News Archive

Electronics work, LOX Engines

Novermber 28, 2004 Notes

Electronics Work

We got one of the main computer boards in, but it has some issues, and the motor drive board still isn’t done. We didn’t find out until this week that some of the PC board mount CPC connectors we had specified weren’t available, and of the few connectors that they actually got installed, one was the wrong polarity, and another was missing hold down screws. Sigh.

In any case, we are finally bringing up the board, with Russ making hand soldered changes as necessary. We have already found a few things we would like to revise in the future, but I’m certainly not going to wait on another rev of the board before flying the vehicle.

Lox Engines

We upsized the hose from our LOX dewar to the test stand, but our flow didn’t increase at all, so we are limited by the dewar dip tube and globe valve. We are probably going to have a custom tank made with 1” or larger plumbing.

We built and tested another all-aluminum preburner that uses a large central burner chamber instead of expecting some of the burning to take place in the radial pancake sections. This design also tried a significant simplification – instead of having an external LOX vaporizer to feed GOX to a control solenoid for the burner, we just drilled three 1/16” holes from the LOX manifold into the burner area. The idea is that the aluminum facing the burner will be hot enough to vaporize the small amount of lox flowing through the little holes, and since we start the lox flow first anyway, everything should just work. The spark plug is also positioned on top, instead of on the side.

We made several extended length runs with it (without a chamber below), and everything worked out well. The first spark plug we used had too much reach and burned off the top, but when we switched to a shorter one, it showed no signs of damage at all. The internal lox vaporizer worked great, all starts were instant. We did notice that at least the top of the burner was frosted over after the long runs, and there was a slow drop in the main gox temperature as the runs stretched towards a minute, so we think it is likely that at thermal equilibrium there is actual LOX getting through the holes to the burner, which messes up the mixture ratio a bit, but it doesn’t seem to be hurting the burner operation other than slightly reducing the total output temperature.

We are probably going to try pulse width modulating a solenoid to demonstrate throttling with the preburner, because the flow is far too small (0.022” injector orifice) for any of our proportioning valves. We’re not sure how that will work out, but we will find out… We are still running all the valves with manual control switches right now (we don’t have any of the tight timing issues that most liquids have, because we bring the lox vaporizer up all by itself to start), but when we get our spare flight computer electronics boards, I plan on using that to bring all the engine actuators under closed loop control. At that point we should finally be able to demonstrate the smooth biprop deep throttling that we have been developing the engine for.

We have built a new channel wall cooled chamber that has the external contour follow the nozzle contour to increase the cooling. As an experiment, I actually used the CNC mill to cut the external contour. It was nice to be able to cut the contour and the cooling channels without changing the setup, but it took 14 hours on the mill to cut (it did fine completely unattended), and even with 3 degree steps, there is a pretty noticeable scalloping on the slopes.

We filled the channels with pressed in slivers of wax, which turned out to be much more labor intensive than expected – pressing them in was pretty easy, but cleaning all the wax off the area in between the channels, especially at the nozzle, took several hours. We had intended to wrap epoxy impregnated fiberglass pipe repair tape around everything, then melt the wax out, but we couldn’t get the tape to follow the nozzle contour worth a damn, so we aborted that operation, and we are going to instead try to wrap epoxy impregnated fiberglass cord around the chamber on Tuesday.

We are going to try making a gun-drilled chamber / hyperboloid nozzle sometime soon, which we have reason to believe may be the best way to go, but likely has a minimum reasonable size a couple times bigger than what we are doing now.

I’m going to be lazy and just dump up all the pictures Matt took for the last several weeks, you can try and match them up with the last two updates yourselves…

http://media.armadilloaerospace.com/2004_11_28/2004_11_02_a.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_02_b.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_a.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_b.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_c.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_d.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_e.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_f.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_g.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_h.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_06_i.jpg

http://media.armadilloaerospace.com/2004_11_28/2004_11_09_a.jpg