MicroSquirt.com read only forum (Jan.15,2018)

Did a bit more testing on this today.

Opened the airgap up to 0.028" and reset my HYST pot about 6 turns CW from full adjustment. Would crank & fire fine. Took it to a friends place and it wouldn't start, LUCKY i had the laptop with me, no RPM signal. Set it back to full adjustment, then 3.5 turns out, fired first go!...

I'm still getting the odd stumble here/there, but I feel without a scope i'm going to chase my tail around in circles.

Would it be worth while soldering in a 220nf cap between the pickup wires at the distributor end to help filter any spark chatter under the cap perhaps?

My next test is to try driving the coil directly and bypass MSD.

kris, can you log some of the stuttering?

see if it's truly a tach spike or dropout (it will be visible on the RPM line in the datalogs), or you might have accell enrichments, or ?? plenty of other things going on.

-scott

That I can do/have done. You can't 'feel' the spikes in the car but the drops are plain obvious, especially when you're under any sort of power! (going up a hill?)

See records:-

5356 (drop)
7132 (drop)
7840 (spike)
7949 (spike)
7972 (spike)
9267 (drop)
9404 (spike)
9433 (drop)
9514 (spike)
10282 (drop)

Trust me, this is a good day

I also messed around with the normal running pulse tolerance. The higher it was the more stumbles I got. The lower it was the better it was! If I got the normal running NPT below about 10%, the car would not fire up again regardless of what changes I made to after start %'s. I remember reading somewhere that the afterstart%'s relate to the ASE and since ASE doesn't kick in once the motor is warm, does this defunct the NPT?

OK... I think I have hit jackpot.

I scoped my Q22 transistor and found it wasn't switching pin2 of the OPAMP to ground properly. Might have damaged it from to much solder heat. I swapped it out with a local BC640 transistor, 80V 1000mA, it was as close as I could get. Initial bench testing is showing me this is firing at <80RPM !!!

On the scope though it's showing me I need to capture on the rising edge though which I thought VR's capture on falling edge?

Output on the VB921 looks perfect now too with max dwell of 25.5 (firing MSD).