Re: X-33 questions

From: Larrison (larrison_at_my-deja.com)
Date: 11/27/04 

Date: 27 Nov 2004 14:56:01 -0800

Earl Colby Pottinger (earlcp@idirect.com) writes:
 [Re: Scaled Composite's cryo tanks...]
>First, you said yourself, they had built them and they
>worked. Second, how much experience did the company that
>did build the tanks have with cryo ones. If none, then
>Scaled is still the one that should have been chosen for
>the very same reasons that NASA is always claiming for not
>using one of the smaller companies out there, ie lack of
>experience.

Hmm.. I can't find any record that Scaled had build
composite LH2 tanks anywhere. I *believe* they have built
LOX tanks or LN2 tanks, but that's a much easier do than
composite LH2 tanks.

The company that build the LH2 test tanks had pretty good
experience, as I know of at least two other companies which
had contracted composite LH2 tanks from them. (Rockwell
for example, bought at least one composite LH2 tank from
ATK.)

And, for your last point, there is always a challenge in
contracting to trade off cost, expectation of success, and
bringing new folks into the industrial base for government
contracts. In the competitive X-33 contracts, going to a
higher risk small contactor with little to no experience
with that type of composite structure (fiber wound,
different materials composite structure) would probably
have been seen as a negative by all involved -- NASA and
the contractors as well. I know at least 2 of the 3 X-33
competitors used Scaled in other composite structures.

>>[ ... ] the X-33 (and other company's designs) for cryo
>>tanks needed to be made in a different manner and of
>>different materials (particularly the expoxy matrix) than
>>Scaled's usual working materials.

>Needed to be diffirent, or choosen to be diffirent? -
>Remember NASA was the contractor and they have the nasty
>habit of telling outside experts what materials or tests
>should be used in a design.

Again, I'm not a composite structures expert -- but I
believe it was NEEDED to be different. This was a
different application than the other things that Scaled had
ever built. There are many different types of materials
and matrixes used in composites, and Scaled is not the
expert in all of them.

And I will also note NASA as NOT the contractor for this --
Lockheed Martin Skunkworks was. First, this was not a
standard NASA contract, but what is called a "Cooperative
Agreement", since the contractor (Lockheed Martin
Skunkworks) was putting in substantial amounts of its own
money (around $190 M in the end, I believe). If Skunkworks
did their standard process, the subcontract to make the
composite tank was bid on by multiple companies to the
Skunkworks requirements, and then a subcontractor was
selected. NASA paid Lockheed Martin for the X-33 program,
but the responsibility for the design and the selection of
the subcontractors, and the manufacturing and test of the
vehicle was Lockheed Martin's. NASA had the right to
challenge the design, but under a cooperative agreement,
NASA could not dictate the details.

>[ ... ] But (a) had ATK built cryo tanks before? (b) the
>X-33 was not a high-G design, it was not doing anything
>that the ROTON or DC-X were not already planned to do and
>both of these used tanks from Scaled.

As I said above, I know that at least one other company had
bought a composite LH2 tank from ATK. I know they have
built hundreds if not thousands of similar fiber wound
pressure vessels, although not for cryogenic applications.

Secondly, I can't find reference that Scaled built a
composite LH2 tank for the DC-X. They build an
unpressurized aeroshell for the DC-X. The DC-XA's
composite LH2 tank was built in-house by McDonald Douglas
(as Henry Spencer stated, and I confirmed with some
additional research in parallel to him).

I don't believe Rotary ever built nor tested a cryogenic
hydrogen tank for their vehicle. The demo vehicle they
build operated on tip-powered hypergolic helicopter blade,
and I don't believe they ever did much testing other than
combuster-level tests for their rotary engines, let alone
the full-scale rotary rocket motor.

>> But it appears that ATK has two critical capabilities,
>> which Scaled didn't have for large cryo tanks; fiber
>> placement and fiber winding machines in large scale, as
>> well as the autoclaves to cure large structures. Most
>>(all?) of Scaled's work is hand layup.

>Fiber machines, yes you are right. But it is my
>understanding that at the time of the contract that ATK
>did not have an autoclave large enough either and had it
>built after winning the contract. (Correct me if I am
>wrong - I may have another company mixed in here) There
>is no reason that Scaled could not do the same. [ ... ]

Hmm.. I'd have to go check. I believe there were
autoclaves around big enough to cure the tanks needed for
at least some of the X-33 designs. But true for the
operational RLV. However, I will note you don't have to
cure the entire tank in one piece. There are means to cure
it outside of an autoclave, or join the pieces together.
But joining makes it heavier and somewhat more prone to
failure to have seams which are joined in some manner, but
it is possible to join multiple pieces of composite
structures, even for cryogenic applications.

