NASA Johnson Space Center Oral History Project
Edited Oral History Transcript

Jack Funk
Interviewed by Jennifer Ross-Nazzal
Houston, Texas – 5 November 2003

Ross-Nazzal: Today is November 5th, 2003. This oral history with Jack Funk is being conducted for the Johnson Space Center Oral History Project in Houston, Texas. Jennifer Ross-Nazzal is the interviewer, and she is assisted by Rebecca Wright.

Thank you again for joining us this afternoon. We really appreciate it. I’d like to begin by asking you about your work with the Space Shuttle Program.

Funk: Well, the Space Shuttle Program, my work started with reviewing the contract evaluation, which we actually went to New Orleans [Louisiana], but the evaluation was at the Michoud [Assembly Facility] test site for Saturn rockets, actually. The contractors, of course, present a contract and a configuration which is based on all the requirements, which are mostly put down by other people up in [NASA] Headquarters [Washington, D.C.] and things like that, and we review it and try to select the best contractor, although usually it’s like a toss-up between two or three of them. But we have to put them in order, and they don’t always select what we say is the best one, either one.

Then I went back to the Center and continued work on a few things, but the configuration they presented, right away I saw two items that I didn’t like. The first one was that they had air-breathing engines in the payload bay, and they weren’t piston engines. They were the jet engines in the payload bay, and they were going to just [put] them out in the air after they came through the entry and got back down into flight status and turn them on to use for landing. You know, it just almost looked to me like they were asking for trouble. They’re going to open up the payload bay when they’re moving at, say, three or four hundred miles an hour. Well, that’s going to destroy the aerodynamics, for one thing.

Two, they’re going to get awful loads on those things. You know, you put a door out in the middle of all that, and you’re going to get awful loads on it. Then they’ve got to air-start [the engines], and if they don’t start both engines at the same time, they’re in deep trouble, because they’re going to get drag on one side and thrust on the other, and they probably [can’t] recover from that.

So when I was flying one day coming home from [a trip], we landed, and I began to think, “You know, I’ve been flying a lot, and I can’t remember once, except when I was flying to Seattle [Washington], that we ever had to go around.” So I went to Bill [Howard W.] Tindall’s staff meeting, who was the Operations Director, and told him I thought it was a dumb idea.

He said, “Well, the astronauts want it.”

I said, “It’s still a dumb idea.” I told him the X-15, the first airplane to fly faster than the speed of sound, they always did a dead-stick landing. They didn’t have any propulsion when [landing]. I suggested that [Bill Tindall] go out there and talk to those astronauts, show them what they’re thinking about doing, and check with them about what they think of the idea and what they think of the idea of landing without propulsion. Well, Tindall didn’t go out, [I think it was Robert R.] Gilruth, [but I am not sure of who made the trip]. When he came back, the engines were gone.

So the other item that right off the bat looked really not too good was that they had—[because of] the fire on the pad—[Apollo added] solid rockets, two big solid rockets, on either side of the payload bay above the wing that [were] here in case they had trouble on the pad, they could light those rockets off and pull the engine out and go out and ditch in the water. But the structure necessary to hold those [rockets] looked to me like it was going to be horrendous, and they had no other further use. They were going to light them off after they got up [to] a certain speed because the weight that they were holding, they didn’t want to take that all the way to orbit, and they were going to fire them to use the propellant rather than staging.

But that’s when I started to figure out what could we do to get rid of those things, and I did some work at my desk with drawings and everything like that about putting [oxygen] and hydrogen inside the Orbiter, just small tanks, enough to fire the main engines. [The main engines] could take it off of [the tanks] once they’re started and use them in the places of solids.

I gave a presentation on that to Max [Maxime A.] Faget and his Engineering Division and other people. Max didn’t like it because of the hydrogen, and I agreed with him, yes, hydrogen is hard to keep from leaking. So, essentially, it got turned thumbs down.

Now, let’s see. How did that go? Then I suggested that they could use—no. When did the OMS [Orbital Maneuvering System] engines come along? Oh, what happened is they decided that the big rockets on the side were reliable enough not to have to back it up with; the probabilities of failure of the system they were talking about abort were probably bigger than the probabilities of the side rockets failing, although we did have one fail during Challenger. But that was never noticed, so those other rockets would have never been used anyhow. So that’s the way that got solved. We just took them off.

