NASA STS Recordation
Oral History Project
Edited Oral History Transcript
Interviewed by Rebecca Wright
Huntsville, Alabama – 20 July 2010
is July 20th, 2010, and this interview is being conducted with Jim
Odom in Huntsville, Alabama as a part of the NASA STS [Space Transportation
System] Recordation Oral History Project. The interviewer is Rebecca
I know you’ve been part of aerospace history for 50 years, but
we want to concentrate today on the contributions you made to the
stack, including the STS design, development and evolution. If you
could start today by sharing with us your experiences when you were
part of the studies of the differences of the orbiter, the Phase A
and the Phase B studies, and then those decisions that were made that
produced the results of what we have today.
be glad to. I joined the studies toward the end of the Phase A period,
and all through the Phase B. Coming out of our original studies was
our most desired concept, which was the orbiter with the flyback booster.
We liked that one a lot because it was the most economical [to operate];
it didn’t throw anything away. It was certainly the most desired
by our team. When that went to [NASA] Headquarters [Washington, DC]
and then to the [Capitol] Hill, it had a development DDT&E [Design,
Development, Test and Evaluation] cost of $10 billion. After the deliberations
with the Hill, we were told that we needed to cut that in half to
a DDT&E of roughly $5 billion.
From that we set aside the flyback booster and came up with the concept
of the tank, the solids and the orbiter. The orbiter didn’t
change that much. It changed some, but that’s when we ended
up with the same engines that the orbiter has today, and obviously
with the reusable motors and the throwaway tank.
During that period also we were dealing with the cross-range requirement
from the Air Force. Not many people realize the impact that the Air
Force requirements had on Shuttle. The 1,500-mile cross-range was
something that they really wanted for the orbiter coming back in.
They also wanted a larger payload bay, and some of the payload requirements
were driven by them. I won’t say they were the outside of the
requirements, but [they were] driven by them. The cross-range had
a lot of impact on the configuration of the orbiter. Max [Maxime A.]
Faget wanted the fixed wing, and that was looked at a lot, and because
of the cross-range it felt like the delta wing—the configuration
that’s been flying—was better suited for that.
Coming from that, we still wanted to get the per launch cost down.
So when we went to the configuration we have today, which is the more
expensive, we raised the launch rate up to a very high number. I think
we finally ended up at 60 a year, and we had to plan for that. We
were very disappointed that we couldn’t have the flyback booster
because it would have been more economical, we believed, than the
configuration we have. But you’ve got to live with what Congress
tells you to live with, so we did.
were on the Source Evaluation Board [SEB], and then served as project
manager for the external tank. You knew of the constraints or requirements
when you developed that production program. Share with us how you
were able to determine what elements now became priority, how you
were able to take such a limited budget—since I know you had
mentioned that you didn’t want to invest into high risk technology
with that tank because it was expendable.
I may sound critical of the results of the configuration that we ended
up with, [I am not]. There were a lot of things done in the program
from a management standpoint, [and] a technical standpoint, that in
my judgment was outstanding. One, that you just alluded to, is coming
out of the Phase B studies a lot of us that were doing those studies
were the ones that ended up managing the elements of the program once
it was implemented. My boss was Bob [Robert E.] Lindstrom. Bob headed
up the Shuttle Office at Marshall [Space Flight Center, Huntsville],
J. R. [James R.] Thompson was the project manager for the engine,
George [B.] Hardy was the project manager for the solids, and I was
given the project manager position for the external tank.
One of the things that Bob Lindstrom did which I think was absolutely
very smart is when we went out with the RFP, request for proposal,
for the tank [he] had already picked me to be the project manager.
When we staffed the Source Selection Board [Source Evaluation Board]
for the external tank, he and I cochaired the Source Selection Board,
which to me was [very] smart, because it gave me the insight from
the [beginning] of the RFP [process]. A fellow by the name of Lowell
[K.] Zoller was my deputy, and he wrote probably 80 percent of that
RFP. He was a great engineer, he was a very good writer. He and I
and the people that we had in the project office, having gone through
the Phase B, we pretty well knew what the program needed, what the
To me, having a project manager on the Source Selection Board, cochairing
it, is the right thing to do, is a very smart thing that NASA ought
to do in the future. It lets you be a part of seeing all these designs
that come in, it makes you a part of the selection—you’re
not the voting, selecting official and that’s good—but
you’re a part of the process, and once your contractor is on
board you’re extremely familiar with that contractor, with what
it has bid. So it makes starting up a project significantly easier
and more productive if the project manager has [gone through] the
Source Selection Board [process]. That’s one thing that I think
was done right.
