NASA Johnson Space Center
Oral History Project
Edited Oral History Transcript
Donald M. Curry
Interviewed by Jennifer Ross-Nazzal
Houston, Texas – 16 April 2014
Today is April 16, 2014. This interview with Don Curry is being conducted
for the JSC Oral History Project. The interviewer is Jennifer Ross-Nazzal,
assisted by Rebecca Wright. Thanks again for taking the morning to
sit down and meet with us.
We appreciate it. Tell us how you came to JSC in ’63.
kind of an interesting situation. I went to work for the Bettis Atomic
Power Laboratory in Pittsburgh, Pennsylvania, where I worked on the
first nuclear aircraft carrier, Enterprise. Also, when I went to work
at Westinghouse, I was selected to go to what they called their Advanced
Design School; they basically paid for my master’s degree at
the Westinghouse Educational Center and the University of Pittsburgh.
After about four years in Pittsburgh, with all the weather, the snow,
et cetera, I wanted to move back to Houston—or I wanted to move
back to the Southwest.
Are you from Houston?
I’m from Oklahoma, but I wanted to move back to the Southwest.
I saw that NASA was interviewing for a new Center in Houston, so I
actually applied. I sent not a résumé, I sent a discussion
question, because they were interviewing in Pittsburgh. I wasn’t
selected. They said I wasn’t qualified, although I had a master’s
degree in mechanical engineering at the time. I still wanted to move
back, so I looked around and I actually ended up working at the General
Dynamics Corporation in Fort Worth [Texas], worked in their thermal
physics branch. At the same time, several people from that branch
were leaving to go to the Manned Spacecraft Center in Houston. One
of my good friends I met in Fort Worth, I worked with him, and I decided
I would like to go there, too, because I wanted to do work that NASA
Basically I had had a really big interest, based on my research for
my master’s degree, on old NASA/NACA [National Advisory Committee
for Aeronautics] technical note publications. I thought this would
be a great opportunity to do that and continue to do the kind of work
that I was doing with General Dynamics. I contacted George Strouhal,
who’s deceased. He in turn got me in contact with a man by the
name of [R.] Bryan Erb, who is still living here in Houston. Within
a couple of weeks, I had an interview in Houston with Bryan Erb. After
that interview, they hired me within about a month’s time, which
was unusual because at that time, there was a lot of paperwork, a
lot of delays, and there were a lot of people being hired. I moved
down here in ’63, to go to work at the Manned Spacecraft Center.
You were hired to work on ablators?
I was hired to work on the thermal protection system, but my first
assignment, which is again interesting to me—when I was at General
Dynamics, I was also doing a little bit of work on entry heating.
We, at that time, thought that NASA probably had all of the ablation
programs that they would ever want, which was a real surprise to me
when I came down here because I did have some computer experience.
That was my first job. Bryan Erb said, “I want you to develop
an ablation program for the Manned Spacecraft Center,” and he
put out a caveat on these, said, “I don’t want one that
runs long.” Mathematic problems can be very long-running, so
he wanted me to do a program that was fast and could be used for design
purposes. Fortunately, I met another man here who had a mathematics
background, and he introduced me to an algorithm, I’ll put it
that way, in a mathematics book that would do that.
That’s what I did; I developed an ablation program. It was an
implicit solution, which I think was one of the first implicit solutions
in ablation. It was fast, and it’s still used today. It’s
been modified, obviously, probably many, many times since that time.
That was my first job, and that was over in the Franklin apartment
buildings, when we were all scattered all over Houston. The computer
work was done at the University of Houston Computing Center, at that
When did you finally move on-site?
a good question. We moved from Franklin to Ellington [Air Force Base],
and were at Ellington, in the barracks, for a while. Building 13 was
one of the first buildings opened, other than Building 1 Headquarters.
It was early on, when the site was first opened. I remember one of
the times, walking into Building 13 after it had rained, and we walked
on wooden planks because we still hadn’t put the sidewalks in.
It was pretty early.
Yes, that is fairly early. After you worked on that computer program,
were you starting to work on testing ablators in different ways, besides
were doing that as soon as the JSC Arc Facility was developed.
That was the original one?
original one, the 1.0-megawatt facility. According to Don’s
[Donald J. Tillian] history, it opened up in September of ’63.
We started some testing back as early as late ’63, probably
Were you involved at all in using that facility?
I was, sure.
Can you talk about that?
was a very interesting facility because it was small, and we could
basically almost be out in the chamber and watch the tests go on.
It was a 1-megawatt facility, and we did a lot of different material
testing at the time. Shortly after that, I’m not sure exactly
what the year was, but then we developed a bigger Arc Jet, the 1.5-megawatt
Do you know the reasoning behind the need for a bigger facility? I
find it interesting that you had a facility that you were using for
Gemini and Apollo, but then there was a need, I guess.
1-megawatt really wasn’t big enough for Apollo. We really couldn’t
get the power we needed, we couldn’t get the heating rates that
we needed, couldn’t get the pressures we needed, the enthalpy
values we needed. Also, we needed to be able to test bigger test articles.
Initially, we tested small test articles, on the order of 1 inch to
2 inches, but we wanted to go to higher-sized articles. To do that,
you had to have more power; you have to have a bigger exit nozzle,
so a lot of equipment had to be changed. At the same time, we changed
from a Plasmadyne facility to the Aerotherm facility, which was another
Would you explain the difference?
Curry: I really
don’t know the big difference between the two. Don would know
Who was put in charge of creating and designing that building? Do
you recall some of the folks?
[David H.] Greenshields was the branch chief for the Arc Jet facility,
so I’m sure he was the major person in charge of getting that
building built, the 1-megawatt facility, and then actually the modifications
after that, too, because he was the branch chief for many years, over
the Arc Jet as well as over the thermal protection group.
