Rogers, Henry

Rogers, Henry

J2 Engine

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_000037_A

Bilstein, Roger E.

Rogers, Henry

[00:00:01] HR: Well, let me tell you what my background is. I came on J-2 when I came here in August of ‘60. The contract was [let?] September of ‘60. As far as my dealings with Lewis prior to that time, I had none. From the time that I got on the contract, it was a matter of…It was already a design that had been bought by the government, so to speak. Of course, the design that we bought as we progressed through development ran into problems, which there were times where we borrowed on past experience. Best of my knowledge, all of the design concepts in that engine, Rocketdyne had certainly utilized those.

[00:01:01] HR: At that time, the only thing we had [hydrogen fueled?] was the RL-10. It was a somewhat different engine in that way it started and, of course, the thrust levels. It was quite a jump from the RL-10 engine to the J-2 in the thrust level, weight, and so forth, and also requirements. I guess mainly what the restart requirements at that time. We had restart requirements. I guess from the…There were probably a couple of areas that…Problems that we encountered, the experiences that we took from the RL-10 program—one was an injector. The concept developed there—the rigid mesh injector—which is nothing more than a transpiration cool of the face by bleeding hydrogen through the backside.

[00:02:08] HR: Hydrogen came into a manifold and dumped in between the LOX side and the combustion side. There was an annulus, which the LOX fluid was in the middle. The fuel fed into that annulus and was then mixed with LOX and, of course, the combustion process took place. A percentage of that, I think would vary anywhere from two or three percent, flowed through the rigid mesh face. Early injectors that Rocketdyne proposed—the old flat face H-1 type—I never had much dealing with H-1 before I got on J-2. They were the old flat face, like on like, ringed injectors—copper rings. We had face burning with them, and also a type impingement they had.

[00:03:03] HR: We weren't able to get the performance specific impulse out of them. I guess we struggled along there for the early part of the program in the thrust chamber injector program trying to meet the C-STAR requirements. A number of the problems ran into was this face burning problem. This is what we did: we pulled that experience out of the [RPM?] program and applied it to J-2. This resolved the face burning problem.

[00:03:38] RB: I was going to ask you how long were you trying to use the H-1 injector face in there before you went to the rigid mesh stuff? How many…

[00:03:47] HR: I have to go back. We got records, but as I remember, it was all…I have to go back and check to be sure, but as I remember, it was up into October. See, we fired in ‘60, ‘62, ‘61, ‘62. We were in ‘61—about the middle part of ‘61—before we made that change. They cranked up—I believe about four months into the program as I remember—from the time they had the contract. The first firing in the thrust chamber stand, which was a water-cooled jacketed combustion chamber, was when we began to find out some of the face cooling problems we had. After several design iterations…You know, a contractor never likes to be asked to use a concept somebody else has used. It's just general nature. He'd rather design around it than somebody stand up and tell him he's got to take something that his competitor has used. That's not for the record, but that's the way it goes. Sometimes it's not invented here, they're not interested, especially when it's the competitor's idea.

[00:05:05] RB: Did you have to start to kind of push that down Rocketdyne’s throat?

[00:05:09] HR: Oh, we had to. I mean, we were way behind on performance. Like the spec at that time was based on a mixed ratio of 5:0 and 200k engine. The spec at that time was 422 seconds. We were running around somewhere around 415 to 419 with those type of injectors. The main reason being was, I guess, trying to get the mixed ratio distribution across the injector face with that type of injectors is pretty difficult. Not only that, but as you go out towards the periphery, you have to be very careful about how your LOX impinges is emitted out of the injector. If you get too much up against the wall, then you're either forced to put more film coolant in, which is you lose performance if you add fuel to the wall to keep it cool. You lose performance because then you're getting out of balanced mix ratio. In this case, you'd like to run around 5:1. That's injector mixed ratio now, not overall engine. If you're running 5:1 in a core, on a wall you're running 3:5:4. That's performance [LOX?]. Ideally you'd like to approach stoichiometric. I would say the thing we talked about, what do we learn out of PAS program, what was probably a major benefit from PAS program, or what technology base did we extrapolate from, I would have to say that probably the RL-10 in the injector area was a great benefit to it.

[00:06:51] RB: Do you know anything…Was it the Pall Company or a guy named Paul who developed that rigid mesh?

[00:06:58] HR: Pall Company.

[00:06:59] RB: Yeah, face. Does that strike a bell with you?

[00:07:01] HR: Yeah.

[00:07:02] RB: Okay, is that the one? P-A-L-L?

[00:07:04] HR: Yeah, I think that's it. I think…You know Dave Christen [sic]?

