Home / News

News Archive

May 15, 2001 Meeting Notes

May 15, 2001 Meeting Notes


In attendance:


John Carmack

Phil Eaton

Russ Blink


New supplies:


½” check valve

another small 12V polypropylene solenoid

More 1/4m to 1/8m fittings

Ceramic scissors

Wire tie assortment

Hose clamp assortment

-6 hose ends

Various AN fittings


On order:


Two liter flasks

½” servo ball valve


To get:


New electronics box

Waterproof connectors and switches

Water hose reel

-10 hose, hose ends, and fittings for running the ½” valve on the test stand





We made a few plumbing improvements today, putting the aluminum –6 hose ends on the test stand hardware, and converting to a –4 swivel fitting on the fill cart from a –3.


We hopped the lander with three 800 ml peroxide loads, but we didn’t have as good of results today due to high winds.


Russ is going to be out of town, so next week’s meeting is pushed to Wednesday unless someone needs it to be Thursday.


First load:






380 psi, which was only just enough to pick it up with basically full throttle. Performance seemed down a bit from the last runs.


You can see the strong wind blow away our leveling pads of foam after liftoff.


Shows the need for active roll control.


Second run ended when the sensors saw a huge rate jerk right as the last of the peroxide was expended.


Making max corrections, and not making much headway. Probably fighting the wind.


In the future, the camera operator should stand off to the side so that all four engines are visible, and the fore/aft rotation can be seen more clearly.



Second load:







Raised pressure to our full 430 psi. The altitude is still easily controllable. Now that the latencies are down, throttle response really isn’t all that much different than in the simulator.


It tips over a fair amount and comes down pretty fast, but the landing gear works perfectly!


The second run is making its maximum correction at every opportunity, but it can’t get it to turn.


The third run displayed the same rate jerk as the last of the peroxide was expended.




Third load:






On the first hop I made some fairly strong manual input on the joystick to try to correct for the wind, and while it probably helped, the attitude correction couldn’t force it all the way over to where I was designating.


The second hop also leaned over fairly quickly.


We noticed a fair amount of leakage out of the engine top closures.





The system did not have enough control authority to deal with the winds. Most of the runs had it doing as many attitude corrections as it was allowed in the same direction, and still slowly tipping away.


On Saturday when it was calm, each attitude correction pulse gave around 15 degrees / second of rate change. For today, I had added one more minimum frame of delay between attitude corrections to account more for the low sensor bandwidth, which reduced the maximum corrections to once every four frames from once every three frames, which


We had some leakage out of the top closures on most of the engines, which may have been skewing the relative performance between pairs.


Engine zero (left side) also had a couple odd things happen. When we first started the water test, it didn’t fire. It turned out that the connector had been bent a bit when we were connecting a manually switch to that solenoid last test. We closed up the connector a bit, and it worked after that. The other odd thing is that the solid state relay driver board shows a faint light in the activity LED even when there isn’t a signal. There might be some kind of a voltage leak, which might make the valve close slower than expected.


The large sensed rates at the time the peroxide runs out are interesting. It seems to be a hard jerk which is shocking the sensors, because it definitely isn’t actually getting to the rates that it indicates. I’m not clear why the final depletion of the peroxide, even if it was uneven between engines, should be a stronger jerk than the normal attitude correction pulses.



Attitude Sensing


While foam mounting the electronics box seems to have solved the jerk behavior of the gyration gyros, the 10hz sensor bandwidth is proving to be an even bigger issue than expected. We really want a sensor that will show the full updated rate within 33 msec of the completion of an attitude adjustment pulse, which seems to translate into a 60hz sensor bandwidth.


While looking for a distributor of BASE silicon gyros, I came across this company:


I have requested information on their three-axis gyro that has 70hz bandwidth and no listed G limit, as well as the BASE parts.


If nothing works out in the next couple weeks, I am going to buy two of the $1500 KVH fiber optic gyros.





We have an electrically actuated ball valve on order from this company:



We will need to drill the ball vent ourselves, but otherwise it should suit our needs. It is a ½” valve, so it should be good for 600+ pounds of thrust as a central lifting engine for the manned vehicle.


