June 15, 2003 notes
We had an all day meeting on Tuesday with three FMC
representatives, who we are still trying to convince to sell 90% peroxide to
us. We showed them our shop and test
site, and went over everything we are doing.
We also learned a few interesting bits from them:
Buffalo Electro Chemical Corp (BECCO) is mentioned in a lot
of early literature on peroxide from the 50s.
I didnt know that FMC had actually acquired them in the early days.
FMCs dislike for polyethylene containers for 90% peroxide
is mostly due to what can happen in a fire molten PE can go into solution
with peroxide and detonate. In normal
operation, there is some danger of embrittlement in long term storage, but
there isnt any actual incompatibility.
There was an incident where a train containing a tank car of 70%
peroxide followed by a car full of PE pellets derailed. After some fires started and some PE melted,
it went boom.
They went over the specifics of what it takes to make 98%
peroxide detonate it has to be contained in a very strong container with no
ullage space at all, then initiated with a blasting cap.
Propulsion grade peroxide is basically equivalent to the
lowest grade of semiconductor peroxide.
In the early days, semiconductor grade was made by diluting propulsion
grade, but they now have a separate process for it. The two highest grades of semiconductor peroxide are even more
pure than propulsion grade.
We had built a drain in the center of our storage shed that
led to an external pit that would be filled with gravel. We were a little surprised that FMC
preferred that we close off the drain and just let any spills stay contained
inside the concrete building.
We hope to hear something from them soon, but we have
another source of 90% peroxide coming on line this weekend, so we expect to get
the little vehicle in the air soon, no matter what.
We set out to do a series of tests to improve the smoothness
of the 50% / methonal combination on Saturday.
We made a spacer plate to fit between the catalyst chamber and the
nozzle, which would allow us to try adding more screens or foam above the
monolith as a flow straightner, or allow us to try sealing the monolith with an
o-ring at the top.
First, we pulled the catalyst and looked at how water flow
exited the spreading plate with the existing plumbing. Not surprisingly, the tiny flow from the
cavitating venturi we were using just sort of flowed along the plate and
streamed down the sides. We changed up
to 6 plumbing, which flows enough for a proper showerhead distribution.
For the first test, we tried heating the catalyst pack with
a hot air gun. We made an extension
tube with a piece of metal conduit and a sheet metal cone held to the gun with
hose clamps, which worked out well. The
gun heated air to about 1100 F, and it should have circulated through the
entire pack very evenly (up in the center, then back down around the sides),
but when we ran the propellant, it pretty much just gushed out.
For the second test, we went back to heating the catalyst
with a propane torch up the nozzle.
This time, we got thrust from the engine (about 75 pounds), and minimal
liquid, but it was just a cloud of vapor without a flame.
We added a 1/8 restrictor to the feed line and tried
again. This made a flame at the very
beginning, but was then extinguished and behaved as the previous test, but with
only 60 pounds of thrust.
For the fourth test, we swapped the nozzle for the water cooled
biprop chamber, which gives a large post-catalyst space for any delayed
combustion to occur. We had to make an
extension for the propane torch to allow us to heat the catalyst pack through
the long combustion chamber, but (somewhat to my surprise) that worked pretty
nicely. The throat on this chamber is
much larger than the nozzle we were using (2 vs 1.25), so at the same feed
pressure it didnt even reach sonic choking pressure, and generated basically
no thrust. There was a little bit of
external burning, but it still wasnt working worth a damn.
At this point, we believed that the catalyst had been
poisoned. We pulled the catalyst out
and did eye-dropper tests with peroxide, and it was clearly greatly reduced in
reactivity from the original. It still
looks the same, so we assume it is a deposition problem, rather than a
stripping problem. We had hoped that
platinum based catalysts would be less susceptible to poisoning than silver
based catalysts, but it doesnt look like that is true. We were running technical grade peroxide,
but it looks like we will need to go to food or semiconductor grade.
We didnt have any nitric acid on hand for catalyst
cleaning, but we had heard from someone at Space Access that vinegar could be
used to clean catalysts, so we gave that a try. It didnt seem to work.
We have several new pieces coming from Catalytic Products: a 6 square by 2 thick block of catalyst, a
6 square by 4 thick block of catalyst, and a 6 square by 2 thick block of
uncoated ceramic monolith that we can use for flow straighteners. We will cut the blocks to round ourselves,
so we can save the corner scraps for control tests on poisoning.
We have a bunch of weld-on flanges coming from Dyna-Turn any
day now, so we will be able to fabricate multiple different length chamber
extensions to fit between the catalyst pack and the nozzle, as well as allowing
4 thick catalyst chambers. It isnt
clear if this arrangement needs additional combustion volume at all, but we
will try a back-to-back comparison and find out.
Strong Enterprises recently finished a set of drop tests on
the new canopy we are going to use for the big vehicle, and they are going to
refurbish their test article for us, so we should have the big parachute system
in hand within two weeks.
In discussing the parachute system for the big vehicle, we
decided to re-rig the drogue ejection system on the small vehicle. Previously, the drogue cannon piston was
attached to its own cable, which was attached to an eye bolt inside the
ejection tube. This made fitting the
piston a little difficult, because the cable had to coil inside it, and it
allowed the piston to swing around on its cable and bash into things while the
drogue was being deployed. We changed
it so that the piston is now connected at the point where the drogue suspension
lines connect to the riser cable, so the piston will stay away from the vehicle
after deployment. It tested out fine,
and seems to be a positive improvement.
The big vehicle will have the drogue ejected perpendicular to the
vehicle to get it in the airstream rapidly, but the small vehicle just fires
I had bought some high alloy chain to use for rigging the
helicopter drop test, but it turned out to be a mistake the 2700 lb rated
alloy chain is quite a bit smaller than even our 800 lb rated chain, so none of
our quick links or shackles that can fit it have anywhere near the rated
strength. Back to cheap, heavy chain in
I am almost done with the data collection electronics we are
going to use for the helicopter drop tests.
We have worked out a nice engine mounting scheme for the big
vehicle. We will probably have that
completed by the time we do the helicopter drop test. We will be doing our initial flights with
four 2 throat engines, but the X-Prize flights will need four 5 throat
engines to fly with a full load of propellant.