Shettles, Mack

Shettles, Mack

Dr. Rudolph, his love of charts, and their impact on the Saturn V project

Interview 1

University of Alabama in Huntsville Archives and Special Collections, Huntsville, Alabama

This material may be protected under U. S. Copyright Law (Title 17, U.S. Code) which governs the making of photocopies or reproductions of copyrighted materials. You may use the digitized material for private study, scholarship, or research. Though the University of Alabama in Huntsville Archives and Special Collections has physical ownership of the material in its collections, in some cases we may not own the copyright to the material. It is the patron's obligation to determine and satisfy copyright restrictions when publishing or otherwise distributing materials found in our collections.

.MP4

en

Interviews

Audio

ohc_stnv_000039_A

Bilstein, Roger E.

Shettles, Mack

[00:00:00] Mack Shettles: See, here's another thing, talking about the complexity in this thing. See, in an orbiter, in the space shuttle, right now they're planning to build at first a couple and later on perhaps five orbiters. But here we had a program where we were going to build 15 of these stages. How many stages are involved when we build fifteen Saturn Vs? Plus simultaneously building a number of Saturn I-Bs, like the S-IV-B stages. On that chart, you'd just be surprised how many S-4-B stages had to be built. Let's see…One, two, three, four. Four stages on every Saturn V times fifteen stages is sixty stages—flight stages. Then you had your ground test stages, you know, like the static firing, the facilities check out, the structural tests. This goes into dozens and dozens of stages for the Saturn V. These things had to follow in some sequence too. Here's one that's taken a different slice of the cake. He's following [nice is nice?]. We get 501 launched, you can't forget that we haven't gotten to the moon. We've got to get 502, 503, 504 down there, delivered, and they have to meet these technical requirements—increased payload, for example. So here's a chart which follows these things and shows the interval and checks overlap and possible problems in transporters and the like. This thing got rather complicated.

[00:02:20] Roger Bilstein: Okay, well, maybe we should...Oh, I know one of the things I wanted to ask about again was these things by Colonel [Sweet?]. Was that his name?

[00:02:30] MS: Oh, yeah. Well, that wasn't really the main part of the Saturn program.

[00:02:38] RB: No, it was just an interesting little sidebar.

[00:02:40] MS: Yeah, well, it was typical of the briefings that we had to give people. We would have generals and admirals and people like that coming in for visits. Occasional congressmen or governors of the state, or just a group of educators, you know, the president of So-and-So University and his group would come in, any number of people would be briefed on the Saturn program. We were told about Colonel [Sweet?] would be here, but Colonel [Sweet?] never showed up. There really was a Colonel [Sweet?], but he simply used his name to send unknown groups. We were always told that the composition of the group was classified, and all you were supposed to do was to brief them on the Saturn program. Sometimes they would come in, and they would be wearing turtleneck shirts and looked a little academic. Then again, they'd come in and half of them would be in uniform with a lot of brass and medals. Occasionally, they would be with an Oriental. I went over to brief one Colonel [Sweet?] group, it turned out the only guy that could speak English was their leader, and the others were all Japanese.

[00:04:09] RB: [Sweet?] was in the Pentagon some place?

[00:04:12] MS: No, he was in NASA headquarters. He was the D.o.D interface. He had a title called NASA Interagency Affairs Coordinator, something to that effect. Interagency Affairs Coordinator.

[00:04:33] RB: But he was acting as representative of D.o.D in that sense then?

[00:04:38] MS: D.o.D, State Department, CIA. See? No matter what agency. Sometimes we'd have people like…guys would come in, another swarthy-skinned individual with a handlebar mustache and sixteen stars. I know this guy's a general from Honduras or someplace or someplace down in South America. You know a whole bunch of people with him and they... I couldn't tell you who they were, but I knew they weren't…

[00:05:15] RB: I just think that it's indicative too of the number of kinds of people that were sent down here because Weber apparently felt very strongly about the success of this particular operation.

[00:05:28] MS: One of the things that NASA was supposed to do was to make public its findings and its results. That's in our charter. We're an arm of the American people.

[00:05:47] RB: Well, I'll guess that’s about [over it?]. I just want to kind of get some of these things back down and talk a little bit about the overhead stuff here. And you did have television, closed-circuit television from both Houston and the Cape? Is that right?

[00:06:01] MS: Yes. Yes.

[00:06:04] RB: So you could cover the launch.

[00:06:05] MS: Launch only, right. Now that wasn't applicable…You couldn't pick up people in the rooms. You could pick up tests, like a wet countdown demonstration test at the Cape. That's where you load the fuel onto the plane…

[00:06:21] RB: You weren't hooked up to [MTF?] by any chance?

