It’s been more than a month since the last time I did a detailed technical update, so for those who are interested here are some more of the gory details on our propulsion development work.
750-2B Testing and Vehicle Integration
At the end of my last post, we had just started firing the 750-2B engine. The -2B was the second rev 750, with a larger throat and longer L*, which had been modified with a bell nozzle. As mentioned, in addition to the bell nozzle, there were several other changes made. These changes included cutting away a lot of material from the outside of the chamber near the throat and replacing it with an aluminum saddle, modifying the cooling groove geometry down near the nozzle end, expanding the cooling grooves in many areas where I thought we had excess cooling to try and cut down on the coolant channel pressure drop. After the series of test, it became apparent that while most of the changes had worked perfectly, I had gotten a little too aggressive on freeing-up pressure drop. While it wasn’t clear that the engine couldn’t steady-state, the thermal margin was much smaller than I wanted, and going for a steady state firing risked damaging the engine.
In the end, we decided that I didn’t want to risk damaging our only flight engine before we had had the opportunity to fly. Our logic was that while we’ve made four different regen cooled engine designs work so far as a company, we hadn’t yet proven to ourselves that we could make an integrated vehicle fly. So, we decided to punt on trying to steady-state the -2B engine, and instead focus our efforts on flying. We had managed to run the engine for 15 seconds without any sign of damage, so we set a maximum flight duration of 9 seconds until we had the time to put the engine back on the test stand to make sure we could go longer, or until one of our other engines in the pipe had been steady-stated.
We finished up the remainder of our test series, including several tests to verify Ian’s throttle controller, and then disassembled the engine, and installed it into the flight chamber. Over the next two days, Orson, Ken Brown (our newest employee), and I started mounting the throttle valves, and figuring out how we were going to run the plumbing for this engine. My original intent for the 750-2 engine module had been to have a single module that had all the valves mounted on it, that would just have four bolts, three fluid connections and 2-3 electrical connections between it and the vehicle. Having the valves closer to the injectors would shorten up the startup and shutdown transients, and reduce the amount of priming we would have to do. However, we decided that it would be better to do first do a “Rev 0″ module first, and then take the lessons learned from that in building the full-capability module. This Rev 0 module has the valves mounted on the vehicle instead of the engine itself, and uses hoses to connect all the valves to the moving part of the engine. The engine layout concept we chose for the 750s uses a heim-joint gimbal ring (much like TrueZer0 and Unreasonable Rocket have done), which allows us to pass the LOX hose, and several smaller hoses and cables up through the middle of the gimbal ring and out the top of the engine frame. Even though the engine components are a bit more spread-out than I wanted, they all fit together pretty well, and we’ve already found several ways that we can improve the packaging for our next engine module rev.
Aluminum Combustion Chambers
While we were getting the -2B engine made late last year, we decided to try making a chamber out of 6061 aluminum, the -2A (for -2 Aluminum), to try out as an experiment. For those wondering, while copper has served us well as a chamber material, and while there are several challenges in making a working aluminum regen cooled chamber, there are several benefits if we can make it work. Aluminum is much cheaper, readily available, and easier to machine well than copper. In the future, if we ever try to use other fuels like diesel or kerosene, aluminum doesn’t catalyze coking and polymerization like copper can. Also, aluminum is substantially lighter–the aluminum chamber/saddle/jacket assembly for the 750 weighs about half what the copper chamber alone weighed for our old 500lbf engines. Lastly, we have a chamber/nozzle fabrication technique we’ve been investigating that could allow for very lightweight aluminum chambers, if we can get the aluminum to work.
Now that isn’t to say that aluminum isn’t free of risk. While a few groups have successfully fired aluminum regen chambers in the past (Bell with their Agena upper stage engine that flew hundreds of times and the Swiss Propulsion Lab), some other groups like Armadillo have run into some problems. Aluminum has a much lower max operating temperature than copper, and also typically has lower thermoconductivity. This makes the thermal design more of a challenge than you have with copper. But the fact that there were existence proofs out there (especially the SPL chamber, which is almost the same size, uses the same propellant, but runs at a higher chamber pressure), meant it was worth trying to demonstrate that ourselves.
So, as soon as we knew about the potential thermal issue on the -2B chamber, we decided to finish making modifications to one of our -2A chambers in parallel with integrating the -2B onto the vehicle.
We took this -2A chamber and had the outside machined to accomodate a saddle (and to factor in all of the cooling channel modifications that had worked for us on the -2B). Because the inside was already thickly anodized, we couldn’t do a bell nozzle on that one, but we were able to make all the other changes. This chamber design is what I’m calling the 750-2AS (-2 Aluminum w/ Saddle), and we got it back quickly and were able to hot fire test it by the middle of May. We steady stated the nozzle on our first day of testing the engine, and worked our way up to a 67 second steady-state firing during our next test session. Surprisingly, the thick anodization layer worked well enough as a thermal barrier coating that the aluminum chamber actually steady-stated at a much lower temperature than the copper chambers did.
Once we had determined that the engine could steady state, we decided to swap it with the -2B chamber on the vehicle, to enable longer duration hovers. We were able to work up to a 60 second hover using the new engine.
After we started flight tests with the -2AS engine, we decided to make the same modifications to our other aluminum chamber, but this time with the bell nozzle. This nozzle had been made as a test-piece by the guy making our copper -2B chamber, but it turned out well enough that we decided to buy it off of him. Unlike the other -2A chamber, this one used brazing instead of welding to join the two combustion chamber halves, so we’ll see how that compares in operations. The modifications to the second -2A chamber (which I’m calling this the -2AB for -2 Aluminum Bell-nozzle) were finished this past week, and the chamber is out at the anodization shop right now. If it gets back soon enough next week, we’ll swap it in before our next flight.
We’re also working on getting a few more injectors made–this time trying a single-piece aluminum construction. These new LOX injector posts have different configurations based on the results of the first set of posts we tested on the 750-1 engine back in February/March. These new injectors should hopefully win us back a few more percent of c* efficiency. Lastly, we’re planning on releasing another major rev of the engines, the 750-3 that will bring in all the various little improvements we’ve come up with as we’ve worked with the 750-2 series. We’re pretty close to meeting all our design goals for the 750lbf engines already, and with this last round of improvements, we may actually beat most of our original design performance specifications. All in all, development is going well.