December 3 and 7, 2002 Meeting Notes
Engine Roundup
We did several big engine tests today. It was cold, with 75% humidity, so all the
runs were cloudy, but we got good data on a couple new configurations. Our initial test was a new cleaning of the
pure-silver engine pack. We had noted
that after cleaning in the already used nitric acid solution, the screens had
not turned the chalky-white color that the last batch had. We got some new nitric acid, and tried
washing some of the screens again. A
10% nitric acid solution did not seem to have much effect, even after a couple
minutes, but a 25% solution almost immediately cleaned the screens. We cleaned half of a disc for a comparison
test, and there was a fairly dramatic difference over the previous cleaning, so
the lesson is clear – use fresh acid to wash all screens. The pure silver screens we get seem to have
a fairly tenacious coating on them, and the acid cleans it better than anything
else.
With the marked difference in activity, we hoped that the
silver-only engine would catalyze well, but it still flowed liquid all the way
through the pack without fully catalyzing it.
We still don’t really understand the behavior of catalyst packs, where
we can have more silver surface area present in this engine, but even with low
flow through it due to a mostly closed throttle, it still doesn’t catalyze
well. Removing some silver screens and
interleaving stainless screens then allows it to work properly.
The second engine we tested was using thickly silver plated stainless
screens. The plating shop needed to do
a nickel flash on the stainless before plating the silver, and we also had them
sandblast the finished screens for increased surface area. We were only able to fit 65 of the thicker
20 mesh screens in the engine, even with removing the later anti-channel rings,
but somewhat to my surprise, it was still a good performer. At the lower flow rate due to the long feed
line to the test stand, the pack performed about 4% better than the interleaved
pack. The peak thrust as the line
cleared was actually lower than the interleaved engine, but we need to repeat
the tests more, ideally with a short-feed line, to make any final conclusions.
There is a few percent more improvement to be had by using a
brazed screen ring instead of the perforated metal plate at the bottom, and
welding the intermediate rings in, but it is a bit surprising that plated pack,
with less total screens of a larger mesh, didn’t have a more noticeable flow
rate improvement.
When our coriolis mass flow sensor and new peroxide supply
arrives, we will probably go through a couple drums of peroxide to fully
characterize a couple different packs over a wide range of pressures, run times,
and throttle positions. We hope to
answer most of our questions before going to the 12” engines.
We have four different pack combinations in engines at the
moment, so it is a good time to recap all the different combinations and
results.
ACR = anti-channel ring, a Smally spiral retaining ring that
fits tightly in the catalyst chamber
Pure silver screens are 32 mesh. Inert stainless screens and silver plated stainless screens are
20 mesh.
Press force in pounds is approximately twice the gauge psi,
and the engine are 5.5” diameter, so the actual psi on the pack is the gauge
pressure divided by 12.
Lander Engine
-----------------
2 ACR
10 inert stainless screens
ACR
20 silver interleaved with 10 stainless
ACR
20 silver interleaved with 10 stainless
ACR
10 silver interleaved with 10 stainless
ACR
10 silver interleaved with 10 stainless
perforated metal retaining plate
After each ACR, the pack was compressed to 5000 gauge psi,
and the pack was under compression when the bolts were initially torqued
down. After running a few times, the
stress in the pack was relieved, and there was no longer active spring force. The engine continues to run smoothly, so
active compression does not seem to be required after run-in, although we have
seen 2” engines start to get rougher after they have run for a while.
Used most commonly at 600 psi tank pressure in the sit-down
lander. After nearly two dozen flights,
we were beginning to see some apparent channeling from the engine, with a small
visible stream under the engine.
We had always noticed that the ACR would no longer be a
tight spring fit after the engine had been run, and sometimes they would have
quite visible gaps around them. We had
previously attributed this to the rings shrinking, but we have since found that
it is a combination of ring shrinkage with stretching of the stainless catalyst
chamber. Welding the spiral rings
closed did not help the rings stay any tighter.
This engine uses one of the old nozzles that was sharp
angled, but it has been smoothed out somewhat to about 1.85” throat diameter. The rest of the engines have 2” diameter
throats with a full 1” radius.
