Home / News

News Archive

Oct 10 meeting notes


We met at Norwood Autocraft, and actually got everything all plumbed together, and were able to load and flow water through the engine on the test stand (without the catalyst bed).  I need to start bringing a digital camera to the meetings, because it looked pretty cool all set up.


Bob Norwood was a huge help, and he is interested in working more with the project. Building things at his shop is a very large benefit.


It was a bit sobering to see that something that takes a couple minutes to draw out on a whiteboard takes three hours to put together at a fully equipped shop.  If we hadn't been at Bob's shop, it probably would have taken us three weeks of ordering a couple new parts each week before we got everything together.


Bob also had some strange metallic foam that had been used as a catalyst base in peroxide dragsters a long time ago.  We may want to do some tests with that, depending on how well the pure silver screens work.


The screens in Juan's motor appear to be silver plated brass.  We will get more exact documentation before we run anything through the motor.


I did a radio-shack switch box for the fill cart valve controls, and Russ and Phil wired it up.  A Good Call on their part was to use different connectors for each valve type, so it isn't possible to connect the switches to the wrong valves.


I showed my current primitive 3D GPS trace visualization program on my laptop.


The flight hardware (which was strapped to the test stand) consists of:


Flight tank

|-- manual bottle valve -- launch valve -- engine

|-- Quick connect  to fill cart

|-- Tank pressure guage

|-- Tank blow-off


The tank manifold is not very optimal, but it is fairly convenient for our current purposes.  Both the manual bottle valve and the launch valve are fairly restrictive, so we may need to run higher tank pressure than would be necessary on a better system.  We have plenty of headroom for pressure with the current tank.


When we get around to mass and flow optimizing a flight vehicle, we will probably get a custom tank manifold made that has a direct shot at the flight valve, and only one side port for the fill connector.


It turns out that a pipe clamp holds the engine to the fill stand quite nicely.  If we put a 90 degree bend on the inlet, it should be able to push against the load cell without any problem.


The fill hardware (which was spread out on a table for now) consists of:


Quick connect

A short length of high pressure hose.

|-- N2 solenoid - high pressure hose - N2 regulator - N2 bottle

|-- H2O2 solenoid - <check valve< -- low pressure hose - H2O2 tank

|-- Vacuum / purge toggle valve - low pressure hose - vacuum pump


The N2 solenoid and the H2O2 solenoid are momentary valves controlled by a single dual throw momentary switch on the launch panel.  There is never a reason during the fill procedure to blow N2 through the H2O2 fill tank, so it was good to make it impossible.


The Vacuum / purge valve and the launch valve are open / close valves that retain their position without power.  This allows us to make sure that both the flight hardware and the fill hardware are never left closed, where they might pressurize from residual peroxide decomposition.


There are several things we are going to improve on the setup.


We intended to have a catch tank before the vacuum pump, but we didn't have the bottle for it.  This will be mandatory to prevent blowing anything into the vacuum pump.  We would also mount a vacuum gauge on the catch tank, and it should have a one-way vent on it to make sure that it never gets pressurized.


We found out that we positively need a check valve before the engine.  I had been thinking that it was a good safety idea, but it turns out to be more crucial than that, because the launch valve is pretty directional, and leaks when there is a vacuum in the tank.  We did our test with Phil holding his hand over the engine nozzle (and getting a bit of a welt from it) while we drew the vacuum and loaded the water.  We managed to not hear Neil making the sensible suggestion of just temporarily closing the manual bottle valve.


We might as well add a check valve to the N2 solenoid as well, although there wasn't a functional need for it.  I have a couple coming with the plumbing order this week, so we should be able to use those.  The check valve on the H2O2 solenoid still doesn't have a spring in it at all.  It can't be very strong, or it will prevent the vacuum from drawing the peroxide through it, but we probably want at least a tiny amount of force to make sure it seats properly.


Our peroxide (water) transfer tank was just a jug with a siphon line dropped in it, but I want us to make a bottom-draining tank for it.


I still have some concern about the current arrangement leaving peroxide in the high pressure hose going to the quick connect.  It would be ideal if we could just be rid of the hose, and have all the fill valves plug right onto the quick connect, but that is probably too much shearing weight for it. If we need the hose, we should arrange for the valve manifold to be suspended above the quick connect, rather than below it.  We may want to have some kind of flexible supporting arm on the fill cart that can bear the weight while it is connected.  We need a good enough solution that will work for both the test stand and the launch pad.


