraints are permitted on
Earth orbit and lunar landing missions, two or three constraints on lunar transfers,
and two constraints on interplanetary missions. Lunar, Mars, and Venus ephemeris
routines are included in the program.
For orbit missions, constraints are imposed either directly on the injection trajectory
variables or on the orbit elements. When the orbit elements are specified, functional
constraint relationships are used.
For lunar missions the constraint input is transfer
time, day of launch and, if desired, transfer orbit inclination. The constraint routine
defines the functional relationships between the injection trajectory variables that produce the required lunar intercept using special closed-form expressions to represent
the transfer orbit. For interplanetary missions the constraint inputs are transfer
time and launch date. The departure asymptote direction and excess speed are
computed from a matched conics routine and the functional relationships between the
constraints and the injection trajectory variables are defined in terms of the excess
speed and the departure asymptote direction.
3.4.2.2
Sequence of Trajectory Interations. The several types of trajectory com-
putations used in this program may be catagorized in three ways. First, they are
either "forward" or "backward." The equations of motion are integrated on a
forward trajectory only; on a backward trajectory, the adjoint differential equations
are solved. The next major categories are "guidance" or "optimization." On a
guidance trajectory, one is concerned only with meeting terminal constraints on the
trajectory variables; in optimization, mass improvement is attempted as well. Since
solutions of the adjoint equations are used differently for guidance and optimization
runs, a backward guidance run must precede a forward guidance run. Similarly,
a forward optimization run is preceded by a backward optimization run. The remaining major categories are "successful" or "unsuccessful," as judged at the end of
each forward trajectory. A successful forward guidance
traj~ctory
satisfies terminal
constraints within acceptable limits, and a successful forward optimization achieves
some increase in terminal mass as well.
3-32
The sequence of trajectory iterations starts with the initial nominal run which, of
necessity, is simply an "open-loop" integration of the equations of motion using an
input thrust-orientation history. This trajectory is then used as the basis for a backward guidance trajectory, which is followed by a forward guidance run. It is always
assumed that a forward guidance trajectory represents an improvement over the previous nominal. Thus each forward guidance trajectory become s the new nnminal and
the basis for a new solution of the adjoint equations. Backward and forward guidance
runs are continued until one is judged "successful." At this point a backward optimization run is made and the magnitude of the initial mass improvement request is computed. A forward optimization run is then made.' If it is successful, it is used as the
new nominal, and backward and forward optimization trajectories are computed. If it
is unsuccessful, that trajectory is discarded and another forward optimization is computed with half of the previous attempt at mass improvement specified. Thus, a succession of successful and unsuccessful
opti~ization
runs are computed either until
the attempted mass improvement is smaller than a specified minimum or until the
count of forward trajectories is within one of a specified limit. In either case, the
sequence is then ended with a backward and a forward guidance trajectory.
3.4.3 Interactive Aspects
3.4.3.1 Cycle of Operations. The following description of the cycle of operations
for the ascent boost optimization
pro~lem
uses the terminology PRESTO to represent
the optimization program, GRAPHICS to denote the display logic program, and EXEC
to denote the decision link between PRESTO and GRAPHICS. The interactions between
these three links and the two graphic display formats to be used, designated REVIEW
and MONITOR, are as follows:
•
The problem is initiated with computer control transferred to the EXE C
program. All punch card data has been read into storage, and this represents
all the basic data necessary to process all cases known at this time.
GRAPHICS presents the "decision-panel" on the display screen.
•
EXEC extracts from storage the data required for the first case.
3-33
•
Using this data, PRESTO computes the initial nominal trajectory for the
case, and determines the terminal constraint errors.
•
If the analyst is confident that the input data and initial nominal trajectory
are satisfactory without reviewing these results, a NO REVIEW selection may
be made on the decision panel, this step is bypassed. Otherwise, selection
of REVIEW on the decision panel activates that display from GRAPHICS for
examination by the analyst. This review of both the initial nominal trajectory
and input data that produced it provides a unique opportunity not only to isolate
and correct input errors but to evaluate the desirability of the initial trajectory. If the trajectory profile does not appear conducive to mission success,
the analyst may change a combination of initial conditions and control parameters and regenerate the' initial nominal trajectory •. This process becomes
particularly desirable for missions that require unusual maneuvers (i.e.,
dogleg turns) or when a new booster system is being evaluated. When the
analyst is satisfied with both data and trajectory, he initiates the optimization
cycle with a light button.
•
The ascent boost optimization cycle
~ses
both PRESTO and GRAPHICS.
PRESTO computes trajectories in a prescribed sequence, and, after each
forward trajectory attempted, certain 'parameters are displayed by GRAPHICS
on the scope working area in the MONITOR format. With no interruption by
the analyst, PRESTO continues the optimization procedure and the GRAPHICS
routines update the scope after each forward trajectory. This dynamic display allows the analyst to monitor key criteria of the
optimizatio~
process
and to interrupt a slow or poorly converging case in order to redirect it.
•
In addition to the information of the MONITOR format, the characteristics of
the last available successful trajectory are being stored in order to display
them in the REVIEW format, if desired. This display would be used as backup
information when analyzing a case with poor convergence.
•
Optimization continues to completion or until the maximum allowable number
of iterations have been computed. Prior to computing the final guided trajectory, the optimization program' will halt and give the analyst the opportunity to extend the optimization cycle or, if satisfied with the present results,
3-34
to complete the case. The final guidance trajectory characteristics will then
be displayed in the MONITOR format, and the REVIEW format would be available upon light button command.
o Satisfied with the results, the analyst would trigger the return to the EXEC
program, which would then extract the data necessary for the following case
from the stored (punch card) input data, and reinitiate the cycle.
o If all of the input cases have been processed, the EXEC would initiate redisplay of the last REVIEW format but with all characteristics resulting from
trajectory integration eliminated, and
a~ait
input
fr~m
the analyst. This
would permit the synthesis of characteristics for additional trajectory
optimizations.
o When the study is completed, triggering the OFF light button would provide a
printed listing of the final trajectories of the cases evaluated using the
standard PRESTO output format. The program would then terminate.
3. 4. 3. 2 Displays "for Ascent Boost Optimization. Two graphical displays have been
developed for the ascent boost optimization problem. They serve tw9 different functions. The REVIEW display provides the means for a more leisurely and thorough
perusal of numerical data, whereas the MONITOR display presents the rapidly changing information in a format where errors, slow convergence, and oscillations in key
parameters should be immediately apparent.
Both displays contain light buttons within the scope working area. In keeping with the
premise of minimizing analyst decision chains, the computer reaction to
a single
analyst light button action should be as complete a logical chain of commands as is
possible. Only when absolutely necessary should the analyst be required to input a
chain of light button commands to achieve the desired result. For example, if the
analyst wishes to change the input data, touching the light button surrounding the
desired data word should initiate machine responses that would include the following:
•
Underline the data word so the analyst can confirm his choice.
•
Accept new input data from the scope keyboard in the correct format, and
display the new value.
3-35
•
Recognize that when control histories are input via light pen a single light
button command should allow the analyst to construct the curve and then automatically manipulate it into the data format necessary for PRESTO.
e Cancel the data input capability for the parameter just selected if the analyst
touched another light button (such as GO) prior to inputting data and terminate
that data acceptance sequence with no further machine action. Touching
another light button after data input would signal acceptance of those data.
o Determine whether those data require a special restart or reinitialization in
the optimization program. (For example, inputting an allowable number of
iterations increases that counter and sets a flag to indicate the end has not
yet been reached, whereas, changing weighting factors requires a backward
integration.) React to all data changes such that all necessary changes are
made, but do not cause duplication or false restarts.
When developing the two display layouts, a :vorking area of 11 x 17 in. was assumed
available. (Illustrations are shown 3/4 size.) An alphanumeric font of seven characters per inch, six lines per inch was used. However, experience has shown that a
smaller font, approximately half size, can be used for the titles whiie retaining full
size for numbers or single letters. This is possible since titles can be easily recognized as a group, whereas numbers have to be individually interpreted. The advantage
of this dual size font layout is that it reduces the apparent amount of information displayed (scope clutter) and differentiates between guides (titles) and information (data).
The decision panel light buttons and a standard alphanumeric keyboard for data input
changes will remain on the scope control surface area outside these displays during
the entire time of program operation. The decision panel layout would be
I STOP I
NO
REVIEW·
RESTART
MONITOR
GO?
OFF
3-36
Some of these decision buttons have already been referred to in the cycle of operations
section; their functions are:
I STOP
I
An immediately identifiable panic button that stops program
operation and lets the analyst make a decision on what action to
take next.
REVIEW
Calls the REVIEW display to the scope, interrupting the optimization cycle if necessary.
NO
Bypasses the REVIEW format pause for the initial nominal and
permits computation to proceed into the optimization mode.
MONITOR
Calls that display to the scope, but
do~s
not initiate the optimi-
zation process.
RESTART
Initiates the computation of the initial nominal trajectory and subsequent display in the REVIEW format (this could be used repeatedly
to refine initial estimates of control history profiles, for example).
GO?
. The question mark appears when the program is waiting for input
of any type by the analyst.· GO after STOP would. continue operation
from the interruption point. GO when the REVIEW display of the
initial nominal is on the scope initiates the optimization computation
and calls the MONITOR display. GO with a REVIEW display of any
other trajectory continues computation from the point of interruption
or as determined by other light buttons and calls the MONITOR
display. GO with the MONITOR display on the scope continues computation from the point of interruption or as determined by other
light buttons.
OFF
Triggers hard copy printout and terminates the program
Finally, any error messages generated during program operation will be displayed in
the control surface area above the display layouts for quick analyst response. Provisions must be made such that analytic program errors don't cause total program
failure, i.e., termination.
3-37
3.4.3.3 REVIEW Display (Fig. 3-2). Details of the working areas of the displays
correspond to the functional layout on the left of Fig. 3-2. Light buttons are indicated
by the dashed boxes on the illustration.
Trajectory Profile. The combination of altitude and aerodynamic velocity provides an
excellent indication of the trajectory profile. For example: Is the terminal altitude
anywhere near the-desired target condition? Did any stage impact the ground? Is each
stage contributing its expected velocity increment? Note that both altitude and velocity
can legitimately be decreasing; for example, descent to lunar surface from lunar
orbit~
Simplicity is the keynote; the altitude scale runs from zero to 100 nm and velocity to
30,000 ft/sec in increments of 10,000 ft/sec, where both these scales can be halved or
doubled by single light button commands (H, D); repetitive commands change scales
once for each command. Curves are plotted only for the time under power; verticals
connect any "breaks" that might occpr at staging because coast stage conditions are not
being shown. The time scale unit length is set by the total amount of time under power
and automatically scaled by the GRAPHICS program. Numerical values are not shown
on the time scale since it is identical to the control history time
scal~
directly below.
However, stage ign,ition points are indicated with short vertical line segments. The
vehicle velocity at the end of the closed-form liftoff is output in ft/sec at the bottom of
the velocity scale and unaffected by scale changes.
Controls. Pitch (theta) and yaw (chi) control histories are plotted starting at the time
the zero angle of attack constraint is removed (i.e., only when meaningful).. The control
history to be used on the next forward trajectory may be input by touching the appropriate light button and then using the light pen to indicate the new data points that the
control history curve should fit. The weighting factor on chi is changed by touching its
light button and entering a numerical value via the keyboard.
The angular scale nominally spans +90 to -30 deg, and may be repetitively doubled or
halved by touching the appropriate light button the desired number of times. The 5-deg
tick mark is always shown to ease rapid estimates of scale height changes, especially
for redrawing. If the upper scale limit becomes 10 deg or less, this mark is eliminated.
3-38
FUNCTIONAL LAYOUT
(3) ADJUST ABLE
PARAMETERS
(1) TRAJECTORY
PROFILE
(4) CONSTRAINTS
I
(5) ITERATION
I
(6) VEHICLE STAGE DATA
(2) CONTROLS
(7) INITIAL AND
FINAL CONDITIONS
I
(8) l-IEADINGS
I
r---------------------------------------------~--_=----------,30
r,-- - -::l
~~~~~~
ADJUST:6:
1___ 1
ALTITUDE
r2.73-.l.-ll
I----~
IL 3.24
_ _ _.I._0-lI
L ____ ..J
100
ADJUSTMT
20
WEIGHT
3.21 - 3
-1.11 + 1
[!~~~J [Y~~£J
r-----I
L_~ ___5_J
rL_____
~ .J
rGAMMA{l
L
____ ..J
L ~N.9~_...J
r-- ----,
r------,
L:':~ ___0...J
3.62 - 7
-3.24 - 4
1.12 - 3
-4.32 - 1
VELOCITY
r-----,
-1-:- =- -,
L _::.:...+_O_J
r----,
L~'_:~~
10
1. 32 - 1
6.23 - 3
r-----,
1-2.21
- 1.....JI
L ____
OK ERROR
CORRECT'N
6.21 + 1
-3.21 - 4
6.21 - 3
6.21 - 3
0~~--~----------~----------------------~--------------------~210
ITERATION 1 CONSTRAINTS
STG
ISP
S
WEIGHT
T/WO STG
SEQ
r -7000000~
I r'--______
2430000: I 1 1.24
L
_____ ..J
6213642.
r-----,
1672000.
r- - - - - I
L~OOO~O~•...J 1_ ':'4~0~~...J 2.13
2
2742295.
666000.
r-----, r-------,
L ~4~0~~.J L _5~700~..J
3
744672.
r- - - - - - ,
~',
1--'
HI
L_-'
"' --- 0
0
• • • • • • • • • • • • • • a.,L • • • • • • • • . . . . .
o'
1_
':'4~4~2.:. -.J 1.62
85692.
r------,
,
5
274000.
r - -----,
303000.
"-----.
1
1_ ~~~~...J
4
1.37
5
L_~~~O.:.._.J
r----,
IL
.02 ...J1
___
0 0
0·0
,0
0'
,00
...
390.2
0·'
,., '
25630.
rINVii-'
STOP L: ___ ..J
r-----,
00
-30
r -~--~~----------~----------------------~--------------------~
-l
243.
124.
396
L~:~2_1
50.
L~~~...J
1.3
183.45
712345.
EYE
BETAP
OMEGAE
90.15
80.30
159.83
21.18
-121.42
QBAR MAX 32482. AT 172.8
TIME
;-.... -72 CHARACTER
CAsEHEADING -LAUNCHSOUTH~ioo NMPERIGEE6CONSTRAOOi
--- --- - -- - - - - - - - - - - - - - - - - - - - - - - - -
Fig. 3-2 Ascent Boost Optimization IlEVIEW
Disp1ay
3-39
The flight path angle (gamma) is shown from the end of closed-form liftoff, unless it
is off scale, in which event the curve does not appear until it returns within scale.
Similarly, when the controls exceed scale height, they are not shown until within scale.
However, when redefining the shape of a control history, the analyst may
wor~
outside
the scales then shown, assuming scale extensions at the current unit rate.
The time scale begins at the end of closed-form liftoff, and this value may be changed
by light pen input. Coast period times are not included, and breaks in the curves
across coasts are fitted with vertical segments. Time scales are identical for the
MONITOR layout.
Adjustable Parameters. Names, input magnitudes, current adjustment, and input
weighting factors are shown only for the adjustable parameters being used. Provision
is made for up to six of the ten possible parameters, since experience shows this
number represents an upper limit requirement. Inputs may be changed by a light
button-keyboard sequence. The number of parameters and the parameters themselves
(numeric codes) may be changed by light button.
Constraints. Names, input magnitudes, error tolerances, and current corrections
are shown only for the constraints being used. Inputs, the number of constraints, and
the constraints themselves (numeric codes) may be changed via light button-keyboard
sequence.
Current Iteration Number and Trajectory Status. Status nomenclature inoludes:
CONSTRAINTS
The trajectory failed to meet constraints.
SATURATE
The control parameter changes necessary to satisfy constraints exceeded limits (saturated controls).
PAYOFF
The trajectory failed to achieve a payoff improvement.
SUCCESS
The constraints were met, and, when optimizing, a payoff
improvement was achieved.
FINAL
Final guidance run (end of case).
3-41
Vehicle Stage Data. A combination of raw input data and program computed values is
shown that ensures a rapid verification of vehicle parameters. The input parameters
that may be changed for each of the five main stages via light button-keyboard are:
DUR'N
Stage burn time duration (sec)
!SP
Stage vacuum specific impulse (sec)
AE
Stage aerodynamic reference area (ft2)
S
Stage total nozzle exit area (ft2)
Stage ignition vacuum thrust (lb) (if this value is changed, the pro°
gram constructs a rectangular thrust curve at the new level for the
entire burn time)
WEIGHT
StOage initial weight (lb)
STG SEQ
Stage sequence, a numerical code indicating the order of computation
of powered and coasting stages
Other parameters that are output for checkout only and cannot be altered are:
T
Net thrust at stage ignition (lb)
WEIGHT
The second value of stage weight is the program computed value of
stage burnout weight (lb)
T /w0
Ignition thrust-to-weight ratio (g)
This data block is sufficient to allow rapid variation of stage parameters to determine
payoff sensitivities, as well as verify inputs.
Initial and Final Conditions. Initial conditions may be changed via light buttonkeyboard; final conditions are output for information on terminal trajectory conditions.
In addition to the six state variables, the stopping parameter and three orbital parameters are output.
Headings - Identification. When a character is touched by the)ight pen, the sixcharacter alphanumeric word of which it is a part is underlined. Keyboard changes
03-42
must then start with the first character of that word, but may continue beyond that one
word until the end of the title or until another light button is activated.
3.4.3.4 MONITOR Display (Fig. 3-3).
Characteristics of the current trajectory are
shown on the MONITOR display; this display changes as fast as a new forward trajectory is computed. Refer to the left side of the illustration for area definition.
Light
buttons are indicated by dashed rectangle s.
Trajectory Profile. Only the altitude profile is displayed, and to the same scale as the
REVIEW display. The scale may be halved or doubled, and changes here carryover
to the next display of the REVIEW format. Placing the indicated upper altitude value at
80 percent of the physical scale height allows for reasonable overboost trajectories
to a given terminal altitude
0
Both the trajectory profile and controls are in the same area and with the sanle format
as the REVIEW display; this should facilitate rapid transition between the two displays
and be less confusmg to the analyst.
Controls. With the flight path ap.gle eliminated, changes in the two controls between
iterations should be more apparent. The angular scale may be doubled or halved, and
any change will be reflected on the REVIE W format. The time scale is identical to
that established for the
R~VIEW
format. If the REVIEW format has not been used, the
initial values are as shown for altitude and control magnitudes. The chi weighting
factor is manipulated in the same manner as the weighting factors on adjustable
parameters, discussed below.
Adjustable Parameters. The adjustable parameters, the indicators, and the constraints
are all represented by "thermometers" or linear indicators. The horizontal segment
of the arrow head indicates the magnitude of the parameters, the direction of the point
shows whether the parameter is increasing or decreasing in absolute magnitude, and
the ± sign grouping is a flashing signal that appears only when the sign of the parameter
changes from one run to the next. These concepts should enhance the analyst's ability
3-43
to detect changes that require a reaction. Since more information is being presented
that can be monitored on each iteration, the display pattern has been designed to
capitalize on the dynamic motion of the error signals.
Thermometer scales are shown only for those adjustable parameters actually being
used, and have a one-to-one correspondence to those shown on the REVIEW display.
The left side of the thermometer shows the magnitude of the current adjustment.
During guidance iterations, the scale is renormalized after each trajectory, and thus
represents a step-by-step progression. Mter optimization starts, the normalizing
value is fixed at that on the last successful forward guidance o Thus the arrows should
represent changes in adjustable parameters caused by mass improvement requests.
However, the scale may be renormalized to the current value during optimization by
touching the light button on the top left side of the respective thermometer scale.
The constraint name appears directly below each scale, and the magnitude of the
weighting factor appears below the name. The numerical value of the gain., or weighting
factor, is output as "a fixed-point number" since exact values are unimportant; only
relative magnitudes between gains are of interest. Fractional values .should be output
with the correct nu;m.ber of preceding zeros, but ~o decimal point. The magnitude of the
gains may be doubled or halved by touching light buttons directly above or below,
respecti.vely, the current numerical value. Each click of the light pen causes an' additional doubling or halving operation, i.e., three clicks above a gain would increase the
value by a factor of eight (3 doubles). Weighting factors, when changed, are introduced
into the program without altering the normal trajectory sequencing. In the. lower left
corner of this block, the D and H serve as reminders of this multiplying capability.
The number above the word ADJUST indicates the number of adjustable parameters
being computed.
Arrows on the right side of the thermometer scales represent changes in the payoff
quantity per unit change in the adjustable parameter, where the computation also includes the effects of changes in control parameters necessary to satisfy terminal constraints. As such, the computation
represe~ts
a "total derivative" type of sensitivity.
The scale runs from 0 to 2 showing only absolute magnitude. It is normalized with
3-44
ALTITUDE
r--'1
10
,-_J
100
FUNCTIONAL LAYOUT
(1) TRAJECTORY
PROFILE
(3) ADJUSTABLE PARAMETERS
O~----~-------L----------------~------------------------~
(4)
PAYOFF
(5) INDICATORS
90r---------------------------------------------------~r~-~-==-~,
--THETA
--- CHI WT
(2) CONTROLS
IL
___ ...JI
300
r------,
,
I
'- _ _ _ J
(6) CONSTRAINTS
r-,
.0.
l._.J
~
,
f' . . . --
-- ------
---
O~----r_------~------_r~------~~----------------------~
-30~----~------~
4
50
________________J __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
124
243
~
396
6.1 - 6
,.- ., ,.- ,
1 :
I
:
L.
I
~
5.1 + 4
r-' ,-- ..
-8.3 + 2
....
I
I
,
-, .... -,
I
I
I
,
I
LJ
I..J
,
,
I
I
I
I
L_.J
,
I
1-_ J
-6.2 + 3
r-,
I
I
I
I
I
I
L_J
I-_J
4.2 + 1
3.3 + 2
r-,
r-,
I
,
I
I
I
I
I
I
....
I
'-_J
I..J
-,,
,
2
I.._J
ALTITUDE
100
ADJ
ADJ
±
6
ADJUST
±
±
±
±
\l
o
1::::,.
~±
±
±
±
±
±
1::::,.
DP
±
1 DADJ
o
o
GAMO
.... ---,
AZM
r- --,
D L ___ 1
GAIN
2000
r- -
H
CST 4, 2
II.. _ _ ...JI
r---,
I
30
0001
,
,
r- ---.
L.. _ _ ...J
--~
L- _ _ _ "'"
CST 4, 1
BURN 4,2
.-- --,
I
IL..
D
1
r---,
___
r- - - - - ,
r---'
L
___ J
I
'---_.I
002
IL.. _ _ ....II
BURN 4, 1
,...---,
IL.. _ _ _ oJI
L... _ _ .JI
,.----,
.J,
I'-
___ JI
... ,
o ~----~--------~------------------~--------------------------~
I
I
1
.J
GUill
SUCCESS
CORR
WT
83640.
,..--,
90~----------------------------------------------------~~~~
- - - THETA
rL -___
-- ~
- - - CHI WT
r-
DW FINAL
...J
:L.. 85.:
__ J
~Q.O_-,
,
,_ _ _ _ J
43.
,---.,
EST REM
I
.,
,.-,
L.. _ _ J
UNIT
MATRIX
-- ------"'"
I
26 - : 32 :
ITRN
-f"'---
.9998
:L ____
88.72
o
'-_..J
DW FINAL
DMASD
WT ERR
DW FINAL
,-----,
GAM
"
lHI
±
I
:
.42
-J
o
.52
.85
r,
,..,,
,
OPM
IND
I
I
L...J
I
I
2
1....1
o ~----4---------4-------~~-------~------------------------~
COR
TOL
-30~----~-------L
4
50
________________
124
~
243
±
±
±
±
±
±
±.
±
DP
±
1 DPSI
________________________-J
396
o
o
6
CONSTRAIN
TOL
ALT
r-------,
I
2.32
-r-
3 :'
L _____ J
GAM
,-------,
! -2.32
- 2 :
L.. _ _ _ _ _ --'
INC
r------,
lL
2.32 -'- 0
I
_ _ _ _ _ ...J
NODE
r----- -,
: 2.32
.!.
0 :
L _____ .....J
LONG
r-----.,
IL 9.21
- 1 :
_____ l
LAT
r-- - --,
:L -8.62
- 1 J:
______
Fig. 3-3 Ascent Boost Optimization l\10NITOR
Display
3-45
respect to the value of the first backward optimization. and may be renormalized to the
value of the last successful trajectory via the light button on the top righthand side of
the respective thermometer scale. The current actual value appears above the scale,
and carries the correct sign. Sign changes are also indicated by the on-off behavior
of the ± symbol.
Note that the current sign is not shown with either the + or - sign,
but rather the combination is used to indicate a changing condition.
If the optimization were proceeding successfully, the behavior of the adjustable
parameter thermometer scales would be a sinking of the left side towards zero and
r
movement of the right side to the center of the scale. Since all scales would exhibit
the same characteristic behavior, deviations should be quite apparent.
Payoff parameters are boxed for rapid visual access.
The current trajectory mode
is indicated by GUID for guidance or OPTM for optimization attempts, and the success
or failure indicated by the trajectory status, identical to the REVIEW display (SUCCESS,
PAYOFF, CONSTRAINTS, SATURATE, FINAL)o CORR WT is the vehicle burnout
weight in pounds corrected for terminal constraint misses and represents the optimization payoff quantity. DW FINAL is the nominal value of the burnou.t weight improvement requested, and diminishes as the optimization nears completion. Analyst changes
in DW FINAL would not require changes in the normal trajectory sequence.
Indicators. Four numerical values and three thermometers are output as key indicators of the degree of success the optimization procedure is having.
EST REM
Estimated remaining weight improvement that can he realized
for this case (lb).
ITRN
Current iteration number and the total that will be allowed. Total
may be changed via light button -keyboard and does not interrupt
the normal trajectory sequencing operation.
UNIT
MATRIX
Deviations from a val,ue of 1.0 indicate failure in matrix
inversion process.
3-47-
Initial flight path angle (deg). Oscillations signify failure of convergence for this key adjustable parameter. If the analyst
changes GAM o ' the trajectory sequencing should be restarted
from the initial nominal logic since changes in this parameter
alter the trajectory and adjoint equation integrations. Control
histories, gains, and vehicle weights from the last successful
trajectory should be retained.
Each of the three thermometer scales are normalized (zero to one range). They are so
designed that with a properly· proceeding optimization, each of the three arrowheads
will sink to the bottom of the scale. The actual numerical values are shown directly
beneath each scale for reference. Only the first scale requires the ± symbol to indicate
sign changes.
Scale 1 shows the weight error due to missing terminal constraints divided by the requested weight improvement. A trajectory that exactly satisfies the terminal constraints produces a value of zero. On the other hand, a value of one means the terminal
constraint errors may cancel any achieved weight improvement, and the constraint
tolerances may have to be reduced.
Scale 2 shows the ratio between the current requested weight improvement DW FlNAL
and its initial value DMASD, and indicates how far the optimization has progressed in
the sense of requiring smaller step sizes in the requested payoff.
Scale 3 is the optimization indicator, actually the inverse of the optimization indicator
currently cOlnputed on PRESTO. The inversion causes the displayed value to start at
1.0 and decrease towards zero, rather than increasing without bound.
Since the values of both scales 2 and 3 can change by several orders of magnitude,
these scales may be renormalized to the value on the last successful trajectory by
merely touching the light button on the top of the scale.
3-48
Constraints. The thermometer indicator scales for the constraints operate similarly
to the adjustable parameter indicators. Scales are shown only for the constraints being
used, and under normal operation the left arrowheads will sink to the bottom of the
scale and the right arrows move to the center of the scale. Direction of motion is indicated by the arrow point. Sign changes would be signalled via the
±
symbols, which are
displayed only when the sign actually changes between iterations. Thus, any erratic
behavior patterns should be immediately evident as the optimization progresses.
Beneath each scale appears the constraint name and the allowable constraint magnitude.
If the analyst changes the constraint tolerance during optimization, the trajectory
sequencing will return to the guidance mode until all constraints are again satisfied
within the specified tolerances. The total number of const'raints currently imposed
appears over the word CONSTRAIN in the lower left corner of this functional area.
The left side of each scale represents the correction required to null the constraint
error normalized to the allowable tolerance; therefore, a trajectory completely
satisfying the constraints shows a value of zero. The right side shows the sensitivity
indicators DP /DPSI, which represent the change in payoff per unit
c~ange
in the con-
straint. This 'scale runs from 0 to 2, and during guidance~ the arrowhead represents
the current value normalized to the value on the previous successful trajectory.
During optimization the current value is normalized to that of the first successful
optimization trajectory, unless the analyst rescales to the last successful optimization
by touching the light button at the top of the scale. If the computed values on either
side of the scale exceed the limits shown, the arrow is located at the top of the scale.
3-49
3.5 REQUIREMENTS SPECIFICATION FOR ASCENT BOOST OPTIMIZATION
3.5.1 General Guidelines
3.5.1.1 Objectives. The five objectives listed in Section 3.4.1 will be
~he
objectives
of the Ascent Boost Optimization Graphical-Computer System.
3.5.1.2 Nature of This Specification. This Specification sets forth requirements on
the performance and capabilities of the system from the viewpoint of the user-analyst.
As far as possible, it is independent of any pa'rticular piece of computer or display
equipment.
3.5. 1. 3 System Operation. The system will be so designed that its operating displays
and instructions will be expressed in the
lan~age
of the user in that it will use the vo-
cabulary of flight mechanics and engineering plotting procedures.
3.5.2 Performance Requirements
3.5. 2. 1 Sequence ·of Operations.
This system will operate in the following sequence
of steps as described in Section 3.4.3:
•
Data preparation by the analyst
•
Program read-in and initialization
•
Computation of the initial nominal trajectory
•
Data display in the REVIEW format:
Data correction
Recycle to initial nominal trajectory
•
Computation of trajectories in the optimization cycle
•
Concurrent display in the MONITOR format:
Interrupt and recycle
Data change and recycle
e Continuation to next case from stored input data
•
Generation of new case from graphic input data
3-50
3.5.2.2 Inputs.
Entries are as follows:
o Program.
The system will read in from magnetic tape or cards either source
or object decks of the PRESTO program suitably modified for graphics.
o Data. The system will read in from magnetic tape or cards all the input data
necessary for program operation using the standard PRESTO data read-in
subroutine.
Data blocks, definitions, and formats are described in detail in
the PRESTO manual (Ref. 3-6).
3.5.2.3 Outputs. Output is as follows:
•
Printed output.
