March 26 and 30, 2002 Meeting Notes

 

In attendance:

 

John Carmack

Phil Eaton

Russ Blink

 

We are getting a batch of 11 1” cat pack diameter engines CNC machined in 316 stainless at DynaTurn of Elk City, OK. The price was very reasonable, so we are going to have two complete sets of attitude engines sitting on the shelf, as well as cat packs for the rotor and a test stand engine. If these come out right, I will probably get a batch of 1.5” diameter engines as well. We are probably never again going to make four copies of something by hand, it is worthwhile to just get ten copies made by CNC. Even making two copies of something by hand is sounding a little dubious in some cases.

 

The supplies for our Really Big Engine are starting to arrive. We have a set of 1” pipe fittings, a four foot length of 1” braided line, a 1” servo ball valve, and our custom order of 12” diameter stainless steel screens just arrived. We will need to order a bit more silver for it, and our anti-channel rings are a custom run from Smalley that we are still waiting for. We are probably going to have the engine rolled and welded out of stainless, because it will be too big to reasonably machine. We will probably make a couple different nozzles for it, but in high pressure form, it should be good for 5,000 pounds of thrust or so.

 

On Tuesday, we did nothing but work in the shop. We are getting a lot of things permanently mounted on the big trailer, including a slick new filling arrangement using big spring-return ball valves. We have everything pulled off of our old fill cart, so when we do VTVL tests, we will have the trailer at the edge of the pad, and use a 30’ long hose to run out to the vehicle in the center. We finally got around to permanently mounting our vacuum pump and adding a nice switch to it. We also have a couple AC outlets permanently mounted on the trailer, which we either connect to a wall outlet when testing locally, or to Joseph’s generator when running in the field. Our field work should be very similar to our shop work, with minimal setup time.

 

Russ and I finished the new hub for our next rocket rotor test. The 7068 aluminum we used is really nice – it has 2.5x the tensile strength of 6061, and machines much nicer. It is about 3x as expensive, and only comes in round stock, but we will probably use this or 7075 for all our future aluminum work, because raw material cost is a very small fraction of what we spend.

 

I made the new tip engine mounts out of 7068 aluminum instead of brass, and I am drastically milling down our right angle nozzle blocks, which will save a lot of stress on the blades.

 

Neil has located another supplier of extruded aluminum airfoils that offer a 2” bar hole, which we are going to move to for high stress cases, but we are probably going to do our computer RPM control at relatively low speeds with our current blades first, and possibly get them on a low performance flying vehicle.

 

On Saturday, we did some more shop work, including mounting a sight glass on our kerosene tank, calibrating a new small load cell, and fixing the wiring on our control box so we can run two arbitrary solenoids at a time.

 

After finishing everything up, we set out to get the biprop working right. The switchbox we had been using to control the kerosene was very messed up, so we hoped that after fixing that, things would work right again.

 

Our first run, with an 0.070 peroxide jet and an 0.018 kerosene jet, didn’t light, and gave the characteristic “rich cloud” exhaust. At least the solenoid was working consistently. This was a combination that did light with older cat packs, but this was the first 36 mesh silver pack we had made, and it had a lot more than our usual amount of compression, and had 80 discs in it from when we were trying random things last week.

 

We tried going to an 0.090 peroxide jet, but it was still rich. We pulled first 10 discs, then 20 discs (pairs of discs, actually) out of the pack, but it was still too rich.

 

We ran the other cat pack from the rotor engines on the biprop, and got it to light. That pack was put together hastily for the rotor test, and wasn’t ever smooth, but it did work. We could then compare the monoprop numbers and see that the new pack was twice as restrictive. As we have always seen before, a rough monoprop engine will always make a rough biprop engine.

 

We started a new pack from scratch, with a new technique: instead of making the first block of screens just 10 silver alternated with 10 stainless, we doubled up the silver, so it went 2 silver, then one stainless. This approximates the 80 mesh screens we had at the top of the other packs. We did minimal compression on the pack, so it filled the entire chamber with just 60 silvers and 50 interspersed stainless, plus a few extra stainless spacers. We weren’t at all sure that the pack would work well, but it ran perfectly smooth. It looked a little borderline in monoprop mode, with a fairly cloudy start and wisps of cloud while running, but that may also have been due to the fact that it was raining during some of our testing.

