(discussion moved from the TRA list & some private emails)
A bit of background - I'm an EE and work in oilfield automation and safety engineering, I have some (dated) background in microcontrollers. I'm also a CAR RSO. I used Adrian's magnetic switches last year on my N1800 L3 project. They work well, though we've found that they tend to be a bit sensitive to stray magnetic fields - got to be careful where the cell phone, radio, pocket with magnet, etc. are in relation the sensor. I understand the concern about potentially launching a rocket with the recovery system turned off, but I think the case of a rocket that won't fly but has armed ejection circuits is also worth consideration. In Canada we require a sterile ballistic recovery zone so odds of anybody getting hurt by a ballistic recovery are low, but we often have incidents where a rocket fails to fly and has to be lowered off the launcher - I consider the danger of accidental deployment during this process a much higher risk since people are close by. Obviously if the rocket has accessible switches there should be no problem, but there is the potential for this to be an issue. Personally I love the idea of remote arming and disarming, beats climbing a rickety ladder and eliminates access hatches / holes. I think that allowing bi-directional comms prior to launch detection would be a valid approach IF a launch detect routine could interrupt a radio subroutine. It could also automatically force the TeleMetrum into flight mode, and the listen mode could be disabled. My question is how well does the processor multi-task? How much room is there for more code, and how fast does it all run? Does the launch detection subroutine run on a timed interrupt basis? The radio routines take a long time to execute due to the need to transmit and receive data at a much lower data rate, but most of the CPU time would be spend in wait cycles as the data is buffered through. Ideally they are not exclusive program calls that block other routines from running, but maybe I'm assuming too much from this particular processor? Regards, Eric -----Original Message----- From: Bdale Garbee [mailto:[email protected]] Sent: Tuesday, February 07, 2012 22:45 To: Eric Weder Subject: Re: FW: [CRA] CRA Twitter Update On Tue, 7 Feb 2012 14:33:23 -0700, "Eric Weder" <[email protected]> wrote: > Once the rocket is set up and TeleMetrum is armed in launch mode, what > happens if you have to take the rocket down? Easy case is that it has > an external power switch, but what if it is not accessible and you > have to leave the altimeter armed while lowering the rocket? Once a TeleMetrum is in launch-detect mode, there is no way to take it back out over the radio. There are two reasons for this. One is that if we happened to detect launch while the radio is in receive mode, we'd lose the ability to collect telemetry briefly until the radio got turned around to transmit. The second is that several early customers were adamant that there be no way to accidentally take a unit out of flight mode over the radio once igniters are installed in a rocket on the rail. So, we strongly recommend having an externally-accessible power switch. Taking a rocket down from vertical with the electronics still armed is not a good idea. Adrian's magnetic power switches from Featherweight Altimeters can be used with TeleMetrum boards, I'm happy to provide wiring details if mechanical switches bother you and you want to go that route instead. Bdale
_______________________________________________ altusmetrum mailing list [email protected] http://lists.gag.com/mailman/listinfo/altusmetrum
