TAPE 1, SIDE 1

MR. COLLINS: I'd like to begin this discussion by following up a couple of points from our previous discussion. In our last interview, we devoted some time to discussing the external environment that North American had to grapple with in its business planning, and one element of that we discussed was understanding the political side of the equation, and the role that Fred Black played in that. The other part of it which we didn't discuss was how North American attempted to get a handle on what the likely technical requirements were going to be, working with those technical elements of the service that had some role in shaping whatthe next projects were going to be looking like. With respect to Fred Black, one question that I had was, how was his remuneration handled? Was he on a normal salary basis or did he have some kind of different arrangement.

MR. ATWOOD: It was an unusual basis. Mr. Kindelberger was our chairman, and as I mentioned before, he was quite concerned, especially after the cancellation of the Navaho, but we had a public relations organization called Braun and Company. They were advisors to us, and we had personal relations and good relationships with Ted Braun, the owner of this group. He was quite an able public relations man, I think.

    He was the political advisor to Nelson Rockefeller, for one thing, and I don't know that that's a big success, but he delved into politics all the time, and he was advisor to us, and that was up until the time I retired from the company. Although Ted is dead and gone, the company still exists here in town, run by a man named Cliff Miller, and Fred Fisher is another principal in the company. They're quite well regarded. They run political campaigns. I mean, they sort of master mind that sort of thing. So I think they're quite prosperous, doing all right. I don't have anything to do with them except I know Cliff Miller, we've done a couple of things together.

    Mr. Kindelberger was quite apprehensive about, we always were, about the proper spending of money in this activity, so what we did was his decision, I mean his work--he dabbled in that quite a bit, to have Black work under Ted Braun. That is, he was reimbursed, and what we did there was, reimbursed Fred Black with a salary and then pay for staff support that he had in Washington, an office, a secretary, and a couple of other people whom he hired from time to time. I don't know how many. And Ted Braun billed us for those expenses like he did for the rest ofhis public relations work. I can't remember any of the numbers, but his salary was a good professional salary, like 50, 60 thousand dollars or 40 or 50, I don't know, and then he had expenses, and I'm sure he didn't skimp on expenses, because that's the kind of a fellow he was, and that's the way we handled it. Mr. Kindelberger was gone then--he died in '62--and I finally, I reached the point where I had quite a bit of confidence in Fred Black.

    I thought that he was helping us with all our relationships, Congressional and here and there, the Defense Department and other places, to the extent that such a person could do that. But after the thing came out about this vending machine deal, I finally let him go. Of course, I talked to Fred, but Charles Pickett, our lawyer, he was with a firm in New York. Of course, Charlie's dead now, but it was to try to get rid of him with as little trouble as possible--and we paid him a severance salary, for I think about a year, and he went on to other things. But he created enough embarrassment for us, I thought.

COLLINS: Were there any special incentive arrangements in his contract?

ATWOOD: No contingency fees or anything of that kind. He was just to operate for us and do the best he could. He couldn't register as a lobbyist, although you might say he should have been, working in that field generally, he was not a registered lobbyist, but he had spent his life cultivating political people, and that's the way he worked.

COLLINS: At the time that Black was working through Braun and Company, did he have other clients?

ATWOOD: Yes, he did. He had the AVCO Company, for a while, and I think he had other people he worked for. They were supposed to be completely non-conflict of interest. I'm sure they were. At least, as far as the competing products were concerned, I think they were. Otherwise, I don't know.

COLLINS: When we were talking last time, we went through this political situation fairly thoroughly. I believe you indicated that it was really equally if not more important for North American to have a good sense of the technical side of the situation as well, understanding what the services were planning or what one might expect from the services, in terms of technical requirements for coming programs.

ATWOOD: Well, I can only tell you what I think. I am a technical man. I'm not up to date on particle physics or radiation, but I think we had achieved a position of real leadership in aerospace technical fields--we built the Rocketdyne on it. We built the automatics and electronics and guidance for the Minuteman missile on it. And we had the technical tools, the wind tunnels, and the people. We tried to concentrate on building technical staff. Iwas going to tell you something about the people involved in the Apollo a little bit later on. But if that's what you mean, perhaps I'm prejudiced, but I think other people were kind of looking to us as the leader in these things.