[00:07:07] RB: He's the one who put me on you. [everyone laughs] Thank him for that later.

[00:07:13] HR: Yeah, I think Dave used to represent them at one time. That was Pall Corporation, I think they’re the only ones who got it. They're the sole source of that material.

[00:07:25] RB: Do you know any more about it? I mean how it was used in earlier programs? Do you have a line on that?

[00:07:30] HR: I think, as best I understand, it got started…The application was for filters. That's how it really got started. It was a filter cartridge or a filter, used as a filter media. Someone had the bright idea, I guess, you know, that one of the things you got to do is keep the face cool, and how do you do it? When you start taking flat faces where you got to drill a hole in the face, you got to depend on heat transfer, which is they use copper. With the combustion temperature we had, it's pretty hot to handle, you know? You got to watch junctures. The copper alone won't work by itself without having steel ends in order to support the copper, so you got a juncture there where you got heat transfer problem. With the heat fluxes that we use associated with LOX hydrogen engines, as opposed to those that are associated with LOX kerosene engines and LOX RP engines.

[00:08:29] Second interviewer: Is it different? Fluctuation in heat in the engine?

[00:08:34] HR: In the combustion chamber, yeah. It gets a little different when it comes to base area because it has to do with the emissivity of the gas, of the exhaust product. In the F-1, you always had that fuel rich shroud around the periphery.

[00:08:50] Second interviewer: And is that effect performance like hydrogen cooling of the walls in the J-2? I mean, does the same principle operate?

[00:08:58] HR: Well, it's not. You're talking about the rigid mesh face?

[00:09:00] Second interviewer: No, I'm just talking about when you're talking about LOX dumping out on the edge and ruining your balance on the wall.

[00:09:06] HR: Well, yeah. Well, see you got to put the fuel…To answer your question, what I was talking about there is that you got to keep the wall. You got a certain heat flux at the throat, and your walls have to be kept cool. You got two things you got to worry about. Right where your combustion is completed—the face—as you progress down, you got to build up a boundary layer. You always have film coolant. You want to minimize that film coolant because it's used to fuel rich, and it's way out of balance. It's the way you ought to be burning to get maximum performance. But what I was telling you there is in flat face, you've got to impinge back in this way with your LOX. You don't dare to get on the side of the wall because of the inertia difference between the density of LOX and those of the fuel. The LOX penetrates through that stream where you've got to, say, relatively low velocities at the outer periphery in comparison to what you got at the center core. Then you begin to burn on the side of the wall at a very high temperature because locally the mixed ratio is real high. You're getting up close to stoichiometric, even higher than that. So what you've got to do is you've got to be able to minimize that film cooling so that the spray, the LOX—the last element—the time it gets over there that you've got a uniform mixture, not a glob going through that sheet of fuel and setting up a very local mixture ratio which are real high. If you do, then you burn right through the wall because the local temperature is really high at that point.

[00:10:44] RB: Did you have trouble finding the right kind of steel tubing? Did you have trouble finding the right kind of steel tubing for the LH tube return post up to keep your wall cool?

[00:10:56] HR: No, [inaudible] uses 347. 347 tube. No, they haven't had any trouble with the tube. We had more than adequate cooling at the throat because we put all the fuel except when we went to the gas generator which was three or four pounds a second. The rest of that flow was, say, seventy-eight pounds a second going through the chamber, so we had more than adequate cooling as far as taking care of the heat fluxes were there as long as everything was uniform. Of course, you get local conditions like sometimes in transit, you know, J-2 was plagued for a while with a fuel pump stall. Whenever you open the LOX valve, about that time, it begins to deprive the system of fuel because of the resistance in the system was so great and there wasn't enough head output at that time when you really needed it in order to open the LOX valve to pour the right amount of fuel in the chamber. You starved the system, in other words, you had almost the fuel flowing down because of the resistance and the fact that your pump was decaying down, you weren't putting out the head. You opened the LOX, which is a good quality, and sometimes we burnt the walls trying to tune the sequence up. This fuel pump stall problem the way we got out of that was going back through the system, taking out Delta P where we could. Initially, I guess you might say it was kind of sloppy as far as their care and design. I'm not assuring that they had the lowest Delta P system. We had to do that, plus we had to add another stage to the fuel pump. We had a six-stage axial flow pump. We had to come back in and add another stage, which went to seven stage pump, which increased the head output. Of course, this had to do too with the tuning of the LOX valve so that it cracked at the right rate, so that the fuel system wasn't overpowered by the LOX system. You got to affect chamber pressure, and they're both seeing the same thing. Now, if the LOX side is running ahead of the fuel side in pressure, then what's creating the chamber pressure is the LOX side. The fuel side now has got to work against it, so you got to tune the LOX valve so that the fuel essentially is leading the LOX in pressure.