Do you recall any interesting tests for Gemini or Apollo that stand
out in your memory, since you said you were more of a user than on
the other side, a test operator?
Probably the most significant one was on Apollo. We were testing the
Apollo material. When you do the test, you have to seal the back surface
of the ablator against the sting arm—we call it the sting arm—we
bonded it to an aluminum plate with an adhesive. The test article
would maybe be two inches in diameter, or two inches thick, or four
inches thick. On one test, we stuck it into the Arc Jet, and within
seconds we destroyed the entire test article, which, as you might
realize, would cause a little consternation among the people. Upon
examination of, we realized that we didn’t get a good seal back
at the interface between the ablator and the carrier plate, or the
aluminum backup structure, we call it. Because of that, we had a gas
flow that just went right through it because the material was porous.
It had a porosity to it, so the gas just went through it and just
basically destroyed it. Pressure of the gas plus the internal pressure
being lower, it just had a great flow.
Yikes. Yes, I’m sure a lot of people were quite scared, thinking,
“We’re going to send this to space? What will happen to
why would this material fail all of a sudden?
Did things change over time, in terms of testing, from Mercury through
Apollo? Were you using different ablative materials?
You’d have to say things changed because we learned from each
test, and we learned more how to build the test articles, design the
test article, design of all of the instrumentation that we put in
a test article, actual shapes. We changed from just basically flat-faced
specimens to what we call iso-Qs. An iso-Q specimen gives you a uniform
heating flux across the surface, so that was an innovation that we
had. Different tests could require different types of test articles
itself, going all the way from stagnation point testing to parallel
flow tests. As we moved into the 10-megawatt facility, that allowed
us to do flat-plate testing, which we couldn’t do in the old
facility, the old 1-megawatt facility. Now, with the new facility
at 10 megawatts, higher power, more flow, more gas flow, we could
test flat plates, we could test plates at an angle of attack, and
we could test larger test articles.
That’s a pretty significant change.
Would you talk about how you prepared for these tests for Apollo?
What sort of requirements were set by the facility for you?
had what’s called test readiness reviews, and we would write
a test request. In writing the test request, you would write exactly
what test conditions you wanted. In other words, what pressure you
would like to test at, what heat flux you would like to test at, or
you could actually specify enthalpy. You couldn’t specify all
three, so you could specify two of the three. That’s one thing.
The second thing is you would specify how you wanted the instrumentation,
whether you wanted just a bond-line thermocouple; whether you wanted
what we called in-depth thermocouples, how you wanted them put in.
The actual installation of the thermocouple was important. We used
what we called land-length type thermocouples, which are little 1-inch
links across the central core of the ablator, as opposed to what we
call a point thermocouple, because the land-length type thermocouple
gave better test data, gave you better data on the temperatures. Design
of the test articles was very important, and different materials required
different types of geometries.
I think one of the things we tested early on, even in Apollo, was
wood. Balsa wood, mahogany, walnut, different types of woods that
you can just go out and cut. Wood’s a very good ablator. My
point is that you have to understand how the ablator works, and I
think I mentioned in the video interview that an ablator is like a
wood-burning log in a fireplace. You put wood in your fireplace, and
you watch it burn. Let’s say you stop the flame for one night
and look at the log, and it’s got a nice black layer on it,
which we call a char layer. An ablator does the same thing, but it’s
more efficient than wood and it’s lighter weight. There’s
a lot of similarities between wood-burning and ablation because it’s
basically the same process. When we have a fire in a forest, forest
fires are ablation of the wood. The National Forest Service actually
uses ablation theory to predict how forests will respond in a forest
fire, believe it or not.
I would never have known that.
people don’t understand that, that’s true.
Tell me, why were you testing wood? What other sort of everyday materials
were you testing in the facilities?
the time, we were interested just in understanding how different materials
perform. Wood is a good material to test and determine its performance;
actually, cork, which is wood, is used on missile systems. Cork is
a material that’s used to protect the missile during its ascent
phase of the trajectory. Of course, for the entry phase of the trajectory,
they use another material, a higher density material. Cork is very
lightweight and pretty weak, but also it has good properties for the
environment that you want to test at.
That’s the other thing: the materials are selected based on
the environment in which they’re going to be used. One of the
parameters that’s very important is temperature because each
material has a temperature limit. You want a design to be as efficient
as possible. So for example, that’s why the Shuttle system had
carbon, which is a very high-temperature material, and had the tiles,
which were a little bit lower-temperature material, and then it had
the blankets, which were even lower-temperature materials. In combination,
all three of them gave you a very efficient thermal protection system,
in terms of weight and performance. The Arc Jet does all that for
you, lets you test all those different materials, so you can determine
When you would run a test during Apollo, were you also there watching
the test being done?
Actually, during Apollo, we could be standing in what I call not the
control room, but they had another little room out by the actual Arc
Jet chamber. For many years, we could stand out in that little room
and actually watch the test, looking through the windows into the
main chamber of the Arc facility. That changed when safety got involved,
and they decided it was too much of a hazard for a human to be out
there, in case of a major accident, I guess, or incident.
When the test was over, then what did you do? You went out, and you
looked at it?
of all, we would watch the test, and when we actually moved into what
I call the control room that we had, we could watch it on the cameras.
If we saw something happening that was bad, we could terminate the
test sooner than we had planned. Each test had a specific test time,
but after the test, absolutely, we could go out. Once the chamber
had cooled down and we had re-pressurized back up to atmospheric conditions,
we could go out and look at the articles in the chamber. After hours,
when they took the articles out, of course, we could look at them
close-up. During the actual test, we could watch the temperature rise.
We could look at the surface temperature, we could look at the internal
temperatures from the thermocouples. There was a lot of things that
you could do and were doing during the test.