Also, I served on the Source Selection Board for the orbiter, which
also gave me a lot of insight into the orbiter and what it was going
to look like because the tank obviously interfaces significantly with
the orbiter, putting it mildly. Those were two things that I didn’t
do myself but management saw the value of it. Afterwards I too saw
a lot of value, and I think the agency should really consider doing
that more in the future.
can talk about your steps in becoming program manager for the tank,
and taking your knowledge that you had learned from the SEB into starting
to design how to make that a reality.
were a lot of changes. The original tank that we went out and procured
was fairly heavy, for the right reasons. One of the things in the
design of the tank [is] it’s pretty big, and it’s got
a lot of metal in it, but you want that metal to be very thin. What
a lot of people don’t realize, a lot of the skin that’s
on the tank that’s been flying now for 30-odd years is like
an eighth of an inch thick.
The tank takes a lot of asymmetrical loads, which really affects the
structural design and capability of the tank. If you visualize—the
tank is stuck between the two solid rocket motors. The thrust from
the solid rocket motor is taken out up at the intertank at the front
end of the motors. The weight [and thrust] of the orbiter is [carried]
at the back end of the tank. So you’ve got the roughly 250,000-pound
orbiter hanging on the back end of the tank when all the [SRB]thrust
after liftoff goes principally in at the front end. The orbiter thrust
also goes in at the back end.
It makes the tank a very asymmetrical vehicle. It looks nice and round
and smooth but the load paths are very complex, because you’ve
got the heavy weight of the LOX [liquid oxygen tanks] on the front
end, you got a long lighter weight hydrogen tank [at the back end],
but yet the orbiter thrust goes in at the back end of the hydrogen
tank. It makes the load paths very complicated, which makes it structurally
a lot more complex than it looks on the surface.
When we went out with the RFP the weight requirements on the tank
were a lot more lenient than it came to be later. When we went out
with the RFP, we picked Martin Marietta [Corporation] as the prime
contractor. They were based in Denver [Colorado]. We moved a number
of the people from the Titan [rocket] program, that had grown up designing
Titan, to [NASA] Michoud [Assembly Facility] to head up the program.
The Michoud plant that’s in New Orleans [Louisiana] had been
used for building the first stage of the Saturn IB and the Saturn
V [rockets], and it had been virtually shut down after the Apollo
program. We had a gap in there from the early ’70s until we
started building tanks, we literally had to revitalize the plant.
The security and the facility contractors had been kept in place so
the place was ready to move into, it’d been taken good care
of, but we had to then lay it out for the production of the tank.
The plant is basically 42 acres under one roof; it’s air conditioned.
It was ideal for building a 25-foot diameter tank. We had to build
some additional high bays and checkout cells, but the basic factory
floor space was already available to us. We had to lay it out initially
to build 60 a year. We never really thought we would do that but that
was the requirement. To get the per launch cost down you had to get
the rate up, which was an interesting maneuver.
We laid out the tooling such that all we’d have to do is add
additional tooling. We spread the production out pretty much over
the 42 acres, but we started off with initial capability of building
up to between 10 and 15 a year. That was the minimum. Our initial
investment in the tooling was about $900 million, which made us have
to really think hard about the design of the tank because we wanted
to build that tooling [only] one time. As the rate went up all we
wanted to do was add additional, like tooling. So the buying and the
designing of the tooling was extremely crucial to the success of the
We wanted to automate the welding process. Welding was the big thing,
and we were using state-of-the-art welding then. That’s been
[upgraded] a couple times in the course of the program. The other
thing that we wanted to do, that to me was very critical to the tank,
was in the structural test program we wanted to really understand
the capability of literally every square foot on the tank. So when
we designed our structural qualification program we put extensive
strain gauges on each article to make sure that we understood the
capability and how the load paths were getting from the solids and
the orbiter into the tank and how they would be accommodated.
We did, in my judgment, a very very thorough structural qualification
program early on because we knew that the tank that we were flying
early on was heavy. We wanted to make sure it was safe, and safe until
we had flown it a few times and knew exactly what the loads were,
how they were dissipated out into the structure.
We instrumented the early flight tanks. I think it was the first four
or six, we put quite a number of strain gauges on the whole vehicle
to understand the loads and understand the aerodynamics of the whole
vehicle. We took a lot of heat in the project for the cost of the
structural test program and the cost of the instrumentation. In my
judgment it was probably [some of] the smartest money that we spent.
did the results of your tests meet with your analysis and expectation?
What it gave us [was] a map of all of the loads, how they got dissipated
into the tank. Having known that, then you could look at the capability
of the metal. We looked at the qualification program [and how] we
loaded the structure. Not only did it verify the design, but it verified
how the structure reacted to the loads. With that information it made
it possible to go through two lightweight [redesigns]. This was after
I left the program. I was in the program from day one and up through
the first six launches and then I went on, did something else. But
it was to me one of the smartest things we did.