[00:13:21] RB: When we were out in Los Angeles, we talked to some people. Paul Fuller, do you know him? He’s out there.

[00:13:26] HR: Yeah.

[00:13:27] RB: Bob Fontaine was working on the F-1. I think John asked the question, “Does engine development and design be kind of a black art sometimes?” that you maybe make a fix on it, and the thing works, and you're not really quite sure what you did to make it work in the long run. Do you ever have that feeling?

[00:13:48] HR: Well, no, not really. I guess the biggest thing sometimes is when you make several changes at one time. That's the thing, you know?

[00:13:53] RB: Okay, that would be that. Right.

[00:13:57] HR: A lot of times you have a problem, and you think that, well, if you do this and that and so forth, it does correct the problem. It's your best estimate through your analysis and observation of hardware. Sometimes you make those changes and then you have another problem, and then you're not quite sure which one of the three fixes you made simultaneously caused the other problem. But I think you all, in fact, I don't know of any case of any problem we ever had that we didn't eventually thoroughly understand it. We knew what the cause was.
We sometimes, like I say, we make too many changes at one time. Sometimes schedule pressures force you to make two or three changes at one time rather than doing one at a time. As you know, it depends on what part of the development program you're in. If you're in the early phase and things are going kind of slow, and it seems like every day you test something's coming loose, you know? It seems like it's never going to end. As opposed to your get on out in the program where you got a lot of testing behind you and the bulk of the problems have gone away or solved. Then you get these sporadic problems like you [cook?] through two damn engines, and you don't even have the problem. In the third engine, there was something you did in manufacturing or there's some little something that you did is all of a sudden now creeping into an engine. Because there are changes going on all the time, really. If you're going to get there, you got to make them. Just, you know, several approaches to developing a rocket engine. You can take that first engine out there, and you just test it and test it and test it and repair and repair. You get all the problems, and you make one big block change. Sometimes that may take you a long time ever getting there or you can make the change, and you make that change in the pipeline over here where the engines are being built. There's pros and cons to developing an engine like this like that. Some people think it's better to do it on a block basis. Some people think it's better to do it when you got a problem, fix it and get it into your line.

[00:16:01] RB: Is that the way you usually handle it then?

[00:16:03] HR: That's the way it’s usually done. It just isn't time because every time you run a test, you don't want to let a component ride on there you know you're going to change. You test your [reliabilities?] based on the number of units and the number of tests, and if you're losing time by not going ahead and making a change, really soon you can.

[00:16:30] RB: I had a question to ask about the failures in the J-2 engines on the S-2 and S-4B stage when you had 502. Now there was something that you discovered after you put it into a vacuum chamber, tested it in vacuum at altitude. Now weren't there vacuum and altitude tests done before or was that basically on just the engine throat areas that you didn't get up into the ASI area? Why didn't you discover that before?

[00:16:56] HR: No. Well, we did testing at Tullahoma. We pulled the sail down, you know, about two tenths [PSI?] or something like that, which would have been adequate except Tullahoma sitting in a big hole, which the whole bottom of the hole was filled with water. Even though there's nitrogen purge, there's still quite a bit of moisture, and that's what this particular problem thrived on. That is, when you were at moist conditions, the minute the fuel started through the line, you started to liquefying the air or the moisture. Or you even froze the moisture on the bellows. Therefore that acted as a dampening device, see? Then when you put it in, say, an environment of helium where there is no moisture, there's no ice accumulation, then the thing is allowed to go through its thousands of cycles right quick.

[00:17:46] RB: Where did you finally do the test with the helium atmosphere?

[00:17:49] Second interviewer: Was that after the failure?

[00:17:51] HR: That was after the failure.

[00:17:52] RB: Yeah, it was after the failure.

[00:17:53] HR: We didn't do any tests. We didn't have any failures.

[00:17:55] Second interviewer: Did them out Santa Susana, that test with the helium?

[00:17:59] HR: Yeah, that was done there in Canoga Park.

[00:18:01] RB: Okay, after the failure.

[00:18:02] HR: After the failure, yeah. I tell you, I think that was just a stroke of luck, really, that we found that thing that quick. That's my personal feeling. Because that's the one you just don't ever, you would never think about.

[00:18:19] RB: Yeah.