COLLINS: The intent of my question was somewhat different. Since most of the decisions about which programs were going to come into the fore were being made within the services now, it wasn't a situation in which corporations, as in the prewar period, developed a craft, a weapons system, built it and tried to sell it on the basis of a prototype. Now it was a situation in the postwar period, especially after 1954, in which the services issued requirements and then the contractors required to those requirements and prepared a proposal. So there was some need to be able to organize your resources so that you could properly respond to those proposals, and there was some need for an advance sense of where these things were heading. How did that element work?

ATWOOD: Well, actually, it was this. In most military projects, let's say, it usually turned out that there were one or two key technical elements that were really critical, and frequently, some of our people at least could anticipate these things. Now, you can go a long way into how close we should be with the military, and we've even found people being prosecuted for it today. Forgetting the money part of it, but trying to understand their problems and what's critical to them was our business, I felt, and on many occasions, pursuing the key technology, spending some of their own money, making tests, and experiments, and mostly laboratory tests of things more than completed airplanes, was frequently decisive in these competitions. I tried to administer that part of the work myself, and a week never went by that some people didn't come in with a plan for spending some money to further master some technology or technical aspect of something they were working on or had started to work on.

COLLINS: You're referring to North American personnel.

ATWOOD: Yes. And I did that without committee. I would sign out the money. For some of these guys, I should have the cash in my desk drawer and dole it out to them. It was practically that way. It was not too bureaucratic, and it was quite successful. We had many many occasions when something like that was the key factor, and people still come up to me at these management club meetings and so forth and say, "Do you remember when you authorized this and we were working on a certain project?" and I really don't, but I can imagine the circumstances, of course. That was the way I thought that corporate money and effort should be applied, if you could make it an arrow type effect or a rifle type effect, you were lucky, and you could make a big impact on the competition.

COLLINS: I'm not clear what you mean by a rifle type of shot.

ATWOOD: Every project has aerodynamic problems, and they involve stability, drag, performance, landing speed, maneuverability. Then they have power plant problems, or installation problems at least. We have electronic problems, as you obviously know, and if you can sort those things out and concentrate on one, that's what I call a rifle shot, rather than parceling out everything in advance and spreading your money over the whole spectrum.

COLLINS: With the thought that this would probably be one of those critical technologies that you were talking about for a particular competition.

ATWOOD: Oh yes. And a demonstration of something that's an improvement in one of these critical technologies is worth a lot, when people are trying to evaluate different technical presentations, as you can imagine.

COLLINS: Was it obvious, what these critical technologies were, what critical technologies the services would be seeking or would you have to engage in discussion?

ATWOOD: Not always, because from the specs, they don't know exactly how much to stretch the technology in every element. They try to stretch it, but they also try to do something that--well, it's all what you might call the art of the possible. No, they don't always know what's the critical thing, although often they have a good idea. See, the relationship with the military is an old long-standing thing.

    As you say, we used to, first thing build a plane and then the Army bought it, and it went on like that more or less until World War I. Of course, most anybody who set up to build airplanes could start in and build something, and then that all collapsed, and during the twenties, it was kind of a bastard arrangement. Wright Field, for instance, and the Navy at Philadelphia designed and built their experimental planes. People like Douglas and others were building planes, too. But the procurement was not very well organized or equitable type of thing.     

    In 1926, the Air Corps Act was passed, and that more or less aimed to take the government out of building aircraft, and established the design competition method of procurement. It so stated, really. And we went into that. Then in '32, '33, '34, the competition was changed to the requirement that a demonstrated article be supplied. So people would then have to build a plane and bring it in generally to specifications that the service had provided.

    At North American, we built a bomber, a two engine bomber, to enter in the competition, with the company's own money, and we lost. The Air Force finally bought it, at the exorbitant price of $555,000. But we didn't win the competition, Douglas got it. Also we built a trainer, which became the AT-6 eventually. We builtanother plane which was a medium bomber. We called it the Model NA-40. We spent plenty to build it. But it crashed in Dayton. Didn't kill anybody, but the pilot made a mistake and it crashed on a hillside. Then--I'm just giving you examples, because I know what you're getting at. The procurement was running on that basis up until about 1939, '38 or '39, and then of course, the wartime thing. We went back to design competition. That's the way we finally got the B-25, an open design competition. And you know pretty well what's happened after the war.

COLLINS: I guess another way of putting it would be, how would you be able to make judgments about which of these technical areas to support? Obviously it would be based on your experience, but I would also expect there'd be some sense of your understanding of what the services, particularly the Air Force, were indicating that they needed?

ATWOOD: I think you're exactly right. All procurement services have and have had for a long time a pretty competent group of technical people to make evaluations, you know. The Air Force has always had, and the Navy, Bureau of Aeronautics. The people there are scientists, engineers, who have devoted themselves to that civil service type of thing.

