Apollo Guidance Computer History Project

Third conference

November 30, 2001

Dave Hanley's Introduction

DAVID HANLEY:  My name is David Hanley. I guess you want a little bit of old history.  I got a degree in physics in '52 and went to work at Sylvania, on semi-conductors.

DAVID MINDELL: Where's your degree from?

DAVID HANLEY: RPI. And, since everybody at that time was making germanium, I had to go from melting my own silicon to fabricate devices, and packaging them.  In '54 I joined the Instrumentation lab and I was there for 33 years, but actually I went back part time so it could be 34 years

I joined the air force department and we made analog computers. The technology at the time was first regular vacuum tubes, then subminiature tubes, and then transistors. I got interested in digital computers and did a survey for the Air Force of all the military digital computers, that existed. It was either late '50s or very early '60s, somewhere in that vicinity. Then on the next system, we used the Veran computer made by Autonetics. So at the time, while I was still in the Air Force department, Eldon would invite me to hear presentations by Hal Laning on the Mars reconnaissance computer and very early thought on theApollo computer. When I joined Eldon’s group I worked on integrated circuits and other components.

Eldon wanted me to tell a couple of stories of how the AGC went to ICs and how it got the multi-layered boards. At the time, Albert Hopkins and Ramon Alonso were designing the AGC with core magnetic logic.. So Eldon wanted me to keep current with their design and to see if it could be implemented with integrated circuits. And after a while, I found it so difficult to convert the core magnetic logic design that I bet Eldon that I could build an AGC faster with integrated circuits directly. So he took my bet, and he provided me with one technician and allowed me to buy the ICs I needed.

I got the instruction set from Hugh Blair-Smith and, of course, that's all I needed, because from there I could put the rest of the machine together. And as the months went by the computer got bigger and bigger and I needed a little bit more help. We did have a deadline. And that's when Eldon got me a Raytheon resident, Herb Thaler, on my machine.

ELDON HALL: Herb Thaler came from Raytheon.

DAVE HANLEY: Yeah, came from Raytheon, as the resident. And that's how eventually, when we had the logic working, Eldon announced that the AGC would be built with integrated circuits, and Herb Thaler would be transferred to Albert Hopkins, and they would build the AGC. I didn't know about Eldon's going to NASA and all that stuff, I just found out a lot of that recently. So I would leave most of that explanation to Eldon.

But I'll say that Eldon did pay off his bet, so that's one thing. This is my version of how the AGC got multi-layered boards. I was playing with multi-layered boards on the side, in the process, but first I want to say how the IC logic was interconnected. We've got a little bit of the information here. The logic modules of the block 2 computer, (which used the flat pack ICs) was being fabricated by Raytheon,  using a nickel matrix for IC interconnections.  The matrix was similar to that used on the block l AGC (which used TO 47 IC) and the Polaris computer. While we were building the AGC, Eldon got a contract to build a computer for the Air Force. I can remember that it was on a Friday just before he was going on a two week vacation--that was the standard thing. He came in and told me that besides the work on the AGC that I had to build this other computer.  First he told me that I had a month to spend all the money, for the components.

I didn't know he had been talking to, it turned out, to Jim Lincoln, and Jim Lincoln turned the job down. I hadn't heard about that till later. And he said the logic had to be identical, mechanically and electrically, to the AGC and it also had to be mechanically set with the AGC. It had to have flat packs and it had to have multi-layer boards. Well, this was a pretty tall order, at that time.  About a year later Hopkins and Taylor had finished the brass board, which, Raytheon had built the modules with nickel matrix and it checked out fine. Shortly thereafter, they received the first prototype of the AGC from Raytheon. Al and Herb found that the prototype had too much noise and would not work. Eldon again was on vacation, so they called Eldon back from vacation.  NASA, Raytheon and the Apollo program office was pretty upset at that time.

When we were alone, Eldon asked me where the components for the Air Force computer were stored, and he took a set of multi-layer boards and gave them to Raytheon and within a week Raytheon had a new prototype all built up, that did work. And I was surprised the air force didn't say anything, NASA didn't say anything, Raytheon didn't say anything, nobody--they just took the multi-layered boards and that was it.

JACK POUNDSTONE: It was the first time I ever heard it belonged to the Air Force.

CLINE FRASIER: I don't think they'd tell you, Jack.

DAVE HANLEY: I have one more thing. A little bit later, you guys made another set of modules, you gave them to Eldon, Eldon gave me to me, and I put them in the Air Force computer.

JACK POUNDSTONE: I see.

DAVE HANLEY: So, everybody was happy.

DAVE BATES: I hate to say this--how did we get paid for that?

__: Ask Cline.

CLINE FRASIER: I don't think we ought to do that.

__: That's why A.C. Sparkplug took over.

__: I see.

DAVE HANLEY: There's a lot of these stories that were hidden away that made me think that things were pushed a little bit.

HERB BRISS: It looks like all informal organizations always get things done.

DAVE HANLEY: Right. It's the people, not the organization.

CLINE FRASIER: Dave, I have a slightly different slant on the story, it doesn't conflict with your facts but it adds a little bit around it, I'll get to.

DAVID MINDELL: We have one more to go, then we can get right into that.

Next: Ed Duggan's introduction

Home: Conference 3