Apollo Guidance Computer Activities

AGC - Conference 1: Cline Frasier's introduction

Apollo Guidance Computer History Project

First conference

July 27, 2001

Cline Fraiser's introduction

CLINE FRASIER: I am Cline Frasier. Let's see, I started at NASA in I think '62, and I came to NASA. My only experience with guidance or control systems or anything, was teaching anti- aircraft fire controls for a couple of years when I was in the service. Then I had spent a few years at Sandia Corporation where they design the nuclear weapons. Dave talks about these ones in a suitcase -- I actually packaged up one of those once in a suitcase and took it. I was amazed to discover it in the atomic museum at Kirkland Air Force base in Albuquerque, so for the first time I could show my wife what I had worked on in those days.

But I think my role in NASA was partially to be a thorn in everyone's side. They mentioned the autopilot.  We were going around looking at all of the various systems and how they were going to do reliability. Because originally the concept was -- either NASA's or North American's or someone's -- the concept was to have in-flight repair of the complex electronics. And it was, I think Scott Carpenter's flight where they got a lot of water in the electronics, in the Mercury one, and that put an end to the idea that you could have in-flight repair because how were you going to seal up all the connectors?

So that then led to well, how are you going to get enough reliability in the space craft. And you had the MIT-designed guidance system which is a fairly conventional overall configuration where it fed steering signals to an autopilot that was in the spacecraft. So North American was going through well, how are we going to handle the redundancy in the autopilot?

On the MIT side, they had put in two computers because Eldon was concerned about the reliability of these computers, and they wanted to be absolutely sure they could do navigation. To do navigation, you needed a computer and the space sextant.

Then if you got a failure in the MIT system, then you had to have some kind of back-up pointing capability for an emergency return to earth in this North American system, plus all the redundancy for the stabilization.

So the thought was at that time, even though I fortunately knew so little as to be dangerous, was gee, we could simplify everything if we would put the primary autopilot in the guidance system, and then North American could just have a much simpler system. And we could also combine the functionality of some of their gyros with what they were going to do for an emergency reentry system, and cut down the number of gyroscopes by a large margin.

Being an old mechanical engineer, well, at the time a very young one, I was always nervous about all of these mechanical parts running around.

So that led to a proposal to, let's see, I was working in the Flight Crew Systems Division at the time, and talked with -- NASA had just brought down Cliff Duncan and Joe Shea to Houston. I talked to Cliff in a meeting and suggested, you know, here is one way that you might solve this whole reliability problem. One thing led to another, and that's how we ended up going with North American.

Still one of my memories of the program that will probably never go away was here I was, I was 30 years old. I didn't know anything about most of this stuff, but I managed to kind of push through this concept, in part because Joe Shea was a program manager. He had also been the program manager on Titan where they used a digital flight control, which I didn't know about. I knew about the Minuteman flight control, digital flight control.

So Cliff sent me to the Instrumentation Lab to go up and tell the guys at the Instrumentation Lab, pretty much out of the blue, that instead of two computers there is only going to be one. Instead of just doing the guidance, they were also going to do digital autopilot, which as I recall, at one point Dave, you said you guys couldn't do it. You had too much other work to do, or something like that.

DAVE HOAG: That sounds like me.

CLINE FRASIER: And even worse, since Milt Tragesor had worked on the space navigation with the space sextant, and at the time, they were going to be the primary navigation, the on-board. So I also had to tell them that guys, you're not the primary navigation anymore, you're the back-up. We're going to do the primary navigation from the ground.

Dave, I think you were in the meeting. I know Dick Battin and a few of the other guys who really knew what they were doing, and senior, and I was up there all alone. And I began to feel just shrinking, and shrinking, and shrinking in the process.

Then we went on to -- North American was fairly easy to push it in. They were told by Joe Shea that this was the way it was going to be, and that was that.

Then it seemed like, that it worked well for the Command Service Module, it will have lots of advantages in the Lunar Module too. Grumman was not so easy. Grumman, I think, deep in their hearts believed that anybody was crazy to think that you could do digital flight control, particularly for a light agile vehicle like that. And they were in exactly the same league as all the astronauts who thought this was the craziest idea ever.

We probably never would have gotten it in, not in both, in either one, if the astronauts hadn't been really busy with Gemini. Because it was a done deal when they found out.  I also remember Pete Conrad coming up to me in the hall one time. Pete, being a navy guy, and I been an old army guy -- he didn't know that -- but I felt like he was putting me in a brace, and he just chewed me up one side and down the other, and said that we're crazy. That his friends at Grumman had told him this stuff wasn't going to work, and it was his life on the line, and kind of on and on, and on.

Then I ended up getting the job of being project manager for the development of this system, or kind of getting it all manufactured. It was all of these people here and the people at AC and Raytheon that did all of the work. I was a NASA person. So Grumman wasn't too happy with me over the digital flight control.

Then there was a time when there was a problem with the radar. There was supposed to be a rendezvous radar in the Command Service Module, and there was a rendezvous radar in the Lunar Module so that you can rendezvous. The one on the Command Service Module had real trouble, and it wasn't getting solved.

So I proposed, or I discovered from some of the work I think that Norm Sears had done at the Instrumentation Lab from the error analysis -- It dawned on me that gee, you could do rendezvous without measuring the range and range rate, since you were under inertial forces, if you could measure the angles accurately enough. And lo and behold, the sextant would measure the angles accurately enough.