If at some point during the test you thought, “This isn’t
what I’m looking for,” you said you could stop?
did what we’d call an abort. Absolutely, we could say, “Terminate
Then you started preparing for another test because you wanted a different
temperature or you weren’t seeing what you liked?
or we could wait, and wait until after we looked at the test article
to make sure that’s what we wanted to do. Generally speaking,
you’re correct, we would just prepare another test article,
maybe change the test conditions. We’d actually do that many
times when we were trying to establish what we called failing temperatures.
If you wanted to do a failing temperature test, then you’d take
it up in temperature, in increments, until you get to the failing
point. Then, as soon as it starts to fail, you terminate the test.
Who would make these test articles? Was that something the contractor
speaking, the company that made the material or fabricated the material
did supply the test articles, yes.
You would just tell them what dimensions you needed and the shape
you would buy the test articles from a corporation like Avco Corporation,
at the time, and we would buy them already instrumented, per our design,
At some point, did you no longer need to test the ablators, since
you had all of this wealth of knowledge from Mercury and Gemini, and
once Apollo had flown 7 and 8? Was that enough, or did you still need
to keep testing?
respect to Mercury, Gemini, and Apollo, as soon as we had test flights,
then we didn’t need to do additional Arc Jet testing, unless
we had some kind of anomaly during the flight. Then, we’d do
anomaly flight test data. With Apollo, we had four separate unmanned
test flights. Two test flights that entered at what’s called
Earth-orbital speed, and two flights that were lunar return speeds.
We had test data from those four test flights, and we had the Arc
Jet test data so we could compare the flight data with the Arc Jet
data. We could also conduct post-test examination of the different
flight materials actually taken from the flight cores with the Arc
Jet cores. We had a way to compare actual ground testing with flight
Were they very similar?
the Arc Jet is a little more severe than the flight test, for certain
types of materials, because the Arc Jet, we always had what we called
a disassociated flow. The gas was not oxygen O2, and it wasn’t
nitrogen N2. It had been broken down into atomic oxygen and atomic
nitrogen. Even at certain conditions for Apollo, we had some ionization.
We tested at high enough temperature to ionize the gas. Generally
flight conditions were pretty much the same but there’s a difference
between the flight and the Arc Jet, so they’re not exactly 1:1.
You would like them to be, but they’re not.
I assumed that it was.
they’re not. Generally Arc Jet testing is more severe than flight,
unless you cannot test at the flight conditions. There are conditions
that you can’t test at.
Can you give some examples?
the pressure’s really high and the heat flux is fairly high,
some Arc Jets can’t reach both conditions at the same time.
The high heating rate plus the high pressure, or you might even have
a low heating rate condition and a high pressure. Sometimes you can’t
match it up exactly, so you do the best you can. Then you use your
analysis to extrapolate the data you like, to go ahead with the design.
During the Apollo program, lots of guys have talked about how they
were working nonstop, seven days a week, they were constantly on travel.
Was that the same for you?
Curry: I wasn’t
on travel because when I was working on Apollo, I worked in the analytical
groups. Yes, we worked a lot of times on the weekends. One of the
biggest jobs we had, after we got into the Apollo program, was the
development of what we call the entry corridor. During the entry process,
it’s hard for people to understand this, but when you’re
at the Moon and you look back at the Earth, you have approximately
a 1-degree angle entry corridor. Think about 1-degrees. At the distance
from the Moon to the Earth, that distance of 1 degree is a huge distance.
Because of that, the entry corridor was extremely important. You can
come in what we call too steep on the entry corridor and burn up because
you’re coming at too high a velocity, you’re hitting the
atmosphere, and the heating rate’s going way beyond what you
had designed for. Or, the alternate is you can come in too shallow
and hit the upper limit of the entry corridor or get above it, and
you skip out. Once you skip out, that’s the end of it, you’re
into endless orbit; you can’t come back.
Being able to hit that corridor was extremely important, and as a
result of that, a lot of work with Apollo, once we had gotten the
ablator material, the AVCO material, and had designed it, was to develop
an entry corridor for MOD (Mission Operations Directorate). At the
time we were doing this on our computer program. Off one single run,
we got one data point. Many times we ran program runs overnight with
priority. Many days, I remember picking up a huge stack of a computer
output, maybe 20 or 30 runs, and we’d get 30 points or 20 points.
From that data, we could develop a plot of the heating and of temperatures
to give to the MOD people to help develop the corridor with respect
to the entry flight angle and the flight conditions and so forth.
A lot of that was done at night, and a lot of it was done over the
As you said, I think most people that worked on the Apollo program
out here worked for no extra pay because we were too interested in
it. It was too much of a challenge because there wasn’t anything
known. When [President John F.] Kennedy said, “We’re going
to the Moon,” well, we didn’t even have the material.
We didn’t have the guidance schemes. We’d never done some
of these things. We’d only flown one Mercury flight, in fact.
Do you remember Apollo 11?
Where were you?
11, I think I was probably at home. Like I said, I wasn’t part
of the project management, so I wasn’t really involved in Apollo
11 that much.
When did you start working on Shuttle?
I became the subsystem manager on the Space Shuttle Orbiter leading
edge structural subsystem, which is the nose cap, wing leading edge
panels, arrowhead plate, and eventually, what we call the RCC [Reinforced
Carbon-Carbon] chin panel.
What lessons did you learn from Apollo that you could transfer to
Shuttle? It’s two very different vehicles, of course.
true, except for one thing. I think the reason I got selected to be
the Shuttle leading edge manager for the carbon system was because
the carbon system was an ablator.
carbon ablates. I had experience in ablation and in the chemistry
aspects of that. We had to develop computer programs to predict the
carbon response, thermally. I think because of that experience I had
on Apollo, it just moved me into the carbon system. I guess again,
another way to say it is your fireplace log—when it forms that
carbon layer, it continues to burn. That carbon layer goes away because
it oxidizes away from the oxygen in the atmosphere. That’s an
ablation problem. I think that’s probably why I got selected
to be the initial subsystem manager on the leading edge structural
In the beginning, they were probably looking at a whole host of ideas.