[00:06:27] MS: I'm not sure about that. It seems to me at one time they were going to be hooked up to [MTF?]. I don't know if it ever came to pass or not.

[00:06:36] RB: I'm just curious—not that it makes that much difference. Do you have a direct hookup with [KSC?] or does that come through the HOSC over here that you pick it up?

[00:06:44] MS: It comes through the, well, the LIEF board, Launch Information Exchange facility is what they call it. HOSC is on the LIEF circuit. Huntsville Operations Support Center is, in effect, a room where all these people meet. They are serviced by this LIEF board network, which is all tied into the overall communications facility—your regular telephone systems, plus your television transmission, and [communicating?] and like that.

[00:07:20] RB: I'm just curious what is stored behind these doors.

[00:07:24] MS: Oh, those are supplies.

[00:07:27] RB:Okay, just pads, pencils, and stuff like that.

[00:07:31] MS: Extra tapes and magnetic gadgets for these boards.

[00:07:36] RB: I heard that Rudolph had problems getting this thing set up. Do you know anything about that?

[00:07:41] MS: A little bit. See, Rudolph was convinced of the necessity of this, and nobody would put out any money for it. So I think he had to sort of scramble around and scrounge. He just took this space here, which isn't necessarily the ideal space, but it's what he could come up with. In fact, at one time—sort of a fun little thing to me—people would come in and look at the paneling, and they'd say, “Oh, these people are living plush! Plush!” So Dr. Rudolph had Harold Price—who at that time was the guy that operated the control room—he had Harold Price prepare a sign that says, “The cost of the vinyl wall covering in the hall is seven cents per square foot. The cost of this paneling is six and a half cents per square foot. We are not wasting governmental money by having paneling in this thing. It's cheaper than the stuff out in the corridor.” He had it right up next to the door there.

[00:08:56] RB: [laughs] Well, was that the main problem? He couldn't get authorization to use the money for this thing?

[00:09:02] MS: See, they needed it in the back back here—you've been in the back?—you know, there's a…A teleprompter came in here and rigged all these, well, carousel slide projectors, view-graph projectors, movie projector, television projector, recording device…Then had this…This all can be remotely controlled from this lectern here. You can start the movie machine, you can operate the slides, you can’t operate the view-graph of course. But you can dim the lights, turn out the lights, and so forth. Teleprompter put all that in. I think the total cost was something like $35,000. That's what I heard for this room. We have had a lot of people ask us what it cost. A lot of people looked at the design. I think I told you, I got a call one day from Raymond Loewy Associates in New York out of the clear blue. To this day I don't exactly know what they were up to, but they wanted to help. They wanted to discuss the layout of this control room. Apparently Raymond Loewy Associates—who I think, I think, “Gee whiz, Raymond Loewy!”—they wanted to, they had somehow heard about room 319, and they wanted to know how it was laid out.

[00:10:30] RB: Why was Rudolph so insistent about getting it? Was it his experience in the Pershing program, you think? Or why did he make such a big thing about getting this thing?

[00:10:40] MS: Well, I think it was just his personal philosophy that if you couldn’t put down information on a chart in a clear, graphic form that you didn't know what you were talking about. He could perceive it better that way too. He was just one of those persons. Some people don't mind reading page after page of writing. He would rather look at a graph. Well, I think a lot of businessmen are that way. They don't see how the sales trends are. We all see this thing. Look at the stock market report, you know, things going up and down, you track the thing. To describe this in words would be practically impossible. In fact, you said it yourself a while ago. So I think it boils down to that. It's a matter of transmitting. It's a way of communicating. It's as old as drawing an arrow in the dirt and saying, “We went this way.”

[00:11:38] RB: Yeah, that's more graphic.

[00:11:40] MS: It's graphic, you know? If your [trial’s?] behind you, and you lay out that arrow or sit stones down, and that's the way you go. This is a little bit more refined. When you get into the business of semantics and communication, you're getting beyond my capabilities. It must be related to that field. It may have been impossible, as you say, I think really it would have been impossible to put in words all that was summarized in graphic form here.

[00:12:27] RB: Well, I think that really kind of covers it.

[00:12:31] MS: I haven't said anything about PERT. I don't know if I should say anything about PERT or not.

[00:12:34] RB: Okay, what do you have to say about PERT?

[00:12:36] MS: PERT was one of our more unsuccessful things. We’ve been talking about success, and I think something should be said about PERT and its unsuccess [sic].

[00:12:45] RB: Okay.