The system will compute and tabulate all the trajectory and
intermediate output from each trajectory iteration using the standard PRESTO
output subroutine.
Data, definitions, and formats are described in Ref. 3-6.
In addition, the system will tabulate a separate output of only the final guidance
trajectories that were computed. This can be done by transferring the standard
final output as it is generated onto a second storage device.
•
Graphics output. The system shall display the data, plots, alphanumeric information and optimization control indicators shown in Figs.. 3-2 and 3-3 of
Section 3.4. 3.
•
Photographic output. No photographic output is required ..
3.5.2.4 Storage and Timing Requirements for Ascent Boost Optimization.
•
Overall requirements.
(1)
Storage - Exclusive of the resident monitor(s) for the computer(s) used
in the system, the core storage requirement for operation
witho~t
lays or chain links is estimated to be:
Code words
33,000
Data words
24,000
57,000 (CDC System words)
Total
3-51
over-
(2)
Timing - During the optimization cycle, the MONITOR display should be
updated every 2 secs with the new trajectory information in less than 1 sec.
The complete REVIEW display should appear in 6 secs.
(3) Data transmission rate - The time specified for updating the MONITOR
display places the most stringent requirements on data transmission rates
between the system hardware components where the trajectory is computed and the display information prepared. It is estimated a data transmission rate of 3,000 words/sec will be required.
e Detailed requirements.
The
comput~r
storage requirements, reaction times,
and data transmission rates have been estimated for the ascent boost optimization graphical-computer system, and are summarized in Table 3-4.
They are
furnished only as an indication of desired performance, and not to indicate the
actual capability Qf any particular hardware.
The sequence numbers in the
table correspond to the steps described in Para. 3. 5. 2.1.
(1)
Storage requirements - The PRESTO computer program, including systems subroutines, currently requires 31,400 storage cells on the LMSC
Univac 1108 Exec II System, using Fortran IV.
This c?mprises 17,670
code ,. 2,200 data, and 8, 600 common words, the balance is system
routines.
These figures were converted to CDC System requirements by
. assuming a one-to-one code correspondence, 30 percent larger system
routines, and, since the CDC System requires two words to represent one
Univac floating point word, the data and common requirements were doubled.
Thus the estimated CDC core requirements becolm 17,600 code, 4,400
data, 17; 200 common, and 3, 800 system words for a total requirement of
43,000 cells.
Core storage requirements may be reduced by approximately 10,000 cells
if an overlay technique is used.
PRESTO was developed by LMSC for the
NASA Langley Research Center under Contract NAS 1-2678. When delivered to Langley RC, the program had to be segmented into a chain link or
overlay program in order to fit on their IBM 7094 computer, and is still
being used in that configuration. The program may be divided into three
overlays without seriously impairing computational speed. Column 4 of
3-52
Table 3-4
STORAGE AND TIMING REQUIREMENTS FOR ASCENT BOOST OPTIMIZATION
Sequcnce
Number
Step
Required Computer
Core Size (a)
(1,000 words)
Operation
Desired
Reaction Times
(sec)
Code
Data
-
-
-
Estimated Data
Rate Requirements
(words/sec)
-
-
1
Preparation
Analyst assembles and prepares
inputs
~
Initiate
PRESTO program loaded and
stored(b)
21. 0
22.0
45-60
9,000 card sourcc
deck at card
reader speed
Punch data read in and stored
2.4
5.0
15
100 to 1, 000 cards
at card reader
speed
10.0
2.0
3
-
2
-
GRAPHICS initialized and decision
panel displayed
3
4
Compute
Initial
REVIEW
Compute initial nominal trajectory
•
•
Data initialization
I
6.0
3.0
13.0
20.5
REVIEW background displayed
10.0
2.0
3
1,000
REVIEW data displayed
12.0
2.0
3
1,800
Analys t edits and corrects errors
12.0
2.0
As
necessary
Trajectory integration
100
Regenerate initial nominal trajectory
•
•
6.0
3.0
13.0
20.5
Update REVIEW display
12.0
2.0
Data initialization
Trajectory integration
I
2
-
3
1,800
1-2
-
5
Compute
Sequence
Compute one backward and one
forward trajectory in iteration
cycle
13.0
20.5
6
MONITOR
Initial display of MONITOR
background
10.0
2.0
1
2,500
Update scope with data for current
trajectory at end of compute
sequence (every 1- 2 sec)
12.0
2.0
1
3,000
2.4
5.0
3
100
7
Next
Case
Retrieve next case data from stored
input data
8
New
Case
REVIEW display
As described above
12.0
Analyst generates data
(a) CDC Computer System assumed. Data include common reqUirements.
(b) Total program size shown for this operation.
3-53
2.0
I
As
nec~ssary
100
Table 3-4 assumes utilization of this overlay mode, and the three links
and their estimated core sizes in a CDC System are:
Overlay Function
Code
Data
Total
Data read-in/retrieval
Initialize case
Trajectory integration
2,400
6,000
3,000
5,000
3,000
20,500
7,400
9,000
33,500
Storage requirements for a typical graphics executive routine,. a graphics
display package and a control program directed by light button signals were
estimated using currently operati<;>nal LMSC CDC 3300 graphics packages.
(2)
Timing requirements - The pacing item in the ascent boost optimization
graphics system is the computational speed of PRESTO. Currently, one
trajectory iteration (backward and forward trajectories) requires less
than 2 sec. When a forward trajectory fails, the next iteratioIl: does not
compute another backward trajectory so the computing time is even less.
In order to capitalize on the PRESTO capabilities, this speed should not
be compromised.
Howev~r,
it is apparent that the G~APHICS system requires a finite
time interval to display the trajectory information and the analyst needs
time for recognition and reaction. Thus, the following sequence of events
and desirable timing was developed:
Step 1 - Initial nominal trajectory computed
1-2 sec
Step 2 - Data transferred and displayed in REVIEW format
6 'sec or less
Step 3 - Backward trajectory integration
Computation
Step 4 - Forward trajectory integration
in 1-2 sec
Step 5 - Data transferred and displayed in
MONITOR format
1 sec or less
The cycle from Step 3 to Step 5 is then repeated, resulting in an overall
cycle time of 3 sec. This timing sequence assumes that the data transfer
3-54
and calculations necessary for the next display will not interrupt the
current display on the console, such that the time interval with no data on
the screen is minimized.
(3)
Data transmission rates - The estimated data transmission rate requirements shown in column 6 in Table 3-4 were generated by assuming that the
data required for a particular display should be transferred from one
hardware unit to another in one-tenth of the time allotted for the overall
transfer-display operation.
3.6 CONCLUSIONS AND RECOMMENDATIONS
Digital computer-graphics systems have been. devised to treat two analytical flight
mechanics problems.
These two problems are representative of a number of dynamics
studies in the fields of flight, structural and- orbit mechanics. Related applications
would include, in part, such topics' as recoverable booster flight profiles, sea-going
tracking ship
plac~ment,
planet surface coverage, interplanetary mission analyses,
locating and designing spent booster vent ports, and analyses of frequency responses
using sand z· transforms and generating· root locus and Bode plots .. The features that
relate the two problems which have been studied to all these applications are operation
with graphical cues in an interactive man-machine mode, the abilio/ to store console
displays and to recall them rapidly for subsequent comparison and reevaluation, and
the need for effective control of the system by means of light button instructions easily
operated and understood by the user.
Furthermore, both the interplanetary transfer
computer program MAOT and the ascent boost optimization program PRE STO are
widely used in NASA.
Recommendation.
Therefore, it is recommended that both computer-graphics systems
based on tl,lese programs be developed and put into operation as soon as possible.
Working with an actual graphics program is an invaluable experience. It provides
virtually the only means of assessing the effectiveness of the human operator; it
3-55
furnishes data on his capabilities, his capacities for pattern recognition and data
handling, and his speeds of response. It is the acid test of the features incorporated
in the system; it quickly points up the inadequacies and superfluities in the software
(such as choices of variables, displays, controls, and programming methods) and
hardware (such as storage capacities, data transmission rates, operating speeds, and
display size and clarity) and can thus guide more future program developments and
equipment acquisitions. In a broad sense, it probes the program philosophy of graphics
operations and shows where major changes in approaches and standards are needed.
Detailed Requirements Specifications have been drawn up to define the sizes and operating speeds needed for the operation of the two studied systelns.
Code and data
storage requirements do not appear to be so large as to exceed the capabilities of the
Univac 1108 computer.
However, the storage and recall requirements for high-
speed data transmission between the
graphic~
console and a remote
large~capacity
data storage device do seem to be one to two orders of magnitude larger than those
presently being contemplated for theMSFC graphics installation.
Recommendation. Therefore, it is recommended that two methods be studied and
compared for meeting these requirements, namely, that currently planned data
transmission rates be increased to about 3000 words per sec or that a high-speed,
large-capacity data storage device be added to the graphics satellite
termin~l,
and
that one of these methods be included in the MSFC graphics system.
3.7 REFERENCES
3-1
B. W. Boehm and J. E. Rieber, "Graphical Aids to Aerospace Vehicle Mission
Analysis ," Paper No. 67-897 presented at the AIAA 4th Annual Meeting and
Technical Display, Anaheim, Calif., 23-27 Oct 1967
3-2
National Aeronautics & Space Administration, Space Flight Handbooks, Vol. 3,
"Planetary Flight Handbook," Parts 1-3, NASA SP-35, Washington, D. C. ,
1963 (Contract No. NAS 8-5031)
3-56
3-3
National Aeronautics & Space Administration, Space Flight Handbooks, Trajectories to Jupiter, Ceres, and Vesta," Part 5, NASA SP-35, Washington, D. C.,
1966
3-4
Space Flight Handbooks, "Mars Stopover Missions Using Venus Swingbys,11
Part 6, in press (Contract No. NAS 2-4175)
3-5
-----, PRESTO: Program for Rapid Earth-to-Space Trajectory Optimization,
R. E. Willwerth, Jr., R. C. Rosenbaum, and W. Chuck, NASA Cr-158,
Feb 1965, also LMSC 4-36-65-1, 18 May 1964 (Contract No. NAS 1-2678)
3-57
Section 4
STRUCTURAL DESIGN ANALYSIS
4. 1 BACKGROUND
While the scope of structural analysis may be simply defined as the assurance of
structural integrity of a structure immersed in a hostile environment, the actual
ways and means of providing and assessing this assurance are many. The field of
structural analysis abounds with "methods" (a blend of
th~ory
and magic). There
exists no unified structural theory of practical value. Each structural problem does,
in general, possess its own unique solution, based on a unique set of assumptions.
Some idea about the variety and, sometimes, multiplicity, of solutions to structural
problems may be obtained from Table 4-1, where a representative selection (not a
universal listing) of computer programs used by the structural analysis department
of one aerospace company is shown.
The list comprises programs which are used by "line" personnel in the solution of
more or less routine assignments. It is this group of programs, and this group of
persons who would benefit most from the Computer Aided Design (CAD) capability.
It is questionable that the engineer engaged in scientific research exploration of
structural problems would get any real benefit from CAD.
Notice, scanning the list of computer programs (Table 4-1), that with few exceptions
they compute the behavior of a specified structure in a specified environment. They
do not define a structure that would be adequate in a given environment, much less a
be st structure.
It appears that the main justification for implementing any of these programs with a
computer graphic capability would be to change this condition, and let the computer
system be a direct design tool, rather than an analysis machine. The ramifications
4-1
Table 4-1
REPRESENTATIVE LISTING OF STRUCTURAL ANALYSIS COMPUTER PROGRAMS
1. STATIC STRUCTURAL ANALYSIS PROGRAMS
1.1
Frame Analysis Programs
1.1.1
WFRAME - WHETSTONE Program For Linear Analysis of
Three-Dimensional Space Frames Subject to Concentrated
External Loads
1.1.2
STRESS - MIT Program For Linear Analysis of ThreeDimensional Space Frames Subject to Concentrated or Distributed Loads, Support Motions and Temperature Effects
1.1.3
FRAN - IBM Program For Linear Analysis of Three-Dimensional
Space Frames Subject to Concentrated or Distributed Loads,
Support Motions and Temperature Effects
1.1.4
PFRAME - Program For' Linear Analysis of Two-Dimensional
Plane Frames Subject to Concentrated or Distributed Loads
1.1.5
. GFRAME - Program For Linear Analysis of Two-Dimensional
Grid Frames Subject to Concentrated or Distributed Loads
Normal to Frame Plane
1.1.6
' SFRAME - Program For Linear Analysis of Three-Dimensional
Space Frames Subject to Concentrated Loads at Joints
1.1.7
GFLAC FRAME - Program For Linear Analysis of TwoDimensional Plane Frames Subject to Concentrated or Distributed Loads
1.1.8
UBEAM - Program For Determining Internal Loads in, SemiM9nocoque Structures by Unit Beam Method
1.1.9
BATMAN ..... Program For Linear Analysis of Three-Dimensional
Space Frame/Panel/Plate Structures Subject to Discrete Applied
Loads
1.1.10
CARR -Program For Linear Analysis of Three-Dimensional
Space Frame and Panel Structures Subject to Concentrated and
Distributed Loads
4-2
Table 4-1 (Cont.)
1.2
Shell Analysis Programs
1.2.1
SADISTIC-I - Program For Elastic Analysis of a Bi-layered
Orthotropic Conical Shell With a Non-Deformable Bond Layer
Subject to Axisymmetric Surface Loads and Thermal Gradients
1.2.2
SADISTIC-II - Program For Elastic Analysis of a Bi-layer
Orthotropic Conical Shell With a Shear Deformable Bond Layer
Subject to Axisymmetric Surface Loads and Thermal Gradients
1.2.3
SADISTIC -III - Program For Elastic-Perfectly Plastic Analysis
of a Multi -layered Orthotropic Conical Shell Subject to Axisymmetric Temperature and Pressure Loading
1.2.4
SADISTIC-IV - Program For Elastic Analysis of Multi-layered
Orthotropic Shells of Revolution Subject to Axisymmetric Thermal
and Surface Loads with Geometric and Load Discontinuities
1.2.5
SADISTIC-V - Program For Elastic Analysis of Multi-layered
Orthotropic Shells of Revolution Subject to Asymmetric Loading
1.2.6
PEANUTS - Program For Elastic Analysis of Non-Uniform
Thiclmess to Axisymmetric Loads and Thermal Gradients
1.2.7
EPSAD - Program For Elastic/Plastic Analysis of Isotropic
Thin Shells of Revolution Subject to Axisymmetric Loads and
Thermal Gradients
1.2.8
PLOCS - Program For Computing Elastic Stress·es in Thin
Cylindrical Shells Due to Localized Lateral Loads
1.2.9
NCCS - Program For Computing Forces and Displacements in
Simply Supported, Non-Circular Orthotropic Shells Under
Pressure Loading
.
1.2.10
SABOR III - Program For the Linear Elastic Analysis of Thin
Isotropic Shells of Revolution Under Axisymmetric or Asymmetric Loading By Using The Matrix Displacement Method
1.2.11
KALNINS - Program For The Linear Elastic Analysis of MultiLayer Orthotropic Thin Shells Subject to Symmetrical and NonSymmetrical Loads
1.2.12
TSLD-2 - Finite Difference Program For Elastic-Plastic
Analysis of Isotropic Shells of Revolution Subject to Axisymmetric Loads and Thermal Gradients
4-3
Table 4-1 (Cont.)
1.3
1.2. 13
RMONIC - Program For Analysis of Axisymmetric Creep of
Circular CyUndrical Shells in Axial Compression
1.2.14
THORNE - Program For Elastic Plastic Creep Analysis of Shells
of Revolution Under Axisymmetric Loading
1.2. 15
SHA ZAM - Program For Computing Stresses and Displacements
In Ring Stiffened Cylinders Subject to Non-Symmetrical Thermal
Gradients
1.2.16
JOSE - Program For Computing Discontinuity Stresses at The
Juncture of Shell Elements
1.2.17
COIC/P - Program For Computing Influence Coefficients and
Fixed End Forces in Shell Elements For Use With JOSE
1.2.18
BERK - General Shell Analysis- Finite Element Method
Shell Instability Programs
1.3.1
VERSION II - Program to Predict the Elastic Buckling Load of
Cylinders Under Axial Compression Stiffened with Rings and
.Stringers of Rectangular Cross Section
1.3.2
GSAP ~ Program to Predict the Elastic Buckling Load of
. Cylinders Under Axial Compression Stiffened \Vith Rings
and/or Stringers of Arbitrary Cross Section
1.3.3
BOCONE - Program to Predict the Elastic Buckling Load of .
Truncated Ring Supported Cones Subject to External Pressure
,1.3.4
SCAR - Program to Predict the General Elastic Instability Loads
of Ring/Stringer Stiffened Cylinders Subject to Axial Compression
and Lateral Pressure
.
1. 3. 5
BOC - Program to Predict the Buckling Load of Isotropic
. Cylinders Under Axially Varying Radial Pressure
1.3.6
BOOC - Program to Predict the Buckling Load of Orthotropic
Cylinders Under Axially Varying Radial Pressures
1. 3. 7
BOCSIC - Program to Predict the Buckling Loads of a Lamtnated Cylindrical Shell Under a Combination of Axial Load,
Bending Moment and Pressure
Table 4-1 (Cont.)
1.3.8
BOSOR - Program to Predict Buckling of Shells of Revolution
With Various Wall Constructions Subjected to Axisymmetric
Edge Loading, Axial Loading and Pressure
1.3.9
SALTS - Program to Predict the Creep Buckling of a Circular
Cylindrical Shell Under Axial Compression and Internal Pressure
1.3.10 . BARSIN - Program to Predict Classic Elastic Buckling Loads of
Simply Supported Cylinders with Various Wall Constructions
Subject to Internal or External Pressure and Axial Load
1.4
1.5
1.3.11
BENCYL - Program to Predict Elastic Buckling of Simply Supported Cylinders with Various Wall Constructions Subject to
Radial Pressure, Bending Moments and Axial Loads Using
Knockdown Factors For Buckling
1.3.12
INTACT - Program to Provide Interaction Elastic Buckling
Curves for Simply Supported Cylinders of Various Wall Construction Subject to Radial Pressure Bending and Axial Loads
Using Method of .BENCYL Program
Structural Optimization Programs
1.4.1
MARK III - Program for S~ructural Optimization of Axially
Compressed Cylinders Stiffened With Rings/Stringers of
Rectangular Cross Section
1.4.2
OPUS I - Program for Structural Optimization of Axially
Compressed Cylinders Stiffened With Longitudinal Trapezoidal
Corrugations and Rings of Angle Cross Section
1.4.3
MARK IV - Program for Structural Optimization of Cylinders
Stiffened With Rings/Stringers of Rectangular Cross Se.ction
Subject to Axial Compression and Radial Lateral Pressure
1.4.4
OPVWHH - Program for Structural Optimization of Thin Walled
Isotropic Cylindrical Pressure Vessels With Hemispherical
Heads
Solids of Revolution and Thick Shells
1.5.1
WILSON - Finite Element Program For Non-Linear Elastic
Analysis of Orthotropic Solids or Thick Shells Subject to
Arbitrary Thermal, Mechanical and Acceleration Loads
4-5
Table 4-1 (Cont.)
1.5.2
1.6
1.7
RAHFEP - Finite Element Program for Linear Elastic Analysis
of Isotropic Solids or Thick Shells Subject to Applied Loads and
. Thermal Gradients
1.5.3
SINC - Program to Compute Elastic Stresses in Bodies of
Revolution Subject to Axisymmetric Forces and Thermal
Gradients
1.5.4
TKCYL - Program to Compute the Plane Strain Elastic Stresses
in a Thick Walled Cylinder Subject to a Steady State Temperature
Gradient Across the Wall
Miscellaneous Programs
1.6.1
SSRING - Program For the Analysis of Circular Shell Supported
Rings Loaded Normal to its Plane of Curvature
1.6.. 2
CBEAM - Program to Compute the Deflected Shape and Stresses
In A Cantilever Beam with Varying Moment of Inertia
1.6.3
HERRMAN - Finite Element Program for Bending Analysis of a
Plate of Arbitrary Shape and Thiclmess Subject to Non-Uniform
Transverse Loading and Thermal Gradient
1.6~4
COMP2 - Program to Predict Mechanical Properties of Layered
. Continuous Filamentary Composites with Three Method Options
1.6.5
COMP3 - Program to Predict Mechanical Properties of Filament
Wound Composites
1.6.6
TSISS - Program to Compute Elastic Thermal Stresses in a
Series of Simple Structural Shapes Subject to a One-Dimensional
Thermal Gradient
1.6.7
TSOC - Program to Compute the Plane Strain Elastic Thermal
Stress in an Orthbtropic Thick Cylinder Subject to a Radial
Thermal Gradient
1.6.8
TSOLC - Program to Compute the Elastic Thermal Stress in
Layered Orthotropic Thick Cylinder Subject to a Radial Thermal
Gradient and futernal Press ure Loading
Programs for Special Problems in Vehicle Re-Entry
i.7.1
SINC-STRESSES IN NOSE CONES - Program Analyzes a Finite
Length Anisotropic Elastic Solid of Revolution Subjected to Axisymmetric Surface and Thermal Loads
4-6
Table 4-1 (Cont.)
1.7.2
INTERFACE STRESS - Program Computes the Stresses Developed at the Interfaces of Any Three- Layered System Due to an
Applied Shock on the Surface
1.7.3
SWAP-8 - Program Describes Shock Interactions in a Multilayered (up to 100), One-Dimensional System - Materials are
Strain-Rate Dependent and Obey Elastic-Plastic Theory
1.7.4
TWIN LEAP - Program Computes by Finite Differences the
Stresses in a Three-Layered, Semi-Infinite Slab Whose End Has
Been Subjected to an In-Plane Velocity Step
1.7.5
LEAP CODE II - Program Utilizes the Finite Difference Method
for Solving One-Dimensional Shock Wave Propagation Problems
With Prescribed Initial and Boundary Conditions
1.7.6
MA TED LEAP CODE - Program Utilizes the LEAP CODE II to
Analyze a Nose Section as a One-Dimensional, Conical Bar and
is Mated to the TWIN LEAP CODE to Determine the Effects of
Coupling
1.7.7
SHA ZAM - Program Determines the Elastic Stresses Which
Occur in a Ring-Stiffened, Circular, Cylindrical Shell Which is
Exposed to an Arbitrarily Varying Circumferential. Temperature
Field
1.7.8
REBEL I - Program Determines the RMS Values and Power
Spectral Density of Acceleration Response of a M\lltilayered,
Orthotropic, Conical Shell Frustum and an Attached Equipment
Package to an External Random Pressure Field due to Boundary
Layer Turbulance
2. DYNAMIC STRUCTURAL ANALYSIS PROGRAMS
2.1
2.2
Ring Programs
2.1.1
RING BOND - Program for Inelastic Dynamic Analysis of Impulsively Loaded Layered Rings With Finite Bond Strength
2.1.2
GIRLS I - Program for Inelastic Dynamic Analysis of a Dynamically Loaded Three-Layered Ring
Shell Programs
2.2.1
Gill LS II - Program for Inelastic Dynamic Analysis of a ThreeLayered Shell Loaded by an Axisymmetric Time Dependent
Force
4-7
Table 4-1 (Cont.)
2.2.2
NESCO - Program to Compute Interface Stresses in a ThreeLayered System due to an Applied Surface Shock
3. MASS PROPERTIES ANALYSIS PROGRAMS
3.1
3.2
Mass Property Programs
3.1.1
SUBMD - A Variant of the MIMKS Program for Submarine
Mass Distribution
3.1.2
MMASS I - Program to Compute Flight Mass Properties of
Multi-stage Missiles
3.1.3
MM.A.SS II - Program to Compute Mass Properties of Multistage Solid Propellant Missiles as Function of Flight Time
3.1.4
STDDEV - Program to Compute Standard Deviation in Mass
Properties Using the Monte Carlo Technique
3.1.5
SHAPE-MASS - ~rogram to Compute Mass Properties of
Various Shell Elements
Miscellaneous Programs
3.2.1
AXIS TRANSFOR:rvl - Program to Compute Axis Transformation
- - - - - - c - f
TO APPLICABLE
DATA BASE
YES
NO
r----~
TRANSFER
LAST CASE
OUTPUT TO
APPLICATION
DATA BASE
COPY APPLICATION
DATA BASE TO
PERMANENT
DEVICE MAG TAPE,
DATA CELL, ETC.
SET
RESTART WORD
IN
GRAPHIC
DATA BASE
CLOSE
APPLICATION
DATA BASE
FILE
COpy
GRAPHIC DATA
BASE ONTO
RESTART DEVICE
DISPLAY
AN
ENDING
PICTURE
CLOSE
GRAPHIC
DATA
BASE
CLOSE
ALL
FILES,
ETC.
Fig. 4-18 Stop Program
4-47
The next case for the same program will be initialized by selecting a next case
command (NXCASE). When the NXCASE command is selected, output data from the
previous case is placed in the application data base and control is passed to the
program selection program. Program logic is shown in Figs. 4-19 and 4-20.
4.5.4 Input Specification
Input data is requested by selecting the input command. When input is selected, a
list of input data requirements for the program whose ID appears in the program ID
register will be displayed on the CRT in the form of commands. Included in the list
will be a command to load all data. The user will select from the input list one
item at a time in any order. Of course, if one item depends upon another the independent one must be selected first; if not a mes sage is displayed. After the item is selected
the user must designate the input mode: card, tape, etc. The mode can be any of
those that appear in Table 4-2 except for command ALL DATA for which CRT input
does not apply. Input data flow for each mode of input is shown in Figs. 4-21 and
4-22.
To effect data input, a set of routines will be required. One will be the main input
program (Fig. 4-23) that will check status of the input requirements, display all input
requirements not yet satisfied, and identify the command selected .. This program
will be executed each time data input is required.
Each input mode will require· a separate routine and can operate independently of the
main input program. These programs must load in all tables from auxiiiary storage,
determine the required formats to be used in reading in the data, remove selected
items from the input list after the data has been read in, and place the data in an
output buffer with the proper labels to identify the data. When the data has been loaded
into the output buffer an output program is automatically entered to display and print
the data. Output is discussed in Para. 4-7.
It has been assumed that all application programs will standardize card and tape
input formats, consequently only one input program for each input mode is required.
4-48
GET PROGRAM
J. D. FROM PICK
TABLE & DISPLAY
IN PROGRAM J. D.
REGISTER
MESSAGE:
PROGRAM
SELECTED
READ CASE
1. D. FROM 1-_ _...-1
LABEL
REGISTER
YES
INSTRUCTION:
t
PUI~ i~~E D. ..-______...1
REGISTER
MESSAGE:
RESELECT
PROGRAM
YES
READ IN CASE 1. D.
TABLE FOR
CORRESPONDING
PROGRAM J.D.
MESSAGE: CASE
ALREADY EXISTS.
nJ'-~~ INSTH: RENAME
OR CONTINUE
INITIALIZE
FOR
NEXT
CASE
LOAD IN NEXT
SEGMENT OF
CASE J. D.
TABLE
INSTR: PROGRAM
READY. SELECT~________________~~
NEXT
COMMAND.
MESSAGE: NO ROOM IN GRAPHIC DATA
~SiE ~~~~R~N ~~:~I~ ~~:A TB~~t~~~-------------------------------~
REMOVING EXISTING CASES
'--.::..;;.:.:..;;.;;...-
Fig. 4-19 Program Selection Program
4-49
ENTER
PUT PROGRAM
NAME
IN .PE REGISTER
WAS OUTPUT DATA
GENERA TED FOR
THIS CASE? REWIND
IOF* 'AND READ PROGRAM ID
NO
READ
CASE ID
MESSAGE: NO
OUTPUT DATA
FOR CASE.
CONTINUING.
NO
GET DATA FROM
IOF AND STORE
IN DATA BASE.
(DIFFERS FOR
EACH APPLICATION '
PROGRAM)
CALL PROGRAM
SELECTION.
RELINQUISH CONTROL
NOTE: This program is similar to Program Library.
*IOF
= Intermediate output file
Fig. 4-20 Program Next Case
4-50
PROGRAM
DATA BASE
(DISK, DATA
CELL, DRUM
ETC. )
CARD
READER
SEPARATE
INPUT ROUTINE
FOR EACH
DEVICE
OUTPUT
BUFFER
OUTPUT
ROUTINE
ON-LINE
PRINTER
Fig. 4-21 Input-Output Flow for Card, Magnetic Tape and Data Base
4-51
GRAPHIC
DATA
BASE
REDISPLAY
IN
INTERPRET LINE
DRA WINGS AND
TEXT. TRANSFORM
IF NECESSARY
AND PROPERLY LABEL
THEM IN GRAPHIC DATA
BASE.
Fig. 4-22 Input Data Flow for Cathode- Ray T~be·
4-52
IDENTIFY INPUT
ITEM SE LECTED.
ENCLOSE COMMAND
IN BOX.
YES
POSITION
WINDOW
TO INPUT
WINDOW
NO
ERASE CRT
AND PREPARE
FOR INPUT
DISPLAY
NO
DISPLAY INPUT
REQUIREMENTS
THAT HAVE
NOT BEEN
SATISFIED
INSTR: SELECT
INPUT
ITEM
Fig. 4-23 Input Program
4-53
Otherwise each application program will require separate input routines. Separate
input routines might be preferred so existing application programs can retain their
familiar formats for card input.
The following "standard" formats are assumed. Card formats will be 6E12. 8 for
floating data, 6112 for integer data, and 12A6 (18A4) for Hollerith data. The latter
depends on computer word length. Tape records are card images. Data arrays must
be preceded by a card containing the number in integer format of single items, pairs,
or triplets in the array. The format to be used for any data item will be determined
by the input routine.
Program flow for card input, tape input, data base input, and CRT input. are shown in
Figs. 4-24, 4-25, 4-26, and 4-27, respectively.
4. 5. 5 Data Modification Specification
Data does not assume meaning to the numerical program until computations are
requested. Until then, data is nothing but groups of lines and text on the CRT with
unique names. Consequently, modification of data is limited primarily by system
hardware and software.
Actual manipulation of lines and BCD text will be performed by basic system software
and thereby limited by it. It is assumed that the graphic hardware and software system being used will have adequate line drawing capabilities and ability to move lines
and text or replace said line and text.
Performing data manipulations using only system hardware and software capabilities
could impose limitations. Some of the drawbacks might be the necessity of imposing
various restrictions on modifiable data or having a piece of data displayed but not
modifiable. Also, unless the system used is highly sophisticated, extra work by the
user is often required for simple modifications like changing one value on a function
curve. To overcome these drawbacks would require another command (MODATA) for
modifying data. This was successfully done in Ref. 4-1 but the program was written
4-54
r-;LATE
MESSAGE:
IU:ADING
DATA CAHDS.