COLLINS: One of the interesting things about working with this advanced technology is that different groups who work on the same problem come up with different solutions, approaches, preferences.

ATWOOD: That's the competition.

COLLINS: But when one of the elements to be considered is the government's own preferences, and predilections, how do you gain an understanding of what those things are?

ATWOOD: Well, some are long-standing, Martin. For instance, I can remember having a conference with two of the Air Force officers back in Dayton maybe 30 or 35 years ago. They, one of them in particular was holding forth on what they wanted and what they needed. In the fighter planes, they wanted a plane that would take off the ground vertically or within 20, 30, 40, 50 feet, whatever they could get, and do its job, come back and land between shell craters. You know, short landing. That was obviously a big desire on their part. That's a military requirement, kind of underlying. How close can they get to that?

    Well, on the one hand, you've got the helicopter, which can't fight much of anything. On the other hand, you've got the high speed landing planes that have the carrying capacity and the range and all that. And I've been watching that go on all the time. Now the British have a Harrier which is closing in on that capability. That's one type of standard requirement. Others, of course, have more detail. But the military requirements are controlling because they're objectives. They don't tell you exactly. How, and that's where the engineers make the compromises, I know, on the technology and the development.

COLLINS: Well, one trend in this postwar period I think is a move from general requirements like the one you've just mentioned, into a more precise statement of requirements. How did you keep on top of understanding when the situation became more detailed like this? Did you have something comparable on the technical side to what Fred Black did on the political side?

ATWOOD: Well, yes, I guess you could say so. The big scandal these days is people getting private information out of government people, isn't it? That's one of them. And that, of course, gets to be kind of a matter of degree. It's absolutely a crime to bribe somebody to come and give you information. On the other hand, if you're conversing with a military officer and he gives you an idea in conversation with him, that's not done anything except help the overall effort of the department. And in between, there's a very large area where it begins to become corrupt at some point.

    In the regime of McNamara as Secretary of Defense, he initiated the policy of preventing military people from accepting ordinary things, such as a meal, a drink, that sort of thing, from contractors. And it's gone on that way pretty much to the present day. I think it isn't the value of the bottle of scotch or the steak that you bought somebody so much, I don't think, it's the fact that maybe they're getting too clubby with some and not equally with others. It sort of runs that way, I guess, and it's a difficult equation to solve.

COLLINS: One interesting element of that is--apart from the ethics, which are very murky--that there seems to be a shift for substituting the technical judgment of the corporation with that of the technical judgment of people in government, and there's something about the full resources that are available within the corporation that may not be brought to bear on certain kinds of problems, technical problems.

ATWOOD: Well, there are so many phrases and flavors. The one that's stuck in my mind, at least for years, is the B-1B bomber. It was contracted for by the military, the systems command in the Air Force. The airplane from the North American people, and then there was a cluster of associate contractors. The engine, of course, the offensive electronics system, the defensive electronics, and maybe something else. But I think that was an error, now.

    Of course, I had no control over it. I was retired when that thing came in. But the problem they've got to work their way out of now is one that should have been solved in the systems workout. For instance, they should have had equivalent airplanes,it didn't matter what kind, an old transport plane or something, that could fly the defensive system, the offensive system, all the electronic gear, to the point of satisfaction, before they put it all into the B-1. Now, it's quite possible that they were reaching for a technical objective that made a parallel development schedule with the airplane itself necessary. Maybe they couldn't do that. Of course, if you could perfect everything before you put it in the final vehicle, you're a lot better off.

COLLINS: You're suggesting that one of the problems with the B-1 was an attempt to do too complex a system?

ATWOOD: No, I didn't make that point. I hope I didn't. I make the point without trying to place any blame, that the project office of the Air Force Systems Command should have had somebody closer to the nuts and bolts integrate these electronic systems. I can picture an entirely different company, for instance, General Electric or General Dynamics even or someone else could have integrated all those electronic systems better than the project office at Dayton, because they are not nuts and bolts people. And when I say nuts and bolts, I really mean electronics and gyroscope people. You know what I mean, so I don't know whether that covers your thought or not, Martin, but, I don't think that the procurement system is anywhere near perfect. I think it's far from it. But I don't know how I could improve it really very much.

COLLINS: This is the kind of general question which over-arches this discussion--how does someone in your position as president and chairman at that time adapt to the situation, and what do they do to make sure that the corporation continues as a viable entity?

