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

Sorry for not reading your top post re ignition drivers, but one of the tips they give you when MicroSquirting is test with a stim, or make a bench model. Live testing on a bike introduces too many variables, and I would recommend you make a test rig. The other thing is try and grab a datalog.

The difference in spacings (time interval) is due to compression slowing the cranking speed, AFAIK, and the moving ignition point I would expect is the MicroSquirt seeing a small ripple in the missing tooth phase, aka "noise".

FWIW, I don't understand why you have such peak to peak variation on your VR sensor trace with your wheel design, it almost looks like the gap is moving.

PS My cranking pulse tolerance is 100% not 75%, works OK me.

Thanks again for your help man. I agree it does seem like there is alot of variance in my VR output. I did extend my cranking pulse tolerance to 95% with no real improvement in reading rpm. *I measured VR peaks from 38ms to 19ms* I will look into a stim and building a small rig, I thought the microsquirt would be more plug and play than the megasquirt... I see why the megasquirt requires the stim.

Here is a data log cranking with cranking pulse tolerance set to 150%. (I get the same results below with 100%)

RPM is coming on and off and reading double the actual cranking speed before it stops reading speed altogether.



The spark advance goes to zero when it receives rpm and back to 26degbtdc when it isn't, does this seem right? I removed one of my spark plugs and it arcs when cranking.

I noticed the spark plug was dry though and the injectors aren't firing using a current probe. I also don't hear the injector solenoids clicking with the injector test??? There is continuity from the AMP harness PIN 9 to Cyl 1&2 inj connectors and from PIN 10 to Cyl 3&4.. There is 12V supplied to all of them as well. They are set up for high impedance measuring 14.6Ohms. Any help with this problem as well?

The datalog and msq files are attached below if you need to see more.

Those teeth don't look very evenly spaced to me. Are they all the same size or does the wheel have some that are bigger than others?

It's the stock Yamaha trigger wheel all evenly spaced and the same size only with 2 missing now.

Here is a picture of it with one missing:

I think you have made things worse by making 2 missing teeth on an 8 tooth wheel. It's ok to have a 36 or 60 - 2 wheel, but 8-2 doesn't buy you anything. You have 2 delay teeth. Tooth 1 is the tooth right after the missing section. Delay 0 means that tooth(1) is your first tach tooth. You have Delay = 2, so you start tach at tooth 3. Then you have no_skip_teeth = 4, so your next tach is tooth 7. But you can't have a tach tooth in the middle of the missing teeth gap. The code wasn't designed to handle this.

If you don't have a spare wheel, then you need to use delay_teeth = 1 and move the sensor or change the trigger offset to compensate.

Very good point Grippo.

I thought 1 missing tooth would be enough but it was hard to see with all the variance in the other VR signals:


I set delay=1 and the trigger offset to 45deg. It gives a longer cranking rpm about 4sec before doubling and cutting out. Cranking pulse tolerance is set at 100%.

Get a new pulse gen wheel, I had to swap my 12-3 for a 12-1 because the gap was too big, but mine is a V-Twin and compression was playing hell. The ECU reject the next pulse because the engine speed changed too much during the gap and make no sense.
You can download the tooth analyser and burn it in your ECU, that tells how the processor see the pulses, mine were very unevenly spaced during cranking


Here's the scope at high engine speed

Man, getting this engine just to start has been killing me these last few months. I went from a 8-1 crank wheel "M-N Crank Wheel w/ No Cam Synch" to a 8-2 which seemed better at finding the missing tooth at the time. Put the harness on the JimStim and it runs fine with a 12-1 trigger except at higher rpm >3000rpm the signal started to cut out on the scope?!? I ignored that for now bc I just want to get the engine running first. I then switched the 8-2 for a 16-1 wheel, which actually started the engine finally but was almost impossible to gain synch during cranking (I just got lucky once or twice out of 40 tries with a 100% cranking pulse tolerance)..

I then tried going "Normal Crank Trigger w/ Cam Synch" with a 2 tooth wheel (formerly the 8-2), using one tooth for each ign event. I wired the existing Cam Hall Effect sensor, supplying 5V, wiring the output into VRIN2+ (4) and grounded to shield (31)of VR crank sensor. Cam Pulses look fine to me on the scope: (A lot better than the crank sensor).. Any ideas for the crank sensor signal conditioning please???



I then downloaded the tooth analyzer and flashed the appropriate software into my us. Is it normal for the tooth analyzer to take extra time after cranking to process the plots. It ends up being very zoomed in, but you can't zoom out?



The crank VR sensor is false triggering a whole bunch of times, any ideas? It looks like its picking up the noise seen on the scope trace above.

Obviously 2 teeth did not register well. I returned to my 16-1 wheel but with falling cam sync.

I found out that I've been struggling most of this time with bad wires off the VR sensor, when the trigger cover is torqued on us loses VR sync sporadically leave it loose and cranking stays steady. The tooth analyzer looks much better as well:





I encountered an odd issue with WUE going on and off constantly once it the engine started causing the engine speed to surge. Has anyone encountered this? I will have to grab a better log to make sure coolant temp doesn't do anything strange once it starts. It's usually constant.