July 19, 2003 Meeting Notes

 

Unequivocal Mixed Monoprop Success

 

On Tuesday, we received our third style of catalyst from Catalytic Products International.  This catalyst is a 3.5” tall roll of platinum plated corrugated foil.  Unlike the random bale metal we tested last week, this material has a solid bond that doesn’t rub or wash off, and it seems as active as the ceramic monolith catalyst.  The catalyst we got on short notice has 200 pores per inch, but it can also be fabricated at 400 ppi.  Corrugations give slightly more surface area to volume ratio than square pores, but the 200 ppi sample should still be presenting less surface area than the 400 ppi ceramic monolith.

 

http://media.armadilloaerospace.com/2003_07_19/corrugated.jpg

 

The physical characteristics of the foil catalyst are nicer than the ceramic monoliths, because it won’t crack or chip.  We had to cut a small ramp into the engines to allow us to squeeze fit it in, but it made a nice, tight seal, and it was plenty rigid to be self-supporting at this diameter.

 

We preheated it and made a test run, with little success.  We were rather expecting this, due to the straight shot from the top of the catalyst out to the bottom.

 

Half the team is at LDRS (the big sport rocketry event) this weekend, but on Saturday those of us remaining set up a test facility to characterize exactly how we were preheating the catalysts.  We put gas flow meters in the propane and air feeds, and drilled a hole in an old catalyst so we could inert a thermocouple deep inside while we adjusted the mixture ratios.  This was very educational.  The temperatures could vary over 1000 degrees F with a twist of the knob, so there was no way we were getting repeatable results in our previous tests, and it wasn’t surprising that we were cooking screens and catalysts sometimes.

 

http://media.armadilloaerospace.com/2003_07_19/controls.jpg

 

We settled on 4 scfm of air and 0.38 scfm (23 scfh) of indicated propane flow (not corrected for density), which produced about a 2000 F pack temperature, and soaked it pretty well after five minutes.  I need to buy one size larger flow meter so we can increase the total flows more to get larger engines heated in a similar or lower time.  There is still a little bit of a procedure to get the preheating started, because if you just turn on the gas flows and light it at the nozzle, the flame actually starts away from the base of the catalyst.  If you take a torch with a long extension and press it directly against the base of the catalyst in a few places while the gas is flowing, the combustion will start near the top, getting everything heated well along the way.  The mixture ratio we are running is very rich, so there is excess propane still coming out the nozzle, which we leave burning.  The flame changes from yellow to blue as the pack gets more thoroughly heated.

 

http://media.armadilloaerospace.com/2003_07_19/burning.jpg

http://media.armadilloaerospace.com/2003_07_19/blueFlame.jpg

 

We noticed something interesting when we were doing the open-air heating tests with the ceramic catalyst.  After I shut off the flow of propane, if I left the air flowing through the engine, a flame continued burning above the catalyst for almost 30 seconds, and the catalyst pack continued to glow red hot in places for another 30 seconds after that.  We believe that the ceramic catalyst must have been absorbing propane during the preheat.  The metallic catalyst did not display this behavior.

 

http://media.armadilloaerospace.com/2003_07_19/propaneSoak.jpg

 

The propane flow control does need to be fiddled with a bit if the propane tank is nearing empty, because the gas pressure drops when there isn’t as large a mass of propane to spread the heat of vaporization over.  We cooked one screen when we let the propane flow drop down, which goes closer to stoichemetric, and a lot hotter.  It wasn’t a problem with a full propane tank.

 

We set up the engine with the new catalyst and the new preheat system, brought it fully up to temperature, and fired it again.  We had the flow jetted way down with a 0.120” orifice, so we could try to get it clean first, then build up the flow.  It again failed to produce any real thrust, and left a pool of peroxide under the engine.  However, looking up into the engine nozzle (with the mirror-on-a-stick we have been using a lot lately), the entire catalyst pack was glowing bright orange, much brighter than it had been from the preheating.  The propellant was definitely reacting on the catalyst, so the liquid seemed very likely to be just channeling straight down the center with a gas cushion along the sides.

 

To combat the theorized straight-through channeling, we cut the catalyst into thirds, and assembled the engine with a few screens between each section of the catalyst, forcing any flow to make at least a couple turns before exiting.  We tried cutting it with the band saw, but it kept grabbing on the foil, so we wound up cutting it by hand with a hack saw, which was no fun at all.  The ragged edges and rolled over cells are also good hedges against channeling, but not terribly repeatable.  It is possible that precisely fabricated slices of the same thickness may perform worse without the ragged edges.

 

It was a little more difficult to get the preheat flame started at the top without the straight shots, but the entire engine still heated up at about the same rate.  Our mixture ratio by volume for these tests is 4000 : 900, which is a bit richer than we were running the last set of tests.

 

This time, the run worked perfectly.  We made a total of six runs, going through 30 liters of propellant, and they all worked great.  Thrust curves were smooth, except for some shaking from the trailer.  Interestingly, there was a pale flame visible in the low pressure runs, but the higher pressure runs had completely clear exhausts, just like a 90% monoprop engine working perfectly.  From the noise, it was obvious that exhaust velocities were higher than pure monoprop.

 

0.120” metering jet restriction 154 psi tank pressure, 60 lbf

Change restriction to 1/8” NPT 162 psi, 125 lbf

Ten second long run repeat of last test

Increase tank pressure to 362 psi, 245 lbf

Change restriction to ¼” NPT 168 psi, 163 lbf

Long run, increase tank pressure, long run 386 psi, 340 lbf

 

I measured Isp on the last run fairly accurately as 140 s.  The best we ever saw on 90% hydrogen peroxide monoprop was 125 s, at a tank pressure over twice as high.  We did see just at 200 s with one of our small biprop engines, even at a lower tank pressure, but the simplicity of the mixed monoprop makes it a clear winner for our current purposes.  The mixed monoprop Isp will increase when we get 50% peroxide that is actually 50% instead of 40%, if we put the mixture ratio at stoichemetric, and when we run it with high flowing plumbing.

 

We are really psyched about these results.  The odds are looking very good that this will be the propulsion system for the X-Prize vehicle.  Cheaper, higher performance, and no availability problems.  Big wins.

 

The next step is to continue opening up the flow rates to see when this amount of catalyst finally gives out, then set up for some very long duration runs to make sure the engines or catalyst don’t give out at thermal equilibrium. We still have a small ¼” ball valve and 10’ of hose on this setup, so there is a large feed system loss that we can make go away by moving to a ½” valve and a shorter hose.  If it is still running clear with that, we will swap the small nozzle for one of the 2” throat nozzles, which should be good for over 1000 lbf.  We will probably be using the denser 400 ppi catalyst roll on future engines, which will give us even more potential.  If we can get by with only 2 x 1” thick rolls of 400 ppi catalyst, we can finally fire the 12” diameter catalyst / 4” throat engine, which may see 5000 lbf.  If we have to fabricate a 1” catalyst extension for that big engine, we might consider a different design.