™
Go UP young man!
On June 18th we fired the hybrid rocket again. Since then I've been busy and haven't had a chance to post an update until now.
The engine lit and burned fine for about 7 seconds. Then it unlit and was spewing a cloud of peroxide. I decided to let it go and dump all the peroxide through the engine since We'd have to dump out the peroxide anyway. This was a mistake. It spewed peroxide for about 14 more seconds, and then it blew up.
It seemed strange that it would have stable behavior for 14 seconds and then suddenly change, until I realized that the time when the explosion happened was about when we were expecting to run out of peroxide and switch to GOX. What I think happened was that the change caused the O/F ratio to transition back into the flammable range and it relit. But since the chamber was already full of fuel vapor it had a hard start.
So this was a classical hard start: too much fuel and oxidizer in the chamber on ignition. It's just strange to have a hard start happen not at the beginning of the firing. The solution to the problem is simple: don't let it unlight, then it can't relight and hard start.
Also, why did it unlight this time and not last time? It turns out there was a difference between the two tests. After the unlight on April 3rd I really wanted the test on June 10th to go well, so as an extra precaution I only pressurized the tank to 140 psi. For this test I went back up to 150 psi. So it seems we were right on the ragged edge and the extra oxidizer flow from the extra 10 psi made the difference.
The thing to do next is to fix the O/F ratio so it won't unlight. In a hybrid this requires changing the injector geometry and/or the fuel grain geometry, and probably the nozzle geometry too if the total mass flow changes. After this test we have to build a new engine anyway so that's what we're going to do.
But in the bigger picture, the reason I chose this design was that with its simplicity it was supposed to be very reliable. With these past two hard starts I'm wondering about that assumption. It's possible that there are only a few ways that it can fail and we have now found them all and can avoid them. After all, it's easy to make a solid with too much burning area that blows up on ignition, but well designed solids are used as commercial upper stages all the time. Or maybe there are more gremlins out there waiting for us. (Rodents of unusual size? I don't think they exist.)