II
~TIIlS
LOGIC APPEAH;; ALL
LOAD IN DATA
HEQUInEMENT
TABLE.
CHECK INPUT
DEVICE STATUS.
I
INPUT~DES.
-- -- -- -- --
GET THE VARIABLE
OUT OF PICK
TABLE FOn
VAlUABLE TO
BE INPUT
EHASE
COMMAND
IN BOX
L _______________
GET THE
NECESSARY
INDICATons
OUT OF THE DATA
HEQumEMENT
TABLE (DHT)
--I
HEAD IN ITEM
ACCOnDING
TO INDICA Tons
AND PUT IN OUTPUT BUFFEH
I
I
I
~
SET
VARIABLE
NAME
NO
SET UP LOOP
Fon ALL DATA
INPUT
1---------------- ---I
I
I
I
I
I
NOTE: DATA IS
NOT STonED
ON AUX STOllAGE
UNTIL A SAVE
COMMAND, LIBRAHY
COMMAND, on NEXT CASE
IS llEQUESTED
POSITION
WINDOW TO
INPUT
FRAME
I
I
I
MESSAGE:
SELECT
NEXT
COMMAND
I
I
I
I
I
I
I
I
I
_______ J
NOTE: THIS pnOGRAM HEADS CARDS FROM THE ON-LINE CARD READER IN A STANDARD FORMAT
smTABLE FOR THE ITEM TO BE READ.
Fig. 4-24 Card Input Program
4-55
GET
NECESSARY
INDICATORS
OUT OF
DRT
YES
SET UP
LOOP
CONTROL
FOR ALL
DATA INPUT
PUT PROGRAM
LD. IN RECORD
WORD.
SET COUNT =0
SET FLAG = 0
REWIND
INPUT
TAPE
READ
NEXT
RECORD
YES
>-----~
..YES
INCREMENT
COUNT
MESSAGE:
PROGRAM LD.
CANNOT BE FOUND
ON INPUT TAPE.
RESELECTINPUT
MODE
YES
t--------------.
REREAD RECORD
WITH PROPER
FORMAT. (CONVERT).
READ IN ITEM
ACCORDING TO .
INDICATORS AND
PUT IN OUTPUT
BUFFER
SET RECORD
WORD TO
NEXT VARIABLE
NAME
PUT VARIABLE NAME
IN RECORD WORD
SET COUNT = 0
SET FLAG = 1.
Fig. 4'"-25 Tape Inp'ut Program
4-56
MESSAGE:
J.---~ DATA BASE
INPUT
BUSY
SET UP
LOOP
CONTROL
NO
GET
INDICATORS .--_ _ _ _ _-'
OUT OF
DRT
READ IN AND
CO:r-.rvERT FROM
GRAPHIC FORMAT TO
ARRAY" AND PUT 1 - 4 - - - - - ,
IN PLOT BUFFER
SET
ITEM
NAl\1:E
Fig. 4-26 Data Base Input Program
4-57
MESSAGE:
CRT
INPUT
NO
NO
LOAD PLOT
B UF F ER
WITH DATA
1-------'-.1
CLEAR DISPLAY
AND
PREPARE
FOR
INPUT
YES
POSITION
WINDOW
TO
PROPER
WINDOW
FOR
DATA ITEM
INTERPRET LINES
LABEL DATA
AND TEXT IN
I--___~AND REPLOT
DISPLAY AND
IF
ABSTRACT THE
NECESSARY
·INPUT DATA
MESSAGE:
SELECT
NEXT
COMMAND
Fig. 4-27 CRT Input Program·
4-58
for specific data and data formats. Many peculiarities of the data were accounted for
and required considerable time to develop. It ~s felt that a general data manipulation
routine could be developed if the data structure were specified.
Precise coordinate input and rescaling of graphs have both been successfully implemented
in Ref 4-1. However, in this case, the data structure was designed specifically for
the two programs' which limited the utility of the coordinate input program and the
rescaling program for other data structures. It is felt that a general coordinate
program and rescale program could be developed for another data structure if adequate
data structure specifications were available.
4. 5. 6 Computation Specification
The computation program is entered by
sele~ting
the compute command (COMPUTE).
When selected, the application data. base is checked to make sure that the case ID
appearing in the Case ID Register has not been computed previously. If not, all input
data are obtained from the graphic data base and input requirements are checked for
completeness. If input requirements are not -satisfied a message is pisplayed and,
.
"
the input program is automatically entered at which time a list of input requirements
that have not been satisfied is displayed.
If the input requirements are satisfied the application data base is checked to see if
input data already exists for the case. If data already exists it is checked to see if it
is referenced by other programs or other cases. If so, a message is displayed and
computation aborted. If no other referen'ces are made, the data in the application
data base is deleted and the data from the graphic data base is placed in the application data base. An override button can be set by the user which allows him to perform
the computations using the data in the application data base rather than that in the
graphic data base.
If the graphic system is operating as a satellite computer, the data is formatted for
transmission to the central computer and all necessary controls are set. Otherwise,
the application program is entered directly and control turned over to it.
4-59
Computed ·items are stored on an intermediate scratch file with each item identified.
This file is opened at sign-on and has a program ID and case ID for a file label. The
program ID and a blank case ID are written each time a program is selected from the
library. The actual case ID is not written until data is ready to be written on the file.
The case ID will then indicate if data has been written onto the file or not. This scheme
would be compatible with a satellite-type system where the intermediate scratch unit
would be an output device for the central computer. This program is shown in
Fig. 4-28.
4. 5. 7 Output Specification
Computed quantities are not automatically. displayed when computed. If this were done
I
the flexibility of graphics would be greatly restricted. The 'chances are that all output
quantities would not be displayed at once so
~
standard set of output would, have to be
preprogrammed, limiting the users access to computed items.
Output is requested by, selecting the output command (OUTPUT). A list of possible
items is
disp~ayed
and the user is given a choice of standard output 9r selecting a
specific item. Included in the list is an option to void output for current case. If the
standard set is chosen, an output routine is automatically entered and the data is
immediately displayed. One item at a time can be selected which automatically'
enters a general output program which displays the selected item. If the user elects
to void the output another program is entered which removes the data.
Computed quantities are not automatically placed into the application data base
This
allows the user to iterate on the input data until satisfactory results are obtained.
Once satisfactory results have been achieved, the user has a complete case and can
proceed to the next case by selecting the "next case" command or return to the
library by selecting the "library" command. If either the next case or library is
selected the quantities just computed are transferred from the intermediate file to
the application data base with proper identification.
4-60
MESSAGE:
COMPUTATION
PROGRAM
IN
PROCESS
PUT
PROGRAM
NAME IN
PE REGISTER
MESSAGE:
DATA
ITEM
MISSING
CONVERT ITEM
AND STORE ON
INTERMEDIATE
SCRATCH.
CAN BE RESIDENT
IF ROOM
CALL
APPLICATION
PROGRAM.
NO RETURN
NO
GET INPUT
AND OUTPUT
REQUffiEMENTS
FOR
PROGRAM
,
TRANSFER
DATA TO
APPLICATION
DATA BASE
FORMAT
DATA
FOR
TRANSMISSION
CHECK APPLICATION
DATA BASE TO SEE
IF
COMPUTATION HAS
BEEN DONE
PREVIOUSLY
SET
NECESSARY
INDICATORS
YES
GET DATA
FROM
~"::":'_-.lAPPLICATION
1----<
DATA
BASE
MESSAGE: THIS
COMPUTATION WAS
DONE PREVIOUSLY
FOR CASE
REQUESTED.
GO TO NEXT CASE.
MESSAGE: OTHER
REFERENCES TO
DATA COMPUTATION
ABORTED.
INSTR: USE
ANOTHER CASE
LABEL
Fig. 4-28 Compute P:-ogram
4-61
Program flow for standard output, general output, and voiding data is shown in
Figs. 4-29, 4-30, and 4-31, respectively. The next case and library command were
discussed previously.
4.5.8 CRT Layout Specification
CRT layout specifications will pertain only to the communication area. The data
display area is rather arbitrary so specifications are not attempted. A table of all
commands required is shown in Table 4-3 with each command cross referenced with
the specification section where the command is described. For illustrative purposes
eight character mnemonics are used.
Those commands with an asterisk appearing in the subsection column were not required
but are briefly mentioned in the
specificatio~.
They are included in the cqmmand
requirement table for completeness. The location of each
com~and
on the CRT is
shown in Fig. 4-32 with only the specified ones shown.
A table of registers required is shown in Table 4-4 with each
regist~r
referenced to the
specification section where it is described. Several registers appear in the table but
are not referenced to any section. These contain an asterisk in the subsection column
and are included for completeness. Some of these will be used with some of the
commands in the command table that also have an asterisk in the subsection field.
;''4-62
MESSAGE:
STANDAHD
OUTPUT
IN
PHOCESS
YES
POSITION
FHAl\1E
NO
EHASE DISPLAY
AND PHEPAHE
TO
DISPLAY
OUTPUT
YES
NOTE: SINGLE ITEM
DISPLAYS ARE NOT
REMOVED IF
PREVIOUSLY EXISTED.
TIllS ALLOWS USEH TO HETAIN
TWO OUTPUTS FOH SAME
CASE. THERE IS A
POSSIBILITY FOR CONFUSION
BUT THE OPERATOR SHOU LD
KNOW WHAT'S WHAT.
REMOVE
STANDARD OUTPUT
DATA PHEVIOUSLY
DlSPLA YED FOH
THIS CASE. IT IS
POSSIBLE THAT
EHASING A GRAPHIC
REMOVES IT FROM
THE GRAPH
DATA BASE.
,
I
I
I
I
_ _ _ _ _ _ _ --l
DISPLA Y STANDARD
OUTPUT SET.
GET DATA FROM
INTERMEDIATE OUTPUT
FILE.
MESSAGE:
OUTPUT
COMPLETED
Fig. 4-29 Standard Output Program
4-63
MESSAGE:
GENERAL
OUTPUT
IN
PROCESS
PUT
PROGRAM
NAME IN
PROGRAM
EXECUTE (PE)
REGISTER
NO
ERASE DISPLAY
AND PREPARE
TO DISPLAY
OUTPUT
REMOVE OUTPUT
ITEM PREVIOUSLY
DISPLAYED FOR
THIS CASE
GET
VARIABLE
NAME OUT
OF PICK
TABLE
GET DATA
FROM IOF*
DISPLAY
DATA
ITEM
MESSAGE:
OUTPUT
COMPUTED
*IOF - INTERMEDIATE OUTPUT
Fig. 4-30 General Output Program
4-64
·
PUT VOID INTO
PE
REGISTER
ENTER
YES
SET THE CASE
LABEL BLANK
ON THE INTERMEDIATE
OUTPUT FILE
FIND EACH OUTPUT
DAT A ITEM AND
YE S
REMOVE FROM DATA
BASE PROVIDED IT
IS NOT HEFERENCED
BY ANOTHER
PROGRAM OR CASE
REMOVE
FROM
GRAPHIC
DATA
BASE
NO
MESSAGE:
OUTPUT
DATA
VOIDED
INSTR:
MAKE
NEXT
SELECTION
WAIT
STATE
Fig. 4-31 Void Output Data Program
4-65
Table 4:-3
COMMAND TABLE
Command
MNEMONIC
Brief Description
Library
LIBRARY
Displays a list of executable
application programs
2.2
Input
INPUT
Displays a list of input
requirements
2.4
Modify Data
MODATA
Activates special data
manipulation routine
*2.5
Compute
COMPUTE
Activates the computation
program
2.6
Check Input
CKINPT
Checks input requirements
for completeness
*2.6
Check Input
and Save
CKNSAV
Checks input requirements and
puts data into data base
*2.6
Output
OUTPUT
Displays a list of
computed output quantities
2.7
Next Case
NXCfiSE
Saves current case output
and prepares for next case
2.3
Stop
STOP
Transfers application data base
to permanent storage device
2.2
Stop and Save
STOPNSAV
Same as STOP, plus it saves
the graphic data base
2.2
Display
Plot
X-Y Input
Subsection
2.5
PLOT
Used to plot individual data
items on request
XY-INPUT
Used in conjuction with X, Y, Z,
and T registers to enter precise
values on a graph
4-66
2.5
*2.5
Table
4~3
(Cont.)
Command
MNEMONIC
Brief Description
Rescale
RESCALE
Used in conjunction with the
X and Y registers to rescale
graphical displays
*2.5
Interrupt
INRUPT
Used to stop numerical
computations at selected points
*2.6
Stop and Go
STOPNGO
Used to terminate computation
or to continue after the INR UPT
command is selected
*2.6
Subsection
* These commands were not included in the requirements but are included in the
table for completeness.
'4-67
MR
INST
PROG. ID
~.~.
__
~
CASE ID
________________
~
LIBRARY
INPUT
STOP
COMPUTE
STOPNSAV.
OUTPUT
PLOT
NXCASE
DISPLAY
LABEL
Fig. 4-32 Cathode-Ray Tube Layout
4-68
Table 4-4
REGISTER TABLE
Section
Register
Mnemonic
Brief Description
PROGRAM ID
PROG' b ID
Displays name of application
program being executed
2.3
CASE ID
CASE ID
b
Contains Case ID
supplied· by user
2.3
MESSAGE
MR
Messages to the operator are
displayed in the MR
2.8
INSTRUCTION
INST
Instructions for the operator
are displayed in the INST
2.8
X-INPUT
X
Enter floating pt number
Y-INPUT
Y
Enter floating pt number
Z-INPUT
Z
Enter floating pt number
TIME
T
Enter floating pt number
LABEL
LABEL
Enter any BCD text
PROGRAM
EXECUTION
PE
Contains the name of the subroutine
on program being executed
t
Appears in various program flow diagrams.
4-69
*
*
*
*
2.3
t
4.6 PEANUTS: AN EXAMPLE OF GRAPHIC PROGRAM REQUIREMENTS AND
SPECIFICATIONS
A comprehensive list of active structural programs (Table 4-1) was presented in the
background section. One of these programs, PEANUTS, is a popular structural program used for performing nonroutine structural design tasks. For this reason PEANUTS was chosen 3:s the structual design program to be conceptually incorporated into
a computer-aided design environment in which man and machine communicate via a
cathode ray tube utilizing some mechanism like a light pen.
The more tractable structural programs to be incorporated into computer graphics
environment are separable into three functions: input, computation and output as shown
in Fig. 4-33. A functional block diagram of PEANUTS as it is used in production is
shown in Fig. 4-34. A close look at PEANUTS (Fig. 4-34) reveals that it is not easily
separable into the ideal function structure. However, it is still well suited from the
logical view point for a graphic environment.
Program requirements and specifications for the PEANUTS program will be presented
here in three sections: Input, Computation, Output.
4.6.1 PEANUTS Example: Input Requirements and Specifications
The PEANUTS Program input requirements are nominal but possible options are numerous. A complete list is shown in Table 4-5. Two types of data are require.d:· tabular
data and single items.. Tabular data is representable as two-dimensional curves which
are normally input in array form.
These arrays define geometry, surface loads, etc.
Single items represent coefficients of various kinds and various boundary conditions.
4-70
START
COMPUTE
STOP
Fig. 4-33 Functional Block Diagram of an Ideal Structural Program
4-71
.
2
1
INPUT:
1. GEOMETRY
2. MATERIAL
PROPERTIES
3. PRINTOUT INPUT
SOLVE
CX=F
FOR DISPLACEMENT
VECTOR X
1. COMPUTE ELEMENTS
COMPUTE STRESSES
AND PRINT THEM
OUT
FOR STIFFNESS
MATRIX C
2. FACTOR C
INPUT:
1. BOUNDARY CONDITIONS
2. LOADS
3. PRINTOUT INPUT
WRITE DATA TO BE
PLOTTED ONTO SCRATCH
FILE
NO
PLOT
ALL
CASES
)YES
Fig. 4-34
Function91~lock
.4-72
Diagram of, PEANUTS
Table 4-5
DESCRIPTION OF INPUT QUANTITIES
Quantity
Description
RECORD
Seventy two characters of descriptive Hollerith information (arbitrary)
to be printed in output headings and on plots
M
Problem number (optional, will be printed in title line)
SS1
Value of s, the independent variable (surface coordinate) at the
"starting" boundary (ignore if IGEOM = 2, 3, or 4)
SS2
Value of s at the terminal boundary B (ignore if IGEOM
NE
Number of "stations" for which the solution is to be obtained (150 maximum) . These will be equally spaced between the boundaries defined
by SS1 and SS2, and need not coincide with the statiol).s used to define
geometry.
IGEOM
An idicator pertaining to geometry input
=
2, 3, or 4)
If
IGEOM = 0, a subroutine containing analytical expressions
applicable to a variable thickness sphere or cylinder
is used.
If
IGEOM = 1, values of geometric quantities are read in point by
point in s' using a matrix format for fifty points
maximum. This permits generation of a general
shell of revolution.
If
IGEOM = 2, specialized geometric data for a toroidal shell of
nonuniform thiclmess and circular section are read.
If
IGEOM = 3, specialized geometric data for a toroidal shell are
read as above.
If
IGEOM = 4, specialized geometric data for an ellipsoidal shell
element are read.
NRITE
The increment in "stations" desired between printout of computed
quantities for the shell interior region.
HN
Total number of input stations for which geometric data will be read
in matrix form .
E
PR
. Value of Young's modulus for the shell material
Poisson's ratio
4-73
Table 4-5 (Cont.)
Quantity
Description
ALP
Coefficient of thermal expansion for the shell material
R1 (a)
Meridional radius of curvature
R2
Circumferential radius of curvature
NA
Number of input stations for which geometric quantities are to be
computed in the region between Sl and S2 below
NB
Ditto for the region between S2 and S3 below
NC
Ditto for the region between S3 and S4 below
so
Value of the shell surface coordinate s defining boundary A of the
geometry
Sl
Value of s defining intermediate point 1 at which the equations for
thickness variation change
S2
Same for intermediate point 2
S3
Same for intermediate point 3
S4
Same for intermediate point 4
S5
Same for intermediate point 5
S6
Value of s for "terminal" point of geometric input data
B6
Derivative of thickness with respect to s for the region between
S5 and S6
HAO
Coefficient in equation for thickness variation
HAl
Same
HA2
Same
HB1
Same
HB2
Same
/
(a) Quantities R1 through EXC2 pertain to the computations for geometry used if
IGEOM = O. If IGEOM = 2 or 4 some of these quantities are used.
4-74
Table 4-5 (Cont.)
Quantity
Description
HC1
Coefficient in equation for thickness variation
HC2
Same
EXA1
Exponent in equation for thickness variation
EXA2
Same
EXB1
Same
EXB2
Same
EXC1
Same
EXC2 (a)
Same
SI(IIN, 1)
A matrix (column vector) of values of s , the independent variable,
for which input geometry is to be read. (50, 1) maximum dimensions
RI(IIN, 1)
Values (or R , the shell radius normal to the axis of revolution
corresponding to s above
RII(lIN, 1)
Values for R1, the meridional radius of curvature
R21(IlN, 1)
Values for R2, the circumferential radius of curvature
HI (lIN ,1)
Values for the shell thiclmess
RA (b)
Radius of revolution of toroidal shell or the equatorial radius of the
ellipsoid
RB
Radius of convolution of toroidal shell or the polar radius of the
ellipsoid
THETA1
Angle (radians) defining the "starting" boundary A of the shell
sector
THETA2 (b)
Angle (radians) defining the terminal boundary B of the shell sector
(a) Quantities R1 through EXC2 pertain to the computations for geometry used if
, IGEOM = o. If IGEOM = 2 or 4 some of these quantities are used.
(b) Quantities RA through THETA2 pertain to the toroidal or ellipsoidal shell
geometry used if IGEOM = 2, 3, or 4.
4-75
Table 4-5 (Cont.)
Quantity
Description
H
Thickness of toroidal shell (used only if IGEON = 3)
TII(IIN, 1)
Values of outer fiber temperature relative to "no thermal stress" or
reference temperature. If IGEOM = 1, the values (if not constant)
must correspond spatially to geometry input stations specified through
the SI matrix. If IGEOM = 0, 2, 3, or 4, the values will be assumed
to correspond to stations equally spaced between the two boundaries
of the shell element.
T2I(IIN, 1)
Input values for inner fiber temperature (Comments for TIl apply)
PZI
(IIN, 1) (a)
Input values for normal pressure;· internal pressure is positive
(Comments for TIl apply)
PSI(IIN, 1)
Input values for surface "shear" positive with increasing s
(Comments for TIl apply)
lSOL
An indicator pertaining to the solution desired
MAXP
If
ISOL = 0,
the program will solve a two-boundary problem
using input influence coefficients for the adjacent
structure at each boundary.
If
ISOL = 1,
the program uses· force input for boundary A, and
adjacent structure influence coefficients for
boundary B.
If
ISOL = 2,
the program computes influence coefficients and
particular solution boundary values only.
If
ISOL = 3,
the program reads force boundary conditions as
input data for both boundaries and solves the corresponding problem.
If
ISOL = 4,
the program solves the axially redundant twoboundary problem using influence coefficients for
adjacent elements.
An indicator pertaining to printout
If
MAXP = 0 ,
printing is restricted to the minimum consistent
with the information sought.
If
MAXP = 1,
complete printout (equation checks, complete
influence functions, etc.) will result.
(a)As a convenience, the program will interpret a single value provided as input for
PZI(l, 1) as a constant pressure acting over the entire shell.
4-76
Table 4-5 (Cont.)
Description
Quantity
Component of stress resultant
parallel to axis of revolution
FVA2
FHA2
Component of stress resultant
normal to axis of revolution
MXA2
Resultant meridional moment
FVB2
FHB2
MXB2
l
I
Analogous to FV A2, etc., above. Boundary conditions to be applied
to shell element 2 at· juncture B.
External line load parallel to
axis of revolution
FVAE
To be applied to shell element 2
as boundary conditions at A for
either the "particular" or final
solution as prescribed by ISOL.
Acting at juncture of shell
elements 1 and 2 (juncture A).
These forces are incorporated
only through the redundant
analysis.
RAE
External line load normal to
axis of revolution
MAE
External meridional line
moment
FVBE
HBE
MBE
Analogous to FVAE, etc., above. External line loads acting at the
juncture of shell elements 2 and 3 (juncture B).
1
CVV
The "reference" value of axial displacement at boundary A
(juncture A) used to compute the "absolute" displacements
NPLOT
An indicator for plotting
For NPLOT = 0, no plots are to be made.
For 1
NFLAG
NEWP
~
NPLOT
~
30, plots will be made as indicated in Table 4-6.
An indicator pertaining to plotting problems in sequence
For NFLAG
= 0,
the current problem is the last or only one to be
plotted.
For NFLAG
= 1,
other problems will be plotted later.
An indicator pertaining to usage of geometric data
For NEWP = 0, the next problem will use the current geometry with
a new force vector as prescribed through ISOL below.
For NEWP = 1, new geometric data will be read for the next problem.
4-77
Table 4-5 (Cont.)
. Quantity
All
Al2
A22
Al3 (a)
A23(a)
A33(a)
CII
Cl2
C22
CI3(a)
C23(a)
C33(a)
Description
Influence coefficients for the adjacent structure (shell element 1) at
juncture A. Symmetry is used to compute values for elements not
specified as input data. For axially determinant cases, ignore
starred quantities (leave blank on data card).
Influence coefficients for the adjacent structure (shell element 3) at
juncture B. Comments above apply.
VIAP
Rotation
WIAP
Displacement normal to axis
MIAP
Moment
HlAP
Component of stress resultant
normal to axis
DVIAP(a)
Axial displacement
FVIAP(a)
Axial component of
stress resultant
V3BP
W3BP
M3BP
H3BP
DV3BP(a)
FV3BP(a)
Corresponding to the "particular"
solution characterizing the
adjacent structure (shell
element 1) at juncture A.
Analogous to above but for adjacent structure (shell element 3) at
juncture B
(a)These quantities are ignored for an axially determinant solution (leave blank on
data card).
4-78
Table 4-6
LIST OF POSSIBLE OUTPUT FROM PEANUTS
NPLOT
Function
Name
Description
1
R
Shell thiclmess
2
SIGXl
External meridional stress
3
SIGX2
Internal meridional stress
4
SIGTI
External circumferential stress
5
SIGT2
Internal circumferential stress
6
SIG!
External "effective" stress
7
SIG2
Internal "effective" stress
8
NX
Meridional membrane stress resultant
9
NT
Circumferential membrane stress resultant
10
MX
Meridional moment
11
MT
Circumferential moment
12
Q
Shear
13
FR
Net stress resultant normal to axis of revolution
14
FV
Net stress resultant parallel to axis of revolution
15
DELTA
Deflection normal to axis of revolution
16
V
Rotation (dependent variable)
17
EXZl
External surface meridional strain
18
EXZ2
Internal surface meridional strain
19
ETZI
External surface circumferential strain
20
ETZ2
Internal f3urface circumferential strain
2!
DR
dR/ds (s is the independent variable, the
meridional distance along the shell reference
surface. )
22
R
Radius normal to the axis of revolution
23
DR
dR/ds
24
Rl
Meridional radius of curvature
25
R2
Circumferential radius of curvature
26
U
Dependent variable (QR2)
4-79
Table 4-6 (Cont.)
NPLOT
Function
Name
Description
27
DU
dU/ds
28
DDU
2
2
d U/ds
29
DV
30
DDV
dV/ds
2
2
d V/ds
To accommodate numerous input as described above using a CRT requires two
different displays as shown in Figs. 4-35 and 4-36. Only the data display area is
shown on these illustrations. The input has been separated into function (tabular) and
single item input. Each occupies separate displays. Item input is divided into
Material Properties, Boundary Conditions, and Influence Coefficients.
Input is achieved by .selecting an input command (INPUT) which displays two options:
Functions and Single Items. Selecting function or single item results in displaying
Fig. 4-35 and Fig. 4-36, respectively.
In function input the geometry should be specified first. The necessa:ry
paramete~s
are input via registers. Thickness, pressure, shear and temperature can be input in
any order and. in any of the desired modes, card, CRT, etc.
Single items are input in groups of material properties, influence coefficients, boundary
conditions, and particular solution. Each group is input by selecting the appropriate
juncture A or B and a group label, e. g., Boundary Conditions. A juncture specification
is not required for material. This will cause the input options to be displayed. Any
one of the options can be selected. However, the option selected pertains to the whole
group, e.g., if CARD input were selected for juncture A Boundary Condition; FVA,
FHA, ... , CVV would be read from cards and displayed on the CRT. The order in
which the groups are selected is arbitrary as long as they are consistent with
PEANUTS input for the particular problem to be solved. An example of input requirements necessary to design the transition region between a thick and thin region on a
4-80
GEOMETRY
CYLINDER:
TOROUS:
SPHERE:
SEC. ELlPSOID:
LOrrn
rn
ANALYTIC
CARD.
CRT
PREVIOUS CASE
~
~
C,,)
H
~
~
I
0
0
MERIDIONAL (IN.)
1.0
1.0r-~
CARD
CRT
PREVIOUS CASE
~
::>
rn
rn
~
~
~
0
I
0
1.0
MERIDIONAL (IN.)
1.0r~
CARD
CRT
PREVIOUS CASE
<
~
~
rn
I
0
0
1.0
MERIDIONAL (IN.)
~
p:;
LOr--
::>
CARD
CRT
PREVIOUS CASE
~
~
~
~
~
~
~
0
I
0
MERIDIONAL (IN.)
Fig. 4-35 Function Input
4-81
1.0
INPUT OPTIONS
MA TERIAL PROP.
E:
CARD
CRT
PREVIOUS CASE
PR:
ALP:
JUNCTURE A
B
BOUNDARY FORCES
FVA:
FHA:
MXA:
FVAE:
HAE:
MAE:
CVV:
FVB:
FHB:
MXB:
FVBE:
HBE:
MBE:
INFLUENCE COEFFICIENTS
All:
A12:
A22:
A23:
A33:
CII:
C12:
C22:
. C23:
C33:
PARTICULAR SOLUTION
VIAP:
WIAP:
MIAP:
HIAP:
DVIA:
FVIA:
V3BP:
W3BP:
M3BP:
H3BP:
DV3B:
FV3B:.
NOTE: A signifies a register.
Fig. 4-36 Single Item Input
4-82
shell subjected to hydrostatic pressure was presented in Para. 4.1. These will be
repeated here for clarity. Geometry is to be a sphere having variable thiclmess. The
thiclmess variation is to be input via CRT. The surface is subjected to hydrostatic
pressure and force boun.dary conditions are applied. Both are described via CRT.
Material properties are also input via CRT. This constitutes all input requirements.
An example step-by-step procedure for a computer-aided design environment is shown
below:
step
Procedure
1.
Select INPUT COMMAND.
2.
Select FUNCTION, Input OPTION.
3.
Input Radius of Sphere, Initial Start, and Stop value in appropriate
registers.
4.
Select SPHERE.
5.
INPUT graph ordinates for thiclmess into appropriate registers.
6.
Select CRT input for Thickness function.
7.
Draw in thickness variation.
8.
Repeat steps 5, 6, and 7 for pressure, shear, and temperature. The latter
two are set to zero in this ~ample.
9.
Select Input command.
10.
Select Single Item Input option.
11.
Select Material Property.
12.
Select CRT input mode.
13.
Select E and input value.
14.
Select PR and input value.
15.
Select ALP and input value.
16.
Select Boundary Forces.
17 .
Select CRT input mode.
18.
Select FVA and input value.
19.
Repeat step 18 for FHA, MXA, ..• , CVV and FVB, FHB, ... MBE. The
juncture does not have to be designated for CRT input since each individual
register is selected.
20.
Select Influence Coefficients.
21.
Repeat steps 17 to 19 but for All to C33.
4-83
Particular solutions for "adjacent structures" was not required as input for this
example. All input has now been defIned. The next step is to perform the computation.
This will be discussed in the following paragraph.
4.6.2 PEANUTS Example: Computation Requirements and Specifications
PEANUTS is a powerful computation tool which employs a finite difference scheme in
its solution procedure. In essence, PEANUTS solves the matrix equation
(C) (X)
= (F)
for X, where C is a banded stiffness matrix, X 1s a displacement vector, and' F a
loading vector. Currently the program can handle 150 grid points with a bandwidth of.
17 elements. Two variables are specified 'at e,ach point requiring solution of 300
simultaneous equations.
The number of grid point~ used in the solution procedure can be specified as input.
