[kickit2]

First off - background on what it is I'm working on. Controlling an engine. Its a standard lawn mower type, except that it has an electric kill switch. The motor runs when its left to float, but when connected to ground it kills the motor. Simple enough.

I want to do two things with this wire:

1. Use a depletion mode FET of some type to control the run / stop function of the motor. I want depletion mode so that, incase of power failure to the control unit, the motor is killed (as another circuit will control throttle, etc - better shutdown then damaged).
2. When the FET is driven into depletion, and the motor allowed to run - pickup the inductive spikes and count them to find RPM.

I have connected my scope to the engine and found that RPM counting is very possible (at least via the scope with human intelligence). The spikes happen once per revolution, so that aspect of things is very easy. Also, it seems that a load of 100ohms or less is needed to actually kill it, so using this as a source shouldn't be an issue (assuming my input impedance is sufficiently high of course). The spikes given are ~130V peak on top of a 12V DC bias and happen 40-60 / sec. They last ~1.4ms in duration each (see scope captures).

So here are my questions.

1. How do I clean this spiking up into a reliable single for an MCU to count - and for that matter at a safe voltage for the MCU?
2. To what degree do I have to take the spiking into account for fet selection? I would prefer 0V GVS on (of course - delpetion here) and GVS Off to be a positive voltage in relation to the source - which will be ground (just for simplicity in circuit design - at least IMO), so I'm looking P channel most likely. Most FETs that I've looked at (MOSFET and JFET) have a max VDS of 40V, and to get larger than that, I seem to be almost exclusivly dealing with devices that want a negative gate voltage. Being that they are so short in duration - can the voltage of the spike be negated to a degree, and if so - to what degree?

Scope is AC coupled and averaged in these captures

 

[kickit2]

Disregard - DPST relays, a voltage divider, and a single-shot setup. Problems solved. Why can't I delete this thread?