Intermittent false high RPM during cranking

[mpcapps]

I've had my megasquirt II V3 board installed in my '78 Datsun 280Z for years and now it's developed a problem.

Sometimes during cranking, it shows very high RPMs (like in the 10k range) and the car won't start. It just cranks and cranks. I keep cycling it off completely and starting it until randomly the problem stops and then the car starts.

My setup is very basic. RPM signal is directly off the (-) end of the coil, controller is running fuel only. Once the car has started, there doesn't appear to be anything wrong when looking at datalogs or anything. I haven't added or changed anything to the car in a while so i'm assuming there's a component failing; i'm just not sure what one.

It's strange that the RPM reading could be so out of whack for a bit and then be completely fine for extended drives. It seems to happen less when the car is warmed up but it's not a great correlation.

a datalog is attached.

[mpcapps]

Intermittent problems are hard to troubleshoot, i know. The car started on the first crank this morning.

My plan going forward is:

-Replace Alternator (lights dim a bit at idle so it probably wearing out, it's 12 years old, it's internally regulated)
-Adjust RPM Lag factor (i worry that this will work well and mask the component failure). I had it at 90 and moved it to 70 - not enough starts to decide if it's improving it.
-Test/Replace both condensers (car is electronic ignition but it has a condenser on the alternator and another between the coil and distributor)
-Get an oscilloscope (i have one but it's loaned out) and monitor the ignition signal right off the coil during startups, try to verify noisy coil signal.

From there I'll look into the battery (during cranking the voltage got pretty low, under 10), and possibly adding some noise supression to the circuit somewhere.

Any other suggestions?

thanks,
Capps

[Matt Cramer]

Another thing you can try is a 20 to 24 volt zener diode inline with the wire to the coil, banded end on the coil side.

[DonTZ125]

This little circuit has a few fans on the MicroSquirt and Extra forums; it gives a nice, clean 12v square-pulse to the Opto circuit. It avoids monkeying around with Zeners etc.

engine_speed_trigger.jpg

If you're running a single coil, then obviously you only need the one diode.

Actually, this could be incorporated into the V3.0 board rather easily; replace D1 with something a little more rugged - I'd suggest at least a 1N4002, if not a 4004 - and flip it around so it's pointing at the Opto pad. Solder in a 1k pullup resistor between D1 and D2 to S12.

V3 Opto.jpg

[mpcapps]

Thanks for the suggestions guys.

The problem has gone away (for now). Car is starting quickly on the first crank attempt warm or cold, and is generally being awesome as usual.

The only thing i've done is adjust the RPM lag factor (it's at 70) and literally drive the car.

I have condensers and a new alternator but haven't put them in yet, and i'll probably replace my battery since it seems to be cranking weaker than it used to.

I'm guessing that the battery being low may be a factor and that driving the car has kept its charge up. condensers could be a problem too since AM radio has significant RPM noise. haven't had time to really test the components since i'm currently redoing the suspension on my subaru and running out of time before the snow...

will do more work when i have my other car back on the road.