A lot of that I give credit to guys like Larry [Lawrence B.] Mulloy,
who was my chief engineer, and a good structures man. The bulk of
the complexity of the tank is in the metal and in the thermal protection
system. The initial design of the tank, the only requirement for the
TPS [thermal protection system] was to maintain the quality of the
propellants and to accommodate the aerodynamic heating up on the front
end, because the tank is the “[tip] of the spear” for
the vehicle, so it gets the highest aerodynamic loading and heating
during the atmospheric flight.
Consequently, the LOX tank that’s on the front end has to protect
against the high heat from the ascent as well as maintain the quality
of the propellant. You got to protect the metal and you got to protect
the quality of the liquid oxygen. When we talk about TPS, what most
people don’t realize is you’ve got a third of an acre
of tank that you have to insulate. We had roughly only requirement
in the early design was, like I say, just to protect the quality of
the propellant, not get too much heat into it to cause it to boil
off too fast, and to protect the metal.
After we got into the design for a couple [of] years, the orbiter
tile issue came on and we had to start protecting the acreage against
ice formation. Not only did you have to protect the metal and the
propellants, but then you had to keep that third of an acre above
32 degrees [Fahrenheit] when it’s -424 just an inch away. That
really added to the complexity of the tank. Then after that we had
to go in and put insulation on all of the lines and brackets on the
outside of the tank.
That really complicated [the TPS application] because all of that
had to be done by hand. We developed the automated spray for the big
acreage on the tank, and that was relatively easy to do because [it
was automated]. Where it really started causing us a lot of labor
was going in and having to hand-insulate all the lines and brackets
that’s on the outside, to avoid chunks of ice coming off and
hitting the orbiter. Of all the changes other than the weight reduction,
the changes to the TPS was probably the most difficult and probably
cost us the most money.
The other thing that happened during the course of the project was
[the fact] it is “bucket chemistry.” It’s a whole
bunch of chemicals you put together and you make a foam. That’s
an oversimplification but it’s true. About the time we started
this was when we had the Love Canal problem [Niagra Falls, New York,
toxic waste dump]. Then, all of the attention to [chemicals] and waste
that go into streams and go into the environment got to be [a major
problem]. All of these foams are made up of six to eight major constituents
and you mix them in the right proportion and you get foam. A lot of
these chemicals were made by small companies, and as the environmental
issues came on, it drove a lot of these small companies out of business.
So the last time I heard, we had reconstituted and requalified the
foam [about] six times just because of the chemicals that they’re
made of and the companies going out of business, and every time you
change a component in a foam like that you have to requalify. We spent
a lot of time and effort over the last 30 years requalifying that
foam, because not only did it have to accommodate the aerodynamic
abrasions, it had to accommodate the thermal, and it had to structurally
stay on and yet be sprayable. So the TPS is a lot more complicated
than on the surface that it looks like.
the time that you were trying to determine what you were doing, were
there other products or were there other vendors trying to come to
you and say, “Use our TPS system?”
it was fairly narrow. There were a few that would come, but this was
a new industry at this point. There were a lot of foams on the market
that were built for just Earth use, and those didn’t require
the structural capability to withstand the high temperature and the
aerodynamics of flight. Those were the things that really drove the
foam, and we were the only users for that so there weren’t a
lot of people standing in line wanting to sell to us.
you have highly specialized technicians that had to hand-paint or
and that was] very critical. One of the things I really give Martin
credit for is they put in place good training programs for their people.
There was another thing that Martin did and they did this with my
[support and] encouragement, it wasn’t my idea—I wish
it had been. We picked about 35 of our most crucial vendors. Most
of our vendors were small businesses. We had Reynolds [Metals Company]
and these kinds of companies that were supplying the aluminum to us,
but a lot of our fabricators and our part vendors were really small
They set up a program that before they signed a contract with a subcontractor—and
we picked 35 of the top most crucial—the Martin manager, myself,
and the head of Martin’s procurement would go sit down with
the owner of that company and understand why they wanted the business,
understand why they thought they could do it. And we would get comfortable
with them almost on a first name basis before we signed a contract.
What’s interesting is a number of those people have been there
for 30 years along with Martin. The contract was never recompeted.
They were a good company.
There was one thing that you might find interesting we did. A number
of us on the project had come through the Apollo program. A number
of us had worked on the Saturn V vehicle. I had spent a number of
years working on the second stage, which was 33 feet in diameter,
and not quite as long but almost as big, as the tank. So we had a
pretty good feel for large structures and building them. When we picked
Martin, the biggest thing they had built was the Titan [rocket], which
was 10 to 12 feet in diameter. As we started into the design, as we
started laying out the plant [and] designing the tooling, I sensed
that a number of the people didn’t really have a good feel for
how big this thing really was.
