December 31, 2005 notes
Since we have determined that the side-injecting throatless engines do not scale up well, we wanted to build
some test engines with converging nozzles (2:1 area contraction ratios for now). I started working on a new design for a
cooled aluminum nozzle, but we decided that the easiest thing to do for testing
would be to try some graphite nozzles, because we could swap different ones out
on the same cooled engine tube.
I ordered superfine isomolded
graphite rods from http://www.graphitestore.com/,
and I also purchased some of the zirconium oxide and silicon carbide based
coatings to try out on the graphite. I
was pretty surprised to find that even high end graphite is cheaper than the phenolic rods we had been using for the ablative engines. Machining the graphite wasnt as big of a
mess as I had feared. Russ and I just
made sure someone else held the shop vac nozzle as
near the cutting point as possible.
Our first engine design used two radial seal o-rings on the
outside of the graphite nozzle, submerged in the cooled aluminum engine tube and
retained by an internal snap ring. This wasnt
a good design for several reasons: It
required more machining work on the graphite (which are expected to be
disposable at some level) for a precise OD and o-ring grooves, you cant pull
the nozzle out past the snap ring groove without tearing the o-rings (and we
had problems tearing the o-rings putting it in until we completely smoothed out
that side of the snap ring groove), the snap ring ears stick in a fair amount and
limit the expansion ratio you can put on the nozzle exit cone, and, most
importantly, the o-rings just wont last very long against the hot graphite.
We applied the zirconium oxide coating to the machined
graphite nozzles, but we didnt have a way to accurately give it the two hour /
200 degree cure, which may have effected how well it stuck. We did a couple test firings, and preliminary
results were encouraging. Our Isp was improved a decent amount
with the contraction ratio / larger L*, but the o-rings cooked off and allowed
some leakage. After a total of about 40 seconds of firing over a 3:1 throttle
range, we pulled it apart to examine it.
The coating was completely gone everywhere. The throat had only eroded by about 0.050 on
each side, but there was some significant erosion along the sides, probably due
to the burning of the o-rings and resulting leakage.
We decided to modify the engine to a different
configuration, essentially hanging the graphite nozzle below the engine just
like we held the ablative chambers, and using a ceramic face seal at the top
instead of o-rings. I was looking for a
fine felt mat style ceramic gasket material, so I was a bit disappointed on
ordering McMaster part number 1687T21 and finding it to be a fairly coarse
cloth weave, but we cut rings out of it to insulate and seal the engine flange
and to insulate the retaining flange. I
want to try machining a retaining ring directly from ceramic in the future, but
the insulated aluminum flange seems to work for our current testing.
I finally bought a lab oven for the shop. We have wanted one for various things for a
long time, but properly curing coatings was finally a good enough reason for me
to buy one. We made a new graphite
nozzle and did the proper cure cycle on the coating.
This design seemed to work much better, giving us higher Isp and thrust, with no visible
leakage, but we only got to fire it for about four seconds before we had a
little mishap with our test stand. We
are building a new test stand with larger tanks to better deal with these
higher thrust motors, and will continue testing of this engine soon. After the short burn, the coating still
looked good except in one place on the top of the converging section where it
had flaked off. The gasket seemed to
have sealed perfectly, with no signs of leakage.
This engine is making around 1000 lbf,
which would be good for a boosted hop with the X-Prize Cup vehicle, or hover
tests with the big vehicle, but we already have all the pieces on hand to put
together a 2000+ lbf engine, which is what we intend
to do most of the big vehicle work with
When we are happy with our thrust and Isp, I will go back to trying to make a completely regeneratively cooled motor that can run indefinitely.
The big vehicle has had all the plumbing completed, and has
been fully leak tested. The only thing
left to do before flight testing is a bit more wiring, getting higher power DC motors
installed in the gimbal linear actuators, and putting
a well-tested engine underneath it.
This vehicle has used hard line tubing for everything except
the hoses going to the gimbaled engine, which is a first for us. Previously we have used flexible hoses for
convenience, but the hard lines are lighter, cheaper, dont require an
anti-chafe sleeve, are probably more robust, and we can make high quality tubes
of any length and bend in the shop. Very
early on we did our own braided hose assembly, but it always involved a lot of
holes poked in fingers and our leak rate was far from perfect, so I moved to
ordering factory assembled hoses from McMaster.
There service is good, so I could always have hoses in hand by the next
shop day, but we have been planning to move to hard line for a while now. Phil is Chief Tubing Bender.
We already have a couple things noted to do differently on
the next vehicle
For some reason, I had always thought quick disconnect hose
couplings had fairly low pressure ratings, but it turns out that there are
plenty of them available with 3000+ psi ratings. Our current vehicles have had an AN fitting backed up by a valve for filling the high
pressure tanks, but I definitely want to replace that with straight through
quick connect couplings next time, to avoid the need for a wrench and the
chance of messing up a threaded connector.
I still want to use a straight-through connector backed up by a valve instead
of one with internal shutoff valves, because it allows us to vent the high
pressure tanks from the filling port if necessary. If we really wanted to save the weight of the
valves, we could use screw-on shutoff couplings that can be connected and
disconnected under pressure, but that would be less convenient to use.
We had discussed capping and pressurizing the structural
cross members in the vehicle that the high pressure manifolds and regulators
are mounted to, and actually using the cross member as a low-pressure manifold,
but we decided not to. We should
have. The side pipes taking ullage down the sides of the vehicle for vent valves, roll
control, burst disk, engine purge, and pressure gauges would have mounted much
cleaner if they could have just punched a hole in the side of the cross members
and been welded in place.