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

hi there,

i need some help here , ive been trying to get spark control for my motorcycle using the bosch igniter for the last month but i still cant get any consistent spark when cranking.

Sometimes it sparks at the right time , sometimes twice in row not at the right time , sometimes doesnt spark at all for a couple of turns, etc.....

im using microsquirt v3.83 on a 4 cylinder even fire wasted spark config with 1 VR sensor on each end of the crankcase.
Crank has only one hole to trigger on each end with 180 deg apart so VR1 triggers coil 1-4 and VR2 triggers coil 2-3.

I ve been using dual tach input with auto trigger enabled and 0 deg offset since engine is even-fire and set trigger wheel teeth to 0.
i ve set dwell time to 8ms to make sure it sparks and i ve read somewhere that it should be set fairly high for my kind of coils like 6ms or more...

When cranking i have the sync stat = 3 , which is "too many teeth before missing tooth (last)".

It does trigger and spark but never at a consistent rate and sometimes offset a lot.
So i tried many things like increasing timing mask , percentage mask and next pulse tolerance which helped a bit but didnt solved the issue.
I tried all cranking trigger methods : calculated , trigger rise , trigger return but still didnt work.
Spark table has been set to 0 so that i could see the TDC mark spot on.
im pretty sure my trigger offset is set correctly cause from time to time it sparks right on timing mark for a couple of turns.

I know a have a mismatch teeth issue because the holes on the crank used to trigger are a lot bigger than the VR sensors themselves but still doesnt explain the effects i am seeing with my timing light.

I tried the timing light with the stock TCI unit on when cranking and it works very well, spot on the mark.

i used the tachRef.exe to check what microsquirt sees when i crank and i got this: Here s my MSQ if anyone wants to take a look at it: On my oscilloscope the signal looks like so:
Im stuck here for now ...

Any help would be appreciated !!!

Do you have a photo of what the VR's are attempting to register with? I'm not quite sure I understand but is it a single "hole" on the crank that is sliding by the VR? On each side of the crankcase?

Heres a pic i made this is the left side of the crank and crankcase.

The right side is the same setup but crank hole is 180 deg apart from the one on the left side.

I'm no authority on VR sensors and not entirely sure I understand your set up - in other words, is the VR registering a single "hole" in the crank rotating past it? If so, I would reiterate what I have come to read here - that being that the more actual teeth on a tone wheel the better sync and trigger behaviour. If it is a hole and not a "tooth", not entirely sure how well it can work. The width of an actual tooth form and the width of the VR tip are critical and when mismatched do not work reliably.
Perhaps others will weigh in.

yes thats what happening.

Vr1 register a single hole on the crank left side to trigger coil 1-4

and vr2 register a single hole on the crank right side to trigger coil 2-3

its not a teeth but a rather big hole very mismatched with the vr sensor tip , which i knew it could cause some reliably issues but my stock TCI unit

works very accurate with those holes , so i thought it shouldnt be so unstable if i tried getting spark control with MS.

The problem is that the VR circuits used by the MS family are all zero-crossing triggers, and while they are extremely accurate with closely-matched tooth / sensor combinations, they are notoriously unstable with mismatches; the bigger the mismatch, the greater the instability.

For the Yam TCI box to have stable, accurate spark control, it would have to use an edge-detecting circuit. Rising or falling, metal approaching or falling away should all work, but one aspect might be significantly better than the others, and the only way to determine that is the old-fashioned way - test runs. Such a circuit would have to be fabbed up and installed between each of the stock VR sensors and the uS VR inputs. The output of the edge circuits would be a square wave, so they would be treated as Hall sensors - connect to VR1- and VR2-, leave VR1+ and VR2+ floating.

hi there,

thanks for the help!

i connected VR1- and VR2- and left VR1+ and VR2+ floating like you said,

i connected VR sensors connector back to the TCI unit or else i had 0 rpm on MS.

I get very stable rpm now !!! I ve never seen my rpm that clean and stable!

I had to change the cranking trigger method to trigger Rise to have a very stable spark on coil 1-4,

but i have no spark on coil 2-3 i dont know why , coil used to spark before.

Looking at a data log , i noticed 3 things :

-there s 0 trigger count.
-Sync stat = 3 is still there.
-cranking state is not showing true when i crank but sparks anyways on coil 1-4.

Any ideas of what i can try next?

I'm not familiar enough with the B&G code; I'm more able to comment intelligently on the Extra code. It strikes me the offset for #2 cyl should be 180, not 0. I have no idea why you're getting a "too many teeth before missing tooth" error when you have NO missing teeth!

i did another tachRef and here s what i got:
i dont understand why there isnt any offset from Vr2.

How exactly do you have the 4 wires from your two VR sensors wired?