It is customary to use all 150. However, far fewer could handle the job adequately in
most cases. The solution procedure uses real numbers with single precision accuracy ~
To incorporate PEANUTS into a man-machine interactive type environment requires
minor changes to the PEANUTS program. The production program; was depIcted in
Fig. 4-34 and' will retain its present character. Howeve'r, mput data flow will have
to be routed through a graphic application program for display and modification purposes and then interpreted by an interface program which converts the data iiito a for- " .
mat acceptable to PEANUTS.
" ."
For simplicity, the interface program will retrieve the 'data from the' Display Console, '
convert it'to ;formatted data in:the form 'PEANUTS nornlally expec~'~,,' aIi~ 'write the
data onto an input'file which is read by PEANUTS as if it were a standard,input file. .
PEANUTS then operates in a normal fashion. Stacked cases must be handled.individu-'"
ally. The capability of iterating on the
bou~~.ary condition~
4-84~
.
and
loa~:
without modifying :~.
the geometry can be handled by proper positioning of the input file. Each case will be
displayed and not stacked as is presently done. Output is discussed in the next
paragraph.
Computations will be requested by selecting a compute command (COMPUTE). The
compute command will have two options: new problem and change loads.
4.6.3 PEANUTS Example: Output Requirements and Specifications
PEANUTS produces considerable output quantities, all of which the us.er is occasionally
interested in. A complete list of possible program output is shown in· Table 4-6. This
list is far too extensive to be displayed as a standard set. A reasonable list of expected output quantities were discussed in Para. 4-1. These will be considered as the
standard set and are repeated here for clarity. Six stresses: SIGX1,· SIGX2, SIGT1,
SIGT2, SIG1 and SIG2, and two displacements DELTA and V , are shown in Table 4-6.
The shell thickness H will be included for convenience.
Output will be achieved by selecting the output command (OUTPUT). This displays
the list of possible output quantities shown in Table 4-5 plus an option for the standard
set of output discussed previously. A standard output display is shown in Fig. 4-37.
External and internal values are displayed on the same graph, e. g., SIGXl and
SIGX2 would be displayed on the SIGX graph. They can be distinguished by different
types of lines, i. e., solid and dashed.
It should be possible to display the results of the previous design on the respective
graphs of the current design by selecting a superimpose command.
4-85
S
1. 0
I
I
G
X
00
1.0
MERIDIAN (IN.)
S
I
G
1.
:I
T
1.0
0
MERIDIAN (IN.)
S
I
1.]
G
0
1.0
MERIDIAN (IN.)
D
I
1.
0
I
S
P
00
1.0
MERIDIAN (IN.)
T
H
1.]
K
0
1.0
MERIDIAN (IN.)
Fig. 4-37 Standard Output Display
4-86
4.7 REFERENCES
4-1
K. J. Forsberg and S. K. Ferriera, Development of Improved Structural
Dynamic Analysis, Volume II, Computer Graphics, Technical Report
AFFDL-TR-66-1S7, Mar 1967
4-2
David N. Keast, "Survey of Graphics for Computer-Aided Design,"
Machine Design, Aug 3, 1967
4-87
Section 5
ENGINEERING DOCUMENT RETRIEVAL
5.1 INTRODUCTION
This section provides the information a computer system analyst needs to perform
the detailed design of an on-line system for the retrieval of engineering information.
First, the analyst must know the scope of the problem. Paragraph 5.2 describes the
system requirements in terms of current and projected usage of the MSFC Document
Repository. The description treats the Repository as a black box and has the form of
an input/output specification.
Next, the analyst must know the environment in which the system is to operate.
Paragraph 5.3 describes the operations of the Repository and the Engineering Release
Center, and the way in which they function.
The analyst must then have a conceptual system design which he can translate into
programming specifications. Paragraph 5.4, Recommendations, suggests a retrieval
system and outlines the programming of some of the major operations implied by the
design. It is worthy of mention that a system similar to the one suggested, using
several of the same functions, is in existence at LMSC.
Finally, the analyst should consider existing hardware that might be used in implementing such a design. For this purpose, a brief survey of the hardware state-of.the-art is discussed in Paragraph 5.5.
Nothing in the proposed system prohibits the addition of further retrieval techniques,
such as coordinate indexing, as auxilaries. The proposed system might be the first
step in the development of a more powerful facility.
5-1
For brevity in description, the system developed here will be called REDI (Retrieval
of Engineering Design Information).
5.2 SYSTEM REQUIREMENTS
5.2.1 Current .Usage
Any system, if it is to be useful, must be responsive to the needs. or requirements of
the user. Responsiveness implies that there is a natural and efficient interface
between the system and the man or group using the capabilities of the system. From
the user's point of view, the "document retrieval system" can be viewed as a black
box-the user inputs information to the function and withdraws information from the
function or process taldng place within the box. The user should not really concern
himself with the inner workings of the
proce~s
or function as long as it is responsive
to his needs.
As a first step in this task of engineering document retrieval analysis, the present
usage of the Document Repository material will be examined.
A thorough understanding of the present file structures - who is using the Enginee'ring Document Retrieval system at MSFC, and how they use the files· - is essential
before any alternate methods of operation may be suggested.
Although this approach is not really directed at the document retrieval requirements
per se, it will bring into focus the present methods of operation. An analysis of the
file dynamics, howe.ver, (input, data manipulation and output) should provide a feeling
for the system requirements, regardless of the detailed functioning of the Repository
Operation. System requirements for a document retrieval system define the following
J
parameters:
•
Input
1. Originating source of information
2. Type of information
3. Rates
4. PhYSical form of information
5-2
•
Processing Function
1. Response time
2. Accessibility of information (file structure)
\
3. File maintenance
•
Output
1. Recipient of information
2. Type of information
3. Rates
4. Physical form of information
Note that user requirements are in no way tied to physical hardware or detailed processing schemes. The requirements placed on the document retrieval system by the
user will, of course, affect hardware selection. Response time requirements, for
example, strongly affect the choice of hardware needed for retrieval of stored material.
A thorough systems analysis would begin with the user asking "What is the data to be
retrieved? How much is there, and how should it be indexed?" The techniques suggested here are submitted to improve rather than revolutionize an existing operation
at MSFC in the form of a Document Repository and an automated Engineering Order
(EO) System under the direction of the Release Center. This study will therefore
build upon the capabilities of the present Repository operation and EO data base and
hopefully lead the way to an improved and expanded service in fulfillment of the mission of an engineering document retrieval system.
5.2. 1. 1 Input. This discussion provides a comprehensive review of the information
that is input to the Repository. The following questions will be explored: Who inputs
material, what does he input, when, and how much, and what is the physical form of
the material?
Originating Source of Information. The following three groups comprise the user
population that will enter information into the Central Repository:
• MSFC Personnel
• Contractor Personnel
• NASA Intercenter Personnel
5-3
The folloWing paragraphs will identify the risers within the groups listed above.
MSFC Submittals. MSFC personnel may input material originally generated by
in-house MSFC 'projects or by contractors. The October 19.67 MS-D Input Chart
(Central Repository) supplied by the Repository personnel shows the following
MSFC groups as having input'material to the Repository during this month:
Laboratory
Branch of Laboratory
ASTR
-BX
P&VE
-SAE
-SSV
AERO
-AB
-AD
-EA
-ESE
-ESS
-GD
-VSD
-xx
-ND
-ME
-VSL
-DA
The input records for the Repository show that the Engineering Release 'Center
(P&VE-VSD) handles a large share of the dOCUtllents input to the Repository from
MSFC activities·.
Data supplied by the Engineering Release Center indicates that the following
Laboratories/Branches submit documentation through P &VE - VSD:
P&VE
Branch of Laborator;y
-p
-S
-VM
ASTR
-E
ME
-ME
AERO
QUAL
-E
-p
TEST
-D
Laboratory
-vv
-M
Contractor Submittals. The list of contractors submitting documentation to the
Central Repository directly or indirectly through a MSFC laboratory/branch monitoring agency is very large. To give a specific example, reference is made to the
October 1967 MS-D Input Chart (Central Repository) which lists the following
contractors:
5-4
Bendix
Hayes International
Boeing
IBM
Chrysler
North American
Douglas
Pratt & Whitney
Fairchild
Rocketdyne
General Electric
Sanders Associates
General Motors
Thiokol
Greer
Walter Kidde
Grumman
Different divisions within the above listed companies also contribute documentation
independently.
There are many other contractors not listed that have at some time contributed to
the data base at the central Repository.
NASA Intercenter Submittals. Intercenter submittals may originate from Kennedy
Space Center (KSC) or the Manned Space Center (MSC) in Houston, Texas. The
content of these data submittals will be discussed in the following paragraphs.
Type of Information Submitted. There are at least 85 different types of documents
handled by the Repository. In order to avoid becoming enveloped in too fine a level
of detail within these claSSifications, only four categories will be discussed here and
subsequent discussions of rate and file structure will be limited to these. Slightly
more detailed subdivisions are shown for illustrative purposes only. These data were
extracted from MS-D Input Chart (Central Repository) for the month of October 1967.
Major Category
Subdi vision
1. Drawings and Control Documents
Drawings
CEI Packages
Engineering Orders (EO's)
Documentation Release List (DRL)
Engineering Master Parts List (EMPL)
Generation Breakdown (GB)
Interface Control Document (lCD)
Interface Revision Notice (IRN)
5-5
Major Category
Subdivision
2. Reports
Test
Trajectory documents
Status
3. Specifications
Saturn-V Display System
4. Standards
Rates for Submittals. The statistics gathered on input rates fall into two categories.
1. Input rates for the document Repository from all sources
2. Engineering documentation sent to the Repository by the MSFC Release
Center (P&VE-VSD).
Category 2 above is included in the rates of Category 1. The release center activity
was analyzed separately since any engineering design/graphic interface would most
likely affect the release center operation. The release center is the agency concerned
with the checldng, formalizing, and controlling of official drawing and control documell
tation in support of the in-house MSFC design groups.
The information used here is extracted from complete statistics normally maintained
by the Central Repository personnel. Figure 5-1 depicts the average monthly input
activity for FY 1965 - 1967 and shows the actual monthly rates for FY "1967.
Figure 5-2 shows the release center activity for the period January 1967 through
October 1967. Four subcategorie"s of information are shown on this chart, namely:
1. Drawings
2. Engineering Orders (EOs)
3. Documentation Release List/Engineering Parts List (DRL/EPL)
4. Interface Control Document/Interface Revision Notice (ICD/IRN)
The extent of automation (computer programs) in the release center operation will be
discussed in Paragraph 5. 3 Storage and Retrieval Systems.
5-6
2.3
ACTUAL MONTHLY WPUTS
2.2
2.1
MONTHLY AVERAGE
2.0
LEGEND
1.9
1.8
....
;>
Po.
.~
CI.l
Cl1
I
.....:]
1.6
....
1.4
1.3
::E
u
;>
1.2
0
0
1.1
~
1.0
CI.l
0 •.9
0
z
0
~
...;l
MICROCARDS
f0J PRODUCTION
LlliJ
JCP UNITS
D
DISTIUBUTION ONLY
JCP UNITS
NOTE: JCP UNIT = AREA OF 1 PIECE OF 8 1/2 X 11 IN. PAPER
1.5
~
Z
I
1.7
0.8
0.7
~ 0.6
0.5
0.4
0.3
0.2
0.1
0
1965
1966
FY
1967
JUL
AUG
SEP
OCT
NOV
DEC
JAN
FY 1967
Fig. 5-1 Documentation Repository Input
FEB
MAR
APR
MAY
JUN
1500r----------------------------------------------------------------------------------LEGEND:
1400 J - - - {t@f~ ICD/IRN - INTEHFACE CONTROL DOCUMENTATION!
311
INTEHF ACE REVISION NOTICE
1300J---I'IlIu'lllIlII DRL/EPD - DOCUMENTATION RELEASE LIST/
ENGINEEHlNG PARTS LIST
_
EO'S - ENGINEERING OIU)EHS
1200 J - - - tif$W.fiMrifll DWGS - DRAWINGS
JIOOr---------------------------------~~~t~~------------------~~~~~------------------~
441
1000 J - - - - - - - - - - - - - - - - - - t ; :--?t-'1-------~I'I11
~
~---liIII'liill------1
1III
900 1---'I;,~;J,·:,:.:i
II
'111 1"
~ 800 t----"it:X~~~f__-----------------;---.......---J'::':\:;.;·.;.;,1----I;>;~ri~-----------"1 "1-----------------'
z
~
t;
~
:!!!I:II
IIIIII
700
t------------I
1i,,;111
::s
~ 600
500
400
300
200
100
o
F
M
A
M
J
J
A
S
MONTHS (1967\
Fig. 5-2 Engineering Documentation Released by MSFC Release Center
and Forwarded to Central Repository
5-8
o
Physical Form of Information Submitted. MSFC Administrative Regulation 25-9
outlines the procedures for submission of documentation to the MSFC Central
Repository. Acceptable media are:
•
Drawings and Associated Documents
1. Original vellum
2. Aperture cards
3. Microfilm roll
4. Black and white vellum reproducible
5. Hard copy black line
•
Reports, Manuals, Specifications, etc.
1. Original manuscript.
2. Black and white vellum re'producibles
3. Printed copy, black line
4. Microfiche
5.2.1.2 Processing Functions. This portion discusses the user requirements
imposed on the processing function or storage/retrieval system operation. The user
is concerned with three main attributes of the processing function ..
• Response time of "function"
• Accessibility of data/information
• File updating (currency of information)
The above three points are very important. The degree of user
confidenc~
that a
repository can adequately meet the requirements inthe above areas determines the
extent to which he uses the service. If users are not satisfied with response times,
cannot find data in the file, or are deluged with out-of-date material, then small
private file stores will build up and eventually annul the benefits of a central repository.
Response Time Requirements. This is a very subjective area of discussion. Some
engineering personnel feel they need repository material immediately while others
are willing to accept longer response times. Project schedules and tight deadlines do,
of course, sometimes force data requests to have high priorities.
5-9
To provide a system that is as flexible as possible and that is responsive to the needs
of its users, two modes of requests should be honored: normal response and high
priority response.
To match these two desired modes of response, the central repository maintains a
24-br response for small requests (five documents or less). Requests for large quantities of data (either many single sheet items or one document having many pages)
may take 36 to 48 hr or longer depending on the workload of the printing facilities
required.
Direct acquisition of copies of documents from satellite microfilm files is possible if
the user has an urgent need for the data. , The ap'erture carqs may be retrieved by
repository personnel and the user may view and/or copy the document if desired.
Mode of document delivery also affects system response time. Various methods of
transmitting the document are available such as the following:
• Call when·ready
• Will pick up
• Delivery
• Mail
Accessibility of Information. This topic is the heart of a document retrieval system.
The user must first know what is in the data base and secondly be able to ask for it in
terms of his goal. There is a major difference between a document retrieval system
and an information retrieval system. It must be recognized that the repository is
primarily a clearing house for documents and therefore is a document retrieval system.
Its mission does not include providing information from the documents it handles, but
rather providing the documents themselves.
Engineering design activities are functionally oriented: for example, an engineer may
be given the task of designing a relay or a motor. The topic of indexing and 'how to
·find out what is in the file relevant to his
~he
s~cific
task will be ,discussed at length in
next sections of this report and in the section on retrieval of design data.
5-10
User requirements for data accessibility will be divided into three areas.
.• Physical accessibility
• Procedural accessibility
• Document content accessibility
To meet the needs of physical accessibility the Repository files should be physically
located near the user. To this end the Repository has provided one main work order
desk at MSFC and five satellite work order desks - four located at MSFC and one in
downtown Huntsville, chiefly for Boeing/Chrysler use.
Procedural accessibility deals with the various methods of requesting services. Two
main classes exist here, namely automatic (regular distribution) and individual
(request). In the automatic mode of distribution, lists of addresses keyed to document
types are predetermined by laboratory, stage, or project management. Entry of these
documents into the Repository automatically triggers the reproduction and distribution
of the required data.
Individual requests may be entered by phone,
m~il,
or in-person via MSFC Form 433.
Document content accessibility is presently limited to document number and revision.
There are master index lists (by document number) that are located at each Repository
work order desk to enable the user to determine whether a given document number is in
the file. This index also contains a description of the doclllll:ent.
File Updating. A file, to be useful, must contain current information. There must be
a minimal time lag between entry of data into the Repository and the capability to withdraw that data. The specifications for the variety of hardware used to expose, process,
index/code, duplicate, and store the documentation are· known but no statistics are
currently available on cycle time (entry to storage)for typical packages of data.
I
Unless superceded data are periodically purged from the file, the data base will continue to expand in size and response time (search time)· will become longer. All
associated support activities such as the preparation of new index lists and automatic
distribution activities will also increase. To keep the overhead support under control,
5-11
some sche·me of file purge or relegating data to "history" files must be implemented.
If the latest revision of a drawing is E, ~for example, it would not be desirable to main-
tain Rev. A on immediate access files (assuming, of course, that Rev. A is truly history and not being used in some current model of hardware).
The Central Repository scheme for file purging depends entirely on the contractor.
Each contractor is asked periodically if certain documentation has been superceded
and if so the data are transferred to the supply warehouse and
~nactive
files.
5.2.1.3 Output. This section discusses the characteristics of the material that is
output from the Central Repository. The method of discussion covers the same building blocks as the input section, namely: who receives the information, what kind of
data is withdrawn from the files, the rates of output, and the physical form of the
output material.
Recipient of Information. As in the case of input material, there are many groups
of personnel receiving data from the files. The Cost Center Usage Report for October
1967 shows that 325 separate groups (cost centers) within the several. laboratories at
MSFC and other NASA centers or facilities requested material from the Repository.
Contractor usage statistics are reflected in MSFC Industrial Operation usage statistics.
Types of Information. MSFC Form 433 is used for requesting documentation from the
Repository. The different tyPes of information as enumerated on the Form 433 is as
•
-·t
!
..
follows:
1. Drawings
2. Engineering Orders (EOs)
3. Parts ,Lists
4. Specifications
5. ,Reports
6. Interfac,e" Control Documentation (leD)
7,.
Interf~ceRevision
Notice (lRN)
8. Documentation Release Lists (DR'L) or Engineering Asse,mbly Parts Lists
(EAPLs)
5-12
Note that the requests (Form 433) are really for documents, not information per se.
Exactly what information is extracted from these documents to accomplish the users'
task has not been examined bec&use that information is not available from the
Form 433.
5.2.1. 4 Rates of Output. MSFC is subdivided into two main groups - Research and
Development (in-house government personnel) and Industrial Operations (contractor
interface). These groups are commonly referred to as R&D and 10. The cost center
usage report for October 1967 shows the usage rates for the R&D and 10 classifications.
The item "JCP Unit" means the area of one 8-1/2 in. x 11 in. sheet of paper. Drawings, for example, may contain several JCP units. Table 5-1 provides a typical
summary of usage.
Table 5-1
SUMMARY OF R&D AND 10 USAGE
User
R&D
Activity
10
April - October
(Year to Date)
46,639
44,139
421,762
JCP Units
956,774
1,548,390
8,948,331
Cards Handled
712,147
506,152
3,574,972
18,893
1,099,577
19,290
920,322
180,133
6,632,853
767,977
353,475
2,097,300
65,532
63,429
.601,895
JCP Units
2,056,351
2,468,712
15,581,184
Cards Handled
1,480,124
859,627
5,672,272
JCP Units
Cards Handled
R&D
and
September
(Last Month)
Documents
Documents
10
October
(This Month)
Documents
Figure 5-3 shows the total output for the period FY 1961- FY 1967.
As a final example of output rates, a tabulation of the six most active requestors for
the month of October 1967 is shown in Table 5-2.
5-13
THOU Mil
AL
FY
I
DOCUMENTS HANDLED
(REPORTS, SPECS., & STANDARDS)
FY-67
I
I
DUPlICARDS REPRODUCED
DISTRIBUTED TO SYSTEM
Fig. 5-3 Documentation Repository Output
~
REPRODUCTION JCP UNITS
~ (DRAWINGS, EO'S, ETC.)
Table 5-2
SIX MOST ACTIVE REQUESTORS - OCTOBER 1967
Documents
Order
of
Usage
Cost
Center
1
2000
Group
R&D
Number
Cost
Center
24.5k
8501
Operations
2
8501
Industrial
Cards Handled
JCP Units
Group
Number
Cost
Center
Industrial
831.7k
8501
2000
Operations
R&D
Industrial
Number
676.2k
Operations
Operations
15.0k
Group
271.5k
6432
Quality
362.4k
Laboratory
Operations
Systems
Analysis
3
5000
3.2k
P&VE
5000
Laboratory
4·
3311
Planning and
1261
Facilitie~
and
2.7k
9520
5134
P &VE Systems
Design
5001
SAT V
2.2k
2920
KSC Launch
121.6k
5102
1261
Facilities and
DesigIi
103.2k
P &VE Vehicle
90.7k
Systems
79.6k
9100
Support
1.9k
P&VE Lab.
Director
Systems Eng.
Design
6
205.7k
Laboratory
Engineering
5
P&VE
Contracts
81.8k
Office
34.2k
8001
Test
Laboratory
Director
39.7k
5.2. 1. 5 Physical Forms of Output Material. There are three basic physical media
output from the Central Repository:
• Hard copy (paper)
• Microfilm aperture card ..
• Microfilm roll
All data contained in the file may' be reproduced on paper.. This is the main output
mode for the repository. Some users may request and receive duplicard distribution
as an output,
,i. e.,
duplicate decks of microfilm aperture cards. A positive micro-
film is sent to the Archives for all data input to the Repository.
5.2.2 Projected Usage
This section discusses anticipated future usage of the Document Repository/..
Engineering Release Center Operation. There are two different assumptions that
might be made
rel~tive
to the mode of Repository operation at MSFC.
e Continue the present mode
• Augment the engineering design/document retrieved operation with on~line
consoles
5.2.2.1 Present Mode of Operation. Under the first assumption above, the type of
data handled, the user population, and physical media of input are expected to remain
the same. New subject matter
~ll
of course be created; the Apollo
Program (AAP) is an example. New
contr~ctors
Appli~ation
will begin to use the Repository;'
the Martin Company, for example, will now, be a prime input source for. AAP material.
,
Because of random flu~tuatiollS' Jt
,
is difficult to extrapoiate. past Repository usage to
arrive at a meaningful projected rate.
In the effort to project usage of the Repository /Engineering'Release activity at MSFC
reference was made to "1968 NASA Authorizations," hearings before the Subcommittee
of Science and Astronautics, U.S. House of Representatives, March 14'to 16, 20 and
21, 1967. Several statistics from those hearings were used for predicting ,a projected
activity rate. The follOwing illustration- (Fig. 5-4) was extracted from the hearings.
5-16
8000
7271
7030
7030
7000
6000
5000
4000
3000
2000
1000
FY-1966
ti~~~i;i; . . . . ·.
SUPPORTING
RESEARCH, SPACE
SCIENCE, ADVANCED
MISSIONS, AAP
FY-1967
FY-1968
I I I I I I I I I I I I I I I ~I I I I I I I I I I I I 1 1 1 1 1 1 1 1
SATURN/APOLLO
ADMINISTRATIVE
Fig. 5-4 Program Distribution of Manpower at MSFC
5-17
The Research and Development appropriation increase for MSFC for FY 1968 is 4.3
percent. This increase should cause a corresponding increase in the documentation
activity of MSFC.
I
Figure 5-5 shows the AAP documentation for the period June 1967 to October 1967
as processed by the Engineering Release Center.
The AAP is one· of the major next generation efforts of MSFC and should produce a
large volume of engineering documentation. The fluctuating statistics for the AAPreleased documentation is an indication of the problem of predicting the rate of
activity for the Engineering Release/Repository operation.
5.2.2.2 Graphic System in Operation.
For the next few years it is forseen that the
computer graphics design effort will have a small impact on the total volume of
activity of the Repository and Release Center.
felt more in the way documents are retrieved.
5-18
The effect of this innovation will be
..-
- 40
..-
30
35
~
28
~
26
~
~
00
~
8,
Z
30
24
8, 24
Z
22
tI.l
~
22
~
20
~
20
0
18
:::>
18
0
0
0
16
<
~
12
::>
~
c:Jl
I
0
0
0
to
'00
~
25
:::>
0
0
0
20
<
~
H
~
15
~
0
~
10
12
~
10
~
0
~
~
p
<
~
~
roil
~
14
H
~
5
~
:g
Z
:::>
Z
J
J
A
S
0
Z
~
0
16
~
00
~
~
26
~
0
00
0
~~
00
~
~
......
~
~
00
0
-
tI.l
~
00
H
~
8
~
6
0
~
4
10
8
6
~
4
:g
2
~
2
14
p
:z;
J
J
A
S
0
J
J
A
S
Fig. 5-5 AAP Documentation Processed by MSFC Engineering Release Center
Between June and December 1967
0
5.3 STORAGE AND RETRIEVAL SYSTEMS
The previous portion of this studY.examinedthe MSFC user requirements with respect
to data input to ~ processing within, and output of m~terial from the Central Repository.
This section will discus s the following storage and retrieval systsms at MSFC:
•
The' Central Repository
•
Engineering Release Center
•
The Saturn V Operational Display System
•
The A'mpex Videofile System
This investigation is reported because, before system integration or modification can
be suggested, an understanding must be attained of what is presently in being at MSFC.
5.3. 1 Central Repository Operation
This paragraph de~cribes the function of the "black box" (Central Repository).
Paragraph 5.2, User Requirements, defined what data was input to and output from the
function and those few constraints imposed by the user on the operations within the
box. The specific data flow within and the attributes of the function will be reviewed.
The following items will be covered:
•
Documentation flow
•
•
File structure
Configuration of facilities
•
Operating procedures/forms
•
File storage media
5.3.1. 1 Documentation Flow. Figure 5-6 shows the documentation flow within the
Central Repository. This flowchart is self-explanatory and shows clearly the internal
operations performed by the Repository personnel. In summary, there are six operations performed by the Repository, namely:
5-20
ORIGINATOR
USER
REPOSITORY
AUTOMATIC DISTRI BUTION
MICROFICHE
AUTOMATIC DISTRIBUTION AS DIRECTED
BY LABORATORY, STAGE MANAGER, OR
PROGRAM MANAGER.
-Contractors
OFFSET
REPRODUCTION
(When Required)
REPORTS
& SPECS
-In House MSFC
-S-IC (Boeing Huntsville)
-S-IC (Boeing Michoud)
·5-11 (North American)
-S-IVB (Douglas)
-HJ, F-l, J2 Engines (Rocketdyne)
-RL lOA3 Engine (Pratt-Whitney)
-Apollo Interface
-ESE (GE)
-IU (IBM)
-A 0110 & LM
REPRODUCIBLE
-In-House MSF C
DRAWINGS &
- SIC (Boeing Huntsville)
ASSOCIATED
- SIC (Boeing Michoud)
DOCUMENTS
• H-l, F-l, J-2 Engine (Rocketdyne)
_ _ _ _ _ _--.J. Apollo Interface
SPECIAL REqUESTS
AUTHORIZED REQUESTOR.
SHELF
STORAGE
COPIES AS REQUESTED.
INITIAL DISTRIBUTION
DIAZO
(White Print)
REPRODUCTION
AUTOMATIC
DISTRIBUTION
LISTS AS
ESTABLISHED
BY
R&D
OPERATIO~S
MANAGEMENT DIRECTIVE 25·1.
HARD COpy AS REQUIRED.
REPRODUCIBLES
I
CONTROL
DOCUMENTS
(MASTER PARTS
LISTS,
GENERATION
BREAKDOWNS,
ETC.) .
COPIES OF REPORTS TO USER ON AUTOMATIC BASIS.
DISTRIBUTION ONLY
DU PLICARD DISTRI BUTION
-In-House MSFC
- All Stage & Program
- Contractors
Bvpassed
When
Microfilm
Is F urni shed
FIRST
GENERATION
APERTURE
CARDS
- 5-1 B (Chrysler Michoud)
• S-IC (Boeing Michoud)
- S IVB (Douglas)
- ESE (GE)
-IU (IBM)
- ST124 (Bendix)
'35 MM
ROLL
MICROFILM
• SIB (Chrysler Huntsville)
Neg.
-SIB (Chrysler Michoud)
*Pos.
• SIC (Boeing Michoud)
*Pos.
-511 (North American Aviation)
Neg.
-SIVB (Douglas)
*Pos.
- H1, F-l, J2 Engine (Rocketdyne) Neg.
- RL lOA3 Engine (Pratt-Whitney) Neg.
• ESE (GE)
Pos.
• IU (I BM)
Pos.
* 'for National Archives Only
MICR OFILMING.
PRINTING &
PROCESSING
AUTOMATIC DISTRIBUTION KEY ESTAB·
LISHED BY ORGANIZATIONAL REQUIRE.
MENTS.
DUPLICARD
PRODUCTION &
DISTRIBUTION
MICROFILM
MASTER FILE/
WORKING FILE
IIp-o-
~~§~
t:
**
..~......v.''''"'''''''.''>Y......:''';'.lCIX.-.c.:''''~~
REPRODUCIBLES
DRAWING
MASTER FILE
(Reproducibles}
ORIGINAL DRAWINtS
NATIONAL
ARCHIVES
GSA ATLANTA
AUTOMATIC DISTRIBUTION OF DUPLI·
CARDS IN ACCORDANCE WITH DISTRIBU·
TION KEY
SPECIAL REQUESTS
AUTHORIZED REQUESTED •
COPIES AS REQUESTED FROM MICROFILM,
TRACING, HARD COPY, OR DUPLICARD,
AS APPLICABLE.
* * SEE
PROCEDURE MSFC 25·2 FOR DETAIL INSTRUCTION
FOR REQUESTING DOCUMENTS.
Fig. 5- G Documentation Flow
5-21
•
Photographing
•
Indexing/Coding
•
Duplicating
Retrieving
G
o Processing
o storing
5. 3. 1. 2 File Structure. There are two main clas ses of documents retained by the
document Repository: Class I - those documents originated by the government following standardized formats and bearing numbers controlled by the government, and
Class II - those documents originated by participating contractors and having numbering systems managed by the contractor.
The numbering system for Class I documents uses eight characters. The first two
specify the Government organization at MSFC that originated the document, the third
denotes the NASA center (constant M for MSFC) and the last five are the serial number of the specific document. Contractors are sometimes requested to supply Class I
type documentation and in those cases the first three digits are 6XY, where the X can
be 0, 5, or 6 and the Y can (for example) be B or C (Boeing or Chrysler).
Following is an example of the numbering scheme for the internal MSFC organizations.
