September 24 and 28, 2002 meeting notes
I made a separate aluminum plate to bolt the CPU board to,
and mounted it to the electronics box with PVC grommets. This should keep the board from flexing, and
provide some vibration and shock isolation.
The memory SIMM is probably still a weak point for us, but there isnt
much we can do with this board.
We now run power and ground to the CPU/PC104 stack through
three independent routes: two separate lines at the power connector, and also
through the PC104 pins at the top of the stack.
We built a manual cutoff box for the manned flights. The box has a switch and a self-contained 27
volt power supply, so if it is switched, no matter what the electronics box is doing,
the main engine valve will close in under a half second.
When it is in auto-land mode, the flight control software
now goes to full down throttle as soon as it detects a 2.5G acceleration, which
seems to be a reliable hit-the-ground determination. Previously, the bouncing accelerations would cause the computer
to hunt the throttle around trying to maintain a steady rate of descent, which
would cause it to bounce back up in the air a little bit. Forcing the throttle down as soon as impact
is detected reduced this quite a bit.
There is a related effect that happens at liftoff, where the
accelerations induced as the vehicle leaves the ground slightly unevenly make
it look like it is accelerating faster than it is, causing the computer to
throttle down earlier than it should, making it waste a bit of time bobbing on
the ground before finally lifting off.
I will probably do something to address this in the future.
Man in the air!
We were finally ready to let someone get on the
vehicle. We were at this point a year
ago with the previous generation (single tank, red tubes) vehicle, but
development always seemed to be pushing us to change and improve the lander in
some way, when we werent working on the other propulsion research. Our criteria was always that we had to have
two absolutely perfect ballasted flights with zero changes before a person
could ride. Usually, after two flights
we would either find a problem, or find something that we wanted to
improve. We finally reached a point
where everything is operating to our satisfaction. I was tempted to do something about the liftoff throttling, but
we decided that it is good enough as is.
Another factor that has kept us cautious is that Anna, my
wife, was our designated first passenger.
This was her rather crafty ploy to make sure that we pay a whole lot of
attention to safety. It would be one
thing for Russ to break a leg in an accident, it would be a completely
different thing to break one of Annas legs!
Although she had been suited up and ready to fly on several past
occasions that we canceled, Anna was scheduled to be out of town this
weekend. Since we seemed to be
operating responsibly, she ceded her position to Russ for the first flight,
rather than make us wait until she got back.
Anna prepping for the canceled flight last weekend:
We had an ambulance on site, just in case. This is surprisingly inexpensive, and should
be considered by anyone doing something potentially dangerous.
The vehicle is chained securely to the ground, with chain
wrapped around the main frame in two places, and connected to two big eye bolts
sunk in the concrete. Nylon rope is
wound through most of the chain links to act as a shock absorber, rather than
let the chain come up hard.
Russ wore goggles, ear plugs, and a filter mask underneath
the helmet, and had a hazmat suit on to protect against any peroxide
spray. We have a 3 energy absorbing
foam pad on the seat, and a 1 pad on the back.
Dry weight was 525 pounds, and we loaded 50 pounds of
peroxide. The flight was perfect. Auto-land was initiated six seconds after
throttle up. There was one bad sample
from the altimeter recorded, but it didnt have any impact. We could have had a couple more seconds of
air time at this load, but it was best to keep everything very conservative.
Russ reported that the flight was smooth, but the landing
thud was noticeable. I have been trying
to get a quote from Enidine for custom wound wire rope isolators to give us a
more cushy landing. If I had a cleaner
signal from the altimeter, I could have the flight control software touch down
almost arbitrarily softly, but right now there is enough uncertainty that
asking for a slower rate of descent would leave the total speed basically in
the noise margin.
A few seconds in the air a few feet off the ground
is not very significant by itself, but the point is that most of the systems
necessary for much more significant flights are demonstrated. Looked at in the best light, it is a
reusable, four axis stabilized, liquid fueled, manned rocket. Our single man suborbital space shot vehicle
is expected to only be about twice the dry mass of this vehicle (but with a LOT
more propellant, including kerosene), and will not be all that much more
It was tempting to fly more, but we are starting to hoard
our peroxide because we are down to our last three drums, and we dont have the
supply resolved yet. We will learn a
lot more flying the tube vehicle to a couple thousand feet than doing yet
another lander hop. When we get a good
supply in, we will do some full-load lander flights with the pilot doing the
Group photo, from left to right: Neil Milburn, Phil Eaton,
Russ Blink, Joseph LaGrave, John Carmack, Matt Ross
We finished all the plumbing and wiring for the high flow
regulated system that we will be testing with the 1,000 lbf regen biprop
We got the new fiberglass nose and tail cones for the tube
vehicle, and mounted the rocket tower to the top, which will pull the parachute
We got seven new 6 engine shells in. These have a slightly larger (2) throat,
and have a full 2 radius of curvature on the throat, instead of our previous
fairly sharp angle. We will be building
the next vehicle frame to use four large differentially throttled engines
instead of a single large engine with four pulsed attitude engines.
Our cavitating venturis from Fox Valve finally arrived,
after we had wrapped up most of our testing on the 2 motors. Sigh.
We may go ahead and do some tests with them on one of the remaining
aluminum chambers that need to be flood cooled.