The next thing that happened was that a note [came] across my desk saying that Bill Tindall didn’t like taking the hydrogen tank into orbit and then getting rid of it by firing a little rocket on it and putting it down into the ocean somewhere, because he thought that in order to make that maneuver firing it, that you were going to have to put an attitude control system on that. That’s something that could pitch it up and hold it while it was firing. That was not only going to be expensive, but it looked to him like a pretty difficult thing to do, and he sent a note over to Mission Planning and Analysis Division that he thought that we ought to look at some way of doing that, of replacing that particular way of doing it.

Well, I went right back to the idea of the abort system and said that if we could use something like that to suborbital stage the tank into the Indian Ocean, and the reason why you pick the Indian Ocean is because a lot of trajectories, this inclination and that inclination, they crossed right there, so that you could put it down there for almost any flight you were going to make. It was a good place—very few boats, very few tankers coming through there. So I started working on that.

Well, Max didn’t like the hydrogen, so I said, “Well, let’s use the OMS system. Let’s increase the tank size and put them down in the wing roots,” where there’s area in there that you could hide double sets of tanks on both sides to fire the OMS engine. He didn’t like that either, but that would have cleaned up the aerodynamics nicely. The thing flies like a brick.

So, well, I made several presentations on that, and they turned thumbs down on that. Then a little note came across my desk, says, “The OMS requirement for the OMS maneuvering onboard,” that’s what the OMS is, [Orbital] Maneuvering System, “has been reduced from 1,000 feet per second to 750 feet per second in order to save the weight of the propellant.” Right now it was already designed to carry 1,000 feet per second with the propellant.

So I shipped back a note to Tindall, told him, “Hey, that 250 feet per second’s enough to suborbital stage the tank.” Then you use it for going into orbit, see. You stage the tank, you use the 250 feet per second to go into orbit, and that increases the payload a little bit, plus it picks up the payload, carrying that rocket, the 7,500-pound rocket that they’re going to use to retrofire it into orbit, and you get rid of the tank weight sooner. That all added up into better performance.

I said, “It doesn’t matter right now. Let them go ahead with what they want. If it doesn’t work, we can drop back to the staging of the tank.” Well, that put the managers in a tough spot. If they go ahead with going into orbit and bringing it back and it didn’t work and they had to go put an attitude control system on there, and at the same time the Martin Corporation in Denver [Colorado], who was building the tank, they come in and made a presentation, said, “Yeah, we will save $70 million on tech [technical] design if you suborbital stage it.”

So from then on, they changed over to the suborbital staging, and we did a complete analysis of—we wrote a program to do the launches, stage the tank, put in all the possible errors in the launch. We had error statements that the engine would not have exactly the right characteristics. A lot of other things would not have the right characteristics. We had those all, and we put it into a program that made the launches for all these different characteristics and made a table of the performance, made a drawing of the footprint of the possible errors in causing the impact points to be changed.

Of course, at that time I had a few people that didn’t have anything else to do. Apollo was doing other things, and they wrote this program. You took it over and you put it in there and pushed a button, out come the report, to show everybody, that was to demonstrate, in terms of simulation, that we could do this. So that was about all I did for Shuttle. [The contract was given to a company.]

Then came the Skylab. It was already up there. [NASA Headquarters in Washington, D.C.] started a project, $20 million, to design an orbiting vehicle that we’d launch into orbit. It would go up there and grab a hold of the Skylab and push it up back higher so it wouldn’t come in. Got a note from Gilruth—[it was a verbal request delivered by division chief]. This project was under Huntsville [Marshall Space Flight Center, Alabama], under [Wernher] von Braun and company. [Von Braun started the Skylab Program. I do not know who was Director of the Marshall Space Flight Center at the time it returned to Earth or who came to JSC to present their analysis of the Skylab trajectory.] Got a note from Gilruth, wanted me to take a look at the project and see what I could do, just keep track of what’s going on.

I got a little document, a little paperlike memorandum, that Huntsville had published, that said that they couldn’t very well simulate the Skylab with their program. They didn’t know any way to take the digital computer and figure out what was going on and things like that. They said they didn’t think they could predict the impact with any confidence.