[00:12:50] MS: The critical path method that this construction firm first came up with…I think they were the first users of it. Forget who it was…Catalytic Construction Company or somebody.

[00:13:03] RB: I had the impression that PERT was the thing that was developed in the Polaris program.

[00:13:07] MS: Well, there's a matter of debate about that. I suspect that this Catalytic Construction Company, which I believe is—I'm not sure—might have been a division of Dupont. Man, the name's all wrong. Anyway, it was a pretty big outfit. In building large buildings and all, they found out they had to have this, this, this, and this, and this. They began to chart this thing out, and then they see what the critical path was. We better be getting the electrical wiring in because if we don't, we're not going to be able to perceive the rest of these things. It's a logic diagram, and you put the time intervals between events on there. Okay. I think that they did that first.

[00:13:55] MS: Now, the Polaris people grabbed it and called it PERT. They refined it somewhat, carried it to a great extreme and said, “Well, this is a really terrific idea.” Now, we used PERT mainly because we were forced into using PERT. Most of the people liked the Gantt chart approach because it had a time scale across the top. PERT doesn't have a time scale across the top. The times are simply written digitally between events. You can't really compare when things are happening. Now, there is a version of PERT called a squared PERT network in which you put a time scale across the top, and you put the events on this time scale. Which improves the heck out of it, as far as conveying the information, then you begin to grasp the thing. But we didn't have that. We required our contractors to have PERT diagrams of all our activities, and I understand this went into the tens of thousands of events. These tens of thousands of events, again, in the same manner as this, flowed up into our project offices. We in turn summarized this thing and put it on this rear wall back there—a little picture of that PERT diagram. I think it may have been helpful in 501, but then it became apparent that simply having the same network… First of all, you didn't do the same things for 502 because a lot of these things were done. Once done, there was no need to show them again. You had to revise the network. You didn't do exactly the same thing in 502 like you didn't activate the test stands every time for every vehicle. There were things like that. You didn't re-qualify the component. You see the things I'm talking about there?

[00:15:53] RB: So there's a lot of useless information that was clogged up the PERT network in a sense?

[00:15:57] MS: Well, on the first one it was there, okay. But on the second one you didn't need all that. Finally it became obvious that you weren't getting anything out of it that you couldn't get out of one of these Gantt-type waterfall charts. So we dropped it.

[00:16:12] RB: [What did you follow?] in the early R&D phase there?

[00:16:21] MS: I think we had more problems than we had problem-solves from it. I'll give you an example. You're supposed to have a critical path. This critical path tells you, “Now get on this subject here.” Well, what happened was we'd have this thing up here, and you'd come out with thirty-seven negative paths. Are you familiar enough with PERT to know? A negative path is one in which you have an end objective set up, and you have a time now and an event. If there's not enough time to take from this event and achieve this objective, then you have what you call a negative path. It has negative time. Now, if from this time event to the end objective you have more than enough time, then you have a slack path. You've got extra time in the path. Okay. Now, whichever path requires the most time between now and the end objective, it may have slack in it. It could still be the critical path. It still hasn't had time to do it all, but it just has the least amount of slack time.

[00:17:32] MS: Now, carrying it on further, if from this event now to the end objective, it turns out to be, requires 15 weeks more than you've got, then you've got a negative path, which is the critical path. You've got a negative time critical path. We wound up—like I say, at one time, I think somebody counted—thirty-seven paths, all with negative time on them. The time that separated them was not great. Let's say that the most critical path had seventeen weeks of negative slack. You had to make up seventeen weeks in order to still meet the end objective. The next critical path might have 16.5 weeks. The next one might have 15.3. The next one 13.5. There was no way of knowing how difficult the effort would be to improve this time so that you could meet the end objective. In other words, if you simply took the one with the greatest amount of negative slack and worked on it, it might have been the simplest problem to solve. Whereas the one way down here, the seventeenth most critical path, might have had a technical problem that was going to be so difficult to solve that you just could not get the slack time out. It may have been a really critical path that you really need to be working on.

[00:19:04] MS: We discussed this with some people who were supposed to be PERT experts. So we kept thinking, “Well, we're learning about this.” We turned to IBM, and we turned to—I forget who all—there was supposed to be experts. The comp lab was real hot on this because it required a lot of computer time. Boy, it ran computers night and day over there on this PERT thing. When it boiled down to it, they didn't know a damn bit more about it than we did. In fact, they never had realized the practical operating problems in the application of this thing. Did I come through to you about this negative path? I've covered it really quickly. And if you're familiar with it, it'll make sense. If you're not, I hope it's made sense anyway. I hope I've been able to explain it.