So one of the smartest things that I think we did, and I highly recommend
it for a new contractor, is we went to the local artists’ guild
in New Orleans and we got this guy that had been building the props
for Mardi Gras parties, and [we] had him paint on canvas a full-scale
tank. It was really interesting, I watched him some. He would get
up on a ladder and his paintbrush would be on a long handle, and he
would paint this thing in one dimension to look like it’s three-dimensional,
to where you could not only visualize the size, you could visualize
the curvature of it.
We built this thing, and we didn’t tell the workforce that we
were doing it. Martin management was in it with me. He painted this
canvas, put it all together, and over one weekend we hung it up on
one of the walls in the plant. On Monday morning I deliberately picked
some of the structural designers that I had [observed]—they
were good but they just hadn’t done anything this large. I walked
out there with them to look at that for the first time. They would
look at it and say, “Oh my goodness, I didn’t know it
was that big.”
The thing is I think we paid $7,000 for that. I took a lot of heat
for having spent that, but to me it was probably one of the best investments
[we made] because the people not only designing the tank, but the
people designing the tools and laying out the plant, it just made
[the size] become real for them. I didn’t have any problem with
them picking up the concept and accommodating the size of it in the
That thing stayed on the wall until just a couple years ago, I think
they had to do some modifications to the wall and [had to take] it
down. But there were a lot of things like that that we did across
the program. There were a lot of things that were done to help people
understand the complexity of [the Shuttle] vehicle. It’s a very
complex vehicle. It looks so simple when it lifts off, but it’s
already mentioned that people think the tank is the most simple, but
it’s very complex on its own. When you set up the facility,
from what I understand the tooling and the processes that were put
in place pretty much have stayed in place all of the years.
[basic] layout didn’t change. They have upgraded some of the
tools or upgraded some of the [tools and] processes, but the basic
tooling is what we put in there at day one.
is an amazing achievement itself. Also the years of ensuring that
every tank was the same as the tank that came before.
critical. A lot of effort went into that, and a lot of effort went
into the non-destructive testing of each tank. In other words we do
everything you can do to one and not destroy it, but still have it
qualified to fly, to give you assurance that the structure is good,
and it’s proved to be good.
you share that whole process of moving it from one set of series of
tests to the other set? Also, if you don’t mind, including the
help of your own knowledge because you worked on the orbiter SEB,
you understood how the orbiter was going to work and how that helped
you understand the tank.
just made a good point. There are a lot of things that were done right
in this program. One of them was each of us helping each other on
SEBs and these kinds of things, which was very good. Also I credit
Bob Thompson. There was a lot of angst in the system relative to Lead
Center. This was one of the first real major tests of a program not
managed out of Headquarters but managed out of a Center. A lot of
people were worried about it.
Those of us in the program really didn’t worry about it that
much, the reason being [men] like Bob Thompson who was the program
manager and Owen [G.] Morris who was the systems engineering lead—those
two people and those two organizations [deserve the credit for the
success]. The three of us that had the projects here, and our boss
Bob Lindstrom—were a very cohesive team. Had that not been the
case, life would have been miserable for all of us. But it wasn’t.
The fact that Bob Thompson was located at [NASA] Johnson [Space Center,
Houston, Texas] and part of the Johnson Center in my judgment was
not an issue, but because of his characteristics.
He was very fair. He was a good manager, and the same with Owen Morris.
A lot of us interfaced with Owen in the systems engineering. What
that helped us do at the project level was understand [the other projects].
In the case of the tank where we’re in the middle of everybody,
being involved with the systems engineering—which was basically
the overall vehicle engineering—and the working groups helped
us understand our element and how it interfaced with all the others.
The way that management structure was set up with the working groups
and the program manager Bob—he treated all the projects very
fairly. That was what everybody worried about, was that Aaron [Cohen]
and the orbiter would get all of the attention and we wouldn’t.
That was not the case.
To me, that’s up to an individual, and I give Bob Thompson credit
for doing that extremely well. The same way with Owen, who basically
led the systems engineering, which was so critical for this vehicle.
As complex as it is, if you don’t do the systems engineering
right then you can have a lot of trouble.
did this during a time when we didn’t have Internet service
and video camera access. You really had to find good ways to communicate.
Odom: We had
an airplane that went back and forth to Houston almost every day.
We didn’t have video conferencing then, but we traveled a lot
back and forth to Houston, we traveled to our vendors, and we traveled
to, of course, our primes [prime contractors]. In my case my prime
was at New Orleans. I had the tank from the beginning through the
sixth launch, which was 11 years, and I went to visit my plant almost
every week. Back then we had our own NASA aircraft here at the Center.
Occasionally we’d have to go commercial but most of the time
we had a little King Air or Queen Air that we could run down to New
Orleans and it was about an hour and a half flight, you could get
a full day in very easy. I went down there almost every week for the
11 years. I got to know New Orleans real well.
me about the first time you saw that first tank come off your process.