So I looked at that and I wondered what I could do with it. I had this report that was written by Dr. Kaplan from Penn[sylvania] State University [University Park, Pennsylvania] about a method of modeling Skylab in a wind tunnel with just three degrees of freedom of pitching to get some idea. He thought they could predict. Well, they did make a prediction of where it would start tumbling, because when it starts tumbling, the drag goes up, comes in much faster. It was already coming in much faster because of their prediction of what the upper atmosphere at those altitudes would be like was a little low, and it just was coming down faster.

I said, “Well, I can take those equations and then just pick up the orbiting [program] and let it produce all the forces for the orbiting program,” so that we could track it, and then it would also show all the motions of the Skylab. So, got the aerodynamics, and we had to put 1,400 points into the table and things like that and model this thing and started running it. And every time I’d run it, it would tumble. I was putting in atmospheric data from the ten-centimeter [radiation from the Sun that JSC was] getting from NOAA [National Oceanic and Atmospheric Administration] on the solar flares and everything else that [I could think of but it] was always tumbling. I looked and looked and looked for an error in that program and couldn’t find it.

Well, one day somebody came in and said, “The Air Force says Skylab is tumbling.”

I said, “Oh, my program’s not wrong after all.” So I said, “The thing for me to do now is to go way back and pick up the Skylab data and get an initial set of conditions to start out the program and figure out where the program says it tumbles.” Well, getting the initial set of conditions, the attitude—I had the velocity and the orbital position and everything like that, but I did not have the attitude. But they had recording gyros on there, which they were recording the attitude before it tumbled and during tumbling.

So I went back, and what I did, I started running error coefficients on the position and do a least squares fit to the gyros. And when I got the position that gave the least error in the gyros, that’s a normal way of fitting data to curves, been there ever since they invented it back a couple centuries ago to do the initial orbits for the satellites. I mean like Mars and those things. Once I got that and put it in and ran it, then it started. It ran along, and after a while it tumbled. I looked at what I was putting in that made it tumble, and that was when a solar flare [which] increased the temperature and densities of the upper atmosphere. It made it tumble at that time.

So I went over and I told Tindall that—I think he was still—it was Tindall or Ronald [L.] Berry, probably Ronald Berry, that I wanted to make a presentation to Gilruth on what I’d found out. [I do not remember who was where. The task of writing the Skylab dynamic model and adding it to an orbit program was very difficult. I know I made a presentation in Washington, D.C. and suggested that they drop the program to put the Skylab in a higher orbit because it was going to come down before they could build the required system to push it to a higher orbit.]

I made a presentation to [Gilruth] and showed him what I’d found out, and then I said, “This idea of going up there and picking it up, getting something that can hold on to it when it’s tumbling or even grab it when it quit tumbling, is pretty—well, you don’t have real confidence that you can do that. Even when it wasn’t tumbling, you don’t have real confidence to do that. The other thing is, it’s now coming down so fast that I don’t think you can get the system ready to go before it comes in. So the thing to do is to cancel that project, bite the bullet, and let it come down.” And they all agreed. So I was there when the impact—and then they did all this. That’s the trajectory of it coming down, from the tracking data.

So then I was sitting around, practically nothing to do, and I had thirty-six years of service, thirty-six and a half, and I had one year of health leave I’d never used, and that gets tacked on as a year of service. Thirty-seven years. And at forty years, my retirement income will never go up. You get 2 percent a year, so I was looking at about 6 percent. Inflation is running at 7 and a half percent, which retirees were getting, 7 and a half percent increase in their retirement income. But the people on [civil] service, the president could cut that 7 and a half percent down, and it was usually cut down to about 5 percent.

I looked at that and I said, “Well, if I retire in a couple years, I’ll be making more money and I will have more income than if I stay here.” So I started looking to put out résumés to companies that were aerospace companies.

Well, Sig [Sigurd A.] Sjoberg had retired, and he was Gilruth’s assistant for quite some time. He had taken a job with a company called OAO Corporation, which supported McDonnell Douglas [Corporation], which supported NASA. So I sent my résumé to him. I actually didn’t have to send my résumé to him. [Laughs]

He called me up as soon he got it and said, “Come on to work.”

I said, “Well, I got to give them two weeks’ notice. Ron Berry says I have to give them two weeks’ notice.”

He says, “No, you don’t. I’ll make a telephone call.” So Sig called Ron Berry and said he wanted me now, right now.

So that’s why Ron Berry [came] in and said, “You can retire.”

