Random RPM spike?
 

So I got my car running fairly well at this point, needs fine tuning, but easily drivable without issues.

I noticed today that while I had cruise control on and driving down the road, the tach would blip instantaneously. It did this about two or three times while I was logging. It's not really consistant, but after looking at the log file, in MS it shows an RPM spike to about 7900 RPM (from the cruising 3100 RPM).

Any clue as to what is causing this? I noticed my battery voltage is fairly noise but the sensors don't seem to be presenting any random noise (that I can see). If anyone can look at my log file and tell me if they see anything wrong (such as noise or whatnot) I would be extremely appreciative.

Thanks!!

BattV doesn't look too bad to me. That noise is only 100-200mv, which is about the same as I see on my car.

The one-sample RPM spike however does look a lot like a fault on the CAS input. If I were to ask you "Have you done the cap mod?" Would your response likely be "yes", "no", or "WTF is the cap mod?"

Yes I have the cap mod. I used a very small ceramic disc capacitor though with +/-20% tolerance, so I may replace it with a better one and see what happens.

hi-z or lo-z injectors? i had rpm signal issues a while back, caused by flyback voltage spikes from lo-z injectors.

They are Lo-Z but with inline resistors.

Well I replaced the capacitor with a higher tolerance one and the situation improved I believe (possibly a placebo effect). I noticed it still has these small spikes every so often, usually only until the engine fully warms up from what I can tell.

Is my only solution to build Abe's input circuit or replace the CAS wires with shielded ones?

Any fresh ideas?

I never could completely get rid of the glitches on mine. Built three totally different input circuits, threw bigger and bigger filters at it, wound up with more signal latency than the software could compensate for, and still had the glitches.

Threw the CAS in the trash, and built a 36-1 crankwheel setup.

Seriously, it seems that there's just something about the CAS such that some of 'em can't be made to work right with a Megasquirt. I guess the stock ECU is coded for some software debounce or other reality checking that MS just doesn't support, as there have been a couple of people whose CASs worked fine on the stock ECU, yet couldn't be brought under control with MS. The solution was to replace the CAS.

Just throwing the idea out there. If you can lay your hands on another CAS to test with, it'd be worthwhile.

Yeah, replacing the CAS is the general consensus I have read to fix this issue. I guess some cars got lucky where some didn't. My 97 has no issues like this, it appears the majority of cars struck with this is is 94/95. Oh well...

Are your CAS wires running near the ignition high voltage wires?