[00:19:57] MS: But at any rate, there was a certain amount of judgment, which did not come out of a computer printout. Now there's another thing. Here comes this computer printout umpteen inches thick. Now that you've got it, what do you do with it? You see, the program managers didn't have time to sit down and flip through this thing. It's very complicated. About every two or three weeks, they'd change something to do in column A to column C and this, that, and the other. It didn't come through graphically. It didn't communicate with you. The machine to man communication was difficult. It took an expert to sit down and analyze the printout and then transfer that into something that could be understood quickly by a busy program manager.

[00:20:46] RB: Well, I've got an impression that sometimes it was useful because since the contractor was required to make PERT reports all the time, it was on occasion possible for a staff manager to find a glitch quicker than a project manager could because they were getting stuff directly off the floor, hardware floor [inaudible].

[00:21:09] MS: If this occurred, you can bet your bottom dollar the next time the printout came around, that wouldn't occur again. As soon as the project manager found out about it, he picks up his pencil, and he says, “Well, really the time between this, fifteen and sixteen, I have here estimated as sixteen weeks. That's really six weeks.” [clicks tongue] Makes that a six. The next printout, it doesn't show up at the critical path anymore. It's that simple to rig it. Just change your time estimate. And they did it.

[00:21:41] RB: The managers here? Or the managers and contractors?

[00:21:43] MS: Everywhere, at all levels, wherever the pressure was put on. The contractors…I don't think our program managers did it because they didn't have the input. But I suspected at the contractor level and at the resident management office level, they'd have a confab, and they'd legitimately say we're going to work out a different plan. You see, this is what you can't argue about. You say, “Well, the reason we reduced that from sixteen to six is we worked out an action plan.” You can't tell the difference between a legitimate action plan and simply a rig in the network unless you go through the whole thing yourself. You see, there's more work involved than there is savings that comes out of it. Incidentally, I say this not as a prejudiced PERT man, because I was one of the first persons involved in PERT back in the old ME lab—manufacturing engineering lab—when we drove PERT networks for SA-5, and I thought it was the most interesting thing I had gotten into. I really went at it in a very positive attitude: “Hot dog! This is very scientific, you know?” All these prints out and all…

[00:22:57] RB: So you didn't really become disenchanted with it until though you got into this program control.

[00:23:02] MS: The more complicated the program gets, the worse that thing gets. The day-to-day application is horrendous.

[00:23:11] RB: What would you have done otherwise?

[00:23:14] MS: What we did: use these summation-type bar charts.

[00:23:19] RB: The guy…The information from the bar chart comes from some guy here picking up the phone out at Los Angeles and saying, “Hi, Kevin, I’m the…”

[00:23:29] MS: Picking up the phone, being on a plane, getting a written report. We required scheduled reports to be submitted weekly, bar chart type. We required technical reports. This is what all these people are being paid for is to put together two-in-two. They see a technical welding problem, yet they see a schedule that says bulkhead number three is going to be completed on so-and-so day. Then you pick up the phone and say, “Look, how are you going to complete bulkhead number three when your welding problem hasn't been solved? In fact, it looks to me like you're going to be running a test on that at the time you're completing the bulkhead. You can't do that.” Then they work the problem.

[00:24:18] RB: Well, that's interesting about PERT. I have to see if I can’t describe that a little better when I write.

[00:24:26] MS: I had a guy on the United Airlines, if I remember, said they were about to get into PERT very heavily one day in a briefing. I just told him very frankly, I said, “I would walk cautiously.” He said they've been told that that would solve all their problems. I said, “Let me tell you a little bit about PERT when you get down to operating that thing.” I said, “In theory, there's no more beautiful system in existence. In fact, it is! It sounds terrific! I've never come across any management thing that sounded better that worked worse!”

[00:25:08] RB: [laughs] Well, is PERT being used increasingly or decreasingly?

[00:25:08] MS: Decreasingly. I very seldom see anything about PERT.

[00:25:14] RB: They're getting back to Gantt charts again [inaudible].

[00:25:17] MS: Well, you'll see logic diagrams of people doing the early planning of projects and all, but I don't think they put times on all these things. I think it's…If I were going to plan the space tug, let's say, if I were a project manager, I might sit down and have my schedules chief fill me out of PERT network or a logic diagram of what we had to do—this and this and this—and how maybe interface with the shuttle had to be available here. I might have this on the chart, and I might look at it and think about it and have it crystallize in my mind the sequence of events. But as far as writing a computer program and start getting a weekly printout on that, I’d never, never do it. It just wouldn’t be worth it.