So I went over there. He sent me to McDonnell Douglas, and McDonnell Douglas put me to working on doing changes to the SVDS [Space Vehicles Design Simulation] Program, which was run by MPAD [Mission Planning and Analysis Division]. [Laughs] I was right back, as an employee rather than a section head, doing work on trajectory programs for NASA. But I didn’t have any trouble. My father taught me, “You do what I tell you, and then when you grow up, you’ll learn enough that you can tell other people.” [Laughs] And I spent five years there, doing [computer program changes]. I guess you don’t want to know anything about that? You want to terminate that?

Ross-Nazzal: If you’d like to share some with us, that’d be great.

Funk: Well, the first problem [I] ran into was being able to run the program. I’d go in there, and it would bomb every time. I said, “Hmm.” So I sat down and I said, “I’m going to run all the inputs here one at a time, putting them in one at a time, until I find the minimum number of inputs that makes a program run,” because if you can’t run it, you can’t debug it. Or I mean, you can’t [add] anything else to it.

I had a users’ guide. It told me all the options and everything else, but it didn’t tell me that one thing. So I started writing a users’ guide, and I got it to running. My job was to add the program for navigation for the Shuttle from people who had done it, you know, standalone into the program so they could run a simulation of rendezvous with the Shuttle with the radar.

Well, first of all, I just took that program and ran it by itself so I knew that when I hooked it up to the main program, if it didn’t run right, it was something to do [with] the hookup, not what was in the program. We hooked it up to the program and, sure enough, it wouldn’t run. So we looked and we looked, and I had an assistant and we looked. I finally looked and said, “Well, every time we run this thing, it points the radar in the wrong direction.”

So I called the people up who wrote the program. I said, “Come on over here. There’s something wrong that’s not necessarily in the program.”

They came over and they looked at the attitude matrix that we were using to mount the radar in the payload bay. They took one look at that and they said, “Well, that’s wrong.” So they give us the right numbers, and from then on everything worked.

Well, let’s see if any other unusual things—that’s about the only significant part of the problem. They did send in Dr. Gottlieb, who came from Huntsville, up to help put in some models of the—was it the OMS system? Yes, into that program. He was working with me. Every time he would come in with a problem that couldn’t run, I said, “Well, that’s because you don’t have the OMS characteristics right.” I knew what they were, but somehow he didn’t have it. I knew. He did that for about a month or two before he got it written right.

That’s why I performed so well. I could read the program. The people you bring in that are program majors, they don’t know the system. They also can’t read the code. When I see, “ISP,” I know that’s an engine characteristic. When I see a lot of other things, they [are systems or trajectories]. I was sitting there, looking at a printout one day, and there’s six numbers for X, Y, Z, and X.YZ, which is position and velocity for the Shuttle, and there was another set and they weren’t the same. Now, first of all, these people wouldn’t recognize what it was, what those numbers were, and that they weren’t the same. I knew why they weren’t the same, that this was computed for a different set of inputs than that one.

I said, “There’s only one way that could happen,” that the [subroutine] called “WHEREAT” is in [the program] twice. Where is it? The printout for [the computer code is in] notebooks that are about that long [gestures]. Where is it?

Well, I had another assistant, a person that was working with me, that was good with the editor, so I told him to take the whole program [and make a copy]. We had lots of storage then. Put it [on]. It would be our program. They’re not going to use it. “I want you to write an editor routine,” and this was a good editor, “that takes and puts the subroutine de-bug at the beginning of every subroutine that’s in that program. [The editor] can do that automatically, and make [debug] pick up the name of that subroutine [it is in] and print it out, and then it returns to the main program.” … [Debug can be programmed to look for errors. When debug finds the errors you know the name of the subroutine that has errors. As an example it found that a program called WHEREAT was called twice in the program and gave different outputs. We deleted one The source code for SVDV required three to four feet of shelf to store the notebooks containing a printout of the code. McDonnell Douglas lost the NASA support contract to UNISYS and that program was lost.]

But that produced an interesting de-bug [technique], which I don’t think I’ve got. I could go into that program, and if I wanted to find out what was going on, I could write a program that would only run when it came to [designated] subprograms and print out all the information I wanted. …

So things went along pretty good until the program got so that it’s too big, so they put a segment in, run it and take it out, then put another segment in it, run it, take it out, and they were gett