Tube Engine
------------
Because the large tanks do not have the pressure capacity of
the tanks on the lander, we began experimenting in hopes of producing an engine
with less pressure drop, as well as improving the channeling issue. Directly welding the anti-channel rings to
the catalyst chamber prevents the engine from being disassembled, but provides
several benefits: it is absolutely
impossible for any peroxide to get around them, it reinforces the chamber to
reduce stretch, it distributes the pressure drop across the pack to multiple
supports, instead of allowing the entire pack to compress down on all the
layers below it, with the attendant pack oscillation problems. When the entire pack is a slip-fit, the ¼”
thick stainless retaining plate would take on a noticeable bow after running
several times. With multiple welded
supports, the only load is divided up among them, such that the we were able to
replace the thick and heavy retaining plate with just a stainless screen brazed
to the last retaining ring before it was welded in. After many firings, it is not bowed out at all.
2 ACR
10 inert stainless screens
welded ACR
10 stainless / 20 silver alternated
welded ACR
10 stainless / 20 silver alternated
welded ACR
10 stainless / 10 silver alternated
welded ACR
10 stainless / 10 silver alternated
welded ACR with brazed screen
Pressed to 5000 gauge psi before welding each ring in, but
pressure was not retained during the welding, so there wasn’t any active
compression in the pack at all.
This engine has been run about ten times, and is completely
smooth. These modifications seem to be
completely positive, at the expense of not being able to disassemble the
engine. If you can’t pull the pack out,
having flanges doesn’t really make sense, so our future engines are likely
going to be top-loaded, with a welded on closure.
Silver Only Engine
----------------
2 ACR
10 inert stainless screens
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
10 silver
welded ACR with brazed screen
Pressed to 4000 gauge psi before welding each ring in, but
pressure was not retained during the welding, so there wasn’t any active
compression in the pack at all.
This was a big hope for a low-loss engine, but we haven’t
been able to make it work yet. A full
pack of pure silver screens gets far too soft, and crushes itself after running
a few times, but by supporting every ten silver screens with a welded ring /
stainless screen combination, we are able to have most of the pack screens be
silver without letting the entire thing crush together. Unfortunately, even though there are more
silver screens than in the interleaved pack, it still won’t fully catalyze.
Plated Engine
----------------
2 anti-channel rings as spacers
10 inert stainless screens
ACR
10 plated screens
ACR
10 plated screens
ACR
45 plated screens
perforated metal retaining plate
After each 10 screens, the pack was compressed to 6000 gauge
psi, and the final assembly with the bolts kept it under fairly near that
compression.
This pack worked, and produced a small improvement over the
interleaved packs, but it wasn’t large, and we are still concerned with the
life of plated screens.
It is tempting to try making a welded engine with silver
plated nickel foam, which we know has spectacular catalytic behavior for a
given pressure loss, but we never did get foam packs that lasted longer than
about 120 seconds, so welding the engines would almost certainly be a bad idea,
even though it would solve the other major problem of the foam packs, which is
that they got compressed with time, and started to run rough.
Quad Engine Setup
The vehicle currently in development is going to be our
first test of our new propulsion direction – four equal sized, canted, servo
throttled engines. This was the layout
for our very first lander that used solenoids, but we moved to
big-central-engine/small-attitude-enginess for the subsequent vehicles because of
some size issues with our our 500-600 lb vehicles, and because we were trying
to mitigate the difficulties of high performance (hybrid / biprop) engines by
only having a single one. Over the last
year, we have become more aware that we can’t get the control authority that we
will need for larger vehicles with solenoid controlled attitude engines, even
if we double them up, and we also found that multiple throttled biprops
probably isn’t something to be all that concerned about.
A test-fit of four engines with (finally!) a master cutoff
valve on a plywood mock-up bulkhead:
http://media.armadilloaerospace.com/2002_12_07/4_engine_mockup.jpg
Russ made a new custom board with four bi-directional motor
drive circuits on it for the new vehicle, and we let it stress test for several
minutes randomly cycling back and forth every 50 msec, without any problems:
http://media.armadilloaerospace.com/2002_12_07/driverboard_test2.jpg
News