ATWOOD: Well, number 1 is technological capability and base. It runs all the way from major effort on processing, like welding, heat treating, finishing and forming and materials research, all the way through to the high speed aerodynamics, radar, and all the elements of technology that you have to bring to bear, or would like to be able to bring to bear. And if you can't afford that kind of thing, you have a real handicap, so you go down to components, you have to do it on trying to get some good results that way. I think during the fifties and into the sixties, we'd come about as near to that balance of technical capability as any company that I know of, with the actual technical people and the products to exemplify it.

COLLINS: What I'm trying to get a better picture of is how one relates that technological base, the technological resources, how they're organized and deployed in the context of the broader environmental situation, your relationship with the services and with NASA.

ATWOOD: You're best off if you actually have some projects under contract that embody these things, because you can't afford to go out and spend huge amounts of money without some support. And it's a big problem. It affects everybody. You don't know how you're going to win competitions. You never know. Some you win, some you lose. I had every hope of winning that F-15 contract, which peaked up in 1969, but McDonnell won it. I think to this day that we probably were better technically, had a more efficient airplane. I think that from just what I've heard and what I see. But they had something awfully strong going for them, that production line on the F-4, which was very highly tooled and in great shape, and I don't know what influence the cost estimates had on the decision. You hear rumors. I know our design was good, as far as I'll take it, and I know that some capable people thought it was superior. But McDonnell did a fine job. They probably got pretty much all the performance out of that engine anyway.

COLLINS: Does this tie in at all to your reflection in our last discussion that you felt that the North American concentration on the NASA business hurt its relationship to the Air Force?

ATWOOD: I don't think it hurt the relationship any, really. I think it probably hurt in this way. We were really in an overload condition, and I had almost a draft on the people. It's like, taking them out of one division and bringing them to another. I think it might have weakened the airplane people's competitive capability. At least in some facets of it, I don't know which ones. But I don't think it was because of poor relationships with the Air Force. They always had a good word for North American. All our planes have been liked by the people, the pilots. And interestingly enough, when I got that Air Force Association Hap Arnold Trophy, it surprised me, but that's Air Force people.

COLLINS: Well, let's conclude that discussion. To set the stage for a more detailed discussion of NASA, I'd like you to sketch out the North American contact and relationship with the NACA in the fifties, up to the time when NASA was formed.

ATWOOD: Well, we were building the X-15, and that had sparked the idea of space in everybody's mind. We'd gotten that contract from, I think the Air Force ran a competition, although it was a joint project, NASA, Air Force, I guess the Navy participated to some extent. It was the first exo-atmospheric manned vehicle that we had. It was done by people like Harrison Storms. Charlie Feltz was the project engineer. They were just airplane people, but the work was quite good and quite successful. One plane was lost, but it was not blamed on any of our engineering. It was a special flight control unit worked out by some of the government people at NASA, they were trying it out, and it caused the plane to go out of control. But it was late in the program and they were just experimenting. It was really quite a program. You know, we had to cover that stability for a very big spread of air density--that is, air density because of altitude, and speed--and then we had to control it in space, with these little jets like they had onthe Mercury and Apollo. So it was a big step ahead. We went to Mach 6, I think, finally.

COLLINS: To take this as an example, what was your working relationship with NACA, during this, because this began before NASA was formed?

ATWOOD: Well, I thought it was quite excellent. In fact, people like Storms and other aerodynamicists were, well, I'd say were very compatible, and they knew each other very well, people like Walter Williams, at Edwards, and all the people at Langley, the people at Sunnyvale. I thought the relationships were, as good as they could be with any

COLLINS: Why don't we move into our discussion of NASA then? We've talked about, at least in a preliminary way, the awarding of the two major contracts, the Saturn II and the command system module.

ATWOOD: The rocket contracts stemmed more I think from the fact that we had existing rockets, rocket engines for the military.

TAPE 1, SIDE 2

COLLINS: With the awarding of the two major contracts, I believe it was in November, 1961, North American had a substantial boost in business with the awarding of these contracts, and I'd be interested to get your perspective on how this posed any special organization or managerial problems for North American in gearing up for this work.

ATWOOD: Yes. Generally, the first award in the program was the F-1 engine, but that was under the jurisdiction of Keith Glennan in the Eisenhower Administration. But the first one in the Apollo structure was the second stage of the Saturn V booster. That was awarded as a result of a design competition. It was awarded in September of 1961, and it was at that time rather vaguely defined. It was anticipated it would use four engines. Eventually it had five. It was changed when the requirements of the Lunar Landing Vehicle were finally worked out in 1962. And of course, we went to work on that, and I might as well give you my summary on the Saturn program. I have some notes and if I can run all the way through it, I think it might be a saving of time and simplify the whole thing.