But think this through -- for the contract to be given to
Scaled, they would have to 1) buy new machines and
equipment, 2) have them made, delivered and installed, 3)
train people to use them, 4) design components using new
materials and designs they had not worked with before, and
5) build these these. This is going to cost Scaled $$$
and time. In a competitive program you want to go with
folks that can do the program with some quality and to your
schedule at a reasonable price. I just don't see how
Scaled could compete doing this without either borrowing
multiple times the value of the company or having someone
in the government dictate a contract to them -- which kinda
loses the idea of "competitive procurement".

[ ... ] >And as for machine vs hand layup - it is clear
>that machine layup did not save the design. If it gets
>laid by machine and fail or laid by hand and works clearly
>using the machines as a reason to award the contract is
>not a good one. No, I am not saying machine layup was at
>fault, I am saying it is just like the present day NASA's
>management to look at the shiny machines first before the
>skills of the hand layers.

I think machine layup was mandatory here -- you couldn't do
this by hand in a reasonable amount of time, and to the
quality required for the large structures required. And
again, it wasn't the composite that failed -- it was the
face *** over the honeycomb insulation on top of the
composite structure that cracked and allowed in liquid air
for cryopumping. My reading of the failure report
indicated there was a structural flaw (a piece of tape had
been left behind when the honeycomb insulation and face
*** had been applied), which I suspect was a hand-
manufacturing defect. So I don't think hand manufacturing
was going to solve the problem.

By the way, go read this link...

http://www.compositesworld.com/hpc/issues/2004/March/379

This isn't the one they've done 9 months of testing on...

[ ... ]
>> Now -- why did the X-33 tank fail? It wasn't the cryo
>> tank that failed -- it was the face *** on the
>> honeycomb insulation that cracked and, coupled with
>> something mis-manufactured (a piece of tape, left over
>> from when the cryo insulation was applied on top of the
>> composite tank structure), that encouraged crack
>> formation, leading to cryo pumping and the failure of
>> the insulation (debonding from the underlying composite
>> structure).

>It was my understanding that the debonding was in the
>composite itself or directly affected the composite
>structure, if it was just the insulation they could have
>just left it loose or reglue it as it was not a structural
>element. At some level the composite tank itself was
>damaged.

No, from my understanding the basic core composite
structure pressure vessel was not breached. However... the
design as proposed by Lockheed Martin required the tank as
an integral part of the structure of the X-33. Having the
honeycomb insulation rip off and debond was going to cause
other failures. The issue was how were you going to
inspect the rest of the insulation/ face*** system so
that you didn't have any more of these problems, rebond the
cryogenic insulation and face sheets back onto the tank,
and how were you going to assure the repairs were good? If
there was a failure like this of the structure in flight,
the vehicle's structural integrity was going to be
compromised, probably leading to failure.

>From what I called, Lockheed Martin came up with a way to
do this, but it was going to cost about as much as building
a new tank system.

Now you can say the composite "tank" was damaged, meaning
the insulation and face*** as part of the overall
integral tank structure -- so I will agree.

>> I dont' find any reference to them making the [DC-XA
>> composite LH2] tank. That tank also had problems with
>>cryo pumping leading to failure of the insulation during
>>test cycles -- although not in as spectacular of a manner
>>(it also wasn't integral to the structural integrity of
>>the vehicle, if memory serves).

>Cite? URL? I only know about hydrogen leaks but not
>debonding.

I'm going to apologise for not having a URL on this. The
source was going to quote -- NASA technical report on LH2
tank testing on the NASA RLV/ X-33 server, is no longer up.
It appears most of the tech demonstration data leading up
to the X-33 competition has been taken down. This was a
NASA Marshall report on cyclic testing of the DC-XA
cryogenic tank using internal insulation. I *think* it
should be available through NASA RECON data base (their
tech report data base) or the IAF STAR data base of
published papers.

>> However, there was a not-highly publicized announcement
>> that Northrop Grumman had successfully completed 9
>> months of testing on a composite LH2 tank at a NASA
>> test faciliity, including scores of fill, pressurized,
>> drain, warm, refill etc cycles.

>Again Cite? URL? Thank you.

http://www.aiada.org/article.asp?id=22462

http://www.irconnect.com/noc/pages/news_releases.mhtml?d=50
053

http://www.capitol.northgrum.com/press_releases/ngpress0113
04.html

Quantcast