What were some of the ideas that were thrown out for the system itself?
the Shuttle? Actually, we studied an ablative leading edge. That was
one system. There were a couple of different contractors on carbon.
One was McDonnell Douglas, and the other one was actually Chance Vought,
at the time. From the competition, the Chance Vought people won the
contract. Of course, they were a subcontractor to Rockwell at the
time. Ablators would have worked fine, except that they were one-use
only. You had to replace the ablator every time, and that was costly.
It easily would have worked.
When did NASA settle on the RCC system?
had these phase II contracts, I’m not exactly sure, but maybe
mid-1970s. I guess we had already decided because when I became subsystem
manager in 1970, we’d already picked the system, basically,
and already pretty much picked the Vought Corporation, with Rockwell.
You guys were testing those materials out at the Arc Jet, as well?
I probably had more test hours at that Arc Jet on carbon testing than
any other person that’s ever used the Arc Jet, yes.
How many tests do you think that you’ve done out there.
or four thousand, easily.
That’s quite a few.
and they were long-term tests because we were trying to establish
the actual oxidation rates of the carbon system. The Arc Jet allowed
us to find a lot of interesting things that we didn’t really
know about the carbon system that was selected. The carbon system
on the Shuttle has a coating on it; a silicon-carbide coating protects
the carbon substrate. That’s what gives you the re-use temperature
capability. Early on in the program, from tests done both at JSC and
at Ames Research Center [Moffett Field, California], we uncovered
what we called subsurface oxidation. Subsurface oxidation occurred
because there were cracks that occurred naturally in the silicon-carbide
coating. Those cracks allowed oxygen to penetrate or diffuse through
the cracks down to the basic carbon. We were finding little pinholes
underneath the carbon interface with the silicon-carbide coating.
The silicon-carbide coating depended upon—because it had been
formed from the basic carbon substrate—that adhesion to hold
the coating on. We had to develop how much oxidation we could have
before we would have to replace the part. At the same time, we didn’t
want even that, so we developed new systems to infiltrate the carbon.
One of them was a material called TEOS, tetraethyl orthosilicate,
which actually infiltrated the carbon, which helped provide oxidation
protection from this so-called pest oxidation or subsurface oxidation.
Later on we determined that we still had some craze cracks on the
exterior surface, so in order to seal those up, we developed a surface
sealant. Arc Jet testing actually was instrumental in the development
of all that and was instrumental in selecting how we did it, selecting
the actual sealant material and how much TEOS we put in. Arc Jet testing
was extremely important; in fact, it was the only way to do it.
As you know, Shuttle was never flight-tested in the sense of a unmanned
flight test. The first test was with a man in it. I guess even the
launch and landing tests were manned. That was different from Apollo.
Apollo had four separate unmanned flights. We were depending and did
depend upon the Arc Jet to give us the right data for these new materials
because the carbon system was a new material, the tile system was
a new material system, and the blankets were all new materials systems.
The Arc Jet was used in all those materials systems.
That’s pretty amazing.
In the ’70s, you were designing and developing this new system,
but then as you began flying, you recognized that things had to be
and also what happened in flying the Shuttle system, I’m not
sure how many people know this, but the weight of the system went
up. Because the weight of the system was up, that meant that the entry
temperatures went up also. The initial carbon system was designed
for a temperature of 2,750 degrees Fahrenheit. We actually learned
from MPAD [Mission Planning and Analysis Division], new trajectories
that they were flying indicated that we had to get to temperatures
in excess of 2,900 degrees Fahrenheit. That caused two problems. First
of all, the carbon system is a shell. Because it’s a shell,
it transmits the energy through the shell, and from top to bottom
there’s a temperature gradient on the lower surface of the shell
to the top surface, which puts a thermal stress into the shell. Then,
the carbon system was mechanically attached to the front spar with
metallic links. Because of the higher entry temperatures and the higher
air loads, we had to redesign the way we attached the carbon system
to the wing spar. So we introduced what we call moment ties, which
mechanically linked the top of the carbon to the bottom of the carbon.
That was one thing.
Then, the second thing we had to do is prove through Arc Jet testing
that the carbon system could, in fact, operate at a higher temperature
than the original design. So, we had to test at higher temperatures,
and eventually we tested and determined that the carbon system could
easily be reused at an operational temperature of 2,900 degrees Fahrenheit.
We established an upper-limit temperature, what we called the failing
temperature, at 3,250 degrees Fahrenheit. With those two limits, then
we analyzed practically every single MOD trajectory. We looked at
all of the flight trajectories, the limits, to make sure that they
stayed within the operational limits of the system, both structurally
That was something I was going to ask you—you obviously helped
to develop, design, and test the system in ’70s, but then we
moved into operations. So you continually were testing this material.
When there was a new mission and there was a new trajectory before
they flew, you were continuing to analyze?
didn’t test, no. Once we established the limits, then all we
did is evaluate the trajectory, took a look at the trajectory to make
sure that by running it through our computer programs, we could determine
the temperature that we were going to reach. We made sure that they
stayed within that limit. That would be almost an endless task. So
to help us, we developed what we called a simplified aeroheating analysis
of the carbon system. We gave them, I’m going to call it a simple
little computer program to run every time they ran their trajectories.
That gave them a preview of the actual temperatures that they were
going to get, so they could modify their trajectories before they
actually made an operational trajectory. Once they made an operational
trajectory, then we could check it with our more sophisticated programs.
There was a lot of interaction between us and the mission operations
group, and we spent a lot of time back and forth with them.
