January 18, 2004 notes
Big thrust gains
Our big order of modular 5.5 chambers arrived from
DynaTurn. We now have lots of sections
that we can bolt together with copper flange gaskets to produce any length 5.5
ID chamber, and swap the sections out to test different things. I milled an aluminum top plate with an
O-ring seal for closing the test engines off, since that part of the engine wont
get hot enough to cook the O-ring.
We closed up the bolt holes on some of the 2 thick sections
and welded this to a 7 ID rolled tube to produce a stepped engine. We would have preferred to make a uniformly
7 ID chamber, but the foil monoliths are long lead time items, so we are
making do with what we have on hand. The
hot section has two layers with 500 grams of rings each. We also used an 0.040 hole spreading plate,
reduced the cold pack to a single 1 thick 900 cpsi monolith, and added a
little more top plenum space above the spreading plate to allow the flow to
spread out easier, so it should be a superior engine in many ways to the
We were surprised to see that it barely made any more thrust
than previous engines:
580 lbf from 234 psi feed, 142 psi mid, 124 psi chamber
However, we noticed that while there wasnt a line clearing
thrust rise, there was a sharp spike at the very end of the run. We guessed that the extra liquid volume
above the spreading plate was giving this spike after the rest of the plumbing
cleared. The ½ ball valve has a Cv of
30, so it should be dropping less than 2 psi, but the 90 degree elbow right
after it might be hurting pretty bad.
We changed the test stand over to the big plumbing we had used for the
12 engine and changed the engine inlet from 10 to -16, and got a BIG
775 lbf from 230 psi feed, 185 psi mid, 167 psi chamber
This showed only 18 psi drop across the hot pack, but still
45 psi drop from the tank to the mid section, so we added a pressure tap at the
top of the engine, above the spreading plate, to measure all the plumbing
losses. At line clear, the top pressure
actually exceeded the tank pressure, which must be a calibration error. This showed that the plumbing loss were
under 10 psi, so it was likely the 0.040 spreading plate is the major restriction.
We then did a higher pressure run (still with the second
pressure tap above the spreading plate), and made GREAT thrust:
1325 lbf from 460 psi feed, 405 psi above spreader, 325 psi mid
At this thrust level, the plumbing is still dropping 45 psi,
so there is a good bit of free thrust to be had with even larger feed plumbing. It looks like about half of the drop is in
the line and half is in the elbow, based on the shape of the line clearing hump. It is interesting that the pressure readings
above the spreading plate were noisier than the chamber readings, perhaps
because the thin spreading plate is vibrating.
(note that the series 2 and 3 pressure taps switched around during the runs)
We cut out the 0.040 hole spreading plate and replaced it
with an 0.045 hole one and got even more thrust:
830 lbf from 232 psi feed, 200 psi mid, 180 psi chamber.
Having a chamber pressure of 77% the tank pressure with a
catalyst motor is pretty damn good! We
also did a long run with 135 pounds of propellant and got a 140 s Isp, which we
are completely happy with. We have
doubled the thrust for a given nozzle / tank pressure over the last two months.
This combination is more than what we needed for flying the
vehicle, so we started building a full ship-set of engines in this style. With any kind of luck, we will be hot-firing
the complete vehicle next week. We have
a lot of re-plumbing to do on the vehicle to upsize all the flow paths, but
McMaster actually had the 1 extra-flex Teflon hoses that I need in stock,
which was the item I was most concerned about.
The cutoff computer integration is all done, but I am having some issues
with the extra serial ports on the flight computer that I need to resolve next
We are also modifying the 12 test engine and test stand
plumbing to be 2 all the way, which should take the thrust over 4000 lbf the
next time we fire it.
All of our weld-on 12 nozzles have arrived from EnTek. The current vehicle will probably eventually
be changed over to using four of these engines, while the cabin-at-the-bottom
vehicle may need a dual-quad of eight engines.
Each quad would be controlled be an independent computer, and either one
could land the vehicle, so it would give us a level of redundancy, and reduce
the deep throttling requirement by only using a single quad at landing. We could add the ninth engine in the middle
if we need the extra thrust.