[00:26:11] RB: They're not using PERT now with the shuttle or with the Sky...

[00:26:17] MS: You would have to check them, but not to my knowledge. I haven't seen anything about PERT. I get a copy of the monthly report schedules book that the shuttle people put out,
and it has nothing about PERT in there.

[00:26:32] RB: [Inaudible] or is that different?

[00:26:35] MS: Well, [SART?] is just a name for this Gantt-type bar chart approach. That's what they use.

[00:26:50] RB: Okay, that’s very good about PERT [inaudible]

[00:26:54] MS: I don't want somebody to jump into that thing. That's one of our replies to this Syracuse University report. They were [knocking?]...A lot of our management techniques were rather expensive. We brought out that PERT was fairly expensive, and we did not recommend its use again. But there were techniques that we thought were worth the money by configuration management. It provided that iron framework that kept things from going to hell in a [wheelbath?]. Things could have really gotten bad without that. They could have gotten bad without all that component qualification programs, all that ground testing—you know—bugs worked out of this thing before they launched. I didn't mean to take it so long. I know I got one of them…

[00:27:53] RB: Okay, that's good.

[tape cuts out]

[00:28:38] MS: ...A rather busy period and production of the latter stages had already started.

[00:28:44] RB: This is early 1966?

[00:28:46] MS: This is early 1966, so I'll get into those charts in a minute. First, let me tell you about the control room, and most of these things…I read your write-up. You've got the hang of it already. But I think what you want me to say is how do I see it, okay?

[00:29:08] RB: Yeah.

[00:29:09] MS: Okay. [Inaudible] I'm going to repeat some of the things you said in order to keep this in my mind.

[00:29:23] RB: That's fine, because I want to really get all this down on the paper for posterity here.

[00:29:27] MS: The Saturn V was…

[00:29:36] Third person: Let me interrupt you a second, Roger.

[00:29:37] RB: Okay, go ahead.

[00:29:39] MS: All right. Dr. Rudolph, I think, will have to be given credit to [inaudible], case may be, for this control room concept. This was inherent in Dr. Rudolph's philosophy of management. We both know that every manager of anything, from the lumber mill to Polaris missile, has his own idea about how to do things. Some people get involved very personally, some people delegate the heck out of things. Dr. Rudolph had a lot of traits, one of them which was he wanted to get all the management data…You’re really breaking in, aren’t you?

[00:30:27] Third Person: Now, now [inaudible]. Roger, [inaudible] be glad to do it [inaudible].

[00:30:43] MS: Dr. Rudolph had a thing about transferring knowledge into graphic form. Some people refer to this as a fetish for charts. But there [inaudible] to be said for that. He was one of the prime movers, although headquarters would probably say [inaudible] in having this control room set up. Each chart in here is a summarization of what went on in a pyramid, as you well know, Saturn V, let’s say, S-1-C stage office, S-1-C contractor, and all the S-1-C subcontractors. This is the hardware [fashion?] now. He had all this data [feeding?]. Parallel to this, he had all these functional things that went on, which are a little bit more difficult to describe. He had a functional manager, you know, for [inaudible] quality, for mission operations, for systems engineering, and that man was responsible for going into each stage, going to other centers, and keeping track of management across the entire stage.

[00:32:39] RB: The vertical…

[00:32:41] MS: The vertical thing. That data was just played on charts in here. Basically at one time, on this side of the room, this western end of the group, these were all hardware or stage oriented charts, and those were all functional charts. The only significance of that is that he attached just as much importance to the functional part that he did to the hardware part.

[00:33:12] MS: Now, in addition to strictly hardware data, for example, in order to know what his hardware problems, progress, trends, and so forth were on a Saturn V launch vehicle in totality, he had to know what was going on on each stage. He therefore delegated to this stage manager the responsibility for accumulating all data related to this stage. Once the S-1-C stage manager [returned?], [went to?] the S-1-C contractor. The S-1-C stage manager, as we both know too, and I don't mean to make this a little complicated, but he had lateral connections with working panels, with the laboratories. This thing was more complex than I'm drawing it right here. I mean, the S-1-C stage manager had to keep on top of how his electrical system was going, for example. He also had to make contact with the engines people. At that time, the engines office was a separate office from Saturn V.

[00:34:29] MS: Anyway, Rudolph said, “All right, you Mr. S-1-C or you Mr. Instrument Unit, go out and get all this data together. [Inaudible] plan to do it. What I want to see is a summarized situation for me. I want this to be summarized in this control room in 319. I want it in a graphic form. I want it [inaudible]. I want somebody specifically responsible,”—as you pointed out, which is one of the key things to it wasn't an office, but a group of people responsible. He had a person—you can still see that person's name on these charts even to this day—that was responsible for making sure that information was correct.