So rather than try to fix this 100 pound monster, expensive monster on the Service Module, we could just throw it off, and then to provide kind of additional baling wire, we could use a scheme that Grumman was using with the rendezvous radar to measure range and range rate. We could add that to the communications link because that didn't require an elaborate, steerable antenna. So we had the back-up system essentially for free. Not for, well, it was for free, because it cost nowhere near as much to put that in the communications loop as it would have to fix up the rendezvous radar for the Service Module.

Then when there was a little weight problem with the Lunar Module, I somewhat audaciously suggested that we replace the rendezvous radar on the Lunar Module, and the telescope that was used to align the guidance system, with an optical tracking system that would replace them all.  Then that would take off I think 75 to 90 pounds of dry weight out of the Lunar Module, which is about what they were over weight.

That, again, was one of the things that was not seen with great favor by a number of people, and it was considered to be impossible to build. But we actually did get one, and got through flight weigh prototypes in a year from scratch, and Grumman got the radar working, and found ways to reduce enough weight on the Lunar Module. So NASA, in their wisdom I think -- and I don't mean that sarcastically, really, in their wisdom -- decided they could live with the weight on the Lunar Module, and keep with the system that everybody was really comfortable with, and confident with.

Then I went on  from Apollo to Space Shuttle.  The one thing I want to mention too is that one of the people that is not here that probably had as much to do with driving, getting the computer actually manufactured and reliable, as anybody was Hugh Brady, who was the program manager at AC Electronics Division. He died about a year ago, but he was a person who really drove things from the manufacturing side.

So then I got assigned to look at guidance and control systems for the space shuttle, and one of the things we did there was we wanted to use a digital flight control and use data bus technology to get rid of a lot of the wires through the length of the vehicle. And we went through all the same arguments again about how the space shuttle was an aerodynamic vehicle. You couldn't possibly think about using a digital flight control for something that was manned, and people wouldn't fly in that kind of environment. The astronauts were against it. All of the traditional control system people were against it.

So I was leading the NASA electronic technology task force, and the folks at Edwards wanted to build, use some technology money to build a quad-redundant analog flight control system to beat the Air Force to having a quad-redundant analog system flying.

And what we suggested to them was that they couldn't have the money for that, but they could have the money if they would do a digital flight control system, and that we'd give them some used Apollo equipment to do that with so they wouldn't have to design reliable equipment. The outcome of that was that the people both in combination at AC and the Instrumentation Lab, largely at the Lab -- Phil Felleman -- We took a returned system out of a Command Module, put it -- they did, not we, they did it -- put it in an F8, which was a fairly high performance fighter at the time, and flew it at Edwards Air Force base.

The first time it took off with that system in it, they had enough confidence that they took off using the digital system. So they had an Apollo, the slow, old Apollo computer, the inertial system, the CDU's, the whole business, and it performed very well.

Then from that point I went on to a number of other things, none of which were ever as much fun or as interesting.

JOHN TYLKO: Can I ask just a couple of questions?


JOHN TYLKO: You kind of glossed over the decision to shift the primary navigation from the Apollo on-board system to the ground. Could you explain that in a little more detail? What were the drivers behind that?

CLINE FRASIER: Initially, let's see. (Laughter)  I will tell you how it initially got to be on board. That's because the people figured out a way, and nobody thought you could do it from the ground, and all of the usual stuff.

What came along was the MSFN and atomic frequency standards, and the realization that by using largely range rate information, you could use a very large base line across the earth and the antenna steering that you really could navigate quite accurately.  To do it, if you said you can't do it from the ground, then you have to have redundant navigation on board the space craft, and we didn't have that, and no way to provide that.  As you can see, you can do it from the ground okay. Why not free up the whole mission planning and everything, and just go ahead and do it from the ground as a primary mode? That then removed the requirement to have the redundant computer on board. It made their life much simpler.

As it turned out, it was really important for the mission because of the work load of the astronauts, the desire to be able to do the toaster thermal control. In fact, without the digital, the combination of the digital flight control and the navigation from the ground, we would have been at least a couple of years later, I don't know how much. The same thing I think is true with regard to the Lunar Module. The flexibility of the digital flight control system gave them, kind of squeezed out a lot of stuff.

Of course, we wouldn't have gotten back Apollo 13 without it, without the crew having put in a somewhat untested ability to guide from, stabilize the thing from the Lunar Module.

I can't remember the rest of the rationale for why we did it, other than it would clearly save some weight and some time. I mean, they had to work.

DAVE HOAG: The intent there was always to have the ability on board, without help from the ground at all, to rescue themselves.

CLINE FRASIER: Right. That was demonstrated on Apollo 8. Jim Lovell demonstrated that on Apollo 8. 
I was on a panel with Jim two years ago, and when he was talking, he had the -- Anyway, he was asked a question about Apollo 13 and was he really frightened or not, or scared, and he said he was never scared as bad as when on Apollo 8, coming out from behind the moon, he punched the wrong keys and the wrong thing, and dumped the alignment on the inertial platform, and then had to do for real, get it realigned. (Laughter)

Now that didn't have anything to do with navigation, but Jim was a person who people said had the golden finger, because he did navigation so well on Apollo 8. 

DAVE HOAG: That's right.

Eldon Hall's introduction

site last updated 12-08-2002 by Alexander Brown