Upthread: More on Program Development (Oct 10, 1966)
Downthread: AS-206 LGC program status (Oct 14, 1966)
See listOCT 11 196666-FM1-124FM/Deputy Chief, Mission Planning and Analysis DivisionProgram Development Plans are coming!!
September 29th shall probably go down in history, at least in my diary, as the day of a major breakthrough at MIT. On that date we had an all day meeting attended by all key MIT management personnel involved in spacecraft computer program development. I expect it to be the first of similar weekly sessions for as long as they are required. The purpose of these meetings is to establish detailed program development plans for the spacecraft computer programs. This basic information is required for the obvious purposes of understanding the schedule situation, of evaluating the impact of program changes and additions, of assigning priority of effort – both manpower and facilities – in the optimum manner, of providing vital information to NASA program management for consideration in their decisions, etc.
I mast say I was tremendously impressed with the cooperative, earnest support all of these MIT people gave to this effort this time and have every hope that it will continue for the four to six weeks of hard, weekly meetings I expect will be needed to reach our objectives.
At this meeting, most of our attention was spent on two items which I will discuss in some detail. First was the availability and adequacy of the computer facilities needed for computer program development, and the second was our investigation into the use of the AS-278 computer programs with minimum change for the AS-503 mission.
At present MIT has two 1800 digital computers on which all program devel- opment and verification is carried out. These machines have been and are currently completely saturated. There are no other facilities in the entire universe, to our knowledge, of proper configuration to relieve this situation completely. This is identified as a major problem area parti- cularly during the months of November and December. However, an IBM 360 is to be installed at MIT very soon and it is currently estimated that it will be on line no later than February 1st. As you recall, we have funded AC to the tune of about $300,000 to develop a facility in Milwau- kee for use on Block I program development, i.e., for AS-501/502. It was emphasized that maximum utilization of this facility is essential.
It was discovered during program development for AS-204/205 that the hybrid facility at MIT was an extremely valuable tool for program debugging. This is apparently because it is so easy to get on and off this machine; in addition, it runs considerably faster than the digital computer. Thus, it is possible for the programmers to check program fixes quickly and determine whether they seem to be working before committing the program to the all-digital tests. Phil Felleman of MIT presented a complete schedule of the tasks currently planned for the hybrid computer through calendar year 1967. This schedule showed that almost continually there are a number of vital tasks which must be carried on simultaneously, or at least on a time sharing basis. This is expected to present serious problems and we are currently looking into the possibility of augmenting the facility to relieve it. In particular, an almost ideal set of hybrid equipment is available at Beckman – a system which had been under develop- ment for MPAD – which MIT can obtain immediately at a “bargain price”. Additional pieces of equipment such as a Block II AGC and a core rope simulator must also be obtained from some, as yet, unknown source. MIT is continuing to formulate plans for augmenting this facility including obtaining for us the influence it would have in improving the computer program development schedule. Specifically, this augmentation would make possible the simultaneous use of the command module and LM cockpit simulators at MIT. In addition, it would give the unique capability of being able to run data flow tests and simulations of these two spacecraft in conjunction with each other, which will certainly be highly desirable for preparation of the AS-278 mission. It was strongly emphasized that the purpose of this facility is not flight crew training, but rather is for the development of the spacecraft computer programs and associated crew procedures.
The second half of the day was spent in discussions of how the AS-278 programs could be used in support of the AS-503 mission. A number of routines were considered for beefing up the AS-278 program, but after lengthy discussions only two candidates were left outstanding. One was the lunar orbit insertion (LOI) program which is certainly not needed to fly the AS-503 mission, but which it might be advantageous to test on it. The second and more important processor which we probably must add to AS-278 is the trans-lunar injection (TLI) steering of the SIVB. This program will probably be needed to obtain the experience of AGC steering the SIVB on AS-503 before it is used for the actual TLI maneuver on AS-504. Of course, it is not yet certain that the AGC will be used for this purpose on AS-504, but its likelihood is great enough that we should be prepared for this important spacecraft systems test.
Our next meeting will be Wednesday, October 5th during which, among other things, we expect to review program plans MIT is preparing based on the following ground rules:
1. Schedules should show influence of augmenting the hybrid facility.
2. They should be based on the assumption that the AS-503 will be flown using the AS-278 programs. The AS-278 programs will be augmented as necessary to do this, but it is expected that no more than the two processors noted above sb11 be added for that purpose.
Finally, I expect we will review open items remaining regarding the “final” definition of the AS-278 program. Stand by for the next exciting episode.