Drawing Prefix
Organization. Code
R-P&VE-VV
R-P&VE-P
R-P&VE-S
R-ASTR-E
R-ASTR-M
R-AERO-E
R-TEST-D
10M
20M
30M
40M
50M
80M
90M
The numbering scheme is more detailed than indicated in the above example; the intent
,.
of the example is to depict the general indexing scheme used by in-house MSFC .
organi zations.
5-23
Technical documentation other than drawings (i. e., procedures t specifications t and
standards) originated by MSFC organizations has the following indexing scheme.
Subject
Prefix
Procedures
Spe cifications
Standards (Book form)
Standards (Sheet form)
MSFC-PROC-(OOO)
MSFC-SPEC-(OOO)
MSFC-STD-(O.OO)
MC-(OOO)
Class II documents have contractor-managed numbering schemes. There has been
no attempt to standardize the indexing schemes originated by the several contractors;
further, in filing and indexing Class II documents, the Repository does not apply any
additional index numbers over and above the contractor index.
The Repository files all drawings and associated documentation (both Class I and ll) in
numerical sequence by vehicle stage within major categories such as S-IC, S-IVB, etc.
Specifications, standards and reports are indexed by numerical sequence and originator
such as MSFC t contractor, etc.
The present size of the files are as follows:
File Type
Size
Drawings (Aperture Cards)
Reports
Specifications
Hard Copy Vellum Drawings
Inactive File (Aperture Cards)
(Warehouse - Building 7207)
2.5 million
6,000.
30,000.
100,000.
1.5 million
5.3.1.3 Configuration of Facilities. Facilities of the Central Repository activity are
located both at MSFC and in downtown Huntsville.
The main repository operation is located at Building 4494 MSFC. Activities at that
loc"ation are receiving, inspecting, photographing, reproducing, and filing. The main
Work Order Desk is also located there.
5-24
The Huntsville Industrial Center (HIC building) is located near downtown HWltsville
and performs the Repository operations of microfilming, producing duplicard, and
filing.
The main reports and specifications file, the Master Microfilm File, and a Microfilm
Satellite File (mainly for use by Boeing and Chrysler) are maintained at the HIC
Building.
There are four other satellite files located on MSFC in addition to the main file
(Bldg 4494) and the HIC Building. Their locations are as follows:
•
Test Laboratory - Building 4666
•
P&VE Laboratory (2) - Building 4881
•
Astrionics Laboratory - Building T-12 (Trailer)
There are two other facilities that contain Repository documentation, e. g., Classified
storage, Bldg 8497 MSFC, and Supply Warehouse and Inactive Files, Bldg 7207 MSFC.
The backup/disaster files are located at the National Archives in Atlanta, Georgia.
Positive roll microfilm is provided to the Archives for all material entered into the
master Repository files.
There are at least 200 items of hardware used in the internal ope.rations of the Central
Repository. A partial list of the major types with some of their characteristics follows
Description
Item
Card punch
Model 026
Card to card printer
Model 041, 2000 cards/hour
Reader /printer
3M and other models
Collator
Model 189, 650 cards/minute
Semiautomatic mounter
(microfilm onto aperture cards)
3000 cards per day
Card interpreter
Model 557, 6000 cards/hour
Contact printer
40 x 60 inch size capability
5-25
Item
Description
Electrostatic copy machine
Flat bed copy camera
Offset master generator
180 masters per hour
Offset press
Diazo reproducer
42 inches wide capability
Automatic drawing folder
Microfilm camera
45 x 63 inches maximum drawing size
Reduction ratio variable12x to 36x
Rotary' aperture card file s
Diebold-200 to 400K capacity
R~mington
Rand
IBM 1401 Computer
Index list preparation master magnetic
tape index generation
Files (general)
Map, letter size, shelf, etc.
Microfiche reader/printer
5.3.1.4 Operating .Procedures/Forms. These items are MSFC administrative
regulations and procedu'res that define the operations of the Central Repository. They
are:
MSFC 25-9 - Input of Documentation to the MSFC Central Repository
MSFC 25-2 - Requests for Documentation From the MSFC Central Repository
Figures 5-7 'and 5-8 are, respectively, "Document Input Record," MSFC Form 2896,
and "Request for Documentation," MSFC Form 433.
Another important item of user control documentation is the "Distribution Data
Acquisition List," MSFC Form 2598. (See Fig. 5-9 . .)' This describes the procedure
to be followed by project, laboratory, or stage managers to establish automatic
distribution lists.
5-26
MSFC 25-9
ANNEX B
MSFC FORM 2896 AND INSTRUCTIONS
FOR COMPLETING IT
DOC 'U MEN T ~ I'N PUT R E COR 0
SUBMITTAL DESCRIPTION:
CONTROL NUMBER:
DATE SUBMITTED:
G)
CD
SUBMITTING ORGANIZATION:
PERFORMING ACTIVITY AND/OR
CONTRACT NUMBER:
SUBMITTAL AUTHORITY:
0
0
0
0
RELEASING AUTHORITY:
0
0
SPECIAL INSTRUCTIONS:
SUBMITTAL INCLUDES:
0)
C;
~~
~
\,~
TO BE COMPLETED BY MSFC CENTRAL REPOSITORY AND A COPY RETURNED TO SENDER
~ECEIVED
DATE RECEIVED:
BY:
WORK ORDER NUMBER:
@
@
@
MSFC Form 2896 (March 1967)
TYPE OF INFORMATION OR EXPLANATION
BLOCK NUMBER
o
G)
o
o
o
Identification of the material being submitted (example - MSFC Drawing package.
Apollo ICD. or 67-FMP2 AS~501 "Joint Oper. Traj.")
Any identification number utilized by the submitting organization (example - Release
Package Number. Documents Requirements List, line item number. etc.)
S elf- explanatory
Office of origin or office of control
Performing activity code as defined in Annex E of Volume II of the Financial
Management Handbook. or Contract Number if a contractor submittal
Reference document that directs or authorizes the submittal of data to the repository,
(example - MSFC Drafting Manual, Engineering Bulletin. Program Directive No.6
from 1- V. Appropriate Contractual Data Submittal Document, etc.)
For Class 1 Documentation of R&D Operations Organizations only. Enter in this
block the official release authority
Any special handling or distribution requirements
Complete identification and quantity of data submitted
To be completed by MSFC Central Repository'
Fig. 5-7 Sample of MSFC Form 2896 - Document Input Record
5-27
MSFC 25-2
ANNEX A
US10MER REFERENCE NUMBER:
(Used as R~quired)
1.1-48
REQUEST
FOR
DOCUMENT ATION
PHONF NO.:
REQUESTED BY:
H. S. Garrett
BL~G.
876-1673
~--O-R-D-ER-/'~~J~ij~~~ _~~'p_O__N_L~Y_ _ _ _ _ I
00 2773
NO.:
01\1 F. REQUESTED:
4746
ORGANIZATION SYMROL OR NAME OF CONTRACTOR:
II /"
C' .. ,lIfI .. rt
rh"rQ""'.'.
MS-D
DATE NE.EDED:
11/28/66
Ihlll Ih .. ,Inrum .."',,
Itf'I' "IIy:
'n. ",- ,.t",,.,.,
It, ..
11/29/66
f .. , tI,p nflld,,' "" .. of. IItId
p'npe,',
~C~O~S~T~C~E~N~T~E~R~C~O~OE=.-.---------r.C~.O~N~T~R~A~C~T~N~U~M~R=rR=:-----------
-------------.
. --::--=-=--=-=-:----:~--------+____________~ _______
s_'r;_'~~I~_o:~~~~F~
T ING A UT Hen IT Y
J-.;:1;..;4...;1~O~___________
t-_P
__
R_O_G_R_A_M_(3_-_0_'G_,_'T_C_Oo_E_''-_f-_ _ _.---;M''-'-'-A_T E R I ~~Q_lLE S T_~______ __
LINE
ITEM
NO.
OOCUMENT
DWGS
,- I SPECIFICATIONS
r -I ORL OR FAf'L
, ., CALL WHfN READY
I -I
[Xl
r.O·S
REPORTS
I _ICMOS
PL OR LM
ICD
(-l'RN
Il{l WILL PICK
r-: 1 DELIVER
,:
C:J
L._J
CI
NUMBER
NO. OF NO. OF
COPIES REPRO.
REV.
FOR -B~PO~LTORY
NOT
SI ZE
FRWn.
RELEASED
Tn
.1.
2.
40M39513
40M39526
A
3.
36-001M-53
C
2
USE
ONLY
BACKORDER
.;
M~ .. n
MAIL
1 OTHER
.NUMBER
FRWO·
V
NC
50M68475
UF'
I
1
/
B
C
I
----+----~---------- ---r-------------r---~
Aero Quip
4.
INS T R U C T ION S
IXl
----- ---4--------.---t·-----------+----~
09-0621
10
B
- ' - ----
------- --
J---r--------------t--- --- ------ .----.- .---- ---.----- ---.
.-. -'-'- --.-- -- - - - - - - - - - - J - - - - I
-- --.--._- - _·_--------1---1
------- -- ------------t-----... --- -
1--+------------~~----+-_+--_1f_--+_----+---t-.
h. ' "
c,'"
1--+--------------:~-----+-_+--___1I__--_t
. -- ------.-------+----1
- ------
-----.. -.- - .. ----- -------... --.-
-
-------.---
-.----- ---- ---------+----1
1--+---------------.-.-1----- ---- ---.- -t--~--------------------,----~--
t--+-----------------.--t--- . ---.-t---+--------.---.----- -----f----..
... ------"-'-
-- -------------
TAKEN BY:
"tAZ,,:
Frances Cobble
Lillian Locke
~_"2..F!?DUCTION INSTRUC T IONS
DISPOSITION OF ORIGINALS
[J
CUSTOMf:R
IXI
FILE
o
r~l
CI
1:]
CARDS
LOGGING
FILE
CUSTOMER
INPUT
TYPE
COPIES
REQ'O
SHEETS
MATS
COPIES
R(~~
UNITt.
TOTAL
INPUT
4
2
1
OUTPUT
9
Fig. 5 -8 Sample of MSFC Form 433 - Request for Documentation
5-28
20
J
t.
2.
DATE:
DISTRIBUTION DATA ACQUISITION LIST
3. TITLE:
4.
Plan, First Article Configuration
Insp. (FACI)
9.
NAME OF RECIPIENT
FIRST
INITIAL
LAST
NAME
SECOND
INITIAL
10.
ORGANIZATION
SYMBOL
IPENT. NUMBER:
5.
NAS7-101
(CM-512)
II.
12.
BLDG.
NO.
CONTRACT NO.:
6
1•
13.
DRL NO.:
021
PAGE
7. DRL LINE NO: B. CONTRACTOR I DENT. CODE
10S3617
028
IS.
14.
ROOM
NO.
- - OF --17.
16.
ZIP
CODE
STATE
CITY
STREET AND NUMBER
I
2
/
3
4
5
6
7
B
~
9
.
10
O~Y"
~~
c,r
II
12
13
14
15
16
17
lB
19
20
21
22
::::::: ..:::'.:::::".;:.:
PROGRAM
OFFICE
WSPC • Porm 25111
';:\':".'::::;:
\'9.
.:;:::::':::'"
·:.'ME:
(P.br_~
; DATA MANAGEMENT COORDINATION APPROVAL
20 CATE:
OATA
121~ "~.~E:
.
MANAGER
1
..
.: ... ;:...:....:.:.>::: .•
..:: ...
[1
1!..1)
Fig. 5-9 Sample of MSFC Form 2598 - Distribution Data Acquisition List
-.:: ........
22.
DAlE:
'.
ITYPE
COpy
lB.
c
0p
I
,s
5.3.1.5 File Storage Media. There. are two basic forms of storage media employed
by the Document Repository:
•
Mic~ofilm
•
Vellum/hardcopy - 10 percent of storage
- 90 percent of storage
Within the class of microfilm storage there are actually three physical forms:
•
Aperture cards
•
•
Microfiche
35 mm roll microfilm (positive and negative)
The microfilm files located at the main work order desk and the several working files
located at the Satellite File Stations are in aperture card form (standard mM cards
with 35 mm microfilm attached). Microfiche storage is used primarily for reports
and specifications. Positive 35 mm roll microfilm is provided to the National Archives
GSA Atlanta, Georgia, for the permanent historical records.
5. 3.~ 2 Engineering Release Center
The Engineering Release Center is an
i~portant
contributor to the Central Repository
data base •. All official engineering documentation originated by in-house MSFC personnel must pass through the release center operation cycle. Some contractors also generate Class I documentation that is release center controlled - Boeing and Chrysler
are prime examples.
5.3. 2. 1 Release Center Operation. The MSFC documentation release system and its
associated, adjacent operational organizations are depicted graphically in Fig. 5-:10.
The MSFC design groups within the various laboratories prepare engineering drawings,
Engineering Orders (EOs), Documentation Release Lists (DRLs), Engineering Parts
Lists (EPLs), specifications, standards, procedures, etc., which are approved
technically within their respective organizations, and then forwarded to R-P&VE-VSD
branch of the PropulSion and Vehicle Engit;teering Laboratory foroffici~l release~
Working groups and panels prepare other documentation, such as interface control
5-30
TECHNICAL REQUIREMENTS
MSFC DESIGN
ELEMENTS
DWG & SPEC
CHANGES ONLY
CHANGE CONTROL
BOARD
(CCB)
SPECIFICATIONS, STANDARDS,
AND PROCEDURES
I
ADMIN ISTRA TIVE
REQUIREMENTS
ALL RELEASES
EXCEPT CHANGES
ALL RELEASES
AND CHANGES
(REVIEW/CORRECTION
CYCLE)
ENGINEEHING
CHECKING
P&VE = VSD
SPECIFICATIONS, STANDARDS
PROCEDURES, DRAFTING MANUAL
AND ENGH BULLETINS
ENGINEERING
RECORDS
P&VE - VSD
COMPUTER INPUT INFO
W/WORKING COPIES OF
DOCUMENTS
COMPUTER PRODUCED
DRL'S, EPL'S,
AND ERROR LISTINGS
MASTER COPIES OF
RELEASED DOCUMENTS
TECHNICAL
DOCUMENTATION
REPOSITORY
COMPUTER PREPARED
EMPL'S RELEASE RECORDS
GENERATION BREAKDOWNS
lNTERF ACE DOCUMENTS
SPECIAL REPORTS
PRINTS, MICROFILM, AND REPORTS
~
NASA CENTERS
MSFC ORGANIZATIONS,
CONTRACTORS
Fig. 5-10 Release and Accounting Procedure Flow
COMPUTATION
LABORATORY
I
PRINT MASTERS
REPORT PRINTING
documents (ICDs), ICD input forms, and interface revision notices (IRNs), which,
after review and approval by the appropriate Change Control Board (CCB), are also
forwarded for release. All such documentation enters the release system at the
release clerk's desk in R-P&VE-V, where the documentation record is posted and
maintained. Engineering documents - DRLs, drawings, EPLs, and EOs - are then
checked for technical accuracy, adequacy, and conformance to all MSFC requirements. Following checking, those documents which lend themselves to a computerized
operation - DRLs, EPLs, EOs, ICD input forms, IRNs, etc. - are forwarded to the
Computation Laboratory for keypunching and processing by computer programs.
standard edits are programmed to prevent the entry of erroneous data into the computer data base. If an error is detected, the input will be rejected automatically with
the reason given for its rejection. If all programmed edits are satisfied, a computerprepared document (output) is furnished, which is.the basis for reports such as Release
Records, Engineering Master Parts Lists (EM,PL), ICD logs, etc. The, computerprepared output documents are checked against the original input forms for keypunch
errors by the release center. The reports are forwarded to Field Printing, where they
are reproduced in sUfficient copies to meet the requirements of the various user organizations at MSFC, other NASA centers, and contractor locations. Distribution is the
responsibility of the Documentation Repository. The repository reproduces as much
as it can of the computer-prepared output documents it distributes.
The automated portion of the reporting system, by using the data processing capabilities
of the Computation Laboratory, provides many needed services which cannot easily be
obtained from an entirely manual system. Some of these services are:
•
More rigid and standard edits
•
Greater speed of reporting
•
Greater variety of reports
•
•
Ease of data search and identification
Generation breakdowns on main assemblies
•
Engineering master parts lists
5-32
5.3. 2. 2 Computer Support to Release Center. Several computer programs are in
existence at the Computation Laboratory to support the release center operation. The
following list contains a description of the computer-prepared publications within the
Engineering Documentation Accounting System.
Advance Release. This publication is a complete listing of Advance Release Documentation Release Lists (DRLs), Documentation Release List and Engineering Parts Lists
(DRL/EPLs), and Engineering Orders (EOs) applicable to Saturn I, Block II Vehicles
Saturn IB Vehicles, and Saturn V, S-IC Stages. The information listed in the report
is obtained directly from advance release documents.
Official Release. This publication is a complete listing of Documentation Release
Lists (DRLs), Documentation Release List, and Engineering Parts List (DRL/EPLs),
and Engineering Orders (EOs) applicable to the Saturn IB program.
The' information
listed in the report is obtained directly from the officially released documents.
Engineering Report. This publication is a complete listing in alphanumeric sequence.
of parts, drawings, specifications, and other engineering documentation applicable to
Saturn IB program and Test Instrument Units.
The information listed in this report
is obtained from officially released Documentation Release Lists (DRLs), Documentation Release List and Engineering Parts Lists (DRL/EPLs), and Engineering Orders
(EOs) •
Generation Breakdown (GB). This publication lists all parts officially released to
define a given assembly for a specific effectivity in generic (Christmas Tree) form.
It is published automatically in conjunction with a GB Cross Reference Index. Both
are published on request; the specific assembly and effectivity to be generated are
supplied by the requestor. The GB Index is an index of all parts called for in a given
GB, listed in alphanumeric sequence. The GB Index also supplies the total numbers
of each component required for the end assembly generated.
5-33
Saturn Interface Control Document Log., This publication is a listing, in alphanumeric
sequence by effectivity, of Saturn interface documents and change documents released
and scheduled to be released for Saturn I, Saturn IB, and Saturn V vehicles. The information listed in the report is obtained from the Interface Control Document Log
Input Form submitted to the Engineering Records Unit by the
responsib~e
Interface
Working Group.
Apollo Interface Control Document Log. This log is published for NASA Headquarters
by MSFC and ,is a listing, in alphanumeric sequency by effectivity, of Apollo Interface
documents and revisions released and scheduled to be released for Saturn IB,and
Saturn V vehicles with respect to Apollo. The information listed in the report is obtained from Interface Control Document Log Input Forms (MSDC Form 2053, Rev.
January 1966) submitted to MSFC by the responsible Interface Panels.
Configuration Management Accounting System. This system of programs use s as input
the Configuration Management Accounting Base Data Input Form (MSFC-Form 2572),
the Engineering Change Proposal Input Form (MSFC
Form~2573),
the Installation
Notice Card (l\I,SFC Form 2490) and Configilration Control Boarp Directives ~
~here
are two main output reports, the Configuration Identification Index and the Modification
Status Report.
5. 3. 3 Saturn V Operational Display Checkout System
The operati,on of the Saturn
y Operational DisplayCh~ckout Syste~ is ou~li~ed.
This system is of interest because of ,the. techniques it u,ses to retrieve 35 ~ni film
data.
The present system at MSFC
Astrio~cs
Laboratory and Quality
figured as follows: ' Remote ,consoles, are connected to
(CCC) data processor (DDP-224)
by~ean~
~
Laqora~ory
is con- "
Computer, Control Company
of a Central Logic Unit, (built by Sanders
Associates). The DDP-224 processor is further connected via a hardware/software
data link to an RCA-110A computer which is in turn is connected to the Saturn V
5-34.
fWlction to be checked out. There is a 35 mm slide holder capable of holding 256
slides located in the central logi,c unit. Provision has been made for expansion to
include a second slide bank at a future time.
Each operator console has multiple fWlction buttons and an alphanumeric keyboard.
Each operator may select anyone 'of the 256 slides directly from his console
0
The
slide holder is activated by the DDP-224 by software command and positioned to the
desired slide. ' A high resolution video subsystem (945 scan lines to the inch) then
transmits the contents of the slide to a cathode ray tube in the operator's console.
In this manual mode of operation, the operator must know the number of the slide of
interest. The slide may contain any item of information that the vehicle checkout
engineer may need such as; tabulated lists, standards or circuit diagrams. The
RCA-ll0A computer can also select slides automatically. If, for example, a certain
checkout test procedure yields results in a range pre-determined as marginal the
computer can automatically display the corresponding slide to an operator's console.
The Saturn V Display System is an example of an on-line real-time retrieval scheme
in which both manual and automatic selection of slides are available. The file size
(256 slides) is admittedly small but, with new, computer-addressable microfilm hardware, an on-line retrieval scheme is quite feasible. Its obvious weakness is in the
means of selection of the slide, which assumes that the user knows the relationship
between slide contents and location in the slide file.
5.3.4 Mass Graphic Storage System (Ampex Videofile)
Approximately two years ago
MSFC obtained a Mass Graphic Storage System, a
system more commonly known as a Videofile.
The unit is located at the Brown Engineering facility at HWltsville and was procured to
support PRINCE (Parts Reliability Information Center) and APIC (Apollo Parts
Information Center).
5-35
The unit converts hardcopy into a videotape recording, and adds index coding to the
video tape to facilitate retrieval. Available information on parts reliability maintained
in the above-mentioned information centers is as follows: inspection reports, failure
reports, environmental test reports, material and process reports, etc. These
I
previously mentioned information centers have no interface to the Central Repository
or Engineering Release Center operation of MSFC. There is, however, some overlap
in the information 'stored in the two facilities.
The Videofile system is currently a developmental effort and consequently no file
(data) conversion, from the existing manual scheme to the video tape recording, has
taken place. As a result, no operating statistics are available. The system is rated
as capable of driving up to twenty remote consoles, but in this initial configuration
only the main input/output console is being used.
Further detail on this system will be found in Para. 5. 5. 2. 3. This unit will be modified with a higher resolution camera.
5-36
5.4 RECOMMENDATION
5.4. 1 Engineering Design Data Retrieval - An Overview
If not only the problem of design data but also that of control of the entire cycle of
design, procurement, and manufacturing (or assembly) is considered, the MSFC
activity lends itself to what is generally called a product-oriented data-base system.
In this overall structure, the design data would have a natural place.
Attention is focused on this concept because, as is well known, engineering design
does not terminate with the original releases. It has been estimated that, at LMSC,
a major portion of the design activity takes the form of changes. These changes in
an established design structure must be reflected in activity in procurement and
configuration control, changes in testing and inspection procedures, alterations. of
the drawing structure, etc. and this interdependence clearly calls for a rapid and
convenient access to all these related and affected d . . . .ta.
The product-oriented data-base system would contain, for example, all the technical
information about the Voyager project as it relates to MSFC; that is, the "product"
will be the sequence of Voyager spacecraft. The data-base would include the complete assembly/component structure for the vehicle (each assembly defined in terms
of its next-lower-order assemblies and so on down to the part level) complete with
tolerances, inspection attributes, pointers to related drawings and to descriptions
of components, pointers to related specifications, procedures, and manuals, etc. It
would also contain schedule/status data, usage data (the assemblies using and drawings
specifying or defining a named component), a map of the drawing structure, etc.
From this extensive and constantly updated data bank, specialized programs would
provide answers to questions posed by the functions of design, procurement and ipventory control, assembly (configuration control, status, etc.), and product assurance (at all levels between part and end-item test and checkout).
This data-base system would be an ambitious project but it is .now both technically
feasible and economically justifiable. Its design is beyond the scope of this study,
5-37
but its potential advantages indicate a strong' possibility that such an approach may
be realized by MSFC within the next few years. The storage and retrieval system
specified in this section would constitute a major part of such an overall data base
system.
There are two storage and retrieval problems within the scope of this study. The
first, which is the' subject of the system design, is the large-Bcale problem of dealing
with engineering information at the stage of completed, formal "packages" of drawings,
specifications, test and inspection procedures, etc. This system does not assume
that all or even a large part of its end-item data is in computer storage. What is
retrievable at the computer console is the identification and location of the data, and
a description of how one package. of data relates to others. This problem was
selected for study at this time because it seems probable that the second problem,
dealing with the retrieval of
machine-readabl~
information to be used in the on-line
creation of a design, is neither well·defined nor acute at this time. The second
problem will, however, be the subject of discussion.
5.4.2 Concept,s for a Program to Retrieve
En~ineering
Design Information
5.4. 2. 1 Introduction. There are two fundamental modes of retrieval of recorded
information. The first might be called document reference retrievai, and consists
of identifying a document (a book, drawing, etc.) which contains the needed information. To select a set small enough to be scanned from the set of documents to be
searched, search keys such as titles, authors, contract numbers, and drawing
number's
~re
used.
The second mode is the retrieval of appropriate documents by reference to their
content. The search keys used here depend on relationships of one kind or another.
This might be the relationship of a document content to a standard index/search
term vocabulary, the relationship of the document to an arbitrary description of lmowledge (e. g., the Dewey Decimal System), the relationship of the document to other
documents, etc.
5-38
The first and second modes differ in several ways. The first demands more specialized information about the document, the second about the retrieval system itself.
The first is oriented toward finding a particular document, the second toward the
desired information independent of its representation. They are not, however,
mutually exclusive; in fact, the more powerful retrieval systems use both modes of
operation.
Traditionally, in engineering, only the first mode has been used; engineering drawings
are retrieved by drawing number and further design information is retrieved using
the drawing. This is understandable as it can be very difficult to assign index terms,
which specifically describe content, to non-textual information. In the following
sections a variation on the second mode of retrieval, involving the identification of
design information with the structure it represents, will be developed.
5.4.2. 2 Discussion of the Problem. In any information retrieval area in which the
data base is not pre specified and the potential user population is not clearly defined,
two obvious questions arise: What is to be retrieved, and how? Possible answers
to the first question include end-item information, or references to sources of enditem information, as well as the definition of information categories. Answers to
the second include retrieval by part of drawing number, which implies that the user
is technically sophisticated in the contents of the data base; or such techniques as
retrieval by tracing a drawing structure tree, which can be achieved by a layman.
To treat, first, the problems of the type and extent of immediately retrievable information, this information will be defined as that subset of engineering design data and
references to sources of such data which are in computer-accessible storage. If
cost were no object, all of the data which could be useful to the designer would be
included; but, in a practical system, much of the data will consist of pointers to
sources. These sources are typically drawings, specifications, operating manuals,
and other collections of data too volwnnious to be stored economically in digital form.
The first problem of the retrieval system designer is to decide what data will and
what data will not be made 'iimmediate." A second question is that of the extent of
the data to be considered: Is the entire engineering design file to be made accessible
(directly in the system or indirectly as a reference) via the retrieval system, or
only the high-activity items? What assumptions can be made about the designer's
ability to locate relevant ancillary information without the help of such a system?
An example of such a question is: Is it worthwhile to include an often -used manual
of component standards in the computer storage, or even a reference to the manual,
or can it be assumed that the design engineer knows about the manual and can obtain
his information as 'required?
The second kind of problem is that of available means of retrieval. Obviously, the
techniques chosen for retrieval depend on what is available; it is not so obvious that
they depend on the background of the system user. Most systems for the retrieval
of engineering drawings (and most, for example, the Central Repository, are manual)
depend entirely on one item: the drawing number. This is usually adequate, at
least for the designer working in daily
contac~
with a restricted area
an~
well-defined
set of problems. It is very difficult; however, for anyone with broader responsibilities
(system integration, test and checkout) to get to the specific data he needs. A basis
for an auxiliary tool, to help the latter class of user, is the next subject of discussion.
The question
o~
what information is to be made' "immediate" will be taken up later,
in Paragraph 5.4.2.5.
5.4.2.3 Relationships in Engineering Design Data. There are several kinds and'
degrees of relationships among engineering data. Three of these types of relationships
will be exploited in this retrieval system design.
The first is that between the elements of a "package, " which might be defined as the
data set required to make a complete and meaningful junction requirement at a given
level of detail. It might, for example, consist of a drawing, a specification, and a
parts list, all of which are required for the fabrication of a given component. The
basis of this relationship must be defined in terms of the use to which the package is
to be put, as, for example, a manufacturing package is usually a different set of data
from that implied by the needs of design or those of procurement. In the 'following
the discussion will be restricted to engineering design packages.
5-40
The second is a relation between packages, usually called an interface. The significance of this relationship is that one component, assembly, etc., is connected
directly to another in a more or less complex manner. This interface may be so
complex as to demand a "package" of its own to describe it.
The third, and perhaps the most obvious relation which exists between packages,
is the component-assembly or "goes-into" pattern. It can be represented by a set
of nested boxes, or better, by a tree showing its partial ordering in a hierarchical
pattern.
The interface and component relations between packages can be represented simultaneously in a three-dimensional pattern. A convenient representation is a pyramid,
the apex of which is the name of the construct being described. Successively lower
"layers, " defined by
p~anes
parallel to the base of the pyramid, contain data packages
at levels of increasingly greater detail.
Voyager
- - - - - - - -
Stages
- - - - - -' - - -Major Assemblies
- - - - - - - - Assemblies
- - - - - - - Subassemblies
- - - - - -
~--------------------------~
Components
The assembly-component relationship is thus represented by a vertical trace through
the structure. For example ,'from the apex,
Voyager
Stages
Payload
S-IVB
S-I!
S-IC
Each horizontal plane, then, represents a set of data at a given level of detail,
represented by a set of data packages or by a ,set of component names. The interface
relation between items at a given level can be graphically represented by component
(or package) identifications connected by lines indicating a physical connection or
contact. In the example. above, the interface diagram for the S-II stage at the "stage"
level of detail is Simply
I
S-IVB
~ s-n ~
S-IC
I
where the rectangles identify the components, and the circles the interface specifications. In the case of interfaces which are not described by docwnentary specifications,
the interface circle could contain a list of the types of interfaces shared by the components ; for example, "elec. / mech/hydra" for electrical, mechanical, hydraulic.
Figures 5-11 and 5-12, extracted from the "Nomenclature and Breakdown Charts for
Saturn Vehicle, SA-5" (M-P&VE-EA-62-1, NASA MSFC), depict the type of structure
identified above.