During the Saturn rocket program, I found it very useful to refer to a NASA publication entitled STAGES TO SATURN by Roger E. Bilstein, dated 1980.1 This book is documented by extensive bibliography and references and is in my opinion a very good source book. In my comments and notes, I will refer to it extensively with selected quotations, and all page numbers will refer to this volume unless otherwise noted. I believe the publication to be quite accurate in everything described and quoted, although I detected one error which might be considered trivial by many, but which might mislead the casual reader.

    On page 192 it is stated that liquid hydrogen, LH 2, is only one half as dense as kerosene. Actually LH 2 weighs only .59 pounds per gallon, whereas rocket fuel, petroleum distillate, weighs about 6 1/2 pounds per gallon, for a ratio of 11 to 1. This misstatement would particularly distort the appreciation of the reader with respect to tank sizes. Liquid oxygen, incidentally, is 9.46 pounds per gallon.

    Since I intend to take the various elements of the program which are discussed in an analytical manner, I will rely on the data from the Saturn V flight manual, dated 15 August, 1969, and amended 5 October 1969. I will include reproductions of certain tables and charts. The few opinions I intend to express are not necessarily those of anyone else.

    First, I will reveal the characteristics of the first stage, S1C, and its influence on the rest of the strike, using some simple calculations and data from the flight manual. No great precision is suggested and accuracies of the slide rule variety. This stage is filled with petroleum distillate RP 1 and liquid oxygen, and was powered by five F-1 engines rated at 1,530,000 pounds each. Since the total stack weight was 6,348,000 pounds, see, the flight manual, and the S1C weighed 5,222,000 pounds loaded, and 288,000 pounds dry, the S1C stage fuel and oxidizer weight was 4,734,000 pounds. It remains that the total stack weight at the end of burnout, the S1C burnout, was 1,614,000 pounds.

    Since the five F-1 engines were capable of burning fuel up to the point of exhaustion, it was apparent that the S1C was capable of accelerating the stack, which included the full second and third stages, and the command and service modules and the lunar excursion module to an acceleration of about 4.7 G. That is, five times the engine thrust over the weight, at burnout, or somewhat more in low atmospheric pressure. This acceleration is actually not obtained, since the center engine is programmed to cut off at just under 4 G, the flight manual, and the remaining fuel and oxidizer is used for the remaining four engines figure 26 attached.

    This explanation is offered to make a point relating to the basic design of the Saturn V, namely that excessive G forces that put a strain on the crew and might interfere with their safety and efficiency also, that it would require increased structural strength in the upper stages. This would be rather self-defeatingin weight control efforts. This will be referred to later in connection with the design criteria of the other two stages. Of course, some additional impulse could have been gained by staging the shutdown of the outer engines if more fuel capacity had been available. The Saturn burst tubes exceed the requirement for heavy space launches before the lunar program was asked for by President Kennedy, and such concept was almost contemporary with the initiation of the one million pound, later one and a half million pound thrust F-1 engine, which was ordered by NASA administrator Keith Glennan during the Eisenhower regime.

    At the start of the Apollo program in May, 1961, contracts were let for a three stage launcher Saturn V, the first stage being ordered from Boeing, the second stage from North American, and the third stage from Douglas. However, it was late 1962 before the lunar mission was clearly defined as lunar rendezvous method with a separate lunar lander. Douglas had been building a stage for the Saturn I which, from an earlier designation was called the S-4, and by enlarging the design somewhat it became the S-4B and served as the third stage for the Saturn V booster. North American started from scratch, and the configurations were developed and sized after the lunar orbit mode was definitely decided.

    The decisions leading to the finalizing of the dimensions of the Saturn V were most critical and depended upon estimates of several key parameters, including the weights of the command and service modules, the lunar landing vehicle, the thrust ratings of the F-1 engines, the G-2 engines for the S-2 and S-4 B stages, and the weights of the booster stages themselves. The efficiency ratings of the engines were also very important, and these estimates cumulatively were really far off the mark, changing the booster dimensions in the advanced design stage, would be unacceptable from a schedule point of view. As it developed, the engine improvements and weight control actions were sufficient to make the mission possible, but not by much margin.