Control is by connecting 12VDC to two wires, reversing the polarity to reverse the valve direction. Feedback is by a potentiometer on the motor shaft. We should be able to run this with two solid-state relays and some diodes, and read it with one single ended A/D channel.


I had been looking for a valve that just took a signal level and opened to that level, but now that I think about it, having binary directional control with feedback is actually a lot better, because it removes a level of black-box computer control in the valve, and lets us actually measure operating response instead of trying to infer it. It is also nice to be able to avoid having a D/A control signal.


With this resolved, the propulsion system for the manned vehicle is now clear: a central lifting engine run by the ½” servo valve, with four outrigger engines run by the medium sized (10 amp, not 30 amp) NOS solenoids. I propose that the outside engines be positioned exactly the way they are on the demonstrator, so that they provide lift as well as attitude correction, which will make motion of the CG (the pilot) a lot less critical. Those solenoids will flow around 50 pounds of thrust at 400 psi, so the platform should be able to lift off and hover without the central engine firing at all, when there is no pilot. The central engine should probably be sized around 600 pounds of thrust.


This basic engine configuration, with higher and higher performing central lift engines, should take us all the way through the vertical dragster and to the space shot vehicle.




The propellant tank is now the number one question for the manned vehicle. One path would have us use 150 psi tanks slightly over spec, starting at around 200 psi. If we target this pressure instead of the 400-450 psi we have been doing all of our work at, we would need to more than double the engine dimensions. Isp would also suffer a fair amount, but that isn’t a critical issue.


I inquired at Lincoln composites about using their NGV division tanks. They would be massively over-built for our needs (3000+ psi operating pressure), but they are built to survive car wrecks, so they would be very safe to use. The response I got was “probably not a qood idea”, but they were vague about “potential contaminants”. If the materials are all good, I may get one for us to work with anyway.


I have a couple other inquiries out for tankage for the manned vehicle. Our desired pressure (500 psi) is in sort of a no-mans land for commercial tanks. There are lots of good 150 psi composite tanks for water processing, and there are lots of 3000 psi tanks for NGV, scuba, compressed gas, etc, but not much in between.


Neil: at one of the very first meetings we had, you had dug out a big NGV tank. At the time, it was far larger than what we wanted, but it might be exactly what we need now. Fifty pounds of tank isn’t that big of a deal when we have a pilot dominating the dry mass.


Our ideal tank has pipe fittings on both ends made of aluminum or stainless, holds between 10 and 20 gallons, has a working pressure above 500 psi, and has a polyethylene liner, or a stainless / aluminum core.



Stuff To Do




Sensor board microcontroller rework: 60 hz synchronous, read axis in reverse order, read accelerometer (After reviewing all the data again, I don’t think we will gain anything by going to 120 hz, because none of our problems are related to integration)

Drill out the engine nozzles to 0.275” from 0.25”

Make another small engine retainer plate so we can run the test stand without stealing it from the VTVL.

Test drill ball vent on our old valve

Make a new custom tank manifold for ½” plumbing (might want to wait on seeing the big tank)




Nitrogen refill

Pick up the rest of our plated foam

Pick up tables

Tool chest and organizing bins

Follow up on the NPT blow off valves




Big tank

1000 psi nitrogen regulator

6” of steel hard line for pressure transducer isolation (we can probably just chain some fittings together if

you don’t find anything handy)




Update 3D simulator to use the new control laws, tank blow down, and the measured angular rates

Investigate the dim signal light on the driver board

Investigate the strange warmup behavior with manual joystick control

Investigate the improper angle exit with joystick input before I commented that check out

Investigate the lost pilot packet exit

Follow up on gyros and tanks.

Gyro integration initialization by accelerometer gravity vector

Instant graphing software for guidance logs


Test stand runs:


Log tank pressure blow down with water, just use the small solenoid with no engine attached.

Run the small engine for 30+ seconds

Run the small engine with the insulated chamber pressure transducer working

Run the small engine with different PWM attitude correction trains to watch recovery times

Run the small engine with the plastic solenoid at 125 psi (I have seen mixed data on the peroxide compatibility of polypropylene)

Run Juan’s engine at 150 psi with the big piloted solenoid

Run Juan’s engine with the super big shot solenoid and –6 plumbing




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