All of your work had come together, your assembly building, your people
were in place. Now you had one. Tell me what that must have felt like.
Odom: It was
a big day. We had spent a lot of time rebuilding a good relationship
between NASA, the Michoud facility, the local government, the state
government, as well as their congressional delegates. Every time we’d
have a major event we would invite them to come in. They responded
very nicely. When we had the rollout the week before last of the last
tank, I was reflecting back on the first tank. We had a big event.
We had a lot of dignitaries there. They had just elected a new mayor,
and he came out. It was quite rewarding to see that thing ready to
go. We had already built some test articles ahead of that, but to
have the first one going to the Cape [Canaveral, Florida was a major
Just shipping the tank and moving it on the ground [required] very
special transporters. We modified some of the existing barges that
NASA had moved the Saturn stages on [and] we had to redo the dock
facility, reset up all the logistics for the tugs and the barge traffic
to the Cape. We spent a lot of time working with the Cape on how to
process the tank. The early tanks we shipped with a lot of open work,
which [we did] not like, but to meet the schedule. [It was necessary.]
Early on we moved a good bit of work to the Cape, which was true for
all the elements. I had worked with the Cape people since the late
’50s. I had worked with them down there through Redstone launches
and Jupiter launches before we even became NASA. So the launch activity
was understood. Of course we had already gone all the way through
the Apollo, so we understood launching big stuff, which was of real
benefit for us.
mentioned about testing, once the tank was ready to be integrated.
Can you give us the evolution of how those tests progressed?
was something that I think current programs and new programs need
to go to school on. We did a very thorough structural qualification
test program. We would test the liquid oxygen tank, we would test
the liquid hydrogen tank. Many of them we would take to destruction
deliberately, some of them would go to destruction when we didn’t
intend for them to, but we understood the structural capability of
the tank extremely well. That’s just from a structural standpoint.
The other function of the tank besides being the backbone, is it’s
the big propellant tank for the orbiter engines. So one of the major
test programs was taking a tank over to [NASA] Stennis [Space Center,
Mississippi] and testing it with the engines. There’s where
we really learned the performance of the tank from a propulsion standpoint.
So far as loading the tank, so far as protecting the quality of the
propellants, delivering the propellants, verifying the pressurization
system—which was really an orbiter system and an engine-supplied
system—but the tank was the user of the pressurants. As you
ascend and as you deplete the propellants, you have to keep the tank
pressurized. It’s a very integral part to the whole propulsion
system. That we verified very thoroughly in the testing at Stennis.
To me, that was probably one of the most critical of the large system
tests that we did.
The second, next most critical, was the ground vibration test. That’s
where we put the tank in the vibration test tower out here at Marshall,
and we put it in concert with a dummy orbiter and dummy motors and
shake it to [simulate] the in-flight frequencies to make sure we understand
the frequency and the mode shapes of the frequencies extremely well.
We had one surprise. When we put the tank in the tower, we were loading
it with water to simulate the liquid oxygen up in the liquid oxygen
tank. I had been in and out each day as we assembled the vehicle,
but over the weekend we were putting the water in the tank, getting
it ready for the dynamic testing. What we didn’t realize was
that as you load the water in and you get the bulkhead at the bottom
end of the LOX [liquid oxygen] tank full of water, there was no pressurization
system and it wasn’t cold. What it did, it just collapsed the
big ogive panels on the front end of the LOX tank.
Dr. [William R.] Lucas was the Center Director, and he was out here
that weekend when they were loading the tank and I was at home. Of
course I didn’t think anything about it because it’s a
lot more complicated to put LOX in it than water I thought. It turned
out that the tank started just collapsing on itself, and Dr. Lucas
called me and he said, “Jim, your tank just caved in.”
I decided I needed to come out pretty quickly. But we learned from
that and [how to] solve the problem. What we did was [keep] pressure
on the tank when we load oxygen at the Cape. We could have gone in
and put additional metal, but that would have just detracted from
the payload. It was a loading condition that we just missed when we
did the structural analysis.
guess it proved that testing [was important and required].
one of those things—it’s the serendipitous, or the things
that you don’t [expect], that quite often are the [most significant]
things that come out of a test program.
there changes in the requirement while the other components were being
made that impacted the tank?
so much. The load profile, the aerodynamic size, the aerodynamic loads—those
were things that matured that had some impact, but not much. We did
enough Phase B to really pretty well understand the vehicle. It was
more the subtle things like the additional requirements for the ice
protection. There were a few things in the pressurization systems
and depleting the fuel out of the bottom of the tank that we learned
as we went along, but they were pretty easy things to fix or to accommodate.