[00:35:12] MS: So all this data fed up, [inaudible] it fed laterally like to the engines office or one of these labs. They summarized it on charts like these, which is a very standard method. This is just a waterfall or just a Gantt-type chart back to the picture of what activities are occurring on which stage. Now, the reason I've got this other set of charts out here is that this entire control room has been an evolving process. As our problems change, what we show changes. This thing is pretty well all done. Now, we're in a storage portion of S-II. All active, all building, all testing and everything else has been done. This will show you that this was all tracked. We had a systematic method of showing whether an event was going to be early or late. If the latter was going to be late, then this thing was followed up. We've got a red arrow [that says?] problem area in addition to having a scheduled slide thing. This was part of a big board over here.

[00:36:42] RB: The problem areas?

[00:36:43] MS: We have two things covering problem areas. We had one that was a problem area board, which listed the problem and who did what to take care of it, and what the status was…The other thing was on the overall schedule, which is still here. This was a whole Saturn V program [thing?] This gave Dr. Rudolph an overall view of things. It went back even prior to President Kennedy’s declaration we're going to make it to the moon [inaudible].

[00:37:27] RB: Master summary schedule in other words?

[00:37:30] MS: Here's JFK’s decision in May ‘61, and here's land on the moon July ‘69. All this stuff [inaudible] vehicle development, test facilities, test programs, showing all of these things. This was the big picture. These large arrows stick out in different places. Everywhere you have an arrow, on the problem board, you had a problem listed. We tracked that thing down so it got solved. I'll show you some charts in here. If the problem were of enough significance, he would make a special chart just for that problem. Generally, just listing it and using words was sufficient.

[00:38:25] MS: This sounds [relative?] and minor also, but it’s important—and you've already written it down—this control room represented the acme of data collection for Saturn V. It corresponded in every way—in symbology, in width of the bars, in everything—to the control rooms that the other centers had, that Washington had, and that the contractors had.

[00:39:03] MS: Dr. Rudolph, in fact, would get on people if they didn't have their triangles just exactly, they had their upside down to his. He didn't like that at all. Rudolph, most people feel like was a nitpicker, a [inaudible], [an imperialist?] What he really wanted, I think, was clear communication. He would just go on and on until everybody he was sure was saying the same thing and knew that they were saying the same thing. You’ve been in a meeting many times, everybody says, “Yeah, this is so, and this is so, and this is so.” Everybody at the center had a different idea. Rudolph would wrestle with this thing, and he would knead that dough until he knew everybody was on board. Then he'd go over to the next point.

[00:39:51] MS: All these control rooms were [inaudible]. A great deal of effort was spent on this. Contractors weren't particularly happy because they had to display their data not only for Rudolph, but they had to display it the way he wanted it seen. A number of cases felt like they had better ways of doing it. You can imagine the Boeing company…They've been building planes, you know, since the old P-26 and all this kind of stuff, so they felt like they knew how to run a program. This didn't matter to Rudolph one wit. He wanted it done like this. And he was right—I think—because it enabled him to go and visit and all the people under him to go and visit from place to place. Immediately the thought process was simplified. They didn't have to go learn all this symbology. Like I say, this sounds like a small thing. This isn't a…In the totality of Saturn V, this is a small thing, but in the matter of management, it sure does save a heck of a lot of time once it got set up.

[00:40:58] MS: Let me show you some of the [charts?] that he had. There were a number of special purpose charts here. He said, “Our number one objective here is to get 501”—the first Saturn V—“down to the Cape and get that thing launched.” So he's got stages of development. He's got activities—see, these are battleship tests and so forth—that must be accomplished prior to the deliverance of 501. These things were all related. See, he's got this drawn out here, and shows how the dynamic test vehicle, all the tests and all…Are we going to get these things? Is this [shaking?] going to get done in time? Will we get the results back? Will it be analyzed? How does that tie into the facility’s checkout vehicle, which was down at the KSC? What sort of problems are we having?

[00:42:05] MS: The whole thing gets back to the fact that this thing was very…It took place in so many diverse geographical locations, with so many different companies, so many different organizations, that tying it all together was the problem. Much, much activity went into this tying together business. I think this is what Jim Webb was talking about when he said he had never seen anything so pulled together. Anybody can make charts, but these charts represented a coherent display of data. Every major item could be tracked back down to its finest points. This isn't so obvious, especially today, looking back on it. I mean, you have to have been here almost to see some of this thing. You can see the gist of this. You can see that he has things other than the 501 vehicle itself. Then he shows what's going on on 501. You can see all these things, you know, like this is a milestone that's taking place early. Here's one that's early. Here's one that's late. He's got major milestones circled around…Intermediate milestones…And it's just all there in a very systematic manner.