    The development and engineering of Saturn V is a story of power versus weight, with the overlay of limiting factors, time and reliability. First, a short discussion of the power element, main engines, five F-1's and fixed J-2's, were designed and built by North American's Rocketdyne division, and the initial specifications were consistent with the weight estimates, with some extra margin. Missile requirements, page 110. The F-1, thrust, 1,500,000 pounds fuel RP 1, kerosene and LOX, LO-2, specific impulse 260 seconds. J-2 page 151. Thrust, 200,000 pounds, liquid oxygen and liquid hydrogen, specific impulse, 4l8 seconds. For reference, specific impulse is a measure of fuel efficiency and can be considered as the time in seconds that a given propellant will produce a unit of thrust.

COLLINS: Resuming after a pause. During this break, I've read over an extended memorandum or essay prepared by Mr. Atwood entitled "The Saturn Program" in which he sets out the NorthAmerican, from his point of view, some of the central issues in the evolution of that contract for the Saturn II rocket. What we'll do is we'll append this document to the interview. Now I'd just like to ask a few questions. One point for clarification, and really the essay I think is kind of an extended comment on it , but I think some summation of this point by you would be useful. This is why the S-II became the critical element in the development of the Saturn V and resulted in the very focussed attention by NASA on North American's performance in this area.

ATWOOD: I tried to make it clear that the technical effort and the time it took to make the weight budget dominated all the schedule problems of the S-II.

COLLINS: One point you make here is that NASA spoke with many voices. Were you referring strictly to your interaction with NASA with respect to the S-II program when you said that?

ATWOOD: There's a strong element there, yes. The performance aspect of it. You see, we stipulated or agreed to a mass fraction of 92 percent when we took the contract. We came out with 92.4, which is exceptionally good, as you can see, by comparison to anything else, and it was designed and redesigned part by part, as a stripping, they say shaving in the book, shaving weight, by means of design, and wherever we think weight can be saved, we go back and redesign it. That means thinner walls in the material, reinforcements put in at a certain point, thickening and stiffeners to save weight by thinning something else. It's an involved technical process, and it's very, very well known. All airplanes go through it to some degree. And as I point out here, we froze the design of the S-1VB against the conditions that were not stringent as far as weight control was concerned, and it's so stated in the book.

    But it would have been most productive if they could have gotten the weight out of the S-1VB because you see, when it burns out, a pound there is the same as a pound in the lunar landing module, as far as acceleration is concerned and velocity, I think it must have been driven by two things, one, the S-1VB was in production, for the Saturn I, you know, for these dummy launches we made. Of course, the first dummy launches were made at White Sands, New Mexico and then down at the Cape for the Apollo, command module launches, orbiter launches and re-entry launches and all that. And the other was that we had specified a pretty high number, a very high target. It was a big reach. We didn't know that exactly, but see, the SIC you couldn't do much with, as I've explained, it wouldn't help much if you could.

    The S-1VB was stabilized, had been used, and I think that they just took for granted the fact that the S-II was going to reach these goals, which were right on the limit of structural capability of the material, and then, once that was established in the minds of the schedule people, it was, "Why don't you finish it?" It was almost that simple. Just get it done. But theother group was pushing for cutting weight the whole time. It's very obvious, it's well described in that book. Nobody has ever taken the time or even the interest to make the comparison of the results. There's even more contrast when you consider that the size of the S-II, I think the engineers at the Cape must have realized how difficult that common bulkhead would be as a structure, because of its size. It's a definite law of engineering that size always penalizes you, in weight as dimensions increase. (square-cube rule) And so, if you wanted to run this right down to the ground, the S-II was even more of an accomplishment in weight saving, and these numbers show it, and it was very, very hard to do.

COLLINS: In this program, as compared to Air Force programs, NASA's philosophy was that the centers--in this case Marshall-- would play an active role in technical direction. What was the nature of the working relationship in which these problems were jointly addressed? Was this question of deciding on a design with a common bulkhead between the two fuel storage containers something that was worked out mutually? I note according to your story here they resisted this design characterization, but how would that kind of design issue be worked on between the two organizations?