There weren’t any major things that we had to do to change the
shape of it or to change the size, because it was pretty well fixed
when we started. It was the attention to detail in the manufacturing,
automating the welding, [and] automating the TPS application. Those
were things that were normal business, but they were significant because
they affected the quality of the tank as well as the productivity
of the tank.
mentioned some about the TPS but we haven’t talked too much
about the welding.
welding was something that we worried a lot about. Michoud had built
the S-IC, the first stage of the Saturn V vehicle, so there were some
people in the area—not many—that had welding experience.
We learned a lot about welding aluminum on Redstones and Jupiters
and the Saturn series, as well as the Saturn V. To me one of the most
significant organizations that helped with this [was] our Materials
Lab here at the Center, as well as our Structures and Propulsion Labs,
because those people had grown up—all of us had worked together
for many many years, even leading up to this. We had really cut our
teeth on welding in the Saturn vehicle stages.
That was something, that the Titan work that Martin had done helped
their people understand the sensitivity [also]. A lot of people don’t
realize that literally every inch of weld, and there are thousands,
[must be quality checked before shipping]. Because if you really look,
at the end of the day when we started looking at what we were going
to need to fly to [International Space] Station, almost half of the
[added] capability to orbit came out of the tank. So you can see why
it was important.
One of the things from a flight standpoint we had to do [was] a lot
of work with [US] State Department as to where we could enter the
tank into the ocean. For the various azimuths that you’d launch
on or the various inclination orbits, you have to pick where you want
the tank to go in. Part of where you pick to put the orbiter in orbit
has to be consistent with where you can drop the tank. We wanted to
put most of them in the Indian Ocean, because that was the least ocean
One of the things that you wanted the tank to do is obviously withstand
all the loads and the heat of ascent, but you also wanted it to start
breaking up as fast as you could, and break up into as many small
pieces as you could. The reentry was extremely crucial to us as to
the breakup mechanisms, the dispersion of the parts, and more importantly
the size of the parts that actually hit the ocean. Most of the skin
burns up, but the heavy structure comes in that’s on the back
end and up in the intertank area. So just because you got it to orbit,
that wasn’t the end of our problems.
the studies extensive and did you monitor where the debris comes down
the first six flights?
yes. Matter of fact, we had a lot of help from DoD [Department of
Defense] in tracking the pieces.
you recover any pieces to study those, or you did it all through analysis?
all analysis. We did not recover any.
were having to have the tank protected from the heat but also ice.
Then when the orbiter finalized its decision for the tiles, I think
there was some thinking that maybe it was going to be the tank that
was going to possibly hold up the first launch. It ended up being
those tiles. Were you able to share expertise from your TPS research?
Aaron and I spent a lot of time together. It’s too bad that
you don’t get to interview him. He was a very capable guy. I
enjoyed working with him a lot. We spent a lot of time together, and
when we would have program reviews we would look at all of the elements
together in these reviews. One of the things that we did here at the
Center—and this started way back even before Apollo—is
we had program reviews with our Center Director, where the project
manager would come in and present the project to the Center Director
and all of his lab directors as well as the contractor would present.
We would do that quarterly. It not only kept management apprised,
it kept us on our toes, and it kept an involvement of all of the technical
organizations very in real time with the project. I think that was
extremely valuable. I saw it all the way through the Apollo program,
and we continued it here through the elements at Marshall. I think
it’s a very good way to keep Center management involved with
guess it also gave you a forum of helping to make the decisions on
what component was going to be impacted.
like any time that you had to save weight the external tank was the
first place [you looked].
Odom: It was
the first place because the other propulsion units [did not have]
much margin. We deliberately put margin in the tank, because until
we flew it a few times you’re not really sure exactly what all
those loads would be—and the loads are very complex, because
it’s such an asymmetrical vehicle. It’s not like a nice
vertical stack like we had on Saturn.
One thing that’s unique, and I’m not sure the public ever
really understood, was if you visualize the tank sitting on the pad,
it’s bolted down by the two solid rocket motors. You’ve
got the tank in the middle, then you’ve got the orbiter hanging
off to the side. So here you got 250,000 pounds hanging on the side
of this fairly rigid tank and solid [rocket motors]. When you light
the orbiter engines first, it actually pushes the tank over, it bends
it about three or four feet. The top of the tank actually goes off
vertical about three or four feet, then you light the solids. Once
they pressurize, it wants to pop it back straight, and you time all
of that such that it lifts off when it’s [again] vertical. I
still [enjoy] just watching, but for the casual observer you would
never notice that detail.
Odom: My point
is all of that is systems engineering and integration between engines
and tanks and solids and orbiters. All of that has to be orchestrated
down to milliseconds of timing relative to launch. But look how many
times it’s worked.
goodness. Speaking of which, I know that you were off on other projects
when [Space Shuttle] Challenger [STS 51-L accident] happened. Were
you pulled back in to help with anything?
