Newsgroups: sci.space.policy,sci.space.history,rec.aviation.military
From: Henry Spencer <henry@zoo.toronto.edu>
Subject: Shuttle-Centaur (was Re: Vandenberg shuttle pad)
Date: Wed, 13 Aug 1997 17:05:00 GMT

In article <EEpq2p.3Dt%spenford@zoo.toronto.edu>,
Henry Spencer  <henry@zoo.toronto.edu> wrote:
>[Shuttle Centaur]  ...The safety crackdown in the wake
>of Challenger killed it, partly because it did have some worrisome aspects
>and partly because it had few users (= inadequate political backing).
>Note that it is a popular misconception that Shuttle-Centaur died due to
>fears of liquid hydrogen in the cargo bay -- every long-duration shuttle
>flight has liquid-hydrogen tanks in the cargo bay! ...

A couple of friends have asked me to elaborate on this...  There were
several reasons why Shuttle Centaur died.  This is my own personal take
on them, although it's based on some fairly good sources.

First, NASA has never liked or trusted balloon tanks -- tanks whose
rigidity depends on pressurization.  (In fact, balloon tanks generally
have had trouble getting respect... the USAF started the Titan program
partly because many of them thought Atlas's balloon tanks wouldn't work.)
They were particularly unhappy about balloon tanks in the orbiter cargo
bay, where it is normally a requirement that there be no single failure
that could endanger the orbiter... especially since some of the Centaur
pressure-regulator hardware was non-redundant.  Centaur's internal common
bulkhead made this worse, because a single pressure-regulation failure
could end up rupturing *both* tanks.

Second, there was concern about the sheer mass of fuels involved, *not*
because of the question of emergency landings -- quite apart from the
emergency-dump plumbing included in the Centaur-capable orbiters, the
once-only emergency-landing weight limit is considerably higher than the
normal-landing weight limit -- but because of problems with weight shifts.
Some classes of failures could move the center of mass around somewhat in
the cargo bay, and that plus sloshing looked like it could cause control
problems in certain situations.

Third, Centaur was lightly insulated using non-rigid insulation (MLI
blankets), and there were unlikely but not entirely negligible scenarios
in which air leaks into the payload bay could lead to oxygen-rich liquid
air dripping into orbiter subsystems.

Fourth, despite being NASA's primary upper stage for the shuttle at the
time, Centaur was treated not as part of the shuttle but as a payload...
which meant that a different, and generally much stricter, set of safety
rules applied.  This decision appears to have been largely because Centaur
was a Lewis project while safety was a JSC responsibility.  Lewis was also
working on a small LH2-handling experiment to fly in the cargo bay, using
a standard shuttle fuel-cell LH2 tank, and eventually concluded that it
was impossible to get JSC safety approval for it under payload rules.
In effect, the bottom line is that nobody but JSC gets to put LH2 in the
cargo bay, period.  Centaur was forced on JSC, they didn't like it one
bit, and when *they* got told to toughen the safety rules no matter who
got hurt, the result was predictable.

The extended-duration pallet uses rigid tanks with extensive insulation
and strong structure, stores the fuels in the supercritical state where
no sloshing is possible, and was done by JSC as part of the shuttle.

A LOX/LH2 upper stage for the shuttle is not impossible.  It would need
rigid tanks, a durable structure with strong cargo-bay attachments, and
substantial rigid insulation to prevent liquid-air problems.  And it
would have to be a JSC project, or at worst a joint project with strong
JSC involvement.
--
Committees do harm merely by existing.             |       Henry Spencer
                           -- Freeman Dyson        |   henry@zoo.toronto.edu

Newsgroups: sci.space.policy,sci.space.history,rec.aviation.military
From: Henry Spencer <henry@zoo.toronto.edu>
Subject: Re: Shuttle-Centaur (was Re: Vandenberg shuttle pad)
Date: Thu, 14 Aug 1997 23:10:54 GMT

In article <5stk5l$57k@lace.colorado.edu>,
Frank Crary <fcrary@rintintin.Colorado.EDU> wrote:
>>A LOX/LH2 upper stage for the shuttle is not impossible [but...]
>
>Of course, that might make the whole idea pointless. You just described
>a major increase in structural/tank mass (relative to a Centaur), which
>would reduce the performance of such an upper stage. It would still
>(probably) be a better than the Inadequate Upper Stage, but perhaps
>not much better. Would a marginal improvement justify the costs of
>developing a new upper stage?