[00:43:29] MS: They wanted to track the activation of a launch facility. Are we getting our GSC down there on time? So he's got all this graphically depicted, see? Howard Burns was a test and checkout project manager. Howard Burns couldn't direct anybody to do anything legally. That's an interesting thing. He was in charge of the test office. He had no directive authority in the classical sense at all, but Dr. Rudolph looked to him to see that this whole thing was being done. He had to get with the [LVGSE?] manager, he had to get with the Cape, Lord knows who all he had to get with…The Sanders display people up in New Hampshire or whatever. Rudolph didn't care, but Burns, in his opinion, was the functional manager to see that all the tests…Well, he had test, and one of the things under test he just allocated to him was this site activation thing here.

[00:44:37] MS: There’s a KSC detailed flow plan, shows how this thing is going to be processed.
Let me just cut through some of these. Here's a blow-up. Remember we showed you the dynamic test program and how it fed into 501? This is just a blow-up of that, of the dynamic test program. One chart, [inaudible]. If you want to see, there were some little bitty tabs back there a while ago says, “We've got a little problem,” flip over the end of the chart so-and-so, and you can follow this problem even more closely. Then here's the verbal problem area board I was talking about. The date the problem was reported, identifying it—that is—what piece of hardware is involved?

[00:45:29] RB: The S-II. [laughs]

[00:45:32] MS: The S-II. And LC-39-500F checkout, see? Which ties into the fact we just had a chart on the LC-39-500F checkout activation. He's got a man honchoing that thing. S-2-T captive firing program, and then it shows the status. Here's an extremely tight schedule for system development facility—red board—in development and debugging of integrated [gets?] operating system programs. The estimates are delivered to KSC 16 March ‘66. Then he's got another thing that says, “See display 40,” which is 500F-LC-39 and SA-501 tape requirements. So we just mentioned LC-39-500F chart. He's got another chart on the 501 tape—that is, you know—the control tapes, the launch vehicle digital control, and the launch site. They have some tapes and all. When a problem arose, it was expanded upon to the point necessary.

[00:46:52] MS: This is another thing. Not only did he identify the problem, he said, “Fine, so we know that, we know the status, and we know that this man here is reporting this thing. Now who are the points of commitment?” He had a...Rudolph—I keep quoting Rudolph because I really feel like he was the guy behind this thing—he didn't believe in systems alone. This looks like a massive system. He tied people into these systems under points of commitment. He says, “Industrial operations” —this is for the S-2-1—he says, “Industrial operations is certainly involved, and R&DO is involved, and the contractor.” Then he says, “Who at each one of these places?” Colonel Yarkin, Mr. Widener, and a fellow named Greer. Now these guys, when he says points of commitment, that terminology in itself was debated, thrashed about, defined and redefined until everybody was nauseous with it because nobody could quite decide what a point of commitment was. Some people want to say, “Well, it's the top man on the pile. He's the only guy that can really commit.” But that really wasn't the case because back in this project component qualification test programs, they got guys down at lower management levels that they could work this thing out. That's all Dr. Rudolph wanted to know. It just tells you whether the problem was something that the top man in the organization—Colonel Yarkin was the head of the S-2 project at that time—or whether somebody at a lower point in the organization can solve the problem. That's all he was wanting to do. He wanted to find out who was working that thing.

[00:48:40] MS: I think I've told you before, I know I've been briefing people who have come in here. So many people from the Army to American Airlines have these control rooms, and they've got about four or five guys or something like that, and these guys run out and try to dig up all this data. They're just scroungers of data, and they come in and they make the best chart they can. So when the vice president in charge of operations comes in, they say, “Okay, now here's a picture on American Airlines progress, so and so, our passenger rate on the West Coast route, so and so on, so and so,” and they give them all this information. That's not the way this worked. There were people who kept up these charts, but they were simply hands of people who were truly responsible. And that's the difference between—in my opinion—an active, efficient control room like this and just whitewash, you know, just hanging stuff out.

[00:49:44] RB: The people who are really involved in the actual operation were in fact responsible for the charts?