5.4.2 .4 Retrieval by Structural Clues. The following discussion, which is concerned
with the definition of a group of operations to facilitate the retrieval of design data,
5-42
MAIN ASSEMBLIES
VEHICLE ASSEMBLY _ _ _ __
-SATURN10 M 10000
Ie
J L Jl
p:JJJ~
PAYLOAD BODY
ASSEMBLY
IOM20002
I
STAGE
m:
ASSEMBLY
-12:
ll-
~
I
I
S I STAGE
10 M 10002
NOMENCLATURE CHART
-SATURN- ISA-5I
aLOCK II
10 M 10030
Fig. 5-11 Saturn SA-5 Main Assemblies
presupposes that the system user will be working at a CRT console on-line to a computer. He will have a light pen and an alphanumeric keyboard as input devices. The
commands, i.e., EXPLODE, are used to directly address or instruct the computer
through the light pen/~eyboard device, and they and available answers are displayed
in real-time on the CRT surface. The computer will have a large direct-access
peripheral storage in which a body of data, organized as suggested in Paragraph 5. 4. 2. 3
is available.
As shown previously, engineering design packages can easily be reiated to the physical
structure they represent, which in turn can be factored into successively greater
levels of detail. The operation of going from general to specific will be called
EXPLODE. At a given level of detail the relation of interfacing and the identification
of the interface specifications will be displayed when the INTERFACE command is
given .. Both EXPLODE and INTERFACE are used, of course, in reference to
specified data packages.
Another function should be available, to permit the user to perform the opposite of
the EXPLODE operation and move upward from components to assemblies. This will
be called USED IN.
The packages will be identified by number (in some cases the drawing-parts list·
number), by drawing name and by generic name. The search through the structure
will start with the use of the START operation (an operation indicating the beginning
of a search) and one of these forms of package identification. The package name or
number input will be displayed and a fourth operation may be invoked - the DISPLAY
option - which causes a list of the contents of the package, by type of information,
to be displayed. If the data base includes such items as inspection criteria, tolerances,
and specifications, these can then be selected and listed on the CRT by a second use
of the DISP~A Y operation • The DISPLAY function is thus a sequential one, the first
use detailing package contents by item and the second detailing the contents of a
selected item.
There are two cases in which finding the desired starting-point package may not be
an immediate result of the input package identification. The first occurs when the
5-44
....
r------,
.
-- ,
-------------------
~I ...rlCT_
IOIIIM)O()f
,,
... ------~
~
OUTSIDI V[m>ORS
_
STMICTUIU WoNCH
~~.W[Qf
c:::J~
~::::ICS
Fig. 5-12 Breakdown of Saturn SA-5 Payload
and Stage I Assemblies
5-45
input was a generic name such as "power supply." In order to discriminate between
power supplies, the list of such components in the structure would be displayed, by
drawing name and number, to permit the user to choose the individual item he needs.
If he is still in doubt after seeing the list, the EXPLODE, INTERFACE, and USED
IN operations will help him to isolate the desired package.
The second case of ambiguity arises when the user identifies a component which is
used in more than one place in the structure as the starting point in. his search.
Since he has just used the START operation, the retrieval system will assume that
the user is not interested in the component designated but is using it only as an entry
to the structure.
(If this is not the case, and the component itself is the end-item of
the search, the user can enter its identification again to get the package.) The next
display will identify all the next-higher assemblies using the given component.
If this still does not discriminate sufficiently between uses of the component (the
same higher assembly uses the same package more than once) the identification of
the higher assembly and use of the USED IN operation will move the process one
step upward. This can be repeated until the appropriate larger assembly is identified,
at which point repeated use of the EXPLODE operation can be used to get to the unique
desired starting point.
A function which permits availability of data to the user about more than one structure
(for example, Saturn V as well as Voyager might be represented in the files) would
be called SELECT. At any time the system userwishes to start a new search,or
change the subject of his search from one structure to another, .he uses the SELECT
operation. The result will be a display of the names of the structures represented
in the data file. He then' indicates the' option he 'desires and uses the START operation
to begin his' search.
5-47
In swnmary, these operations, operands, 'and their consequences are as follows:
on
Operation
Operates
SELECT
Total data base
Selects the pyramid
relevant to a given structure
START
Structure data
base
,Selects a starting point in
the structure
.
DISPLAY
Data package or
item in package
Displays a list of items in
package, then, if desired,
details are of the items
listed
EXPLODE
Data' package
Displays component packages'
USED IN
Data package
Displays identification of
next-higher assemb~y
INTERFACE
Data' package
Provides a graphic display
of packages with direct
interfaces with that given,
and interface control
information if, applicable
Results
Several further operators are needed. The first is not strictly a part of the retrieval
function'but serves to record the interim or final results of a search. 'The PRINT
operation causes the contents of the display to be transferred to a permanent
~e,~iwn.
Two other operators, not strictly needed. to carry out a search but without, which ,
searching is .extremely tedious, are called RE~TORE and TRACE. The RESTO~E
function brings back the.; dis play seen just prior to the current one. The. TRAC E
,
,
operation displays.;a list of. the last ten steps the, searcher has made, identifi~~ by
package nwnber and name, to permit the searcher to return to a ~revious step. !his
is extremely useful when the user discovers he has taken a wrong turn in tracing
through the data base.
5.4.2.5 File Contents. In Paragraph 5.4.2.2 two prinCipal problems were discussed;
the problem of selecting the data to be placed in the retrieval file, and that of
5-48
specifying a retrieval technique. The ensuing description of a solution to the second
problem indicates some of the constraints which must be placed on the solution of
the first.
The file must be organized by structures, and its elements are what have been called
data packages. It must be complete in the sense that there is an uninterrupted linkage
from the most general item (the name of the structure) to the most specific level
included.
The data package, identified by a number and a name, contains at least a list of components (if it is not itself a single indivisible component). It will contain, if appropriate, the locations of the associated drawing, specification, standards, tolerances,
inspection criteria, etc., if these items of information are not included in digital
storage. The decision as to which elements should be in literal form in digital
storage and which should be represented by identifying information is an economic
one. Among the parameters which will influence this decision are:
•
The number of data packages needed to describe the structure
o The volume of data per item of information (e. g., specification)
•
The cost of converting these items to digital form
•
The cost and size of the peripheral storage de~ices needed to maintain
the files
•
Whether or not each file has to be continuously available (that is, .whether a
storage device such as a data cell can be loaded with specified Jiles at specified times)
Among the data will be packages specifying interfaces as well as those specifying
components. Their retrieval can be achieved directly if they· are identified by
drawing number or name, or by association if they are identified by finding the components they relate to and the INTERFACE operation is executed.
5.4.2. 6 File Creation and Maintenance. The file would be large· for structures of
the scale of Voyager. Its size would depend on the amount of direct data included,
as opposed to the number of data references, but even at a minimum it would be large
5-49
enough that initial on -line input of the entire volume of data would be extremely expensive. On the other hand, the file maintenance could well be performed on-line.
The initial data base would then be loaded as a batch process, and changes to the
file would be performed using a graphic console. To assure that only authorized
changes are made, only designated consoles would be used to update the file.
The update operations, which would be implemented in such a way that they would be
available for either on-line or off-line use, would be ADD, DELETE, REPLACE, and
INSERT, and they would refer to one of three categories of data - DATA, ITEM, or
PACKAGE.
In each use of one of the operation/data-category pairs, a data-set identification
is required. If the data set is an entire package,
the package number, is s¢ficient.
.
In the case of smaller sets the first step in updating is to retrieve and DISPLAY the
package which contains them. If the data set is an ITEM, the corresponding item
identification number i~ then used to point to the item to be modified. If the set is
smaller than ~he ITEM level, the DISPLAY operation is used again, .this time in
relation to the item containing the data set to be altered. The character at which the
alteration is to start is indicated by light-pen, the corresponding operation button is
pressed (e. g., INSERT), the data-set button DATA is pressed, and the entry is' made
In summary:
Operation
ADD
DELETE
REPLACE
INSERT
Data Set
DA T A - Character in item in package
I
ITEM - Item in package
PACKAGE - Package I.D.
The update operation would include a safeguard to protect users of the file. This
might take the form of actually inhibiting all references to the part of the file being
updated, or might merely inform anyone examining a package being altered that
alterations are in progress.
5-50
5.4.2.7 Retrieval Equipment Configuration. It has been implicitly asswned that
retrieval would be performed using a CRT console with light pen, having an alphanwneric keyboard and a few function buttons (which might be selected from little-used
characters on a standard keyboard).
In this configuration, after an initial alphanwneric entry to the data structure was
achieved, the rest of the search and data extraction process could be carried out
using the light pen to select the appropriate items and the function buttons to perform
the desired operations.
The only consoles which must have graphic capability are those which are assigned
to update the files. Consoles used strictly for retrieval do not need graphic-input
capability and are called upon to display only very simple patterns (boxes connected
by lines).
5.4.2. 8 Notes and Comments on RED!. There has been no discussion of "effectivities" in this conceptual outline, but they would certainly be included in the data
packages. This would not only be a tool for the designer but would make the retrieval
system a major tool in configuration control, as well as to facilitate historical reviews
on 'previously assembled structures.
The system as outlined is not complete in the sense that it supplies the designer with J
all the information he might conceivably use but it could be supplemented with other
systems' supporting' ancillary data sets. Some of these are discussed in the next
subsection.
The previous descriptions imply that the structure to be factored, the "top assembly",
might be a complete vehicle. It should be noted that this idea is not an essential part
of the plan, might be preferable to start at the "main assemblies" level, as for
example, one sequence of Voyagers might use a different booster from that used by
another.
5.4.2.9 Retrieval of In-Process Design Information. This problem, cited in
Paragraph 5.4.1 as the "second problem", is that of supplying the user of an on-line
machine-aided design system such as Digigraphics with the data he needs to work
efficiently. This data includes conceptual versions of his drawings, related drawings,
tolerances, standards, inspection attributes, preferred parts lists, etc., and is
similar to that which can be handled by the RED! system and its retrieval can be
accomplished in the same manner.
Other data i~ involved however, and the nature and structure of this further data
indicate that other retrieval techniques will be needed to deal with them. Among the
more obvious categories of design data are those contained in engineering handbooks,
drafting manuals, etc. The retrieva~ of this kirid of data is most naturally performed
by the display of a back-of-the-book type of alphabetized index. This technique, if
not entirely adequate, could be supplemented. by coordinate indexing. This. would
require that skilled indexers assign descriptive terms from a controlled vocabulary
to major topics covered in the handbook.
The current u~ers of the LMSC computer graphics system have indic.ated some more
specialized requirements calling for more exotic .approaches to retrieval. The.
group doing orbital analysis foresees the storage of Venus swingby (to an~ from Mars)
flight parameters and summaries of past performance studies. The structural .
a~alysis.
group specifies a system which will permit the d~sign engine~r to retrieve
reports on performance data. by structure type (honeycomb, waffle, etc.), ~~ well as
the kind of data to be handled in the RED! system. The net~prk analysi~ gr~up would
like to have a Preferred Parts Handbook in computer storage. They require that
the results of network analyses, as well as the parameters of derived equivalent
circuits, be stored and retrievaQle. They stated, incidentally,
th~t in. their opinion
. ..
.
,
,
~.
".
. '
the retri~val of their data ,co~d be adequately pet10rmed by reference to, the functional use or discussion of the circuit.
. .
t.:
Many of these requirements can be met by a "structural" system such asREDI. Mqst
of the remainder can be met by a system designed to retrieve data. from a ha...'1dbook.
The balance indicate a need. for a library type of reference retrieval system based on
5-52
coordinate -indexing. The first and third of these techniques have been implemented,
the third widely, but methods fqr handbo9k data retrieval are still in the research
stage.
A final note on this problem concerns archival rather than work-in-progress storage
of this design data in digital form '. References to designs having a direct bearing on
structures could be added to the corresponding files in the REDI system, but separate
indexes should be prepared for such special data as swingby orbits.
5.4.3 Program Description
Following· is a description of a set of program segments designed to implement the
retrieval system developed in Paragraph 5.4.2. It is assumed that these program segments would operate under the control of a -real time, multi-processing monitor such
as the Univac EXEC8, the IBM system/360 OS, the CDC Digigraphics system, or
the LMSC LACONIQ (for System/360). A program segment is a computer program
which is called into main storage by the monitor, performs one or more functions,
and then returns control to the monitor. Real-time programs typically consist of a
number of program segments, each of which is relatively small.
It is assumed that the monitor will permit a program segment to communicate with
the monitor and with other program segments. For example, a program segment
should be able to designate its successor segment and it should be able to transmit
data and program flags to other program segments servicing the same input-output
terminal.
The following items are terms and abbreviations used in this section:
•
Package number (PKN) is defined as the number used to identify a package
or item in a package.
.
•
Basic PKN is defined as the portion of the PKN to the left of the "dash".
•
Data sent to an input-output terminal as a unit shall be called a message.
5-53
The program logic was developed with the assumption that data would be organized
into four files. The data need not necessarily be segregated in the storage device
but each file must have its own directory. Furthermore, each structure shall have
its own set of directories. The following is a description of the files.
I
a. File A: This file contains lists of the "dash number" versions for
each basic PKN. Entry to the file is by basic PKN. Individual records
contain a description of the item and a listing of the dash numbers with
their effectivities.
b. File B: This file contains the PKN detail. If the PKN is a document
such as a specification, the record will contain the text of the document.
If it is a hardware item, the record will contain parts list information
such as description, unit weight, a list of the next level assemblies
of which this PKN is a component, with quantity per assembly, and
effectivity. In addition, the record will list the items in the package
and such information as description, quantity per assembly, drawing
releas~
status, etc. Entry to the file is by PKN .
c. File C: This file contains interface information. Entry is by PKN.
q.
File D: This is an "inverted file" which provides a
cross~reference
to PKNs by their generic names. Entry is by generic name. Records
contain basic PKNs and descriptions.
5.4.3. 1 Pr.ogram Segment Descriptions.
Segments SLl and SL2
Function: SELECT
The purpose of this segment is to permit the user to select a structure. Once lhe
selection 'is made it is stored in an area associated with the input-output terminal
where the selection was made . Each time data is retrieved by other program segments this area
is examined to determine the structure.
selected an error message would be sent to the console.
5-54
If no structure had been
The user may change to another structure by pressing the SELEC T function button
at any phase of his inquiry. Selection of a new structure will cancel all requests for
data on the previous structure still waiting to be processed.
Only one structure may be selected at a time.
Segment Outline
Segment SL1 (Ref. Fig. 5-13)
1.
Notify the monitor that program segment SL2 is the successor segment.
2.
Send a pre-formatted message listing the various structures and their
codes to the input-output terminal.
Segment SL2 (Ref. Fig. 5-13)
Initialize the request queue. This is a request queue maintained by the
program, not the monitor. Initializing the queue will cancel any requests
waiting to be processed.
Segment S
Function: START
The purpose of this program segment is to permit the user to initiate a search. The
user may enter anyone of the following type of parameters in conjunction with the
ST ART function button:
1.
A PKN with a "dash number"
2.
A basic PKN
3.
One of a selected list of package descriptions
4.
One of a selected list of generic names such as fuel pump, heat shield,
bracket assembly, etc.
5-55
SEGMENT
S~t
.
NOTIFY MONITOR
THAT SEGMENT
SL2 IS SUCCESSOR
SEGMENT
.
. .
SEND PREFORMATTED
MESSAGE LISTING
STRUCTURES IN FILE
SEGMENT SL2
INITIALIZE REQUEST .
QUEUE, CANCELLING
ANY PENDING REQUESTS
MOVE NEW STRUCTURE
CODE TO COMMUNICA TIONS AREA
Fig. 5-13 Logical Flow Chart Segments SL1, SL2
5-56
The segment responds by displaying a list of PKNs on which the user may operate
with such function buttons as START, DISPLAY, EXPLODE, USED IN, INTERFACE,
etc.
If the user enters a PKN with a "dash number, " the segment will display the descrip-
tion and effectivity of the PKN. If the user enters a basic PKN, the segment will list
all of the "dash numbers" (including the original, which has no "dash number") with
their effectivities.
If the user enters a package description, the segment will display the basic PKN for
that segment. If he enters a generic name, the segment will display the basic PKNs
indexed by that name and their descriptions.
Segment Outline (Ref. Fig. 5-14)
1.
If the input parameter is not a PKN, with or without a "dash number, " go
to Step 7.
2.
Separate the PKN into its basic PKN and "dash number" if it has one.
3.
Retrieve the basic PKN record from File A.
4.
Format header portions of message. This will include the description.
5.
If the input PKN was a basic PKN, go to?
If it was not, position the data
for the PKN in the message, send the message to the input-output terminal,
and exit to the monitor.
6.
The input parameter was a basic PKN.
List all of the "dash numbers" in
the message, send the message to the input-output terminal, and exit to the
monitor.
7.
The input parameter was not a PKN. If the input parameter is in the
selected table of package descriptions.
Pick up the PKN and transfer to 3
for further processing. If not, continue to 8.
8.
By elimination, the parameter is assumed to be a generic term. Retrieve
its record from File D.
9.
Format the header portion of the message.
List the basic PKNs catalogued
under the generic term with their descriptions, send the message to the
input-output terminal, and exit to the monitor.
5-57
YES
0r
2
SEPARATE PN
INTO. BASIC PN
ND "DASH NO.'
RETRIEVE BASIC
PNRECo.RD
FROM FILE A
Fo.RMAT
HEADER
PORTIo.N o.F
MESSAGE
POSITION
ALL ITEMS
IN MESSAGE
NO.
c.n
I
c.n
o.SITION DATA
Fo.R THIS PN
IN MESSAGE
PICK UP PN
FRo.M TABLE
ex>
RETRIEVE
FILE D RECo.RD
Fo.R THIS TERM
Fo.RMAT
HEADER
PORTIo.N
o.F MESSAGE
POSITION THE
BASIC PN'S
IN MESSAGE
t----------------l
SEND MESSAGE
TO. TERMINAL
Fig. 5-14 Logical Flow Chart, Segment S
Segment D
Function: DISPLAY
The purpose of this segment is to display the File B data for a given PKN. The output
format will vary with the nature of the PKN. For example, if the PKN were a document, and the document was in digital storage, the text of the document would be
formatted and displayed. If it were a parts list, the header and item detail would be
formatted and dis played.
Segment Outline (Ref. Fig. 5-15)
1.
Retrieve the File B record for the input PKN.
2.
Determine the type of record and format the message in an appropriate
manner.
3.
Send the message to the input-output terminal and exit to the monitor.
Segment E
Function: EXPLODE
The purpose of this program segment is to select and list the components of a given
PKN.
Segment Outline (Ref. Fig. 5-16)
1.
Retrieve the File B record for the input PKN.
2.
Format the header portion of the message. Select an,d position the
items in the message.
3.
Send the message to the input-output terminal and exit to the monitor.
5-59
RETRIEVE FILE B
RECORD FOR
INPUT PN
DETERMINE
TYPE OF
RECORD.
FORMAT
MESSA'GE
SEND MESSAGE
TO TERMINAL
EXIT
Fig. 5-15 Logical Flow Chart, Segment D
5-60
J--------~
RETRIEVE
FILE B RECORD
FOR INPUT PN
FORMAT
HEADER
PORTION OF
MESSAGE
t--------~
c.n '
I
(j)
I-'
POSITION
HARDWARE ,ITEMSt-----~--~
IN MESSAGE
r-_______~~SENDMESSAGE
TO TERMINAL
Fig. 5-16 Logical Flow Chart, Segment E
t--------.l
EXIT
Segment U
Function: USED IN
The purpose of this program segment is to list next higher level assemblies of which
a given PKN is a component, together with the quantity per assembly and effectivity.
Segment Outline (Ref. Fig. 5 -17)
1.
Retrieve the File B record for the PKN.
2.
Format the header portion of message. Select and position the "next
assembly" data in the message.
3.
Send the message to the input-output terminal and exit.
Segment I
Function: INTERFACE
The function of this program segment is to format ·and display a diagram in response
to the INTERFACE function button.
Segment Outline (Ref. Fig. 5-18)
1.
Retrieve the File C record for the PKN.
2.
Initialize index used to step through chain of interface.
3.
Construct first box and position the PKN in it.
4.
If the last item has been processed, determine the section of the construction grid in core which will be sent to the
input~output
terminal, send the
"message", and exit to the monitor.
5.
Position first circle. If the record contains data describing the interface,
such as a document
~umber
or descriptive term, position the data in the
circle.
5-62
RETRIEVE
t - - - - - - - - - - a... FILE B RECORD
FOR INPUT PN
POSITION NEXT
ASSEMBLY
INFORMA TION
IN MESSAGE
t - - - - - - -..........
t----------8-'
SEND MESSAGE
TO TERMINAL t---.,.----~
Fig. 5 -1 7 Logical Flow Chart, Segment U
FORMAT
HEADER
PORTION OF
MESSAGE
EXIT
RETIUEVE
FILE C RECORD
FOR INPUT
PN
SET
i = 1
POSITION
BOX iON
CONSTRUCTION
GRID
POSITION
PNIN
BOX i
YES
CONSTRUCT
LINE CONNECTING
BOX i TO
CmCLE i
POSITION
CmCLE
i
CONSTRUCT
LINE CONNECTING
CmCLE i TO
LOCATION OF
BOX i + ·1
DEFINE
DISPLAY
GRID
YES
POSITION
DATA IN
CmCLE i
10
_I~~TI
SEND
DISPLAY TO
TERMINAL
-0
Fig. 5-18 Logical Flow Chart, Segment I
6.
Construct lines between first box and first circle and between first circle
and the boundary of next box.
7.
Increment index and transfer to 3 to process the next item.
Segment A
Function: ADD
The purpose of this program segment is to add DA TA to an item, add an ITEM to a
package, or add a complete PACKAGE. The segment is called by the ADD function
button. The scope of the operation is determined by the choice of one of the following
DATA SET keys: DATA, ITEM, or PACKAGE.
Segment Outline (Ref. Fig. 5-19)
1.
Identify the data set key. If it is PACKAGE, go to 7; if not, go to 2.
2.
Retrieve the File B record for the input PKN.
3.
Identify the data set key. If it is DATA go to 4. If it is ITEM, add the new
item to the record and go to 6.
4.
Data set key was DATA.
Locate the item in the record.
5.
Add the new data to the item.
6.
Return the updated record to File B and exit to the monitor.
7.
Data set key was PACKAGE. Analyze input PKN. If it has a "dash number"
go to 11; if it does not, go to 8.
8.
Create a new File A record for the basic PKN.
9.
Add the new or updated record in File A.
10.
Create File B record, deposit it in the file and exit to the monitor.
11.
Retrieve the File A record for the basic PKN. Add the new "dash number"
to the record, return the record to the file and transfer to 10.
5-65
r---_~
RETRIEVE
FILE B
RECORD
LOCATE
ITEM
5 t---~ ADD DATA
TO ITEM
ITEM
RETRIEVE
FILE A
RECORD
RETRIEVE
FILE A
RECORD
ADD NEW
"DASH NO"
TO RECORD
ADD NEW
RECORD
TO FILE A
RETURN
RECORD
TO FILE A
f-:\
--\V
ADD ITEM
TO RECORD .........
CREATE
FILEB
RECORD
SEND
RECORD
TO FILE B
Fig. 5-19 Logical Flow Chart, Segment A
RETURN
RECORD TO
FILE B
Segment DL
Function: DELETE
The purpose of this program segment is to delete DATA from an item, an ITEM
from a package, or a complete PACKAGE. The segment is called by the DELETE
function button. The scope of the operation is determined by the choice of one of
the following DATA SET keys: DATA, ITEM, or PACKAGE.
Segment Outline (Ref. Fig. 5-20)
1.
Identify the data set key. If it is PACKAGE go to 7; if it is not, go to 2 ..
2.
Retrieve the File B record for the PKN.
3.
Locate the item to be operated on.
4.
Identify the data set key. If it is DATA go to 5. If it is ITEM delete the
item from the record and go to 6.
5.
DELETE the selected data from the item and reformat the remainder of the
item if necessary.
6.
Return the record to File B.
7.
Data set key was PACKAGE. Retrieve File A record for basic PKN.
8.
Determine whether input PKN has a "dash" number. If it does not, go to
10. If it does, delete the "dash" number item from the record and return
the updated record to File A,
9.
10.
Delete the record from File B and exit to the monitor.
Input PKN does not have a "dash" number. Delete the File B records for
each of the "dash numbers, " including the original PKN, which would have
no "dash number. "
11.
Delete the File A record and exit to the monitor.
5-67
./"-..
ISDATA'N~
SET KEY __
'PACKAGE"
?
>
2
RETRIEVE
FILE B RECORD
FOR INPUT PN
DELETE
DATA FROl\I
ITEM
LOCATE
ITEM
DELETE
ITEM FROM
RECORD
RETRIEVE
FILE A RECORD
FOR BASIC PN
~
~
DOES
INPUT PN
HAVE A
"DASH NO?"
........"
1--------..
RETURN
RECORD
TO FILE B
DELETE
"DASH NO"
FROM FILE A
RECORD
RETURN
UPDATED
RECORD
TO FILE A
DELETE
RECORD FROM
FILE B
NO
DELETE
EACH OF
THE FILE B
RECORDS
DELETE
THE FILE A
RECORD
Fig. 5-20 Logical Flow Chart, Segment DL
EXIT
Segment R
Function: REPLACE
The purpose of this program segment is to revise DATA in an item, replace an ITEM
in a package, or replace an entire PACKAGE. The segment is called by the REPLACE
function button. The scope of the operation is determined by the choice of one of the
following DATA SET keys: DATA, ITEM, or PACKAGE.
Segment Outline (Ref. Fig. 5-21)
1.
Identify data set key.
If it· is PACKAGE, go to 7; if not, go to 2.
2.
Retrieve the File B record for the input PKN.
3.
Locate the item which is to be revised or replaced.
4.
Identify the data set key.
If it is DATA, go to 5.
If it is ITEM, delete the
old item from the record, insert the new item, and go to 6.
5.
Data set key was DATA. Reformat the item, substituting the new data for
the old.
6.
Return the updated record to File B and exit to the monitor.
7.
Data set key was PACKAGE.
Create the new File B record and transfer
to 6.
Segment IN
Function: INSERT
The purpose of program segment IN is to create or revise an interface record
(File C). It must be preceded by a "clear screen" or INTERFACE command.
Figures, such as boxes, circles, lines, etc. may be "drawn" by the light pen and
alphanumeric data may be entered by the terminal keyboard.
5-69
RETRIEVE
FILE B
RECORD FOR
INPUT PN
REFORMAT
ITEM
SUBSTITUTING
NEW DATA
LOCATE
ITEM
ITEM
6
en
I
-:J
o
CREATE NEW
FILE B
RECORD
DELETE OLD
ITEM FROM
RECORD
INSERT NEW
ITEM IN
RECORD
Fig. 5-21 Logical Flow Chart, Segment R
SEND
RECORD
TO FILE B
Segment Outline (Ref Fig. 5-22)
1.
Determine whether there is a record for the input PKN in File C.
If there
is not, transfer to 6 to create a new record. If there is, go to 2.
2.
Retrieve record for input PKN from File C.
3.
Identify changes to be made. The technique for analyzing the graphic input
w ill depend on the particular hardware and software used.
4.
Reformat the record to reflect these changes.
5.
Return the updated record to File C and exit to the monitor.
6.
New record. Create a new record from the input data.
7.
Add this new record to File C and exit to the monitor.
5.4.3.2
Typical Display Formats. Figures 5-23, 5-24 and 5-25 show a typical con-
sole display screen representation resulting from the operations described above.
Figure 5-23 shows a typical interface structure diagram resulting from the INTERFACE
operation. Figure 5-24 could represent the final response to an inquiry for a digitized drawing and Fig. 5-25 shows a typical response to an inquiry for a data package
structure.
5.4.4
Implementation: Data Base
5.4.4. 1
Introduction. The creation of a data base to implement the REDI system
for the retrieval of design documents of MSFC could be performed in two steps. The
first depends on a maximum use of existing machine-readable information and the
second is that of adding the further data needed for the complete implementation of
the retrieval system.
5.4.4. 2
Data Base Requirements. In order to prepare a data package it will be
necessary to examine the master documents to identify their supporting documents,
such as process specifications, test specification procedures, inspection attributes,
etc. It is then necessary to relate these data packages sequentially in their componentassembly chains and laterally in terms of item interfaces.
5-71
RETRIEVE
RECORD
FROM
FILE C
IDENTIFY
CHANGES
TO BE MADE
REFORMAT
RECORD TO
REFLECT
CHANGES
RETURN
RECORD TO
FILE C
CREATE
NEW
RECORD
01
I
...::J
~
EXIT
ADD NEW
RECORD TO
FILE C
EXIT
Fig. 5-22 Logical Flow Chart, Segment IN
STAGE S-II INTERFACE
S-IVB
STAGE
S-U
STAGE
Fig. 5-23 Typical Display - Interface Structure
5-73
S-IC
STAGE
MAIN ASSEMBLIES - SATURN VEHICLE
I
PAYLOAD
ASSEMBLy--1
INSTRUMENTATION
UNIT ASSEMBLY
lOM20000
Fig. 5-24 Typical Display - Digitized Drawing
5-74.
PACKAGE BREAKDOWN
PKN 486 - STAGE IC SATURN V
STAGE IC
ASSEMBLY NO.
10M46504
I
HARDWARE
ASSEMBLIES
PKN 220
I
I
I
I
HYDRAULIC
ASSEMBLY
PKN 406
ELECTRICAL
ASSEMBLIES
PKN 420
FUEL
ASSEMBLIES
PKN 419
I
I
I
DOCUMENTATION
PKN 106
SPECS PROC
INTERFACE
PKN 106
I
DRAWING
PKN
104
I
TEST AND
INSPECTION
.PKN 100
Fig. 5-25 Typical Display _. Structure Breakdown
5-75
DRAWING/
EPL
PKN 101
5.4.4.3 Existing Machine -Readable Data. The pertinent machine readable documents
used by the MSFC are identified in Table 5-3. The "next assembly" data entry enables
the creation of the "used in" file and the separate engineering parts lists provide the
basic "explode" capability. The existence of a manufacturer code identification may
point to interface documents that exist as the result of the documentation requirements
imposed on the contractors.
Not all the documents shown are pertinent to the retrieval system implementation; the
full list is given for completeness. Some documents of special interest are the
Generation Breakdown or "Christmas Tree" list which provides direct identification
of the drawing level, the specific program, and the vehicle number. The Saturn and
Apollo Interface Control Document Logs provide Contract End Item (CEI) identification
plus release status. The CEI entry enables the retrieval process to proceed directly
to the highest assembly level. The Engineer,ing Parts List and Separate Parts List
(EPL/SPL) identify the operating division responsible for preparing the drawing as
well as naming the des ign engineer.
Since the engineering function is predominantly one of processing
sy~tem
changes, each
drawing in the system typically includes several Engineering Orders (EOs), so that a
majority of the supporting documentation consists of EO transactions. The machinereadable document generated when such a transaction takes place represents a
primary source of data for entry in the REDI system.