Were you ever called on during a mission, when they were looking at
the vehicle and they were concerned something had been hit, prior
Curry: I had
never been called into the Mission Evaluation Room, the MER Room,
except prior to the Challenger accident [STS-51L]. I got a phone call,
I don’t know, I think it was, like 1:00 a.m. in the morning,
that I needed to come in and look at the temperatures on the Shuttle
system, from the carbon perspective. The carbon system can take low
temperatures and had what we thought was good impact resistance, so
I came in and looked at the temperatures and looked at the system.
In conjunction with the subsystem manager out at Rockwell at the time,
we determined that we could launch the system from the carbon perspective.
Obviously, that turned out to be a bad decision from the solid rocket
booster perspective. That was the first time I came in.
Then, of course, when we had the Columbia accident [STS-107], we were
involved with evaluation of that. The group at the Structures and
Mechanics Division, where I worked, we specifically requested inspection.
It was turned down by the project management, but we didn’t
know where the foam had hit. As you know, we were turned down. Many
of us felt like—and to this day, I feel like—had we known
where the actual impact was, I feel like we could have fixed it sufficient
to make the re-entry, not sufficient to re-use the vehicle, necessarily,
but that we could probably have made through the entry cycle. We almost
did, but we didn’t quite make it. I think there was ways that
we could have probably fixed it. Now, that’s an opinion. Many
people do not agree. It was a disaster that some of us feel like could
have been avoided.
There’s some other things about it, which again, are interesting
to me. As the subsystem manager, at the time, on the carbon system,
I guess it was prior to that flight, maybe one or two flights prior
to the actual Columbia accident, they had had a piece of foam come
off the size of the foam that hit the carbon system, but it hit some
metal. I don’t know where it hit, exactly. It must have hit
on the solid rocket booster somewhere because they saw, on the post-inspections
of some parts, that it had bent an aluminum part. The carbon system
had been hit many times by foam. We could see it. It had been hit
by little micrometeoroid impacts because we inspected for all this,
after every flight. I, as a subsystem manager, had never been informed
that we had had this huge piece of foam come off and hit the aluminum.
It’s kind of like, in my mind, the Challenger accident; we knew
about blow-by on those O-rings for many flights, but we didn’t
do anything about it. Here we’ve had damage on the tile system.
If the tiles have been hit, we know the carbon’s been hit, but
we’d never seen any real damage on the carbon because it wasn’t
big enough, but had that one piece that came off that would have.
If it can cause damage to metal, it can damage carbon, there’s
no question about it. I think that we didn’t learn anything.
Both accidents could have been prevented, but they weren’t.
Engineering tried their best, at least on Columbia, I feel like. That
still haunts me, today, because it was my system, it was my panel
nine or panel eight that got hit. Yet, I still feel like we could
have fixed it enough, and it wouldn’t have taken much.
What would have been the fix?
of all, we had to get an astronaut out there to do it, that’s
the first thing. That’s a safety problem, I realize that, but
we have on board all kinds of materials—blankets in the payload
bay, rubber shoes that they wear, and things of this sort—that
they could have filled up that hole to ablate those materials first.
Because of that hole, we actually ingested gas temperatures probably
on the order of 10,000 degrees Fahrenheit, maybe higher. No wonder
it burned right through the aluminum front spar. It wasn’t going
to last long at those temperatures. Now, whether we could have done
that or not, we’ll never know.
I was going to ask if you had studied that because I came across an
article that talked about how you and some other folks had studied,
like in ’99, 2000 sometime, the impact of a micrometeoroid or
orbital debris on the reinforced carbon-carbon. I was curious about
had some damage levels on that, but they were small particles. We
really never did do any real test of really large particles hitting
the carbon system.
it’s very interesting.
You were involved in the accident investigation?
Will you talk about that, and how the Arc Jet played a role as well?
sure. Of course, I went down to the reconstruction at the Cape [Canaveral,
Florida], to see the parts and be a part of that investigation. Probably
the main thing out of that accident was that, again, the Arc Jet turned
out to be a very interesting facility to use, a major facility to
use, in fact. The test data from the wing leading edge parts that
they had picked up at the Cape, when they reconstructed it, the carbon
parts had what’s called a knife edge to them. In other words,
they were shaped like the edge of a knife, going from thick to thin.
That’s the best way to say it, I guess, to you. We had done
some tests in our damage assessments on the carbon system. Once we
failed a coating, we let it go ahead and do some more damage. We could
actually form a knife edge in a stagnation point test. The wing leading
edge was not a stagnation point; it was what we called an angled flow
test. We developed another system where we could test at an angle
and a wedge. When we formed little damages, we actually were able
to form a wedge test with a small damage, and then do an Arc Jet test.
When we examined the part after the test, we had a knife edge. It
looked pretty much like what we saw on the Columbia. We actually were
able to simulate it in the Arc Jet, and that was a request by the
CAIB [Columbia Accident Investigation Board], for us to do those tests,
to prove that we could halfway simulate what the entry’s heating
would do to the carbon system. I think we did do that. As a result
of all of that, we developed what we call a damaged threshold. That
was a big activity. We established exactly what size particle or what
size damage in the coating we could have and re-enter safety, and
which size that we could not have. Every single flight after that,
we were in the MER, and we helped developed the inspection system
that looked at the carbon, looked at the tiles, also, and we evaluated
every single spot that we could see. As you know, we never did really
find any major damage after that, but we looked at every single one.
One of the interesting things happened was that since we had developed
these temperature limits for the carbon system, I don’t know
if you remember the incident where we had the gap filler sticking
up, up near the nose cap area, right behind the nose cap, near the
nose landing gear door? Aerothermal came up with heating on that,
and they said that we are predicting temperatures that could indicate
that the carbon’s going to hit temperatures in excess of their
failing mode, so we did a lot of analysis on that and actually made
some presentations to the management council that met downstairs in
Building 30. One of our recommendations to them was, “Right
now, we can’t tell you that we can enter safely if you leave
that gap filler sticking up and the flow is tripped to a turbulent
condition, like the Aerothermal group says it’s going to be,
so we recommend that we take it out.” Several other people did,
too, but that was our recommendation. Every flight, we spent hours
over there. After the Columbia accident, we manned the MER almost
24 hours a day, definitely two-thirds of the 24 hours. We had real
specific inspection techniques developed, inspection requirements
that we had to adhere to.