ATWOOD: Well, actually I think this was an unusual case. Let's just take the philosophy, conservative philosophy, which is expressed in STAGES TO SATURN in a number of ways, in a number of cases. If it didn't work the first time, we were dead as far as the date of 1969 was concerned for a lunar landing. And they did attempt to apply conservative factors. On the other hand, the people who did the orbital mechanics and did the sums on thrust, weight, impulse, what you had to carry and all that. They were kind of, I think, boxed into a corner, to make us perform on the basis of the weight estimate, and it was just a difficult thing to do as you can see. Now, you mentioned airplanes and the Air Force. If you had two airplanes of the same general specifications, where the difference between the useful loads was 20 percent in between the two of them, you can't imagine anything that grossly out of line. Two transport planes, two bombers. And that's exactly what we had here. Even more than pertained to these numbers. But the conservative philosophy ran through the place. Von Braun and his people were conservative. They were also realistic. They had to make this weight or start over again, really, re-sizing boosters, maybe put another engine on the S-IVB, God knows what?

COLLINS: If we could look for a comparison, when you ran into technical problems on an Air Force program, my understanding is that it was primarily your responsibility to resolve the problem, find a solution. In the case of either the Saturn or the command module, when problems occurred, it was I think meant to be at least a kind of joint undertaking in finding a solution. Is that a fair characterization?

ATWOOD: You put your finger exactly on the unusualness of this circumstance. You're exactly right. NASA is--I don't think of good comparisons right off, but trying to do two things. And it results in a very strange situation. If things are done well, NASA succeeded; if things are done poorly, the contractor failed. And it is like the hatch on the Apollo command module. Originally they had an explosively blow-off hatch that was postulated. Then about that time, the sinking of Grissom's (Gemini) capsule came along. He claimed, in fact, he said he did not touch any control to make it blow off. And I guess he didn't. But it blew anyway. And so the designers in the Houston group--of course, I'm not speaking for them, but they told our people that they didn't have to have a blow-off hatch, they didn't want the crew to have access to some switch or circular lever that would blow the hatch.

    Charlie Feltz who was the project engineer on that command module, I've talked to him at some length, he lives down in the country here. Of course, he didn't go very far up the line, but he was worried about all these things. But NASA decided that the hatch should be opened from the outside, really, and it was designed that way. Of course, we never did go back to a blow-off hatch, but we had one where you could pull one lever and it disconnected about six lock points and pushed out. Before, you see, it was kind of a plug type of thing where they couldn't get out unless somebody helped them. Well, I don't know. When it came right down to it, who was responsible for that? NASA approved every design. The mockup people went through everything with a fine tooth comb again and again and again, inspection after inspection, from a boiler plate wooden mockup all the way through. And the only thing you can say is that nobody connected the over-pressure of oxygen with the fact that these lads were locked in. For some reason, I don't know why. They'd just gotten away with it on most of their flights before.

    So the NASA did have, to me, an anomalous position. They were not exactly the builders, but they were the control designers, and they could go as far into detail as they wanted. It just depended on the person and the problem. The NASA engineers could look at something, go over the drawing and approve it, or they could work their way down to the screws and the bolts and have them changed. But the Air Force doesn't do that kind of thing, and the Navy doesn't do that sort of thing. As long as you meet their general specifications.

    So you're exactly right, NASA did go into the detailed design. But they didn't have, of course, the number of people that the contractor had, and they couldn't go back and build it from the ground up, in design. They only could look at what was coming through and request changes, which they did. Informally. You see, as that paper you showed me a while ago shows, NASA never came up with a specification for the CSM.

COLLINS: Just for the record, let me read the reference to thisthing in here. This is a memorandum for Mr. Webb from John Biggs who was apparently one of his staff assistants. The memo is entitled "Negotiation with North American Aviation," dated April 20th, 1967, and one of the attachments is entitled, "Possible Criticisms Which May Be Aimed at NASA by North American Aviation."

ATWOOD: Well, that was an interesting paper, all rights almost cynical. It might be connected with something I'm going to give you later, anticipating a defense from the company.

COLLINS: Right. Well, I guess one of the interesting problems in understanding the NASA-North American relationship in this period is this question of the complexity of both of these contracts, and hence the amount of time that elapsed from the awarding of the contracts to more definitized versions of what precisely was to be done.

ATWOOD: They were open ended. And Marshall tried to keep a running skeleton type of specification, as you can see. They wanted an open ticket for changing. Later on, of course, people like Phillips tried to not maintain--change control. But I've got, in my other paper here, I've got some comments on that and personnel too, which I'd like to give you.

COLLINS: Given this new kind of working relationship with the customer, how did North American as an organization respond to this different way of working, from your perception? And how did it have to adapt, to work in this different fashion?