[00:49:51] MS: That's right, and they have their name on the chart. The head of the office has his name and then the man in that office—it didn't matter what his position was either. I mean, first of all, they had to have an office symbol up there so they could, you know, you just had a name, you said, “Well, what office is he in?” So they had the office chief to get you zeroed in to that degree. Then they had a person, and he could have been just a plain engineer, or he could have been a branch chief, whoever it was, that man was responsible for that data on that chart. When I say he was responsible for that data that meant he had to collect it. He had to verify it, and he had to do this periodically. He had to really know what he was putting on there.

[00:50:44] MS: Well, you came across those management matrices. That's the same thing.
Rudolph wanted to show who was responsible for structural integrity of the S-2 stage within the Saturn V program office, within the S-2 project office, at North American Rockwell, in the R&D lab. He wanted to know who is responsible for structural integrity on the S-2 stage. We had that thing plugged out. It had a structural subsystem. It had a person. If you wanted to talk that subject, you'd get these guys together, and boy, you could talk that subject. There wasn't any doubt about…“Well, gee whiz, I don’t know, I think the laboratory does that.” Then you have to call the laboratory chief and go through some rigmarole. There was none of that. It was very specific. These are quite a…Problem area things…Let's see…

[00:52:02]: MS: Now, these problems I was just talking about, that could be anything from a technical problem to a financial problem to a contractual problem. There's no limitation on what these problems are. Now, here's a specialized chart. He had his engineering manager—Lou Bell at the time—Lou Bell picked out the major technical problem areas that he as the systems engineering chief…Here's a thing called “Uncontrolled Instrumentation For Captive Firing Tests, MTF, MSSC, SACTO, and Santa Susana.” Apparently, they had a lot of different types of instrumentation for captive firing at all these places. You see, a stage might be fired at Santa Susana and then fired at MTF, and they couldn't correlate the findings because the instrumentation was different. They may have it located in different ways or it might have had a different range of pickup or you couldn't compare a thrust on the S-4-B stage with thrust on the S-1-C stage because they had it instrumented differently. The project manager for the S-4-B and the project manager for S-1-C could [sic] care less. They were doing it the way they thought best.

[00:53:29] MS: So here's where this functional manager comes in. He says, “Look, you fellas are doing this differently. We can't compare it. We've got to get this thing coordinated.” The same thing would occur in reliability. One project office would be testing a LOX relief valve to certain criteria, and another stage office would be testing it to a different set of criteria, and you couldn't correlate the things. Dr. Rudolph, as the program manager, says, “Look, this has got to be integrated.” That's why he put so much emphasis on the functional managers. I don't know that we'll ever have that much emphasis again. Maybe so. But as you think about it, you'll see why he had to have not staff members but functional managers: because they had so damn many pieces of hardware that had to fit together.

[00:54:23] RB: The functional manager, in a sense, had the authority to say to the S-1-C and S-4-B, “We've got to get the instrumentation together,” and they had to do it. They could get together because if you had the right matrix chart, the S-4-B instrumentation manager could pick up a chart and see exactly who his counterpart was in the S-1-C [stage?]. They could just get together and talk the same language and say, “Okay, we'll do this.” The systems engineer was the action manager overall for that, and he would [inaudible] them until he got it all…. And if he wanted help from R&DO, the systems engineer could pick up another matrix chart and find out exactly who R&DO was to help the two guys solve their mutual problems.

[00:55:05] MS: Right. And see, this is extended in a number of things from instrumentation to welding problems between stages. You see, I mean, why…If one guy's got a welding problem on one stage, then you need to get the people that are specialists in welding. Maybe another stage has got this problem licked. You said it very well.

[00:55:34] MS: Here's another example of a functional thing that was so necessary. This shows the mission requirements. For example, it shows the payload required for 501. It shows that 501 will be an unmanned vehicle. It says it's a launch vehicle and CSM development, 501 and 502. That's their primary mission. It shows how the payload commitment has to increase. I have to stop a minute here and think in terms of, I think, more like a mini historian. See, the [Manned’s?] building 501, and it's so far down the track that you can't go back and, let's say, reduce the weight of the S-II stage. But a report comes from Houston that the CSM is growing in weight, and fellas, the CSM is going to weigh five thousand pounds more than we thought it was going to weigh. Now, the first one doesn't, but later on we find out we're going to have to add these systems to the third and fourth flights and subsequent, so you're going to have to boost five thousand pounds more. So while you were building 501, you were changing 503 and 504, perhaps going to a different material. This did occur. It wasn't in those particular stages. They would upgrade the engine thrust, and all this sort of a mishmash. The more you think about it, you wonder how they kept it straight. This was configuration management can. We even have charts that show which vehicle, so-called first-effectivity charts, shows when this change first…

[tape ends]

0:57:42