Sounds like a big effort from your group.
was a big effort, yes.
What was the size that was small enough that you could enter?
have to go back to look at it. I’m thinking it was like, eighth
of an inch? About an eighth of an inch damage in the coating.
not much, and of course, it was panel-dependent. Each panel had a
little bit separate requirement, because the heating was different
and environment was different.
You had tested each individual panel?
did a lot of testing at different damages in the coating to see how
fast we could burn through, or if it would burn through the coating,
to the substrate. We analyzed a lot of different test data. When that
flight changes or something happened during the flight, we worked
on additional calculations.
Were you ever doing any in-flight testing ever at the Arc Jet?
didn’t have to actually do, as I recall, any real in-flight
testing. The tile people had to do a lot of in-flight testing at the
Arc Jet, but I can’t remember us doing any really Arc Jet testing
When you were doing all these, I guess, life cycle or impact tests,
were you doing them all here at JSC, or were you also doing some at
was done at JSC. During the Shuttle program, we had Ames do some tests.
Their test data did not agree with ours. In fact, their test data
indicated we would have had failures. We would have a failure in a
normal flight. Just for a normal entry, we would have exceeded some
of our temperatures, had we used their test data. The only time we
ever used any Ames data was when they helped find the subsurface oxidation,
way back in the early ’70s. For Shuttle, almost all of the carbon
testing was done at JSC. I can’t remember really going out to
use Ames, because we just didn’t get the same results. Our data
seemed to match flight data because we had on-board measurements of
temperatures, and so forth.
That says a lot about the JSC facility itself.
it does. We’ve now closed it. Now we have one facility. I was
on an NESC [NASA Engineering and Safety Center] board that studied
the two facilities, and we found there’s differences between
them, and we couldn’t resolve the differences. From the NESC
report, which is restricted, we recommended a test program to be done
between JSC and Ames to try to understand this problem. It was not
approved. We’ve now closed the JSC facility, and we don’t
have the answer.
Tell me about that assignment. I wasn’t aware of that study
that was done. How did you get assigned to the board?
Curry: I got
assigned as a subject matter expert on ablative thermal protection
systems and Arc Jet testing. I was what they call a national expert
on ablatives, and the NESC hired me. I had retired, by the way, from
NASA by then. I retired in 2007. I was a NASA expert, or an advisory
expert, with that system. I think that came about because at that
time, they were still trying to close the JSC facility, so the NESC
did this study. We made these recommendations to the management of
the NESC, we signed the report with them, they signed it, but the
recommended Arc Jet testing was never implemented. Along with a lot
of other people, like Don Tillian, we both have the same opinion about
testing at Ames. We don’t know what the data means—totally.
It’s good data, but what does it mean? Whether it is not correct,
or is it correct? I don’t know.
Yes, that’s a shame.
is; it’s a shame. The government destroyed a valuable resource.
I spent two years writing letters to Congress on this, but to no avail.
I’m sorry to hear that.
Curry: I am,
too. The worst part about it is, for your information, the people
at NASA Headquarters [Washington, DC] did not provide answers to some
of the questions to the congressional people.
That’s too bad.
it is. It is too bad because my early years with NASA, I used to spend
a lot of time at Ames on Arc Jet testing and ablation analysis. I
spent weeks at Ames and at the Langley Research Center, both of them
were really training grounds for me at the time. Those early scientists
were really very smart.
Tell me a little bit about some of the advanced technology that you
studied out at the Arc Jet. I saw you worked some on the X-38 and
the Aeroassist Flight Experiment?
Curry: I was
the manager for the Aeroassist Flight Experiment. That was another
really good project. The Aeroassist Flight Experiment was going to
be taken up in the Shuttle Orbiter payload bay, and then we would
re-enter the aerobraking experiment at a little over geosynchronous
orbit speed. At geosynchronous orbit speed, we could get the temperatures
up and the heat flux up, so that we had both the convective heating
and radiative heating on the forward portion of the heat shield. That
was mainly a tile system because we wanted it to be re-usable, and
we didn’t want any ablation products, because we wanted to have
a pure re-entry environment. Ablation products will mess up the boundary
layer and change the chemistry, so we didn’t want to do that.
I was a manager on that, here at JSC. We actually built the aerobrake,
and we were in the process of installing a tile system on it, which,
at the time, was basically the Shuttle tile and blanket system, to
deliver the aerobrake to Marshall Space Flight Center [Huntsville,
Alabama]. They were the project management of it. It got canceled
because of cost. The project management costs killed it, but we actually
had built it here at JSC—in house. We bought the tiles from
Lockheed, and we could install them ourselves. Our own technicians
built the structure out in Building 9. It was wonderful. We had a
great time on that.
But it never flew?
never flew. It would have provided data, aerothermodynamic data, for
Mars entry that we were trying to get. It was the perfect experiment.
This was the precursor to what the Arc Jet had developed in terms
of that CO2 landing that Chris [Christopher B. Madden] and Steve [Steven
V.] Del Papa talked a little bit about?
yes, that was for Mars entry. You need to have a CO2 environment,
so JSC developed the ability to test materials in a CO2 environment.
What about the X-38? How different was its thermal protection system
the Germans provided the carbon system on X-38, the nose cap and the
wing leading edge. Basically, all I did on X-38 was just monitor their
work. I did some work on their seal system; we developed a seal system
for the body flap area and did some work on that.
Were you are the Arc Jet for the final test that happened?
Would you tell me about that moment and your memories about what happened?
final test that I was at was for Boeing. I work part-time for Boeing.