[00:18:20] HR: I have to give Rocketdyne credit. They did a fantastic job of taking what test data and flight data and piecing all, you know, putting all the pieces together. If you took the data, it would tell you two things that went on. One was when the line failed, the start fuel to the injector, to the ASI. Well, when they, we didn't know it was flying, but as you start the fuel, you begin to roll the ASI out. Just begin to burn the whole center of it out. When you did that, then of course that changes your C-STAR. Now, if one takes the flight data and tries to go back and says, “Well, you know, I know something happened. I know I'm losing fuel.” We could see that in the environmental data. You could see temperatures. It had to be a fuel leak. Then you say, “Well, wherein a world. How much fuel am I losing?” Well, if you go back and assume in the calculations that you've got no damage to the injector, you know what the C-STAR is. You got hundreds of tests to tell you what that is. Then you try to balance the engine out with that C-STAR, not knowing if a damn big hole in the injector. It comes out in terms of you losing so much fuel. It came out like six or seven pounds of fuel going someplace. You try to balance it out that way. Well, we're talking about an ASI line that was flowing something like a pound or two at the most. That kind of fogged the issue there, trying to use the balance of what was going on that flight to say, where in the hell could we lose that much fuel? You know, what sources and start taking instrumentation, track it through the system. Someone got the bright idea out there, you know, you better start trying some of these lines to see what they do under flow condition. By doing the test and the environment they did it in, they found out right quick, you know? It’s amazing. I don’t know if you've ever seen in pictures or not. The actual failure.

[00:20:15] Second interviewer: I heard there were pictures.

[00:20:16] HR: They turned the valves on to start the test, and the line fails like that practically. What amazes me is how we went through all these other flights up until that time. Then had two failures, one on S-2 and one on S-4B. [laughs]

[00:20:32] RB: Well, that first one was a suborbital flight, though. Of course, the S-4B is on the S-1…

[00:20:37] HR: It's in a vacuum, always in a hard vacuum to start.

[00:20:40] RB: Yeah. Were there any other problems with the J-2 engines that you recall?

[00:20:46] HR: Any other problems?

[00:20:48] RB: Yeah. You mentioned the injector face and that little problem with the ASI fuel line.

[00:20:54] HR: Of course, that was the pogo problem. I think that was more of a…I would say it was the engine’s problem. It's just more of a structural problem.

[00:21:05] RB: What about some of the materials that were used in the J-2 like Rene 48 [sic] and Inconel? Were those around at the beginning?

[00:21:14] HR: Yeah.

[00:21:15] RB: Okay. Could you tell me where you used some of those things specifically?

[00:21:18] Yeah, 718 was used in the injector billet. The injector's made from a big billet, which is eloxed out. In other words, the post of the injectors use an eloxed graphite plate. It's electrical discharge machining. You just take it, make the post. Later on, they drilled it first, then they came back and did elox on the post with a drill with elox process. The injector assembly, manifold, pumps—that's 718.

[00:21:56] RB: What about the Rene? Was that primarily an F-1 material?

[00:22:00] HR: Rene was an F-1 material.

[00:22:02] RB: Yeah. What are the Kel-F lines? I just don't know.

[00:22:05] HR: Huh?

[00:22:06] RB: Kel-F.

[00:22:07] HR: Kel-F liners?

[00:22:08] RB: Yeah.

[00:22:09] HR: Kel-F? It’s just…

[00:22:10] RB: I have no idea what that is.

[00:22:14] HR: We have a Kel-F liner LOX pump.

[00:22:18] RB: In the LOX pump?

[00:22:20] HR: [Inaudible] area. All that was for was to protect the blade surfaces from the walls of the housing in case there’s ever a part that came in and got lodged between they [wouldn’t?] be scrubbing against the surface. You have a plastic protection, you know, keep from building up any heat or if an inducer ever kissed [inaudible] bearing or something suddenly went bad, you have a certain amount of motion then you’d be kissing plastic for a while before you finally got to the metal. Well, then you could build up heat and blow the LOX pump.

[00:22:56] RB: There was another question I had…Oh! About the gimbaling system…Did you ever build a mechanical screw system?

[00:22:58] HR: Mechanical screw?

[00:23:03] RB: Yeah.

[00:23:04] HR: No.

[00:23:05] RB: Rocketdyne was testing that on some test stands

[00:23:06] HR: The program did some work on a pneumatic actuator, hydrogen driven turbine ball screw.

[00:23:14] RB: But it never worked out very well?

[00:23:15] HR: They never did much on it. It’s what they proposed. It’s a proposal they made. They built one or two and tried them out, but they didn’t have enough umph to them. They didn’t have power to give the rates that we required. Very sluggish.

[00:23:32] RB: Okay, another question: how do you go about upgrading an engine when you went from 200 to 230k? When you start with a 200k engine what do you have to do to it to meet the desired thrust level? The higher one?

