On Tuesday, December 22, 2015, David McMillan <skyefire@xxxxxxxxxxxx> wrote:
On 12/22/2015 8:28 AM, William Claybaugh wrote:
Marcus:I'm not really sure the SRBs are the correct comparison here, though.
As I recall, two SRB's sank. The remainder were recovered and reused.
There is enough remaining hardware to do the first couple of SLS launches;
I believe the pacing item is rear fulcrum's.
In our enthusiasm for SpaceX's impressive achievement we should not
overlook that reuse of the SRB's never made economic sense: flight rates
needed to hit about 18 per year to reach break even. While the recovery
costs were around $300k per SRB, refurbishment cost many millions--more
than the cost of a new motor. That is the issue SpaceX--and Blue
Origin--now confront.
After all:
1. Water recovery vs dry-land (or barge deck) recovery -- saltwater
corrosion was a significant cost driver on SRB refurb, IIRC
2. Solid vs liquid -- As I understand it, the inspection and cleaning
process for an SRB is much more involved than what it's likely to take for
a liquid stage (once multi-flight MTBF numbers have been established with
some reliability). And the refueling process is *definitely* simpler.
3. Re-testing. Test-firing one of the SRBs left you with just as big a
cleanup and refurb task as a full-up STS flight. SpaceX can (and has, for
static test-firings) just keep re-fueling and re-lighting. About the only
thing they can't ground-test is the hypersonic reentry stresses. In fact,
one rumor I heard is that this stage will be moved to NM and put through
multiple "hop" flight tests, Grasshopper-style, precisely to see what
performance post-recovery looks like relative to baseline.
4. Infrastructure. An RTLS F9 stage recovery is nearly as simple as
driving the TEL over and grabbing it -- just like the trip from Hangar S to
SLC40, but in reverse. The SRBs required substantially more.
Of course, the STS orbiter itself isn't a great comparison either,
although I've heard it claimed that near the end of the STS life cycle, the
latter marks of the RS-25 were coming close to the original re-usability
specs. But even if that's true, there's so many other differences as to
make direct comparison problematic.
What good baselines *do* we have for a near-apples-to-apples
comparison? The closest thing I can think of is the DC-X, although that
never went to the kinds of altitudes and velocities that the F9 booster has
to endure. Are the DC-X data for engine life and MTBF public?
I have them.
Of course, the *closest* data is probably the Grasshopper and F9R results,
but SpaceX has held those close to their vest, AFAIK. And, again, no
hypersonic phase.
Still, SpaceX has already demonstrated that the first-stage Merlins
can be re-fired multiple times, and clock up several flights' worth of
firing time, without requiring substantial overhaul (so far as the publicly
available information indicates). They've also demonstrated (IMO) that
landing the first stage is entirely within reach, although barge landings
are still questionable and they still need to prove they can do it
reliably. The only major unknown, from my seat in the bleachers, is what
effect flight stresses (particularly Max Q and the hypersonic reentry
phase) have on the vehicle life. My guess is that the tentpole items there
are metal fatigue on the vehicle as a whole, and possible damage to the
engines from taking re-entry heating and aerodynamic stress "in the teeth,"
as it were.
Given what we *do* know, I don't see any reason to expect that F9
first-stage re-use costs would be anywhere close to that of an STS SRB. I
also think there's good reason to be cautiously optimistic about the re-use
costs, relative to the costs of a new-build F9 first stage. The good news
here, of course, is that we (well, SpaceX) now have actual flight hardware
in hand to get hard data from, and little stopping them from getting
another 2-4 recovered boosters over 2016.