 

The important point was that the biprop run was PERFECT. Instant light, dead smooth. This was with the 0.070 peroxide, 0.018 kerosene jet combination, at 250 psi regulated pressure.

 

We then made another, longer run, and everything continued to work great. A lot of the subtleties don’t come out well on video, but it is still very nice looking. There was a brief green flare midway through the burn, which turned out to be a small chunk of our brass retaining plate being consumed by the engine. I’m going to make one in stainless tomorrow.

 

http://media.armadilloaerospace.com/misc/goodKeroseneBurn.mpg

 

The thrust graph is picture-perfect, with only a couple percent thrust variation:

 

media.armadilloaerospace.com/2002_03_30/goodKeroseneBurn.xls

 

Thrust went from 12 pounds in monoprop mode to 18 pounds in biprop mode, which highlights how much pressure drop we have across the catalyst pack in monoprop mode at these low pressures. The chamber pressure jumps from around 100 psi to around 150 psi when the kerosene is burning.

 

With this success, we set up to try and light the ethane, which we had yet to get to burn. We put an 0.090 jet on the fuel port for the gaseous ethane, and started with 200 psi ethane pressure. There was a faint flame and some burning going on, but nothing dramatic. We increased the ethane pressure to 250 psi, and the ethane lit right up, making a great looking plume of blue mach diamonds. It started and stopped instantly on multiple pushes.

 

http://media.armadilloaerospace.com/misc/goodEthaneBurn.mpg

 

We ran a couple more tests, and observed that the ethane combustion seems to be a lot hotter, because we could watch our cooling water starting to steam and bubble a bit very rapidly, while the kerosene ran for 20 seconds without a problem. The ethane was making a little more thrust, so the chamber pressure was a bit higher, which contributes some, and the combustion was likely a lot better with the gas/gas mixing, but we also thought we might be running on the lean side, since we had just come off the “too lean to burn right” level. We had never been able to make the kerosene too lean, because we are at our smallest jet size, and we are likely still rich there.

 

We sort of botched our last ethane test, because we didn’t wait for our nitrogen pressure to get to the full 250 psi on the peroxide side before starting the test, which resulted in extreme richness on the ethane side. Still, we got to see what an over-rich ethane condition is like. When it was extremely rich, it would light briefly, then “blow out”. When the ratio got down to just very rich, it gave a puffy, blurry blue exhaust plume.

 

We also noticed while running the ethane at higher pressures that the regulated ethane pressure dropped during operation. The ethane was only self-pressurizing to 395 psi this cool evening, so when we were drawing gas off, it could drop all the way to 320 when it is self pressurizing. Instead of raising the ethane regulator pressure, we should increase the ethane orifice size and stay with 250 psi regulator pressure. The mixture ratio will still be changing somewhat, because the cooler ethane later in the run will be more dense. Not sure if we should worry about that.

 

We will do more rigorous testing of the ethane next time.

 

We are having the current cooled chamber nickel plated, in preparation for possibly running it in a regeneratively cooled manner. We will repeat some of our tests from today, making sure to exactly measure the cooling water flow rate, before trying to cool with peroxide, and that will be a test we run at the 100 acres, from behind our earth berms. There have been reported problems that 90% peroxide, when used as a coolant, picks up enough extra heat that the decomposition temperature melts silver catalyst. Our pure silver / stainless packs may or may not suffer from this. We will see. We have noticed a bit of silver plating on the stainless screens at the bottom of the packs, which is probably a good thing for activity, but it does mean that there is some erosion from the silver screens. We still haven’t seen any hint of wearing out the pure silver screens in any of our packs.

 

Right now, our engine is in three segments: catalyst pack, fuel injector ring, and cooled chamber/nozzle. The fuel injector ring is getting pretty hot, warping a bit after we run it hard. The next cooled chamber we make will have the fuel injector directly in the chamber section. While it won’t have any coolant circulating around it, it will at least have a better conduction path to the cooled areas. We may try making the next cooled chamber out of aluminum. That will provide much better heat transfer than the brass, but still seems rather more likely to melt if anything goes at all wrong. Something else to test at the 100 acres.