August 21, 2008 notes:
Things are going well
I just cant talk about much that we are working on. Armadillo has been running at a slight
operating profit so far this year, three team members are full time, and
everyone is getting paid. However, most
of our effort is going to the Rocket Racing League project, and all information
about the project is going through them.
With the Lunar Lander Challenge coming up, there should be more publicly
available information bits soon.
We are prepping the same three vehicles for the Lunar Lander
Challenge that we took last year. The
only real changes that have been made are to move to the film cooled engines
and new startup / purging sequences. The
electronics boxes have had pull down resistors added to all the A/D channels
for some extra ability to detect wiring faults, but we are pretty happy with
how the rest of the systems performed last year.
We had a weld crack on an injector, and we have made a
couple changes based on that: We now
just use 310 alloy for the injector face that is exposed to combustion, and 304
for everything else. 310 has better high temperature properties, but is more
susceptible to weld cracking. This saves
money as well, since 310 is many times more
expensive. Welding our internal cross
drilled rings used to be just a fusion weld, because it was a fairly tricky
position, but James now uses a smaller torch head, fine welding rod, and an
automated rotary table to allow the parts to use a good bevel and filler
material. We saved some money by just
cobbling a motorized rotary table together out of some old valve actuators and
a manual rotary table:
(the piece clamped down is just an aluminum test scrap)
We have poured a pad for tethered hover testing at our new
location, but there was a recent FAA re-interpretation of the law that absurdly
states that testing under a tether, as we have been doing for over eight years,
is now considered a suborbital launch, and requires a permit or waiver just as
a free flight would. This is retarded
and counterproductive in so many ways, and the entire industry is lashing back
over it, but it is an issue we have to deal with in the next couple months.
We are still struggling with the methane engine for
NASA. We have fired five different
injector combinations, but we still dont have a combination that can run
stably at a high performance level without burning either the chamber or
injector face. We are making progress,
with the last engine running at only a couple percent roughness
and lasting 15 seconds before eroding some of the face, but it definitely isnt
working as well as the alcohol engines, which have less than one percent
roughness and run indefinitely.
We have been fairly unhappy for some time with the Parker
cryogenic hose couplings that we use for lox due to significant leakage during
filling operations, but when we tried using them for LNG (differently keyed, of
course), they were completely unacceptable.
The LNG isnt quite as cold, but it leaks a lot easier. We are now using couplings from Carter
Cryogenics, which work perfectly so far.
We are going the other direction now, using a Carter coupling for lox,
and it also seems to work well. The high
flow couplings are very expensive, so we probably arent going to retrofit all
of our vehicles, but future connections will be Carter.
In general, cryogenic fuels seem to be much more troublesome
than cryogenic oxidizers. A little
leakage around a valve stem from lox isnt an issue at all, but with methane in
can easily catch fire during an engine burn and give you an almost invisible
fire to deal with. We lost some sensor
wiring after one test due to this.
Conditioning the plumbing and injector for startup is similarly more
troublesome if you get hot lox or a gox bubble, the
engine generally doesnt mind running very rich for a couple seconds, but
getting a fuel gas bubble can result in damaging lean conditions. You cant expect both propellants to go
through gas / liquid transitions at the same time even if they are both cryo, due to differences in the plumbing and manifolding. In a
vertical configuration, flowing some propellant
through each side in sequence is probably a good idea, but horizontally that
would be unacceptable due to mixed propellant pooling.
The large difference in element sizes seems to make it more
difficult to get a good high performance injector pattern. For our split-triplet alcohol injectors, the fuel
and oxidizer holes are within 20% of the diameter of each other, but a similar
methane injector requires oxidizer holes 50% larger. We are currently working with an O-F-O true
triplet design to keep the holes closer to the same size, but that has forced
changes to the manifolding and deck thickness to get
the exit points close enough together.
Somewhat to our surprise, a methane engine is also more difficult
to ignite. We had a couple frightening
tests where the igniter was operating with a strong pressure signal, but the engine
was just streaming out propellant for the 400 milliseconds between igniter
check and first combustion chamber pressure check. We have doubled the mass flow through the
igniter to account for this.
The promise of methane is still as a self-pressurized upper
stage propellant that burns all the way to vacuum in the tanks. I cant recommend it for booster applications
at the moment, although letting it self pressurize to 200 300 psi and
avoiding helium in the logistics chain may still be relevant.
We have modified our horizontal test sled to be bolted down
to one of our anchor chain boxes to allow us to conduct test firings while the
crane truck is off at a remote airport for rocket racing league work.