Kevin wrote:
>>>>>
You're close to describing ours. What we did is have an aluminum rod that sticks out of the booster, into an aluminum block inside the main airframe. The fit is a bit loose, and there's an o-ring on the rod, to seal it. We drop a #10 nylon bolt down through the block and rod, to hold things in place.
At the bottom of the well that the rod fits into is a very small BP charge and an electric match, wired to the sequencer which handles igniting the boosters and firing their separation charges.
<<<<<
Yes, I think yours must be one of those I was thinking of, details were hazy. Plus, somewhere, a couple of years back, there was a guy in Germany working up a large shuttle, who did a pyro-type method too but I cannot recall exactly what he was doing for the mechanism.
>>>>>>
One concern we had is what would happen if a booster's recovery system deployed when the booster was still connected to the main body of the rocket.
To prevent this, we have a separate timer inside of each booster. The timer is connected via wires to the sequencer, for a simple "I'm here!" circuit. The timer does not begin its countdown until the booster separates from the main body, at which point the connection breaks, and it loses communication with the sequencer.
<<<<<<<
Sort of what I have with the shuttle SRB’s. For limited mass purposes, I went with converted wind-up Tomy Toy type timers, and a mechanical system so when a retaining disc rotated to open up a slot, it would allow a rubber-band driven dowel to push the nose off and let the chutes out.
As with your “I’m here” sensors, the wind-up timers have a rod pushed in by the ET to keep the timers from running. When the SRB seps, the stop-rod is no longer pressed, so that lets the wind-up timers start to rotate their discs. About 2 seconds later, they pop the chutes out.
This was about 10.5 years ago, there was not an electrical timer and battery system light enough (in the 5-6 gram range) that I would have trusted, though in 2000 I came across an ideal timer that could run off a 1.5v watch battery and fire a very low current e-match, but it was too late by then.
If I was doing any similar project like that now, I would definitely use an electronic timer, since there are few light and reliable enough now.
Although, I’d be really careful on the “I’m here” sensor switches since a false trigger can be so bad (at least with the wind-up timer, you could hear it start to run and then at worst I’d have to repack the chutes. No "Bang" of accidental pyrotechnic ejection). Add an arming switch for each, of course, with a remove before flight type of pin. But I’d probably use two switches, in parallel. When I did the shuttle model, I had an “I'm here” switch for the orbiter, on the right aft ET to orbiter attach point. But the micro switch was so tiny, and its level motion so limited I was concerned it might false trigger open. So, I added another switch to the left ET to orbiter attach point too. So, in parallel, both would have to detect the orbiter gone before the flight computer (in the ET nose) would run the “Orbiter has sepped” part of the program (the Orbiter was sepped by me manually by R/C, usually waiting till after the SRB’s were auto-sepped by the computer, but on a couple of bad boosts I sepped the orbiter first to save the orbiter and start an “abort” over-ride sequence in the flight computer to sep the SRB’s and fire ejection). That was important since sensing orbiter sep caused a 1 second countdown to firing the ejection charge in the ET (There were two other sensors switches in the ET, an on-the-pad sensor, to avoid false triggering in case for example the orbiter came off by accident, and a “thrust” switch pressed only when the engine was thrusting, to detect a truly valid liftoff.
>>>>>
One additional benefit to this is that when the rocket is on the pad, we can hook a laptop to the sequencer, and check the status of the entire system -- we can make sure the sequencer is talking with all of the timers, that the timers have continuity on their electric matches, that the separation matches have continuity, and that the booster igniters have continuity.
<<<<<<
Also some parallels with what I was doing with the shuttle. The “Flight Computer” was by Jay Marsh, made of two Basic Stamp PIC’s. He had developed this for his FAI Saturn-IB models, which eventually was 3-staged, and had R/C override to either fire stages early if it got into trouble early while pointing up, or stop the firing of stages if it get into too much trouble, otherwise the computer ran staging on auto. For my shuttle, he made a physically similar version, but with a lot of different code to do what was needed. If the computer was “Happy” it would put out one single beep every 2 or 3 seconds. If something went wrong it would do a lot of quick beeps as an alert, then settle into beeping out a number which was a code for the problem. Such as liftoff sensed without thrust sensed. Or orbiter sep sensed without liftoff. Or continuity bad for the two ejection charges. Any show-stopper condition sensed. Things like that. Of course, it had no idea of the status of the “dumb SRB’s” the shuttle model used, since it did not need to. Once it commanded the servo to sep the SRB's, they sepped, or if not, the comptuer coudl do no more about it anyway (they never failed to sep once bench-tests de-bugged the latch system of the boilerplate).
- George Gassaway