And that was deliberate, because I had been a part of the project.
When that happened, I was the director of engineering here at the
Center. I’d already moved off of hardware management, and they
asked me to serve as the head of engineering. I basically took care
of the engineering of all the other programs, but I was enjoined from
working on the Challenger accident, which is right.
you got to watch the external tank still do its job.
were also involved in the Hubble [Space Telescope], the program itself
and how the Shuttle had to integrate with it as well.
that was a very challenging program that I thoroughly enjoyed. I was
the project manager for Hubble for three years, the last three years
of the development, right up to the flight. We were going to launch
the Shuttle in [August 1985], and then the accident happened in January.
We had it all ready, and I’d served my commitment to that so
I moved off after the accident. But I thoroughly enjoyed Hubble.
One thing that helped me in managing the Hubble was I knew the Shuttle
very well. The Shuttle and the Hubble could not have been integrated
better in my judgment. The thing that surprised me the most was how
the crews—especially on this last repair—developed the
capability to do those repairs. We had designed the Hubble to be repaired,
but we didn’t design it for them to take 110 #6 screws out.
That just still blows my mind how they could do that with all the
heavy gloves, but they did. I went to the Shuttle launch, I went to
two of the repair launches, and I went to the last one, which I thoroughly
is the tank impacted depending on the payloads?
at all. You build the tank for a payload capability that covers all
the eventual things that the orbiter can do. The only real impact
was making the tank lighter to make room for more payload. Because
the tank virtually goes to orbit, it goes just a few hundred feet
per second short of the orbital velocity that the orbiter needs. That
says the tank is the one to go after weight, because it’s basically
pound for pound. You take a pound out of the tank, you get almost
a pound of payload.
you stopped painting you saved thousands of pounds.
yes. That’s an interesting [experience]. We painted the first
three tanks because we didn’t know how long it was going to
be on the pad, and that foam is very susceptible to ultraviolet light
so the longer it’s on the pad, the foam will start to deteriorate,
and little minute surface [particles] will start to shed off. To avoid
that, we painted the first [three] white. It was like 1,500 pounds
of paint we put on it, basically a flat latex paint. You wouldn’t
believe the ugly letters I [received] when we took the paint off.
“That old ugly colored tank.” Most of them were from ladies
that just thought it really looked good before.
wasn’t pretty anymore.
pretty anymore. I wouldn’t have thought the public would have
paid [that] much attention to it.
you do tests for the elements of nature, how long the tank could be
exposed on the Cape? Or was that mostly done with analysis?
We did a tremendous amount of testing of the TPS, and we did a lot
of it right out here at the Center. Every time we’d have to
change the chemistry of it, we would go all the way back through these
tests and expose them to arc jets, we would expose them to structural
tests, we would expose them to environmental testing. I’m sure
we went through that probably six or eight times in the life of the
program. TPS quality and qualification testing was a big component
of the program, and it was right up to the end.
had a range safety system. How were you involved with that? Tell us
about the importance of why, then of course if you were part of that
when it stopped.
we ended up not having to do too much of it, because we had to have
a linear shaped charge that you could cut the tank [with]. Basically
in our case, if you had to abort a flight, we had a linear shaped
charge that would cut a big gash in the LOX tank and a big gash in
the hydrogen tank. Once you destroyed the integrity of the pressure
vessel then it crumbled very quickly so it wasn’t that hard
to accommodate. It was a critical system, but it was pretty benign.
It just went along for the ride, but it had to work. We knew that
from the beginning, so that wasn’t anything that surprised us
because we’d [worked] that on Apollo.
there discussions about including it, or was it always part of the
have liked to not include it, but it really wasn’t a choice.
there other safety measures that you had put in with the tank? I know
we talked about the debris and making sure it was in the ocean.
most critical to a pressure vessel is the factor of safety. That’s
how much capability the structure has over and above the maximum anticipated
load. We typically liked to have a 1.4 factor of safety for structural
components. In the tank because we understood the capability, we understood
the loads—it was I think one of the first where we ever took
the factor of safety down to one and a quarter. Leaving you only 25
percent margin is getting pretty precise for a vehicle as complex
as this. But we did that because we had enough testing to really understand
the loads and how they dissipated through the structure.
seems like it was a system that one piece impacted the next piece.
Odom: Oh yes,
you believe that the TPS was the most critical system, or is it hard
to consider one?
Odom: It was
probably the most demanding because it had so many different requirements
on it. It had to adhere to the tank, it had to withstand the cryogenic
temperatures at the surface, it had to withstand the [aerodynamic]
loads on it, and it had to keep the surface above 32 degrees. That’s
tough—and it had to be light. So I would say from a complexity
of requirements, the TPS is probably the number one from a tank standpoint.
