Edge triggered timer

Thread Starter

poopscoop

Joined Dec 12, 2012
140
I'm building what is basically a skydiving micro controller, and I want a backup parachute release mechanism.

This is to release a chute on project, and in no way related to the safety of a person.

My intent is to reset a timer on an edge, produced by the uC. I've looked at missing pulse detectors, but the ones I see don't reset on the edge, but rather on the state.

I'm considering a 556, with one timer reset on the uC LOW state, and the other on the HIGH, so if it hangs in either state it would time out and deploy the chute. Obviously a self contained hardware solution is a requirement.

EDIT:::

Better description:
I seem to have expected you guys to read minds.

The arduino should handle chute deployment, however, being a low cost microprocessor there is a chance it will hang, or my code will have a bug. If this happens, I want a totally separate fail-safe chute deployment. The arduino will output a 'life message' in the form of a pin = !pin pulse every complete code cycle. If it hangs, this pulse will stop.

If the pulse stops, the timer, or whatever else I can use, will time out. When it time outs, it will override the electromagnet holding the chute in the stowed position, thus deploying the chute and saving my $80 worth of parts.

This will not solve every possible code problem, as the loop can continue to output an 'alive' message without ever executing my chute deployment routine. This can, and will, be remedied with a 555 calibrated for worst case drop time.

Ideally I can have multiple hardware fail-safe devices on a PCB. I'm thinking of the following implementation:
1. 556 or another device to read 'alive' pulse.
2. 555 time for worst case drop. Activated when the device detaches from the balloon.
3. 555 for emergency balloon release. Should the balloon break free of the line and start climbing, and the arduino fails, it will cut the device free.
4. Power supply short circuit protection. Should there be a wiring problem inside the device, the arduino could shutdown and the coils wont have the power to release from the balloon (Balloon fails connected, chute fails deployed. Lose power on ascent and I can't release the device). If the voltage across a series resistor reaches X value, one MOSFET disables the primary current path and another activates a secondary. Secondary path releases from balloon and deploys chute. Not sure on the implementation of this one, but I will probably mirror it after rudimentary power supply short circuit protection and possibly some logic.
 
Last edited:

Thread Starter

poopscoop

Joined Dec 12, 2012
140
I should mention that I'm already using an Arduino (have it on hand), and cost is a concern, as there's a chance it will smash into the ground.
 

AnalogKid

Joined Aug 1, 2013
10,986
Your description is a bit vague. The arduino produces an edge that resets the timer, and then the timer times out. Then what? What does this do, what is the output of the timer connected to, and where does a retriggerable circuit fit in?

ak
 

Thread Starter

poopscoop

Joined Dec 12, 2012
140
Arduino produces a pulse -> hangs -> timer times out -> mosfet goes LOW -> magnetic field decays -> chute deploys
 

AnalogKid

Joined Aug 1, 2013
10,986
One question would be why not implement the timer within the arduino code, but I'm still curious about the retriggerable. Also, from your description it sounds like the chute deploys when something is turned off (decays). Does this mean that the release mechanism has to be powered constantly to prevent it from releasing too early? Seems backwards...

ak
 

Thread Starter

poopscoop

Joined Dec 12, 2012
140
I seem to have expected you guys to read minds.

The arduino should handle chute deployment, however, being a low cost microprocessor there is a chance it will hang, or my code will have a bug. If this happens, I want a totally separate fail-safe chute deployment. The arduino will output a 'life message' in the form of a pin = !pin pulse every complete code cycle. If it hangs, this pulse will stop.

If the pulse stops, the timer, or whatever else I can use, will time out. When it time outs, it will override the electromagnet holding the chute in the stowed position, thus deploying the chute and saving my $80 worth of parts.

This will not solve every possible code problem, as the loop can continue to output an 'alive' message without ever executing my chute deployment routine. This can, and will, be remedied with a 555 calibrated for worst case drop time.

Ideally I can have multiple hardware fail-safe devices on a PCB. I'm thinking of the following implementation:
1. 556 or another device to read 'alive' pulse.
2. 555 time for worst case drop. Activated when the device detaches from the balloon.
3. 555 for emergency balloon release. Should the balloon break free of the line and start climbing, and the arduino fails, it will cut the device free.
4. Power supply short circuit protection. Should there be a wiring problem inside the device, the arduino could shutdown and the coils wont have the power to release from the balloon (Balloon fails connected, chute fails deployed. Lose power on ascent and I can't release the device). If the voltage across a series resistor reaches X value, one MOSFET disables the primary current path and another activates a secondary. Secondary path releases from balloon and deploys chute. Not sure on the implementation of this one, but I will probably mirror it after rudimentary power supply short circuit protection and possibly some logic.
 
Top