Apollo Guidance Computer Activities

AGC - Conference 1: Simulation and training

Apollo Guidance Computer History Project

First conference

July 27, 2001

From left: Fred Martin, Margaret Hamilton, and Eldon Hall

Simulation and Training

DAVID MINDELL: How about simulation and training? How did these issues play into the process of teaching the astronauts, the operators, to use the system? How were changes or even decisions made in the design process then embodied in a system which can be used to instruct the users?

MARGARET HAMILTON: I remember one time when the astronauts came up and actually ran a simulation, and we went through an entire simulation, which ended up crashing at the end. I don't even remember why that happened, but there were some steps we had to take to prevent that crash from happening on a real mission. And I don't know if it would have been found if we hadn't had such simulations.

DAVID MINDELL: Crashing the vehicle or the computer?

MARGARET HAMILTON: The vehicle in the simulation.

Then there was another time when there was an error, actually during a mission. I think it was a hardware error. And Don Eyles had gone off to run a trial simulation at Draper before we sent the instructions to Houston for the real "run." Then in the first test of Don's it didn't work. Then he'd try something else, just in time and it worked, and sent it to Houston and it did work, just in time. But simulation definitely played a role during that mission in real time.

FRED MARTIN: Of course, the astronauts had an incredibly elaborate simulation facility down in Houston. I mean, six degrees of freedom. They had computers in there connected to mainframes. They had out the window displays, and they had a lot of stuff that they did with it. This works into this alarm that we talked about, that I was going to get into a little bit later, which I'm sure everybody is familiar with.

But I wanted to bring up something before I forgot it. Every once in a while you got into a finger pointing session. I remember this particular one -- this won't help me with my Grumman friend down the end there -- but this was Apollo 6 I think. I may be wrong. This was a test firing associated with an engine and other tests to do with the LM at the Cape. The firing went off and shut down in I think four seconds, or three seconds, and it had to be destroyed.  It was an unmanned mission, of course. Maybe I've got the number wrong.

The engine was shut off by the guidance computer because of an engine constant having to do with a specific impulse number. We were absolutely sure we put in the correct number, and the Grumman folks absolutely were sure that we didn't.

So the scene that I remember, though, was an interesting scene. It was all very disappointing, and of course a lot of people there were down  --

The amusing scene was we were sitting in this room afterwards and there was Doc Draper with his nose in a computer listing, which, perhaps I don't know,  might have been the first computer listing he ever looked at.  He's getting an explanation from people down there about where this, you know, these four lines of code are in the middle of this thing that's this big, that actually shut off the engine, and how that was calculated, and so on.

There was a Grumman fellow there, and I don't remember his name, insisting that it wasn't his company's fault, and it was ours. And there was us, insisting that we don't know whose fault it is, but we used the number that you gave us. Back and forth it went, with Doc looking at these instructions. This was after a test firing that was supposed to last, I don't know, a minute and a half or something, and it lasted four seconds.

CLINE FRASIER: This had to be Apollo 9, during the earth orbital unmanned flight.

DAVE HOAG: That was Apollo 5.

FRED MARTIN: Was it 5? It was something very early on.

CLINE FRASIER: Okay, 5. 9 was a manned flight.

JOE GAVIN: The thrust build up was slower than anticipated, and it was cut off prematurely. We looked at that at great length afterwards, and on the basis of understanding it, we concluded that we could go along without a second unmanned flight. That led directly to Apollo 9 which was the first manned flight.

CLINE FRASIER: But as I remember in Tom's book or something, that the slow build up was due to somebody did what was planned, which was a lower level of pressurization, or something.

JOE GAVIN: A change in rocket pressurization?

CLINE FRASIER: Yes. So it was an interface mismatch that didn't get taken care of. 

Then once the computer got out, the primary computer got out, because of the limited communication time in earth's orbit, they couldn't get it kind of hooked back up or restarted. So they went ahead and did the rest of the mission on the abort guidance system.

Management issues


site last updated 12-08-2002 by Alexander Brown