And the structures would obviously be second.
stated a couple times about working with the vendors and getting to
know them so well. This was somewhat of a change from what you had
done in Apollo because so much had been done here.
reason being, the tank was the first [real] production program NASA
had ever done. We got a lot of special attention with the tank because
NASA had [typically] built maybe a dozen articles, like Apollo we
did 15, [only required] one or two or three of a kind. NASA really
had never done a production program [as large as required by Shuttle],
and still hasn’t. The tank is still the biggest production program
as far as quantity. The solids are obviously close to that. The engines
are close to that because we built a lot of engines and we built a
lot of motors, but those [were reused]. So far as a throwaway, the
tank is the largest production program that the agency has ever done.
Consequently, it was unique in that that was not characteristic of
the longest running. It’s been going on for years till just,
as you mentioned, recently the last one. A lot of tanks came out of
that facility, haven’t they?
have, [and it required] a lot of metal.
many people did you have working for you when it was at the peak during
those first flights?
Odom: In the
project office I had probably 25 or 30 people. I had probably another
200 or 300 here at the Center, not full-time but almost full-time,
in all the engineering disciplines. You’d need to get from Martin,
how many. We probably had I would guess 1,000, 1,200 people by the
time we got into production at Michoud.
easy now to look back on the decision of putting so much money into
low development cost compared to operational cost. Can you talk about
how much time you did or didn’t spend on looking at the overall
operational cost compared to trying to move into development at a
the tank was a high production thing, we spent a lot of time—and
we should, because like I say, early on we put almost $1 billion just
in the tooling. The other key thing that I wanted and the project
wanted was to build all of the test articles on exactly the same tooling
that we built the flight articles, because [we] wanted to know that
I was testing exactly what [we were] flying. That is so critical.
If you don’t do that, if you don’t know what you’re
testing, then you don’t know that your test program is qualifying
your flight hardware. That’s why we put so much money at the
front end of the tooling. DoD would not spend as much money on tooling
on the early development articles, and then once they had flown a
few, they would go into a production contract.
What we did differently [was starting] our production contract at
day one. And that was gutty. That was out of the government’s
experience. It was another thing that was quite unique for NASA, to
basically start a production program at day one. What [we] wanted
to do was make sure that we were testing what we were going to fly,
and use the same welding techniques, the same TPS, all of that the
same that you would have on the first flight.
had a pretty daunting 20 a year, 400 in that facility.
was big then.
it’s not germane to the building of the tanks, there was one
thing that NASA did that to me was very smart. When we were getting
ready for the first tanking test and the first launch, all of the
management in the program had to divide up our time and meet with
the press at the Cape or wherever they showed up. I don’t know
whose idea it was, but the agency hired two men that had been writers
for [Walter] Cronkite [broadcast journalist], and they’d worked
for almost all of the networks writing news stories. They contracted
with these two guys to come around to the Centers, and those of us
that were selected to have to interface with the press, they would
train us how to do that.
When I first heard that I was going to go—I think it was a three-day
class—I first thought well gee, I’d worked with the press
some but not a lot back in Apollo. The press was different in Shuttle
than it was in Apollo. What they taught us is something I still use
today. They taught us to take a negative question and give a positive
answer to it. That’s the most disarming thing you can do to
an adversarial press person. To me, it was probably the most valuable
training that I’ve ever had. I wish I knew who to give credit
for it. It was sure worthwhile, and it paid off dividends.
When we were doing the first tanking test on the pad a few months
before the launch we were all down there, and I had drawn the straw
for one particular day—we had to just mill around the press
site and be available to answer questions. I saw this one cameraman
giving the Cape guys just fits. They had this great big, basically
a big revetment where they put all the TV cameras. It was a good shot,
but he wanted his around on the other side and they kept explaining
to him, “Look, we put this safe place where you can get to it.”
He wanted to go on the other side. They called me in to help them
argue with the guy. We kept asking, “Why do you want to be on
the other side?”
He finally said, “I’m not here to film a success, I’m
here to film this tank when it blows up. That’s all I’m
interested in.” They didn’t prepare me for that. That’s
the closest I’ve ever come to hitting a press [person]—I
really wanted to but I didn’t. My point is another integral
part of any manager’s role in NASA needs to understand how to
deal with the press. That’s probably even more critical now
than it was then. But it was a very good thing. There are a lot of
things that were done right in the Shuttle program that records will
probably never hear of.
I’m glad we’ve gone over that. Is there some more that
you can think of or some other aspects?
Odom: I think
a strong systems engineering and integration organization is an absolute
must. A must is involvement of the Centers and their [engineering]
laboratory capability. To me that’s been one of the strong assets
of NASA. We certainly saw a lot of that here at Marshall because that’s
where [we] grew up. You’ve got to have the confidence and the
trust of management at the Headquarters level, at the program level,
and the project and below. Without that, life can be [very difficult].