[00:23:44] HR: Went from 200 to 225 then went 230k a second.

[00:23:49] Second interviewer: In two separate steps?

[00:23:50] HR: It was two separate steps. That’s kind of a hard one. First of all, the engine, when it was designed, was designed at a thrust point of 200k and a 5:0 mixed ratio. It had a PU valve, which allowed an excursion of five mixed ratio units on either side of that. In other words, that would’ve taken you to 5:5 mixed ratio or up to 4:5 from another 5:0. That’s if you vary the mixed ratio. In this case it was bypassing LOX around the LOX pump. Then your thruster is going to vary. You’re either taking out LOX or you’re putting more in. One or the other. In this case your thrust is going to rise. Well, in J-2 when it was bought it was a 5:0, 200k, and that’s the way it was going to be acceptance tested with the PU excursion. Since the engine was orificed [sic] 5:0, then you had no control of the exact thrust that you’d get when you’d go to either end of the excursion. It just so happened that when you went to the 5:5, you’d go as high as 238k.

[00:25:13] RB: What was the factor….

[00:25:14] HR: As high as, okay? So as the design of the S-2 stage progressed, and they got thinking about the mission and what the requirements were, they came up with a scheme of flying the first portion of it at 5:5. When they needed the high thrust then the last third portion of the burn going back down to the lower mixed ratio with the specific impulse. As soon as you go up you lost gain [inaudible] like this occur second impulse. As you go out towards the higher mixed ratio, you don’t get the specific impulse but you get the higher thrust. That’s more important than the trajectory equations at that particular time in the S-2 boost. Later on in the flight you’re not so concerned about thrust anymore, you need Isp. You switch it back the other way, so your thrust is going down, but your Isp is going up. That was kind of the way the engine was burned in the S-2 and the S-4B. It had that same profile.

[00:26:33] RB: Do you have to start upgrading stuff though like the turbo pumps and everything else when you start doing that?

[00:26:37] HR: Well, you got to start testing that way.

[00:26:39] RB: Yeah.

[00:26:43] Second interviewer: Okay, what about the structures that these interface with? Has that all been taken care of beforehand that it will stand the stress of another...

[00:26:51] HR: Well, so yeah, when we went to 225k, we orificed [sic] the engine at the 5:5 mix ratio in. They gave us very precise thrust at the end. We calibrated the engine 225 plus or minus about 6k. Usually the engine ran within 2 or 3k of the value. Every firing would be within that dispersion about 225. Well, wherever it was calibrated, you had about…Sometimes it was calibrated 227k.

[00:27:20] Second interviewer: How do you go about calibration?

[00:27:22] HR: Just change the orifice. Bouncing out there.

[00:27:26] Second interviewer: Where do you get your calibration standards? Just accumulated data?

[00:27:31] HR: Well, you got a computer program that models the engine. Before you have a flow test on various components, you have this data. You got pump data. All your pumps are green run before they go in the engine, so you know what their performance are. You take all this data, and you put it in the computer, and it comes out and tells you where the first cut to make. What the orifice should be put in there. It's not exactly the first run that you make. Then you come back and change the orifice, and you make another cut at it. Usually about two runs,
I mean, you get really...Further along you get in the program, you get pretty good at it. You usually make it the first cut. But until you develop that skill and learn to get enough data, I guess, that's what it amounts to on valves. What kind of spurs, what influences the valves have on the balance of the engine. The chambers have various Delta Ps in them, and pumps have different efficiencies. They're very narrow. But when you start talking about hitting something within 3K out of 225, that's a pretty close shooting.

[00:28:49] RB: So one other question. We're about out of the tape here, and we want to ask for a half an hour, so that we’ve done it. What does the J in the J-2 mean?

[00:28:56] HR: The J?

[00:28:58] RB: You know where they got the F in the F-1 and the H in the H-1?

[00:29:02] HR: I don't know. I never really stopped to think where they got it. It's a series of A, B, C, D, H-1, F-1, and J, H. I don’t know. I never thought to how they…

[00:29:07] Second Interviewer: It's kind of a series, Roger. [Inaudible] started off with…

[00:29:13] RB: [Inaudible] start off with F.

[00:29:19] Second interviewer: It's kind of in the series. It’s not like anything else. It’s not uniform.

[00:29:23] RB: It's like playing the SA-203 before the SA- 202, okay. [laughs]

[00:29:27] HR: I don't know.

[00:29:29] Second interviewer: That's what it comes from.That's what people at Rocketdyne told me.

[tape ends]

0:29:32