ATWOOD: Well, I didn't get much of a feeling of it, of a problem, so much. People in our company were just as enthusiastic as the NASA people, and Harrison Storms was a very aggressively minded type of a technical man. And he was a real designer and promoter? of technical progress, and exceedingly good at it. But when it began to get very sticky was about, oh, 1965, '66, '67 period--well, take the stack. The whole stack in the period of 1967, '68, '69, was all electrically controlled. All the instrument readouts and controls, and activities of everything, were electric, and that meant that starting at the console in the blockhouse, and the engines at the rear end, and every action that took place in the staging and launching of the vehicle had an electric wire, and in most cases, these wires ran to the command module. So a large fraction of the changes that were made in all these things affected some wire connection, instrument, wire routing, switch, circuit breaker, something in the command module. You just couldn't stop the changes, because the changes were the result of something that the makers of the command module had nothing to do with, in so many cases.

COLLINS: So this is in part some kind of interface problem.

ATWOOD: It was, and of course the command module would only tolerate about four or five people working at one time inside. Itgot to be quite a difficult thing. In other words, to answer your question, it was very, very difficult. It grew that way. It didn't start out that way. It just got worse and worse. Sam Phillips put on a very heavy drive to control changes, did everything he possibly could. But when something had to be done, it had to be done, and its roots were fairly long.

COLLINS: When you say its roots were fairly long, you mean its ramifications.

ATWOOD: Yes, really, its branches were fairly long, rather than roots. So that's one of the relationship things that you're trying to get a feel for. I didn't sense any real problem with relationships with NASA. In fact, in some areas, I never did have any problems. But there was a kind of an anvil chorus that kept getting louder and louder and louder. And you'd be surprised, things that were absolutely reasonable, like the static test of the S-II, it was just about where it ought to be, and yet that was one of the biggest negatives that we ever had. So it can build up that way. But I don't have any feeling that there was any cause other than natural explainable reasons, and that's what I'm trying to do here, is to explain it.

COLLINS: Before work was initiated on the Saturn II contract and the command module contract, what regimen did North American have in place for managing large projects of this type? Was there some codified set of procedures, or was it more just an informal regimen based on experience?

ATWOOD: Well, of course, we were in the airplane business, and we'd produced maybe 50,000 airplanes, counting the wartime production, and there was pretty much a regular method of doing these things. I thought our system was really quite good. I think earlier in my tape recording, I talked to you about how during wartime we developed methods of controlling and improving and changing and all that. And after the war of course we had the F-86, the F-100, we had the XB 70, which was a very remarkable project comparable to Space Shuttle in a way, as far as complexity of advancement was concerned, maybe even more advanced in some respects. Of course, just as a run of the mill thing, we had production contracts and we had experimental contracts, and they were different, but they all grew by certain steps and stages and rules and controls. For instance, engineering--we had a method of course where the manufacturing people worked with the engineers, the engineers were designing, the building people would suggest a better assembly method or something. There was a feedback continuously, but as you go along, you'd get your design pretty well stabilized, and you'd go and finish it, finish building it, and the results always involved some deficiencies, after you'd test, you'd make some changes, always some improvements. Then of course we had the regular factors, as any business has, we had material procurement, manufacturing, inspection and so called quality control, we had accounting, we had reporting, and all the treasury functions and personnelfunctions, all those things that go with any business.

COLLINS: What I'm leading to here is the question of whether or not the approaches used for developing your sophisticated aircraft and for handling your experimental projects like the X-15 and for R and D projects like the Navaho provided an adequate basis or were directly applicable to the type of work that was done for NASA on these two contracts.

ATWOOD: Well, of course the interface is really an engineering and--the interface with NASA, the interface with any customer, but with NASA there's much more detailed interface, because the NASA engineering people come and really park, you know, for long periods of time. They'd deal with your soldiers and your desk men and other people keeping up with what's going on, and progress reports, and generally reporting every time anything looked out of line to them. It was really quite a surveillance then. They came in teams. Of course the astronauts had a great deal of clout in these things, they all had some engineering background, one way or another, at least something. But it wasn't the kind of engineering background that goes with detailed stress analysis which everything has to go through, the drawings and the fit of bolts and the fit of rivets and the techniques of welding and all those detailed things that the company has to know and practice. The inspectors, of course, are ranged too from those that are trying to make sure the controls and instruments and all that are placed in a good place for the pilots, all the way through to inspection of the fit of parts and the accuracy of machining and all those things. But very rarely do the NASA people get into inspection problems so much as they did into the engineering.

1 Roger E. Bilstein, STAGES TO SATURN, (Washington: Scientic and Technical Information Branch, National Aeronautics and Space Administration, 1980)

Rev. 08/13/96