On the whole, probably not.  The real killer now is the lack of payloads
for such a thing, given that NASA is not planning any more "Battlestar
class" planetary probes.
--
Committees do harm merely by existing.             |       Henry Spencer
                           -- Freeman Dyson        |   henry@zoo.toronto.edu

Newsgroups: sci.space.policy,sci.space.history,rec.aviation.military
From: Henry Spencer <henry@zoo.toronto.edu>
Subject: Re: Shuttle-Centaur (was Re: Vandenberg shuttle pad)
Date: Thu, 14 Aug 1997 23:24:25 GMT

In article <maury-1408971541040001@199.166.204.230>,
Maury Markowitz <maury@softarc.com> wrote:
>> Some classes of failures could move the center of mass around somewhat in
>> the cargo bay, and that plus sloshing
>
>  I'm not so sure I can see a class of failures here that doesn't result
>in catastrophic failure of the orbiter anyway.  The only one I can think
>of is a single tank rupture with a _slow_ leak combined with an aborted
>launch.  What else would result in "sloshing"?

Remember, the tanks in the stage generally won't be entirely full -- there
has to be a bit of headroom, if only to allow for thermal expansion due to
slight changes of conditions.  Any substantial liquid-fuel tank has this
problem, actually; the first Iridium launch had attitude-control problems
during satellite deployment, tentatively attributed to fuel sloshing in
the satellites.

I confess I'm slightly unclear on exactly what classes of problems would
cause a CM shift, although I suppose a partial structural failure in the
support hardware would.

>> In effect, the bottom line is that nobody but JSC gets to put LH2
>
>  Considering it's a LH2 engine, if weight considerations are important, I
>must assume that it's a _large_ booster.

Well, most of the "fuel" mass involved is LOX, which is not particularly
light (and is probably where most of the trouble comes from).

>> The extended-duration pallet uses rigid tanks with extensive insulation
>> and strong structure, stores the fuels in the supercritical state where
>> no sloshing is possible, and was done by JSC as part of the shuttle.
>
>  What is the supercritical state?  Pressure frozen?

No, the other end of the spectrum.  When you heat a liquid (keeping it
under pressure to prevent it from boiling), it starts to behave a little
bit more like a gas -- the molecules start bouncing around more vigorously
and there starts to be some free space in between, so it becomes somewhat
compressible.  When you compress a gas (keeping it warm to prevent it from
condensing), it starts to behave a little bit more like a liquid -- the
molecules start noticing each other more as the space between them starts
to close up.  Push either of these trends far enough, and the difference
vanishes.

On a pressure-vs-temperature graph, the phase-transition line that
separates "gas" from "liquid" goes up for a while and then simply stops,
at the so-called critical point.  Beyond that, there is only a smooth
change in properties as you change temperature or pressure -- there is no
sharp transition between gas and liquid.  The supercritical state has been
described as "sort of a dense fog".

The advantage of storing LOX or LH2 this way is that there is no liquid
surface within the tank, and hence no problems with sloshing or with
getting gas instead of liquid into the plumbing in free fall.  You just
raise the temperature a bit to keep the pressure up as the tank empties.
Gemini, Apollo, and the shuttle all use supercritical storage for their
fuel-cell fuels (which is what's in the extended-duration pallet).

The disadvantage of supercritical storage is that it requires relatively
high pressures, which means heavy tanks.  This is why rocket stages do
*not* use supercritical storage.
--
Committees do harm merely by existing.             |       Henry Spencer
                           -- Freeman Dyson        |   henry@zoo.toronto.edu