5.4.4.4
Use of Existing Data Systems as Retrieval Aids. The use of existing re-
ference documents to provide entries to the REDI files could be a signifi'cant tool.
One such reference is the Apollo Document Index, Section 2 , TALK (Titles Alphabetically Listed by Keywords), which provides direct reference to interface documents,
system reports covering the evaluation of systems and subsystems, and test reports.
5-76
Table 5-3
MACHINE READABLE DOCUMENTS
DATA ENTRY
OTHER
DOCUMENT NAME
ENGINEERING MASTER
A 10 (AN)
5 (N)
20 (AN) 12 (N) 4 (AN) 3N 3 (AN) 3 AN
18 (N)
A
A
5 (N) 6 (N)
6
PARTS LIST (EMPL)
x
ADVANCE/OFFICIAL
x
X
x
7 (AN)
x
x
(AN) 2N
3N
x x
x
x x
X
x x
x x
A 6N 6N 3N 6N
x x
x x x
x x x x x
x x
x x x
x
DOCUMENTATION RELEASE
LIST (DRL)
GENERATION
x
x
x
x
x
x
x
x
x
x
x
x
x
2N
x
x
2N 3
BREAKDOHN (GB)
CROSS REFERENCE LISTING
x
x
X 10 (AN)
OF GENERATION BREAKDo,vN
SATURN INTERFACE CONTROL
x
APOLLO INTERFACE
X
X
DIV
40 (AN)
DOCUMENT LOG
x
x
x
X
X
DIV
40 (AN)
CONTROL DOCUMENT LOG
ENGINEERING PARTS LIST
HEADER
PART lD
X
x
x
x
x
x
x
x
x
x
x
x
x
:20 (AN) X
DIV X
X
X
X
x
X
x
X
DItA\\'I~G ZO~E
x
DOCUMENTATION RELEASE
LIST
HEADER
x
x
x
x
x
x
x
x
x
x
x
x
x
DIV X
x
x x x
x
AUTOMATED ENGINEERING
P.-\HT BEING
ORDER SYSTEM (EO)
CIL\~GED
SHEET A
x
x
x
X 9 (AN)
x
x
x x
~------------------------+-~----~~---+----~----+----+--1----4~--~---r--+---~~---1--1-1----+--~~-+--r--1----+--r--~-+~-
-
-----~+-----------~
SHEET A
X
X
X X
X
,
r-------------------------+-~----~~--_+----~----+_--_+--~--~r_--r_--_+--+_--r_~--_1--~~--_+--r_+__+--~_1----+__r--~_+_1--r_--~
SHEET B
X
X
X
x
x
IDELETE
ADD
}
,
ITE~IS
I
5.5 HARDWARE STATE-OF-THE-ART
5. 5. 1 Mechanized Pictorial Storage and Retrieval
In surveying the equipment for mechanized pictorial storage and retrieval, a startling
variety of microforms and means for retrieving these forms is found.
For example,
the tabulation given in Van Dam (Ref. 5-1) includes 20 different systems.
Other
exhaustive listings can be found in Refs. 5-2 and 5-3. The systems described here
are typical of some of the larger ones, and have been selected to' represent a variety
of microforms and retrieval approaches. Good general reviews of the microfilm
field can be found in Refs. 4 and 5.
5.5.1.1 FMA-Filesearch Device. The Filesearch searches a 1000-ft roll of 35mm
film with a maximum of 32,000 document pages per reel. The pages are recorded at
a reduction of 25:1, and optically encoded indexing information accompanies each
film frame. A 32, OOO-page reel can be searched at 200 ft per minute (6400 pages),
and output may be displayed, printed out as hard cOJ?Y, or recopied onto roll film.
The search operation consists of typing a search descriptor or nUll1ber in a punched
card and then using the card to set up the search. pattern in the machine.
5.5.1.2 Eastman Kodak Minicard System. The Minicard subsystem of the Defense
Intelligence System, one of the largest and most expensive systems developed, uses
a unit-record film chip, 16 by 32 mm, as the storage medium for up to 12 pages of
input and associated photographic indexing data. The index codes are at. one edge of
the film chip in a matrix of from 252 to 2730 code bit positions. The film chips are
arranged by their respective subject areas, which divides the file into pre-ordered
categories which can then be searched separately without resorting to a sequential
sort of the entire file. The file may be searched at 1200 chips per minute (up to
14,400 pages). The question statements are supplied on paper tape, and logic is
stated by plugboard wiring.
5. 5. 1. 3 Mosler Selectriever. The Selectriever provides random access to anyone
of 200,000 unit documents - tab cards, aperture cards, or·microfiche cards - in
5-78
less than 10 sec. The 3-1/4 by 7-3/S-in. card is coded with 35 round holes along the
bottom edge, and holes are notched out to identify the card for automated handling.
A
cartridge holds 100 unit documents and is housed in one of two parallel honeycombed
walls, each holding 1000 cartridges. A lightweight cartridge retrieval mechanism
operates between the two walls, moving at high speed to retrieve cartridges in response
to commands entering the system via keyboard, punched paper tape, or computer
interface. The time from command to readout averages 6-1/2 sec; the card is automatically returned to its cartridge; and the cartridge is returned to its storage location in 3 sec.
5.5.1.4 Houston-Fearless Microfiche Retrieval-Display Unit. The Houston-Fearless
card retrieval-display unit stores 750 microfiche cards containing 70, 000 pages of
information. Ten category pushbuttons, each corresponding to a "book" of information, select the classification topic. A" section" set of keys selects the section of
the book. The section index is automatically displayed on the screen so that the
actual page of information may be selected. The entire selection cycle will not
exceed four sec for an operator who has minimal training.
5.5. 1. 5 Magnavox Magnavue System. The Magnavue system of the U. S. Army
Missile Command uses a 35 mm by three-in. -long mylar base film chip with a code
field that can store SO alphanumeric characters (560 bits). The chips are stored
300 per magazine in 300 magazines, providing line random access to 900,000 chips.
The control unit can handle two such files for a total of 1, SOO, 000 chip capacity. The
photographic codes on the request card are matched against the codes in the master
file, and the "hits" are photographed onto a working film or copy card. Response to
"random" inquiries on the Magnavue is performed in 60 sec.
5. 5. 2 New Storage Concepts
The advantages of microfilm as a storage medium include an established and relatively
inexpensive technology; compact storage; many shapes, forms, types of film; easy
reproduction; and permanance. The single disadvantage, hO'Yever, is the nonerasability of the medium. In an effort to overcome this liability and to create an even
5-79
larger stOrage capactiy (or real-time processing), new media have been investigated,
including thermoplastic tape, video (magnetic) tape, and photochromic materials.
5. 5.2.1 Thermoplastic Film. For several years the General Electric Company has
been investigating a type of film lmown as Thermoplastic Film. This film, like the
Photochromic Film, requires no separate development process and is erasable and
reusable and can" accommodate high reduction ratios.
Recording on this film is accomplished by an electron beam operating in a vacuum.
The electron beam forms a charge pattern on the thermoplastic, and the plastic's
insulating qualities cause the charge to dissipate slowly. While the charge pattern is
on the surface, the thermoplastic is"liquified by heating to permit the electrical
attractive forces between a conducting layer in the film and the electrons in the charge
pattern to form distinctive impressions in
t~e
thermoplastic material. When the film
cools, the impressions are solidified virtually instantenously to form the image.
Xerox Corporation is conducting research with a material that works in much the
same manner as the G. E. Thermoplastic Film and has the same basic characteristics.
Xerox calls its process "Frost. "
5.5.2.2 Photochromic Storage. Photochromic materials (PCMI) are light-sensitive
organic dyes placed as a molecular dispersion in suitable coatings and applied to a
surface. With photochromics, the molecular coating can be exposed with ultraviolet
radiation and erased with white light. It is this reversible characteristic that makes
PCMI attractive. Furthermore, it is grain free, thus permitting the recording of
very high density micro-images.
A reduction ratio of 200 to 1 allows for the storage of 2500 pages on one three by five
in. transparency. At present, the PCMI process to produce micro-images is available at National Cash Register on a service bureau basis. A three by five inch master
plate costs $250 with contact prints made from the master at $0.50 to $1 each depending on quantity. Thus, there is the ability to mass produce micro-image dissemination
files economically. Thus far, document retrieval systems have tended in the direction
5-80
of providing a central store of images that could be placed on-line through remote
consoles. PCMI would allow nonreturn copies to be sent out to the remote stations.
5. 5. 2.3 Video Recording. The Videofile system, an automated image storage system
using magnetic tape developed by Ampex Corporation, has the capability of storing a
large amount of graphic material with fast access to any particular required document.
The stored images can be added to, updated, or purged under the application of
external controls with tape images immediately ready to view without processing.
They have an archival permanence as long as they are needed, and they are of high
quality on viewing. The techniques employed in this system are based on principles
similar to those used in commercial television recording. The documents put into
the system are scanned by a high resolution television camera and recorded electronically on two in. wide magnetic tape as a video signal.
Each image occupies either a l/3-in. length of tape (for ll-1/2-by l4-in. documents)
or a l/6-in. length of tape (for 5-l/2-by 8-l/2-in. documents) depending on which of
two standards are employed. For the larger documents (or higher resolution of
smaller documents), 1280 scan lines at 15 frames per second are used. For smaller
documents, 640 scan lines at 30 frames per second are used. The standard involved
determines the number of document images that can be stored on each reel.
With each image written on the magnetic tape is a digitally coded address to identify
that image. The address can consist of a combination of either 12 alphanumeric
characters or 18 numbers.
On recall, the tape with the stored images is traversed at high speed until the location
of the required document is reached. The image is then read from the tape in the
form of a video signal and can be displayed on a monitor (soft copy) and/ or printed out
(hard copy).
5.5.2.4 Dynamic Updating of Filmed Base. An interesting new concept which has
been fabricated into prototype hardware is reported by the MITRE Corporation
(Ref. 5-6). Two magnetic tracks stripped on roll microfilm are used to allow
5-81
dynamic updating of the filmed base. Consecutive search capability (i. e., store the
present position and search forward or reverse from it) provides a significant decreas
in the retrieval response time.
A basic problem with this approach seems to be the reliability of the magnetic stripping for storing digital codes.
5. 5. 3 The AMACUS System
The maintenance of engineering design data in hard copy media is time consuming
and costly. AMACUS is a system that enters drawings to the computers via microfilm operations cards and an optical manner. The Automatic Microfilm Aperture
Card Updating System (AMACUS) is currently being developed by General Precision,
Link Division, for the U. S. Army Weapons
~ystem
Command.
AMACUS is used primarily as a system for incorporating design changes into existing design. Graphic data are entered into a computer via an aperture card read by
an optical scanne,r. The image is displayed simultaneously on an aperture card
viewer and on a CRT.' The CRT unit is equipped with a light pen for altering the
geometric pattern being displayed on the tUbe. The selection of one of three line
widths permits matching the alteration with the original design. Ail alphanumeric
keyboard and
~ ~eybo~rd
containing standard geometric patterns and electrical
symbols is 'used for 'entering applicable data.
AMACUS is currently scheduled to be installed at the U. S. Army Weapons Command
,
.
during Decembe~ 1967. An advanced AMACUS with storage and remote retrieval/input
.
,
,
capability is currently being develo'ped as part of a total data system for the U. S.
Army Weapons Command. This development is directedby"Mr. M. D. Silkiner,
Chief, Management Science and Data Systems Office.
,5-82
5.6 REFERENCES
1.
A. Van Dam, A Survey of Pictorial Data Processing Techniques and
Equipments, University of Pennsylvania, AD 626 155, Aug 1965, 160 pp .
(plus extensive tables). This study was performed for the Bureau of
Supplies and Accounts of the Navy, and covers the entire range of pictorial
data processing including display, transmission, storage and retrieval,
optical pattern recognition, and man! machine interaction.
The study
covers the principles of operations of the various devices, and then
presents extensive tabulations of equipment characteristics for the various
categories of device. The scope of this report is quite broad, and should
be required reading for analysts working with pictorial data systems.
2.
Libraries and Automation, Library of Congress, 1964
This is the Proceedings of the Conference
o~
Libraries and Automation
covering seven topics: the library of the future, file organization and conversion, file storage and access, graphic storage (current status of
graphich storage techniques with application to library mechanization),
output printing, librarary communications networks, and automation of
library systems.
3.
C. P. Bourne, Methods of Information Handling, New York, Wiley, 1963
This book provides an illustration of the tools, equipment, and methodology
that can be applied in the design of information systems. Of particular
interest to retrieval sciences are the chapters "Classification and
Indexing" and "Microfilm and Image Handling Equipment."
4.
E. J. Menkhaus, "The Many New Images of Microfilm, "Business
Automation, Oct 1966, 11. 32-43
A good general review of the microfilm field.
5.
G. M. Lear, "A Review of the Status of Existing and Advanced Microimage Systems," Proceedings of Air Force/Industry Management Symposium, published by Ballistics Systems Division, Norton Air Force
Base, Calif. , N 66 38742, Sep 1965, pp. 257-266
An excellent, concise review of the. microimage state of the art and its
relation to Air Force operations.
5-83
6.
G. Barboza, Summary of Efforts Expended on the Concept of Dynamic Update
of a Microfilm Film, The l\HTRE Corp., AD 620 279, Jul 1965, 84 pp.
Description of the prototype hardware that was designed and fabricated to
prove the feasibility of combining magnetic and photographic techniques
of information storage.
5-84
Section 6
STUDY OF MAN/MACHINE SOLUTION OPTIMIZATION
6.1 INTRODUCTION
The man/ machine relationship must be a complementary one with each assisting the
other in the things each does best. The relationship can be thought of as a chain
formed of the basic links:
I
COMPUTER
H
MAN
H
COMPUTER
I
in which the man makes decisions which determine succeeding computer operations.
It may be helpful to think of the human nervous system and the electronic system as
two computers with vastly different characteristics. One of the distinctive qualities
of the man in this context is that he requires far less detailed instructions than a
machine and can make decisions which, if they were to be programmed into "a machine, would lead to an immensely complicated decision logic.
The man, nevertheless, does require "programming" or instructions to aid in fulfilling his role. It is assumed that he is instructed in the techniques of dealing with
the machine but, in addition, he must have a plan or program for dealing
w~th
the con-
tent of his work and the alternate decisions he may make as a result of the outcome
of computing operations.
6.1.1 Objective of the Program
The purpose of the work reported here is to establish the criteria of cooperation between the man and the machine to maximize the effectiveness per hour of machine
utilization. For this purpose it appeared essential to use an already established
.6-1
program in which the role of the operator was not so involved as to obscure the essentials of his performance, and yet simple enough so that personnel unfamiliar with the
program could learn the procedures and perform the cooperative tasks.
6.2 DESCRIPTION OF PROGRAM
The 144 Curve Fit Program was chosen for the man-computer experiments because it
is straightforward enough so that the principles of investigation could be recognized
without being obscured by excessive detail. This program fits experimental data by up
to fourth order polynomials; the X and Y variables can be transformed by log,exponential or reciprocal transformations. This amounts to 144 possible curves. The program computes the mean square fit for all forms simultaneously and lists the 10 best
fits on the surface of the CRT. The man can then choose any of the 144 curves for the
graphical display CRT. The curve chosen for permanent record can be reproduced
(i. e., by a Polaroid camera or by tape onto an SC-4020 for compute prepared plot); the
set of points can be corrected on the CRT at the display console; and further runs can
be made if desired by the operator. A user's flow diagram for subsequent experiments
is shown in Fig. 6-1. The program handles 31'points.
6.2.1 Boxes D and J (Fig. 6-1)
The man can predict that the plots will follow certain functional forms on the basis of
physical know ledge or inspection of a graph of the data, for example:
~
1.
For chemical rate equations, log rate
(l/temperature).
2.
Scaling laws are frequently straight lines, in suitable coordinates.
3.
Inspection of Fig. 6-8 (shown later) suggests the form Y = constant + I/X. *
*If high confidence is held for the correct form, the computer could be instructed to
consider this form only.
6-2
A (SEE FIGS. 6-2 AND 6-3) (COMPUTER)
PRINT A IS DATA ON "BEST FIT"
C (SEE FIG. 6-4)
(COMPUTER)
(COMPUTER)
COM1?UTE
144 CURVES
-
F
IS DATA ON "SELECTED" FIT
PRINT B IS DATA ON CURVE NOT IN BEST 10, BUT DEVELOPED ON SPECIAL OR "EXPECTED" FIT
INSERT DATA
B
OR ON REPEAT C
~~
D1SPLA Y FORMS
OF 10 BEST
E (SEE F1G. 6-5)
(COMPUTER)
D (MAN)
--
SELECT DESIRED
CURVE
r---
DISPLA Y GRAPH
OF SELECTED CURVE
G (COMPUTER)
F (MAN)
---
DECISION
-
PRINT
A
II
REPEAT C -
..
I (SEE FIGS. 6-6 AND 6-7)
(COMPUTER)
DISPLAY FUNCTION
SELECTOR
J
r--
r---
DISPLA Y GRAPH
OF SELECTED CURVE
MACmNE AND MAN
N (l'vIAN)
REPEAT A TO G
ELIMINATE BAD DATA POINTS
Fig. 6-1
1\1 (COMPUTER)
L (MAN)
K (COMPUTER)
(l'vIAN)
SELECT
CURVE
F
---
Flow Diagram Used for the 144 Experiment
DECISION
~~
PRINT
B
6.2.2 Original Instrumentation Errors'
The man has information on original instrumentation errors based either on knowledge
of the measuring instrument or inspection of the graph. This can influence his acceptance of a given machine plot. For example, if the residual variance is not much
greater than that expected from instrumentation error, a curve can be accepted even
though it is not a best fit.
(In principle, decision rules could be formulated and pro-
grammed into the machine, but these rules might depend on the type of data.)
6.2.3 Boxes F and L (Fig. 6-1)
With the possibility of alternate forms, the man might choose a form more physically
acceptable even though it was not the best fit, for example:
1.
Many physical laws can be expressed as straight lines in suitable coordinate
systems and the resulting intuitive tendency is to accept a linear solution
over a higher order form.
2.
If the formulas are going to be used
~n
complex systems analyses, the ad-
vantages of a simple formula may outweigh considerations 'of accuracy.
6.2.4 Box N
The man can .uf:!e his knowledge of instrumentation error or inconsistencies of plotted
data to recognize wild points.
For example, the circled point shown later
~
Fig. 6-12
is obviously suspicious. This point could be deleted prior to entry into the machine.
(Decision rules could be programmed for such conditions if these situations were frequent enough to warrant the programmatic complexity.)
6.3 EXPERIMENT DATA AND OPERATIONS
These runs were made to verify the adequacy of the instructions, to establish characteristic times for the functions shown in the boxes of Fig. 6-1, to see what type of
6-4
decisions were best made by the man, and to use the results in formulating suitable
criteria for instructions to be generated for making decisions in the general case.
For these experiments sets of pairs of data points (X, Y) were put on punch cards.
These sets were collected from sources having known relationships and statistical
error magnitudes in the point determinations. The three trials reported used three of
these sets of data. Plots of the three sets of data were made, and, with estimates of the
best fit curves, are shown in the following figures.
Each data run presented the displays
having items similar to those shown in Table 6-1. The X variable is resistance and the
Y variable is transconductance. They are expected to be of the form Y
= F
(l/X).
6.3. 1 Discussion of Table 6-1
Item 1 is a machine operation of accepting the cards and displaying the pairs of numbers in X and Y (Fig. 6--2). The recorded time is from insertion to display. The man
can take as long as is needed to make a comparison to some other data list. (Usually,
comparing the cards with the data list should be done in advance, and the later point
display in Item 4 is a better comparison. )
Item 2 is a machine operation ordering the cards in increasing value of X (Fig. 6-3).
Item 3 is a machine operation displaying the 10 forms of the equations giving the best
fits (Fig. 6 -4).
(This will not always be a quartic as might be expected since these
are ordered in increasing standard deviation in which the sum of the squares of the
individual residuals is divided by the number of points minus one, minus the order of
the curve.)
Item 4 is the display of a curve selected from the best 10 (Fig. 6-5). A small part of
this time, e.g., five sec, is used by the operator in selecting the curve. The greater
part of the time is used in computing, storing and erasing the display, and displaying
the curve.
Item 5 consists of recalling the function display, five sec, selecting a new curve form,
10 sec, and generating and displaying a second curve (Fig. 6-7).
6-5
CATA
....
• • • • • OOOOOOE
03
5.02'7'000000£
02
..'.1000000E
03
... '7'20000000E
02
•• 3.0000000£
03
5.126000000£
02
•• '7".000000£
03
".3 '7'SH'iO 0 tHHi £
02
••••• 000000£
03
... '7'6'7'000000£
02
3.055000000£
03
".2 .. 600ClliOO£
CiZ
'P.
,.5.30000lH1£
03
". Z 51 tH'iCitititiE
t12
8.
3. '7'55t'll10000E:
03
".5190titit'itiCl£
t'2
3. '7''7' .. t1(10000E
03
".22'7'000000£
OZ
5.5."000000£
03
3.9"3000tH1Cl£
tiZ
5.'7'.5000000£
03
3.8ZZ0tHHiClOE
t'iZ
.2.
5. "2 00 0 OtHlt1E
03
3.9"1000ClOO£
t'iZ
.3.
6.6160(1(1000E
03
3.5650(10(10(1£
oz
..196000000£
03
3 ..... I. CiO CltitiO£
ti2
..5"5000CiOOE
ti3
3.
1 tjt;~jt;~jtjE
liZ
1 .1120(HI000£
tj.
:5.:5 52 l1Cjti CHHi£
OZ
,.
..
..
..
s.
.0')
I
0')
1 ....
...
...
...
.0.
.5.
5~
• '7'.
1 •• 31000000£
0"
:5.513(1!H'itiOti£
liZ
.8.
..233000000E
0"
Z.ilZ'7'1100(100£
OZ
19.
1.336(HH1000E
0 ..
Z. '7'1 OCHH1000£
OZ
20.
1.363 !HHitiOOE
O.
Z.97CHitil'itiOO£
l1Z
Fig. 6-2 Complete Data Display
CURVE
SET
"'T
FROGItAN
!'
c:
r . :
I
363(lO('l('lO('l~
0.
2
I
336!'lO(lOOOE
0.
2
T 1 [iC;ti
!i~;!itiE
t~2
1 .Z:53tH'ltHl('10E
CiA
2
it2T!ili!.iGr,CiE
G2
1
13 1 O(ltH'OOC
!i4
3
513!iOGli!iOE
G2
I
1 12~'.'Of~lit~E
Ci4
3
352!i!i!H'!i!iE
~~
50nt't-:t'7t;~
'.;3
3
Stl1 l1LitiOtiG£
tj2
9. I 9 E-t'
r: r; titi!; C
t'3
3
4 J14
£.. £.1 S t~
t~ t~
ti ti ti E
ti3
3
:- Eo ': !': ~;
:3 • 9 .t !
~~
t: !.: '.': '.'; ~-; E
.
: ..
!'7 E
,..
9. 5t'
97(i!ir.ti!i!i(iE
ti~;
~.
:' Z !oi !o; '.' ti f'i ti ti E
~'l
5
79 5tititit,,-it"i£
1.";:5
:3
:.."'!
Z2
~'7
SO
~3
3
!i2
2
ti!i ti
t~
e:
~-;2
~~
t: E
'.::E
!.i t~
~. ~~ ~i
..
,
5. 594 tHHHHHi C
n3
:
(t" tAt-
t-t':e:
~"2
:!>
77.0(ll,r,(lOC
t'l3
I! • 2 2 7 t~ t~ t': ~-:
t: !.i E
~-,
~
75 5titi'.'t'10tiE
ti3
I!. ;: 19'.:'.-:t"'t-:r:t":E
t'"'Z
:5
5 43tH' !HHH'l E
t; :5
4.2 51 ti t':!i ti
e:
'.';2
:5
!' 5 50t' OOl'lO C
{;3
4 • 2460 ti !i!i!i ti E
~'2
Z
991 O!'lOO!'lOE
!i3
4. 767 ti ti tiGtitiE
tiZ
z.
7:590!1!1!100C
!;3
4. 379 tit; !.i !i
Z
! 2 !oj '.1 ti ti ti t"i ti E:
t;3
5
12
z
:5 1 1
O('t'Jt;~'tjE
!i3
4
72titi!.iti!ititiE
z.
1 9 6- t' ti ti ti !.i !.j £
~i
5. !i2
:3
Fig. 6-3 Display of Data Before Acceptance Into Data Set
t~t;
~jtjE
~i2
e. ~; ti !.i ~j ti !.; E
:,2
7!itjt;~jtjtie:
ti2
'.'2
ElEST
CURVE
0')
CURvES
TRANS"
"UNCTION
,. (X)
O"N
CEGREE
Y=l1'LG(P(x»
Y.S I'LG
y= 11''' (x)
Y.S,.,,(X)
"',X).",X)
v=£ (F (X) )
Y:E (F' (xJ )
F 'X) .... 'X)
3
r=F (x)
Y:F(X)
F
X) = .. (X)
,3
Y:F'(X)
F (X) .... 'LG
'1'=£ (F C'X»
Y:E (F' ex»
FCX,=",X)
Y=l1''' (X)
y=s,.F'ex)
FOO .... 'X)
Y=l1'LC (F (X»
Y=Sl'LG
Y:P (X)
Y=F'ex)
F (X) = .. (X)
Y=F'(x)
F 00= .. (II'
V=F (LC
x)
"(X).",X)
X)
3
I
(X)
Y:F(l.1LC
x)
F (X)
•
F (X) = .. 'X)
Fig. 6-4 Display of Functions of 10 Best Curves
LG
X)
3
R[SICUAL
Z • 1 9 611(1 00011 E
03
2.
Z • 3 1 I 1111(1111111 £
03
3.
Z • 3 2 11111111 111111 [
03
4.
2.1'390(1l1011l1£
tl3
,.
2 • 9 9 lOll t1 !1t1t1 E
t13
6.
3 • 11 5 5 1111 Ill1 lH1 E
l13
-2.512J29514[
til
-G.532116465[
till
1'.
3 • 54 3 11I1111ll1l1 [
113
.
3.755111111Ill1l1£
l13
9.
3 • l' l' 4 Ill111 !ll1l1 £
113
Ill.
5 • 5 9 4 l1l1lllll111 £
03
II.
5 • 7 9 5LiLH1I1I1I1 £
l13
~
8.1315ZJ635£
lHl
y:: I /L C; IF I X) )
00
B=
-1.53331756tl[
Z.5IH433Z5G£
-2.6l168Il1293[
2.427174433£
2.690tll15215£
-1.728311844[
6.7383119965£
-2.744741976£
STANCARC
nI
til
6
5110
X
\
'\
l1l1
03
03
-1.9435181611£
01
14.
9 • I 9 6111HHHHl E
03
-8.914993420£
llLl
15.
9.585111HHlLH1E
03
8.786885761£
llLl
16.
I . I 121lLHH1I1I1E
114
6.li98946534E
011
11'.
I • I 3 I IH1I1I1I1(1 £
n4
2.527477429£
111
a.
1 • 2 3 3 11Illlt1(1(1 £
l14
-2.452767854£
111
19.
I • 3 3 61HH1IHHl £
04
-1.365543430£
111
ZO.
1 • 3 6 3 11IHHHiLl E
04
nl
6
1111
5 • 82 l1llLHlL1IHl £
CEvrATrON
1.957329741£
lltl
6 • 6 I 6lll11lLlLlll £
1.875371826E
C"
111
13.
9.467393957£
1.072511'610[-06
til
OJ
12.
I
COEfF'ICIENTS
'UII liE
I.
)C
11Ll
\
,
)(
\
\\
\
\\
x
lC
x
to
~\
~
111
"",
41Hl
~•
It
~~
I:
~ r-.....
•
4-
N 61
-.......,
I-----. ~
~
k
............
1'-" ,
.
,
t
.
. .
I
-4111111
6000
.,,""
10000
12000
[y
Fig. 6-5 Display of Best Curve
= ~g (A + BX + CX2 + DX3 ))
6-9
Table 6-1
TIME ELEMENTS
Function
Item
Operator
Time
(sec)
1
Enter Cards and Compare to Data List
Computer
2
Accept Data and Reorder Cards in "X"
Computer
3
Compute 144 Fits, Select and Display 10
Best Fit Forms
Machine
10
4
Select Specific or 10 Best and Display
Curve (1)
ManMachine
10
134
5
Select Specific from Function Display,
Compute and Display Curve (1)
ManMachine
35
135
6
Enter Title
Man
120
7
Print Out (2)
Machine
5
4
(1) The curves of which Items 4 and 5 are examples can be recorded on
magnetic tape and printed out on a curve plotter, i. e., SC-4020.
(2) The equations of the curves and the 10 best curves can be determined
in 10 sec since printout can be done anytime after Step 3. The time
after this step involves about 2 min for each alternate called up for examination by the man.
Item 6 consists of selecting letters, spaces, etc. , on the display tube using a light pen.
The time to display a full title is large and the man would do well to use a short coded
identification, filling in the title later (Fig. 6-6).
Item 7 consists of recording on paper tape the 144 curves, the 10 best, and any others
of interest (Fig. 6-7).
6. 3.2 First Trial Runs
The first trial was designed to establish the completeness of the set of instructions
written for the operator of the 144 Curve Fit Program. Three separate operators,
6-11
F' UNC T ION
y
=F'
(
CP' CX) )
l( )
F'
(Ie)
Y
=I
jr
ex)
TFt
0
0
0
~
0
U
Z
~
~
U
300
~
~
0
Z
0
U
00
Z
~
~
200
100
o
2
4
6
8
10
RESISTANCE (X)
X x 10- 3
Fig. 6-8 Run 1, 144 Curve Fit Program, Preplot and Curve
6-16
12
600~------------------------------------------------------~
500
y
/
1
3
log (A + BX + CX2 + DX )
o
300
Y = A + B
X
o
200~----~~----~------~------~------~------~------~--J
2000
4000
6000
8000
10000
12000
RESISTANCE (X)
Fig. 6-9 Run 1, Computer Output
6-17
14000
16000
0
0
201
-,
C'f:)
I
0
t"""f
X
-!
:>t
15'\
-
:>t
~!
Oi
Zi
<
E-i
10!
