Notes from the MIT program development plan meeting of April 12

See listAPR 28 196767-FM1-35FM/Deputy ChiefNotes from the MIT program development plan meeting of April 12

A few things came out of our program development plan meeting at MIT on April 12 that might be worth recording for my memory and your information.

1. We have decided not to produce a unique LM program for its first manned mission. What this basically amounts to is that MIT will continue with program development pretty much as they have been for AS-208, but instead of freezing the program at some point in time (probably in June) and flight qualifying it for rope manufacture, they would begin to add in the extra programs required to support the lunar landing mission. This would con- tinue until the complete lunar landing program has been developed (probably in December 1967) or until it became apparent that the projected release date of the complete program was no longer consistent with the scheduled launch date. If the latter happened, we would freeze the program in its most suitable configuration, flight qualify it, and release it to Raytheon for rope manufacture. This procedure should provide the best, most com- plete program for the first manned LM and should also result in the least work required by MIT. In addition, recognizing that “shelf life” of a pro- gram is really quite short, we remain in the most favorable position to correct deficiencies as they are uncovered without having to perform extensive reverification or, even worse, having to remake rope modules.

2. In line with that last remark, I would like to emphasize that no one involved in the development of programs as complex as these would ever be so naive as to claim a program was completely free of bugs. In fact I would like to emphasize just the opposite – that it is almost certain that defi- ciencies will exist in the program we ultimately fly. Of course, extensive effort will have been spend in checking the program to prove that it is safe and should perform adequately in the mission. It will always be our respon- sibility to describe accurately the quality of the flight programs to manage- ment, highlighting known deficiencies and describing, in whatever detail is required, the verification procedures which have been undergone in addition to stating some measure of our confidence in flight worthiness. We must not mislead management by overstating the quality of the flight program, since to do so is to implicitly take it upon ourselves to make the ultimate deci- sion that a program is flight worthy – and that is not our job. This subject was discussed at MIT, and the reason that I mention it here is that I'm not sure this position is clearly understood by everyone.

3. I had thought, and have told others, that Sundisk – the old AS-205 program – was a true subset of the AS-504 command module program. This turns out not to be the case in that not only must unique processors be added to Sundisk to support the lunar landing mission, but also some of the Sundisk processors themselves must be modified before they will be capable of supporting both earth orbital and lunar landing missions. Our current plans are not to make those modifications prior to the Sundisk release.

4. A number of modifications and additions are being made to the Sundisk AS-205 program in accordance with the wishes of various elements of MSC, such as use of the reticle for aligning the platform, providing the pilot the “mark” input capability from his couch position via the DSKY, and some others. There is at least one big one, however, which will not be added because its impact on release of the program is too great. This is the capability of performing rendezvous navigation (i.e., state vector updat- ing) utilizing the reticle observations.

5. We have requested MIT to provide us an MDRB for changing the program such that it will recognize depression of the Standby button as a signal both to “Standby” and “Proceed”. We expect to make this program change regardless of whether or not the computer itself is modified, since that is immaterial to the program and the likelihood of its being implemented in the computer is quite great.

6. We had been asked to have MIT investigate and propose RCS jet select logic capable of handling individual jet failure as opposed to total quad failure. This would obviously require considerable amount of MIT effort and significant changes in the spacecraft computer program if it were to be implemented. As a result, we told MIT to do nothing on this until a much more precise requirement had been established by MSC.

7. We have recently had discussions with MSFC regarding the use of the LM spacecraft computer servicing their Apollo Telescope Mount (ATM). There was some concern expressed by MSC people that it might be necessary to pro- vide coupling of the LM and command module guidance and propulsion systems, particularly to carry out the terminal rendezvous in the docked configura- tion. Without commenting on the advisability of this cross coupling for this purpose, I would like to remind everyone that according to MIT a data channel has been provided on these computers to handle that sort of an interconnection. I must confess that I didn't realize they existed.

Development of the spacecraft computer programs seems to have become routine enough that our program development plan meetings should occur no more fre- quently than once every three or four weeks, and it's our intention to schedule them that way. The next meeting is now set for May 10, in Beantown.