The little solenoids get pretty hot while they are operating.  We may want to experiment with adding some resistors in line with them.  The peroxide solenoid won't have to open against any pressure, although it must stay closed when the fill manifold is pressurized.  It may not be an issue once we get a bottom draining tank and aren't screwing around with keeping the siphon line off the wall.


We have a pressure gauge visible on the flight tank when it is on the test stand, but we won't be able to see that when it is in a rocket.  We could add one to the fill system, or we could just rely on the regulator gauge from the N2 bottle.


We were still using the AC vacuum pump.  I mounted a switch on the fill controller for a pump control, but we still don't have a 12V solution yet.


The first time we pressurized the system, we were leaking all over the place. We weren't sure before, but it turns out that both types of valves really do need to be installed with high pressure on only one side.  They don't seal backwards.


The quick-disconnect works extremely well under high pressure, with no drips or spurts.


Phil suggested an additional safety switch for the N2 solenoid in addition to the momentary actuator.


The nylon fill line we had going into the water tank kept sucking itself to the wall during loading.  If we don't get a good bottom draining tank for peroxide / water loading, we should cut notches out of the end of the siphon line so it can't seal itself to a flat surface.


We pressurized to 400 psi.


We didn't have a switch box for the launch valve (which is intended to be a separate box from the fill cart control panel), so we just tapped a power jumper directly to the open leads.


We were all set for a water-rocket like jet out the engine nozzle, and were disappointed when the water mostly just flowed out the bottom of the nozzle.  We then remembered that the front catalyst retainer was fixed in place, so the water didn't have a direct spray exit.  If we want a jet next time, we will disconnect the line from the motor.


All in all, it was an extremely good night.


Here is the full fill procedure:


Check N2 bottle for sufficient pressure to fill the engine tank.


Set the vacuum / purge valve to open.

Nothing should happen, the port should have been left open.

If it had been left closed, there might be a venting of N2 or decomposed



Set the vacuum / purge valve to closed.

Just in case the engine tank was accidentally left pressurized.


Connect fill cart to engine fill port.


Set the vacuum / purge port to open.

If there was anything in the engine tank (there shouldn't be!), it will purge.


Set the launch valve to closed.

Needed before we fill the tank.  It should have been left open.


Start the vacuum pump.


Wait for the desired vacuum level.


Set the vacuum / purge valve to closed.


Stop the vacuum pump.


Hold the H2O2 valve open until it is all sucked into the flight tank.


Hold the N2 valve open until the flight tank is pressurized.


Disconnect the fill cart from the engine fill port.


Set the vacuum / purge valve to open, relieving the N2 pressure in the manifold and safing it for any H2O2 decomposition.

Up next:


We need to get all the improvements into the fill system, get it all secured into it's own cart, then practice and polish the filling checklist until we consistently do it without spilling a drop of water.  We also need to practice purges as well as engine firings.


We need to do peroxide compatibility checks.  We may have some synthetic rubber in the solenoids that need to be replaced.  We can run some lower concentration peroxide through the system a few times before trying the concentrated stuff.


We only have a small quantity of concentrated peroxide left, but we will probably run it through the engine to get a short, but gratifying blast of steam.


A bunch of Earl's plumbing fittings, hoses, and accessories will be here by next week.


A bunch of radio and telemetry stuff should be here by next week.


The heavier load cells are still on order.


The silver screens are still on order.


I have handed off the FMC 90% contact to Russ.  We may have a sire inspection before next week.


Russ or Phil should pick up some stabilized 70% for the hybrid test track, in case we have a delay with the 90%.


Did the safety equipment get ordered yet?  Before we touch even the lower concentration peroxide in the fill system, I want everyone wearing at least goggles and gloves.


I'll pick up another little project box and materials for our launch valve controller.  I'll go ahead and get a long spool of wire for it, so it will be appropriate for launching as well as static testing.  How far away do you think we should be for the launch?  I will try it with straight wiring first, but we might need a relay system if we have an extremely long run.


We could either make the launch valve controller self-contained with it's own batteries (probably one or two RC car batteries), or require the launch valve controller to be hooked up to the battery on the fill cart.  I hate having multiple batteries, but it is probably better to remove the temptation of opening the launch valve on the test stand while the fill cart is beside the engine.



© 2001-2011 Armadillo Aerospace, LLC. All rights reserved.