0'
:::>
~
J
z'
0'
0
00'
Z
<
0
~'
E-il
5
\
0
0
8
0
0
0
O\~~------~----------~----------~--------~
01
20
30
RESISTANCE (X)
40
Fig. 6-10 Run 2, 144' Curve Fit Program
6-18
o
o
20
Y
= A + B log X + C
(log X)
2
• D (log X)
3
0)
15
10
o
o
o
5
0
o
o20~----------------~----------------~----------------~
40
30
RESISTANCE (X)
Fig. 6-11 Run 2, Computer Output
6-19
4.4
4.0
3.6
~
~
(.)
Z
~
f-iM
(.),
:.:J
Cl
Z
0
u
;::
a.2
)<
0
WILD SHOT
>t
en
Z
~
f-i
2.8
2.4
~------~~------~--------~--------~--------~--------~--------~
, 2.01.6
3.6
4.0
2.8
3.2
2.0
2.4
RESISTANCE (X)
(X x 10- 2 )
Fig. 6-12 Run 3, 144 Curve Fit Program, Preplot and Curve
6-20
5000
-
®
.-y
WILD
SHOT
3000
BCD
= A+-+-+X
X2
0
2000--------------------------~----------~------~--------~
100
200
300
RESISTANCE (X)
Fig. 6-13 Run 3, Computer Output
6-21
400
excluding the point Curve B. This trial'demonstrates the difficulty of programming
the machine to ignore bad data as compared with the capability of the operator.
From these results it can be seen that the man can be more discriminating concerning
items which are not programmed into the machine. He can make a be.tter selection of
"wild points." He can force the acceptance of curves which might be rejected by the
computer and he can choose whether or not to use data at all. The time results also
show how the user can save time by exercising discipline in choosing which curVes
to display.
6.4 INSTRUCTIONS
The operator should prepare himself for the particular program before working at the
machine. In the present program, he should have plotted the points estimated for the
. best fit curve, its form and degree, ·and ask for this immediately from the machine,
thereby reducing the total display time. The data should be prepared for input so that
the fastest approach is used. In the present case, the use of punched cards reduced
the input time rrom 10 min (console/ light pen input using A/ N Register vs cards) to
about four sec. In general, this preparation should include (at the minimum):
a.
Before Going to the Graphics Console:
1.
Graph data
2.· Mentally review physical considerations
b.
3.
Decide on most likely functional form
4.
Select possible alternates
5.
Estimate accuracy of data
6.
Clean up obvious wild points
Use of Computer:
1.
Punch cards (Do not use light pencil for original input. )
2.
Run through steps A - F
6-22
3;
If first choice appears in top ten, call for display:
(a) If this curve is obviously acceptable, print and leave.
(b) If curve is not obviously acceptable, call for alternates.
(c) If one or more alternates are obviously acceptable, print and leave.
(d) If some alternates cannot be obviously accepted or rejected, print
out and go to desk for further consideration.
4.
If first choice does not appear, either call for first choice and look at it
or call for alternates.
5.
If no alternates appear on display, go back to desk and consider possible
reasons.
NOTE:
Only decisions which are obvious (i. e., can be made quickly) should be
made at the console.
6.5 SUGGESTIONS FOR MORE EFFICIENT DIVISION OF LABOR BETWEEN MAN AND
COMPUTER
The original designer of any man-machine program should keep in mind the speed of the
computer, the speed of the mechanism which communicates with the man, and the speed
of the man in making decisions as well as the speed with which the man can communicate with the machine.
This implies that the computations should be done by time-sharing if a high-speed computer is used since a very small fraction of the time is generally spent in calculations.
As can be seen from Table 6-1, the longest times are associated in generating a change
in the display, i. e., for generating curves as in Items 4 and 5, or of generating a title
as in Item 6. Part of this time is utilized in erasing the existing display; however, as
can be seen from Item 2, this is not too great an interval. A major portion of the time
consumption is in generating the displays of curves as shown in Items 4 and 5. It appears that this is because we have copied too slavishly man's method of operations.
6-23
Since the psychology and physiology of the computer differs markedly from man's, it
can be expected that suitable strategies differ correspondingly. Also, since the man
is more flexible than the computer, he should expect to adapt himself to the limitations
of the computer.
With respect to curve plotting, the man does not like to compute points. He has a good
interpolator in the French curve. He also needs coordinate paper as an aid to placing
points. On the other hand, the computer is extremely efficient in computing points and
does not need paper. It does, however, have difficulty in stringing them into a continuous curve. The curve generation could be shortened by dropping the coordinate grid
and printing only discrete points (50 might be enough) and allowing the man to do the
interpolation visibly as part of his evaluation of fit. Part of man's adaptability is this
capability to visualize the connections between close points.
6.6 SIMULATION
Lockheed has prepared a simulation code, LOMUS I, which simulates operations of a
complex "job sh,op" where long predictable runs' are not realized. A large computer
facility has such a character. The output can answer such questions as:
1.
Will the cost of adding another machine, component, subsystem or operation
be more than offset by the increase in service?
2.
Should a system structure of type A or type B be acquired?
3.
Could system performance be improved through the use of differe~t sched - ,
uling rules and information networks?
4.
What procedures should be followed to minimize the disruptions due to a
computing machine failure?
5.
What is the probable impact of a priority scheme upon system performance?
6.
How do manpower variations affect system cost?
7.
How are the dynamics of queues affected by work load variations?
8.
How does the average system response time vary with work load?
9.
How does the input job. mix affect machine utilization?
6-24
The LOMUS I program requires information based on operating experience. This information include s data tables and parametric input. A list of these from LOMUS I
is as follows:
Data Tables
Manning
Workload initialization
Software
Transit time
Function volume relationships
Distributions
Arrival rates
Machine name and cost
Parameters
Workload initialization and rate
De bug percentage
Scheduling rule
Routing control
Manpower loading
Constraint complexity
Priority allocation
Value control
Application area
The addition of the man in the loop would emphasize the variability of estimated time
to complete a job due to choices which a man can make during the running of the program. The other inputs would be the same. Therefore, no modifications to the program are recommended at this time.
6-25
Section 7
HARDWARE ANALYSIS
7. 1 INTRODUCTION
It is the purpose of this section to discuss the compatibility of the programming speci-
fications contained in this report with the Univac-supplied Advanced Mariner Display as
described in the Univac Product Description S-90060. Because the Univac specifications
were not available until late in the contract period, all specifications were written assuming a generalized hardware and software system. The Control Data Corporation
(CDC) Digigraphic 270 system typifies most of the desirable features of a computer
I
graphics configuration. It was used as a standard system with which LMSC had had
experience and it could be easily written about.
Therefore, this section will describe the
CDC system, discuss the influence some of the assumptions have had on the programming speCifications, and evaluate some of the changes which will be required by the
use of the Univac system.
7.2 HARDWARE SYSTEM CONFIGURATION
The assumed hardware system as configured in Fig. 7-1 comprises the following types
of units:
•
Three CDC 273 display units and a controller
•
A CDC 3300 computer with 32,768 words of memory
•
Random storage on two, 2-million word disc drives
•
A high-speed printer (1000 lines per minute)
•
Four tape units
•
A card punch
7-1
TAPE
CONTROLLER
DISC
CONTROLLER
PRINTER
~-.......,
CHI
CH3
CDC 3300
COMPUTER
CHO
CH2
CARD
READER
GRAPHIC
CONTROLLER
...........---tCARD PUNCH
3 DISPLAY CONSOLES
Fig. 7 -1 Hardware C6nfiguration
7-2
Assumption of this system will influence such decisions as the follwing: a. Should
the display console operator be given the option of entering parameters on cards or
must he enter all parameters through the display interface? b.
Can the operator
get immediate access to printed output or must the display console software provide
for paging through lists of. output parameters? c. Will the program fit into the mem0ry of the machine in one segment or must it be partitioned?
Upon configuring the target Univac system, each of these questions and others will have
to be asked again and the answers reevaluated.
7.3 SOFTWARE SYSTEM
The assumed software support is based on the CDC Functional Control Package. This
. software will support up to three display consoles on a time-sliced operating basis under
which the consoles may share application and software support programs. The system
provides several features described below. All of these features have been assumed in
writing the programming specifications and all of them are considered to be necessary
for harmonious multi -console operation. The features are as follows:
•
A basic software operating system comprising system and application programs
to lend homogeneity to the· operation. This permits any application programs to
be run from anyone or several consoles Simultaneously with any other program
functioning at another console, and with a minimum duplication of programming
effort and computer memory content.
(I
A basic package of programs that can be called from any console to perform
certain universal operations, e. g. entering parameters, drawing lines, collecting groups, transforming or erasing graphic entities, storing and retrieving
displayed graphics, etc.
•
A standard model structure which will provide a link between the herrarchy
represented by the world as shown on the display window and the universe as
defined within the computer. On the CDC system, the model defines a "universe'
eight-million addressable units square while at the window of the display (only
7-3
four thousand units square) the viewer may examine any part of this world with
a useful range of display magnification of 1 to 32,768. For example, working
at an accuracy of 1 x 10-4 in. per unit, the universe may contain a drawing
70 ft long by 70 ft high.
•
A package of programs which can be called by an application to p.erform manipulations of the model and of the display such as: entering a graphic entity or
other data into the model, displaying a light button, or copying and moving a
template group.
The basic specification decisions influenced by these assumptions are as follows: a.
What is the operating environment of the application program, and how is it called?
b. Which standard console operations may the application program take advantage of?
c. How should the problem be structured by the program to best use the model provided? d. Which operations are easy to
perf~rm
on this model and which are difficult?
7.4 DISPLAY OPERATIONS
The facilities
~iscussed
here are assumed to bOe available to the display console opera-
tor. The following paragraphs describe the functional features and areas of the console
shown in Fig. 7 -2:
•
Display Working Surface: A section of the display designated for presentation
of the. window through which the window is viewed. It is on the window area
that the lines or other graphic entities are displayed as the operato:r:
vie~s
and/or operates upon them.
•
Display Control Area: A section of the display designated for presentation of
general operator controls such as light buttons or registers for entry or display
of parameters.
•
Light Pen:
The pen is used to select graphics, registers, or light buttons to
perform specific functions.
The pen is also used to select and move the track-
ing cross.
7-4
Fig. 7-2 Display Console With Permanent and Programmed Controls Shown
•
Tracking Cross: A special graphic entity in the form of a cross which may be
moved with the light pen and is used to designate an area of the working surface
which is not illuminated and thus not selectab.le with the light pen.
•
Function Keyboard: A hardware box of keys that the operator may press to
indicate an operation to be performed. The definition of each key is changeable under software control.
•
Light Button: A special graphic entity which is used like a function key. When
selected by the light pen, an operation is called.
•
Alphanumeric Keyboard: A keyboard for entering alphanumeric data. On
the CDC system this consists of an array of light buttons. No difficulty is
foreseen in substituting a hardware keyboard.
7. 5 CONVERSION CONSIDERATIONS
The following discussion will give consideration to the differences between the CDC 270
system with the Functional Control Package software and the Univac Advanced Mariner
Display. These differences will be evaluated in light of their infiuenceonthe programming specifications depicted in this document. It should be noted that a philosophical
difference makes direct hardware comparison difficult; this difference is the trade
between software display generation on the CDC system and hardware display generation
on the Univac· system. For this reason the two systems should each be evaluated as a
hardware/software combination. Since little or nothing is lmow of the Univac software,
this is difficult. Table 7 -1 compares the display characteristics of the two systems.
7 . 5. 1 Hardware
It is readily seen that hardware for both systems is comparable in size and speed.
Any speed differences are considered insignificant because both systems can display
a satisfactory quantity of data in regeneration cycle time. Size considerations, however, are more severe since many of the programming specifications make use of
light buttons and registers. (The 12 -in. -square display area does not allow much space
7-6
Table 7-1
HARDWARE COMPARISON - CDC 273 VS. UNIVAC MARINER DISPLAY
Parameters
CDC 273
Univac
~--------------------------~--------------------------------------~----------------------------------~
Basic Display Generating
Scheme:
Hardware displays as commanded by
drum memory. Display synchronized
to drum revolution. Nonlinear analog
display characteristics compensated
for by software when commands are
written. Regeneration fixed at 30 Hz.
Hardware display as commanded
by core memory. Memory
synchronized to display characteristics; next display instruction
waits for completion of previous.
Nonlinear display characteristics
do not manifest thems~lves. Regeneration rate set by software in display
controller.
Display Dimensions:
CDC 273 (with standard software)
Univae
Usable Display Area:
Shape
20.5 in. diag'onal round
Area
323 in.
2
12 x 12 in. square
144 in. 2
11 x 17 in.
Full screen
Display Control Area
Circular sectors outside of the
11 x 17 in. construction area
Minimal(b)
Addressable Units
4,096 x4,096
1,024 xl, 024
Inches Per Unit
0.005 in.
0.010 in.
25 J1.s full screen, 20 in.
15 J1.s full screen, 17 in.
Display Working Surface:
Display Timing(a)
Random Reset
Positioning
Long Line (17 in.)
44.08 J1.s
32
Short Line (0.5 in.)
4.04 J1.s
2 J1.s
Character (average)
10 J1.s
5 ps
Small
0.14 x 0.1 in.
0.14 x 0.095 in.
Large
Not programmed
0.28 x 0.190 in.
J1.~
Character Generation
Characteristics:
Character Size:
Characters Per Line:
Small
146
80
Large
Not programmed
40
Character Rotation
Not programmed
90 deg
Generation Method
Software vectors, variable font
Hardware generator, fixed font
Function Keyboard
(Program Definable)
24 keys
35 or 40 keys
(Text says 40 j figure
Alphanumeric Keyboard
Software feature:
Variable font displayed in lower
control sector
Hardware keyboard fixed
alphanumerics
Light Pen
Standard
Optional
Light Buttons
As desired in Control surface
sector
Very little room for light buttons.
but usable(b)
Console Controls for
Operator Action:
shows 35)
(a) The times shown for the CDC 273 are average times for the specified length and are dependent on how the
software package chooses to generate the beam driving' instructions to produce a smooth line.
(b) If graphic entities are displayed simultaneously with control functions, minimal usable area is available
for control (light buttons and registers) .
7-7
for other than graphic data presentation). The user of the Univac system would have
to make greater use of the programmable function keys as a less satisfactory substitute for the light buttons. With respect to the alphanumeric keyboard and the hardware generation of characters, the Univac equipment would seem to have an advantage
unless a variable type font is required. In summary, the hardware considerations
cannot be fully assessed without a detailed hardware/software description, but it
appears that there will be no major problems.
7.5.2 Software
The software system assumed in these specifications for the various applications discussed in this report is considered to be near minimum level for operation of a display
system for a diverse group of users ·operating simultaneously. Without such a basic,
comprehensive software package, each
appl~cation
must provide all distinct operating
procedures, console controls, model manipulations, etc.
The graphic programming language outlined in the report of Adams Associates to
Univac would, be a most satisfactory basis for such a
comprehensiv~
software system,
however it is not all clear from this report or from the Univac documentation that
such a language will actually be provided. It is a requirement that if a comprehensive
package is not provided by Univac with the delivered hardware, and differing and distinct applications such as those outlined in this report are to share the graphic facility,
then a basic package of software and manipulative routines must be specified and written to ensure homogeneity of all efforts. It is estimated that a· software effort of this
size will require eight to nine man-years of programming effort.
7-8
Appendix
GLOSSARY
AC ANALYSIS: The analysis of the behavior of a circuit which is subjected to steady,
single-frequency, sinusoidally varying waveforms.
ACCESS, RANDOM: Access to storage such that the next position from which information is obtained is not dependent on previous location.
ACCESS TIME: The time interval between the instant at which the arithmetic unit
requires information from storage or memory unit.
ACTIVE ELEMENT: A source of current or voltage.
ADDRESS: A label, name, or number.
ALGORITHM: A fixed, step-by-step procedure for performing some data processing
function, usually a simplified procedure for accomplishing a complex result.
ALPHANUMERIC: A generic term for alphabetic letters, numerical digits, and some
special symbols.
AMPERE: A unit of electric current, abbreviated as amp.
ANALOG COMPUTER: A computer which calculates by using physical analogs of the
variables.
APPLICA TION PROGRAM: Existing analytical engineering program(s) or one to be
developed.
ARITHMETIC UNIT: The section of the hardware of a computer where arithmetic and
logical operations are performed on information.
BATCHING: Grouping a number of similar input jobs (tasks) together for processing,
thereby using a single setup or program for a longer period in a single run.
BIT: Contraction for binary digit.
A-I
BYTE: A set of consecutive binary digits, usually representing either a character code
(BCD, BCC, etc.) or some easily manipulated set of digits or portion of a machine·
word; e.g., an eight-bit or six-bix byte.
CHORDING: The act of depressing more than one keyboard key at a time.
CIRCUIT: A communications link between two or more points.
CODE: A system of symbols for representing information in a computer and the rules
for associating them.
COMPILER: A program-making routine which produces a specific program for a
particular problem.
COMPUTER: A machine which is able to calculate or compute, i. e., one which will
perform sequences of reasonable operations with information, mainly arithmetical
and logical operations.
COMPUTER WORD: A 24-bit word (for discussion purposes) .
CONTROL SURFA.CE: The area on the CRT display surface used for light buttons and
light registers. That area on the CRT not reserved as working area.
CONTROL UNIT: That portion of the hardware of an automatic digital computer which
directs the sequence of operations, interprets the coded instructions, and initiates
the proper signals to the computer circuits to execute the instr-uctions.
CRT: Cathode-ray tube.
CURRENT: Refers to an electric current which is the time rate of flow of electric
charge across a surface.
DATA BASE: A data storage scheme.
DATA ITEM: One input variable. It can be an array of values or a single value.
DC ANALYSIS: The analysis of the behavior of a circuit which is subjected to steady,
direct current voltage and/or current sources.
DEPENDENT CURRENT SOURCE: In ECAP, a transistor. A dependent current
source is always associated with two branches; it is placed in one branch and is
A-2
controlled by the other. The assumed direction of positive current flow for the
dependent current source must be the same as that of the branch in which it is
placed.
DESCRIPTOR: The first three 24-bit words related to an entity stored in the Computer
Graphic List. The descriptor contains the key, category, reference, and parental
pointer fields for the specific entity stored therein. Every entity in the list is
headed by a descriptor.
DESIGN, SYSTEM: A plan or outline for a system covering its aims and' purposes,
environment, functions, the phenomena to be processed, and the partitioning of
the system into subsystems and components.
DIGITAL COMPUTER: A computer in which information is represented in discrete
form and which calculates using numbers expressed in digits, with yes and no
expressed in l's and O's to represent all the variables that occur in a problem.
DIRECT CURRENT: Unidirectional current as produced from batteries, from dynamo
machinery equipped with commutators, or by means of rectifiers.
DISPLAY: Presentation of information in a form c'omprehensible by a human being,
as a chart or graph, hard copy, cathode-ray tube pictures, letters or figures,
meters, dials, etc.
DISPLA Y, CORE: A method of graphic display using information stored in the computer core memory. Core display is synonymous with on-line display.
DISPLAY SURFACE: The 20-in. -diameter area on the CRT face utilized for manmachine communications.
DRUM, MAGNETIC: A rotating cylinder coated or impregnated with magnetic material upon which information is stored by polarized magnetic dots.
ELECTRONIC: Pertaining to the motion, emission, and behavior of currents of free
electrons, especially in vacuum, gas, or photo tubes and special conductors or
semiconductors. Contrast this with electric, the flow of currents in wires only.
ELECTRONIC COMPONENT: A device used for controlling current and/or voltage
in an electronic circuit, e.g., a diode, tube, or transistor.
A-3
ELEMENT, ELECTRICAL: An electr,ical device which has only one electrical proper1
e. g., an inductor has only the inductance property and no resistance.
ELEMENT, GRAPIllC: Copies of the templates displayed on the working surface as
part of the circuit schematic.
ENTITY: The smallest addressable information unit within the Computer Graphic
List. It is 3: variable-length block containing a range of data specified by the
entity type. There are five classes of entities and within each class there are a
discrete number of entity types. (See ENTITY CLASSES.)
ENTITY, ALPHANUMERIC,: A record of alphanumeric data for placement anywhere
on the display surface.
ENTITY, APPLICATION: A record for storing general purpose information.
ENTITY, CHILD: An entity which has been designated as a member of a logical unit
called a group. Each child entity is related to the parent entity by its parental
pointer field ..
ENTITY CLASSES: Ther.e are five general classe,s of entities: (1) control surface,
. (2) alphanumeric, ,(3) graphic, (4) linkage, and (5) application. Each class consists of a discrete number of entity types.
ENTITY, CONTROL SURFACE: Within this class four types of entity, are defined:
(1) tracking cross, (2) frame, (3) light ,register, and (4) light button.
E~TITY,
GRA,PIllC: Within this class five types of entity are defined: (1) dot,
(2) line, (3) circle, (4) circular arc, and (5) polys,tring.
ENTITY ,LINKAGE: Within this class one type ofentity is defined: group control.
ENTITY, PARENT: An -entity consisting of a group of child entities.
FIELD, CATEGORY: The last six bits of the first 24-bit word of the descriptor for
an entity stored in the Computer Graphic List. The category field has been,
established for use by the application programmer.
FIELD, KEY: The first 12 bits of the first 24-bit word of the descriptor for an entity
stored in the Computer Graphic List. The key field defines the entity type.
A-4
FIELD, PARENTAL POINTER: The third 24-bit word of the descriptor for an entity
stored in the Computer Graphic List.
Zeros in this field indicate that the entity
is not currently grouped. When the entity is grouped, the field contains the list
address of its parent entity.
FIELD, REFERENCE: The second 24-bit word related to an entity stored in the
Computer Graphic List. This
fi~ld
may only be used by the application program.
FIXED CURRENT SOURCE: A current source that has no provision for varying its
output.
FIXED POINT ARITHMETIC: An arithmetic notation or representation in which all
numerical quantities are expressed to the specified number of digits, with the
point implicity located at the same specified position.
FIXED VOLTAGE SOURCE: A voltage source that has no provision for varying its
output.
FLOA TING POINT ARITHMETIC: An arithmetic notation taking into account varying
locations of the base point, writing each number by specifying its sign, its coefficient, and its exponent affecting the base.
FLOW DIAGRAM: A graphical representation of a sequence of operations.
FRAME: Usually a rectangular display on the CRT display surface which encloses the
working surface; however, the frame dimensions can be changed by FCp· common
parameter s .
FREQUENCY: The rate at which a current alternates, usually measured, in cycles
per second.
GRAPHIC APPLICATION PROCRAM: A computer graphic program written for an
application program.
GRID, CONSTRUCTION: An area consisting of 8,388,607 addressable points in the
X and Y planes. This is the area upon which the user constructs his model. The
construction grid can be scaled by the user for his particular application.
A-5
GRID, DISPLAY: An area consisting of 4,096 addressable points in the X and Y planes.
The display grid circumscribes the display surface so that any combination of
points in the X and Y planes can be given on the display surface.
HEURISTIC: An intuitive trial and error method of attacking a problem (as opposed
to the algorithmic or set procedure method) .
HIERARCHY: A
~pecified
rank or order of items, personnel orders, nested sub-
routines, etc.
IMPEDANCE: The term expresses the relation between a sinusoidally varying quantity
(such as force, pressure, voltage, electric field strength, temperature) and a
second quantity (such as velocity, current, magnetic field strength, or heat flow)
which is a measure of the response of a physical system to the first.
INDEX, ZOOM: Binary logarithmic ratio of display grid units to construction grid
units provided for use by the operator to magnify the display on the working surface.
The magnification can be increased or decreased in
int~gral
increments of powers
of two.
INPUT: Information transferred from
outsid~
the computer, including secondary or
external storage, into the internal storage of the computer.
INSTRUCTION: A machine word or a set of characters in machine language which
specifies that the computer take a certain action.
INTEGER: A whole number; not fractional or mixed.
INTERFACE: A common boundary between systems or parts of a single f?ystem.
I/O EQUIPMENT: The set of devices used to present information to and receive
information from a computer, e.g., card readers and punches, printers and typewriters, paper tape readers and punches, magnetic tapes, data channels, CRT
and console displays, mechanical switches, and magnetic drums.
JOB: An operation or series of tasks to be performed on a computer or in a computing
facility .
A-6
KEYBOARD: Consists of 25 manually operated keys and is part of the Computer
Graphics Console. The functions of the keys are defined by the software but can
be modified or completely changed by the user's application program.
LANGUAGE: A system consisting of a carefully defined set of characters, rules for
combining them into larger units (words or expressions), and specifically
assigned meanings, used for representing and communicating information or data
among a group of people, machines, etc.
LIGHT BUTTON: Software-defined functions displayed on the control surface. There
are primary and sec oodary light buttons; a primary may have up to 16 secondaries
associated with it.
LIGHT PEN: A pencil-like bundle of.fiber-optics and associated hardware. It is used
by the operator to interact with data on the display surface. It does not write on
the display surface; rather, it senses the current X and Y display grid coordinates
of the light beam and, under appropriate program control, the display can be
updated with a light-trace of the light pen's path.
LIGHT REGISTER: There are several FCP-implemented light registers displayed on
the control surface. They are provided for operator input of data and for output
of program -generated values or data.
LIST, COMPUTER GRAPIDC: A data base containing the parameters which define the
entities created by and related to the Computer Graphic System. It is kept in
book form.
LOGIC: The science that deals with the canons and criteria of validity in thought and
demonstration of fact; the formal principles of reasoning.
MAcmNE LANGUAGE: Information in the physical form which a computer can
handle.
MEMORY: Any device into which information can be introduced and then extracted at
a later time.
MODULATION: Process by which certain characteristics of a wave are modified in
accordance with a characteristic of another wave or signal.
A-7
OPERATION, REAL TIME: The processing of data in concert or concurrent with a
physical process so that the results of the calculation are useful in the physical
operation; continuous data processing in synchrony with an attendent upon another
operation.
OPERATOR: The person operating the graphic console. Same as USER.
OPTIMIZE: To arrange phenomena to achieve a maximum operation or effect with
regard to some variable.
OUTPUT: Information transferred from the internal storage of
a computer to
secondary or external storage or to any device outside the computer.
PAGE: Sequence of N consecutive computer words where N is arbitrary but fixed
for a particular application.
PICKING: The FCP function which identifes specific graphics or points on the graphics
through interpretation of light seen by the light pen.
PICK TABLE: A table in which the item pointed to on the screen is placed for program
interpretation and execution .
'PL UGBOARD: A removable board holding many electric terminals ..
POINTER: lvlechanism for obtaining an address; usually the address itself.
PQI~T
WRITING: A dot displayed in the upper left quadrant of the tracking cross. Its
j".
•
purpose, as the location of the tracking cross, is to enable light pen definition of
specific points such as the center point of a circle or the starting and end points
of a'line.
PRINTOUT, DYNAMIC: A dump or printout taken during the' course of processing.
:
.
.
PRINTOUT, HARD COPY: The 'printing of infor'mation onto paper from another storage
medium or from a processor.
PROGRAM: A precise sequence of coded i~~tructions enabling a digital computer to
solve a problem.
A-8
PROGRAM, UTILITY: A program necessary to support the development and maintenance of other programs or program systems, or to support a computing and
programming facility.
PROGRAMMING: The act of planning, flowing, coding, debugging and testing a
program.
PULSE: In general, a sharp difference, usually over a relatively short period of time,
between normal level of some physical variable corresponding to the average
level of a wave or waves and a high or low level of that physical variable corresponding to the crest or trough of the wave; often a sharp voltage change.
RANDOM ACCESS: Access to the memory or storage under conditions where the
next register from which information is to be obtained is chosen at random - in
other words, it does not depend upon the location of the previous register.
SCANNING, OPTICAL: Reading information from cards or papers via a light-sensitive
device.
SCISSOR: The act of dropping out an entity from the display when, as a result of demagnification due to the zoom index, its coordinate parameters exceed the range
of the display grid. This is a software function.
SIGN-ON: The last part of the FCP initialization procedure which initiates Computer
Graphic System operation by calling the SIGNON overlay, transferring control to
a display system executive routine, and enabling all keyboard and light button
controls. The frame, light buttons, and light registers are then displayed.
STRING: A sequence of groups of conseutive computer words in which one word in
the group is a pointer to the next group of words. The string is terminated with
a null pointer.
SURFACE: The area reserved for light buttons and light registers. The area on the
CRT display surface exclusive of the working surface.
SURFACE, DISPLAY: A 20-in.-diameter area on the CRT screen utilized for manmachine communications. ,A light blue, field-free display is presented to the
operator due to components of the P7 phosphor coating deposited on the inside
surface of the CRT screen.
A-9
SURFACE, WORKING: One of two divisions made on the CRT display surface.
The
working surface is enclosed by a frame (viewing window) which is a displayed
graphic.
TABLE, PICK: Used to store references to specific parameters selected for subsequent graphic construction or processing by either FCP or application programs.
TEMPLA TES: Graphic models of components. A circuit diagram.
THESA URUS: A collection of words or other stored items arranged by concept rather
than by alphabetic order, rank, or sequence.
TIME SHARING: The use of a computer or other device for two or more purposes
during the same overall time interval through multiprocessing, interleaving,
alteration of tasks, etc.
TRACKING: The FCP function wl1ich maintains cognizance of the position of the light
pen as it moves across the display.surface. A core-displayed tracking cross is
used as the light source for the light pen.
TRACKING CROSS: There are two tracking crosses displayed, one at the lower left
and one at the lower right of the display surface and just outside the working sur.
face.
I
They are an integral pat:t of the frame 'and serve as a light source for the
light pen while tracking.
UNIT, CONSTRUCTION GRID: The spacing between the 8, 388,607 points on the X
and Y axes of the construction ,grid.
USER: The person operating the graphic console. Same as OPERATOR.'
VOLT: A unit of electrical potential difference, abbreviated V or v.
WINDOW CONCEPT: Method
us~d
whereby selected portions of
~.picture
can be
displayed on the CRT.
WORD: A set of characters which occupies one storage location and is treated and
transported as a unit by
t~e
computer circuitry. Words may be fixed or variable
in length depending on the construction of the particular computer.
WORD SIZE: The number of bits or
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-c041 52.342996, 2008/05/07-21:37:19
Create Date : 2015:09:01 13:00:58-08:00
Modify Date : 2015:09:01 12:32:37-07:00
Metadata Date : 2015:09:01 12:32:37-07:00
Producer : Adobe Acrobat 9.0 Paper Capture Plug-in
Format : application/pdf
Document ID : uuid:dcf99900-20e5-6e4b-bb33-fab1c80f3369
Instance ID : uuid:89e2cf95-5c95-734c-925d-5c8c391e33f3
Page Layout : SinglePage
Page Mode : UseOutlines
Page Count : 404
EXIF Metadata provided by EXIF.tools