Boeing is developing an ablator for their CCDev [Commercial Crew Development]
capsule, which is their capsule to take men to the Space Station.
We were doing the final set of ablation tests on the Boeing material;
we’ve been testing that material at different conditions of
temperature and pressure to determine its performance. The last test
was one of those tests. I can’t really tell you much about the
test, other than it was a Boeing material. It was sad to see. We got
the test done fine, no problems, and got the data.
We’re actually analyzing the data now, but to know we’re
sitting in that facility for the last time, that that’ll be
not there anymore is kind of sad, I guess. In some ways, angry. Like
my wife said, “Well, Don, everything ends.” That’s
true, it does, so let’s end it. I feel sorry for the guys here
Yes, they’ve lost a good resource.
they have. Now, they’ve got to go to another place. One thing
about our facility is that when the contractors came in to do the
tests, we’ve always allowed them free access to the facility,
been willing to make changes to the facility, arc test conditions,
whatever. Plus, our costs have been about a factor of three less than
Ames was. There’s a big difference in our approach to testing
and Ames’s approach to testing.
I wanted to ask you, you mentioned that the test you were doing for
Boeing, it’s a Boeing material, but it’s an ablator. I
know it’s probably proprietary, but is it similar to AVCO, or
is it completely different from what was used during Apollo?
is different than Apollo material. It’s hard for me to say it’s
similar to Apollo because it’s a different material, it’s
a different-based material, but it has to operate at non-lunar conditions.
It only has to operate at Earth-orbital conditions, which are significantly
different than lunar conditions.
Have you run into any challenges, or were they similar challenges
to the one you faced for testing of Apollo ablators?
Curry: I don’t
think they’re much different. We have to go through the same
process, we have to get the data, we have to analyze the data, and
determine if the material’s going to work or not. We’re
in that process. It can meet all the design requirements, that’s
what I was trying to say.
Do you think, having worked on the Arc Jet for all these years, that
testing has changed or anything changed over the years, since you
we’ve become more efficient. We’ve been able to develop
test article designs to meet certain requirements that we couldn’t
do before, initially. We understand a lot more now about the actual
Arc Jet itself and its flow characteristics. Sure, we learned a lot.
Remember, when we first started, back in ’63, this was one of
the first Arc Jets every built. From nothing to where we are, yes,
sure, been a lot of our work done.
That was another question I wanted to ask. I know that there are other
Arc Jets around the country. Langley [Research Center, Hampton, Virginia]
has that small Arc Jet, and there’s Ames, and I think Boeing
has one. How was JSC’s Arc Jet facility unique, when compared
to all of these other facilities?
of the biggest differences is that we could run continuously, for
hours, if we had to. When we were doing Shuttle testing on the carbon
system, we used to test for one hour, continuously and at high temperature.
Ames couldn’t do that, and I’m not sure they can now.
We could run long run-times. We could do low-heat flux cases, which
some Arc Jets can’t do, do testing at low-heat flux. That’s
That’s another thing about ablators: at low-heat flux, ablators
don’t necessarily perform as well as they do at high-heat flux.
High-heat flux, you really get all the ablation process going, which
is what you want. Low-heat flux, you get it going, but it’s
just a little bit different. Ablators can’t be quite as efficient
sometimes at the low-heat flux, but you have to test down there because
you’re going to go through that environment. Our facility could
test low-heat flux, high-heat flux; it could test different size articles.
Others can do that, too, but I think the biggest thing, at one time,
was the ability for us just to test long times, continuous. Most facilities
can’t do that.
Like now, you’re right, we have the Ames facility, which is
the closest thing to the JSC facility—or the JSC facility is
the closest things to the Ames facility, either way of saying it.
The facility at St. Louis [Missouri], which is the Boeing facility,
they can’t test big test articles. They’re limited in
the size of the test article. You like to have a big enough test article
that you can get a good representation of the heat flux and the way
the material responds. The smaller the test article, the more, I guess,
error you can have for performance. I’m not sure that they can
test for long periods of time, and I’m not sure they can test
some of the configurations that we can. The other facility that’s
available is the one at AEDC, Arnold Engineering [Development Center,
Arnold Air Force Base, Tennessee], and their biggest problem is that
their pressures are normally too high. They’re a ballistic missile
facility more than anything else. A lot of times, their pressures
are too high to get the test conditions that you want for the entry
conditions that we have, which are lower pressures, basically. So
now we’re down to Ames or AEDC. We can test there, you just
got to realize what you’re doing when you test there.
As you were leaving, or maybe you were discussing this with your colleagues,
did you mention that you did a test for the Russians out at the Arc
Jet, at one point?
Can you share that?
I can’t share it because I don’t have any of the data.
We did the test—and Tillian knows more about this than I do—there
was a test report written on the response of their material. It was
secret, and it went to our division office, who kept it locked up
for a while. Then, they transferred it to NASA Headquarters.
This was a test for Soyuz? Or you’re not sure what it was?
was the ablation material for Soyuz. I don’t think they’ve
changed. That was a long time ago, and people ask for those results.
They may exist, but I don’t know where they are.
Was this before ISS [International Space Station] or after?
way before. We tested this during the Apollo program.
Now I see why it was locked up. I wondered if you could talk about
the camaraderie out at the Arc Jet over the years, since you worked
it for so long.
was wonderful. Everybody was dedicated to that facility, first of
all, and when you asked them to do something or wanted to be out there,
they were very open to helping you design the test articles. A man
by the name of Jim [James] Milhoan, who works out there, designed
several test articles that helped us tremendously on the carbon system,
particularly on our damage assessment analysis of the carbon. Developed
a whole other system that allowed us to actually simulate a cavity
behind the carbon system, so that once we actually burned through
the carbon, we could exhaust that gas out and let the carbon continue
to burn. That’s how we’re able to form these so-called
knife edges. They’ve been extremely cooperative. There was a
great respect between the engineering group, where I worked, and the
Arc Jet group. We were the same group, as far as I was concerned.
There was no difference, we all worked together. I can’t remember
us every having any problems at all, of any kind.
What significance do you think the Arc Jet has had, in terms of human
Curry: I think
without the JSC Arc Jet, we wouldn’t have flown the Shuttle
the way we did. We may have flown it, but not the way we did fly it.
During Apollo era, that’s another interesting story, we had
a lot of Arc Jets. There were Arc Jets out at Boeing, there were Arc
Jets at General Electric in Philadelphia [Pennsylvania], at that time,
Langley had a big Arc Jet, not a small, little Arc Jet facility, a
big Arc Jet. Of course, Ames had theirs. A company called Aerotherm
had Arc Jets. Plasmadyne had an Arc Jet. We had seven Arc Jets around
the country during Apollo, so we were testing all over the country.
Apollo was so different from Shuttle, in the sense that during Apollo,
because of the mandate to land a man on the Moon by the end of the
century, funding was available to do all this. We had Arc Jets around
the country, we could test at the same conditions, and we could check
each other. We had really good crosschecks against each other, plus,
at the time, Langley and Ames both had real active analytical groups,
too. So we could crosscheck analysis with each other. That’s
what we’ve lost when we closed the JSC Arc Jet facility. We
don’t have a crosscheck now. We’ve gone from a lot of
crosschecks during Apollo to a couple of cross-checks during Shuttle
to no cross-checks.
That’s a big loss.
is. If you reduce the Arc Jet facilities down to one, you don’t
have a crosscheck. There is no way to understand any potential errors
in the tests, test results, or the test analysis. Now, let’s
say Orion launches. They eventually go to the Moon or they go to Mars,
then have to analyze some conditions that they hadn’t anticipated.
Now, they’ve got to go back to the same facility that gave them
the data that said the design was adequate. Where’s the crosscheck?
There isn’t. That’s a bad situation.
What has the Arc Jet Facility meant to you, over the years?
Curry: I guess
it’s my career because I’ve used it during Apollo, started
using it actually with Gemini, a little bit, Apollo, and all of Shuttle.
I’ve done a lot of advanced technology work with the Arc Jet,
developing new ablator materials. I used to have SBIR [Small Business
Innovation Research] contracts. Had one with another company, and
we developed a whole series of ablative materials beyond Apollo, which
I thought were really good materials that could be used for other
programs. I think some of them actually are being used. But it was
an SBIR, so as you know, SBIRs are proprietary to the company, so
that’s where it stayed.
One thing I noticed when I was out doing research, you have a Ph.D.
in engineering. Was the Arc Jet in any way related to your doctoral
not really, because my Ph.D. was related to my analytical work I did
in ablation. I just extended it a little bit in a dimensional way,
to study some effects of multi-dimensional mass flux and flow in a
porous media. It extended some of the equations that I had done in
ablation. It just worked out that way, but I actually didn’t
do any tests at the Arc Jet.
Were you working on that during Apollo?
Curry: I actually
got my Ph.D. degree in ’70, so yes, I guess I was.
sorry, end of Apollo, the start of Shuttle, in that interim period
between the two. I got a full year’s leave of absence to do
That’s a nice perk.
it was, it sure was, to go to school for free and get paid.
Yes, that doesn’t happen every day, does it?
doesn’t. We did that a long time here at JSC. We’d still
be doing that. In fact, we used to offer University of Houston graduate
courses in Building 13.
I’ve seen the memos; I guess now you go over to UHCL [University
of Houston-Clear Lake].
I finished my courses at the University of Houston main campus and
then applied for the year’s leave of absence to do my dissertation,
and I got it.
I know you brought some notes.
they are not notes. These were Tillian’s in case you wanted
I think that we have hit on the questions I had. I’m just taking
a look. Is there anything else you think that we might want to talk
about, in terms of the Arc Jet Facility, Shuttle, Apollo, advanced
I think we’ve hit on it pretty well. I guess my concern is that
on these advanced programs—I was noticing this latest thing
that the [NASA] Administrator put out for NASA, Visions, or whatever
he called it, new projects. I just wonder how he’s going to
accomplish this with less facilities. I was on a NRC [National Research
Council] team; we did a 20-year look at NASA [called] Roadmaps [NASA
Space Technology Roadmaps and Priorities]. One of the things that
our group came up with was a lack of test facilities; NASA seems to
be closing down test facilities. How are we going to advance technology
to do these missions if we don’t have the test facilities to
prove the concepts? There was a quote by an Air Force general, he
said, “I’ll take one wind tunnel test to 10,000 CFD [Computational
Fluid Dynamics] solutions,” because the wind tunnel test tells
him exactly what he wants to know. CFD solutions tell him something,
but they’re analytical.
If the guy running the program doesn’t know what he’s
doing or doesn’t understand the grid, is the answer good or
not? Generally, they are, but it’s like anything else. Tests
always validate what you’re doing, or they don’t validate
it, one or the other, but at least you get an answer. I guess my biggest
concern is for what’s going to happen at NASA. We’ve done
a lot of good, good work, in the medical field and communications
and computer—you name it. NASA spin-offs have been fantastic.
I’m wondering, when you shut down some of these things, how
that’s going to be affected, and how are you going to meet these
goals? One facility in the country can’t do all the testing.
It can’t do the Air Force testing, and it can’t do non-Air
Force testing, or military testing. There’s just not that much
That’s a good point.
made those points in this NESC report, here. That’s what’s
so frustrating, to me. If it had been totally political, I can understand
it, but this was totally from an engineering perspective and a scientific
Well, thank you so much for coming in today. I enjoyed it.
I hope it was satisfactory. I don’t know.
[End of interview]
to JSC Oral History Website