AtomX on Suzuki F6A - Trigger issues on cranking

[cappo_nz]

I've got a Suzuki Cappuccino with the stock F6A and AtomX. It has a 12-1 trigger wheel on the crankshaft, reluctor sensor mounted with a <1mm air gap to the sensor. Trigger 2 is the standard cam position sensor with 3 teeth, 1 per TDC, so I am using cam pulse window for sync, the window being 40 degrees before tooth 6.

It used to start second or third try, which I put down to bad cold start tuning. In theory, nothing at all has changed since then.

I now have a no start fault where I cannot make the car fire on the starter. It will bump start fine. Cranking on the starter, the trigger scope shows trigger 1 status as "test gap" and changing to "verify gap" on each trigger 2 pulse. Unfortunately being a new, but not new enough AtomX (V2.4) I don't have a voltage trigger scope. Any ideas as to what I can change or fiddle with to improve this? It's been stuck in my driveway all week..

Cappuccino no start 2.llgx

[Adamw]

Your trigger scope does not show 12 teeth.  Is the 12-1 wheel custom or factory?  That could possibly be that arming threshold too high so some teeth are not reaching the threshold and therefore are ignored.

1mm air gap is a lot for a reluctor - 0.3mm is more typical.  What specific sensor do you have and what are the trigger wheel teeth like?

[cappo_nz]

The 12-1 wheel is custom, normally the engine uses the 3 pointed cam wheel and a distributor. I've tried with arming thresholds down to 0.2v. The trigger wheel is currently square teeth, and uses a Subaru EJ sensor. Should I recut the trigger wheel to use sawteeth?

[Adamw]

For a VR sensor the length of the tooth should be about the same as the diameter of the pole inside the sensor.  Often this is about 4mm.  If the pole is not externally visible (they often poke out the end) then you can get a feel for the diameter of the pole by dragging a paperclip across the end of the sensor and the magnetic bit is the pole.  But the length of the teeth I dont think would stop it from running - it will more likely just stretch out the wave form and give you poor accuracy as per the pic below.  

So I still suspect it is more voltage related.  Maybe there is some wheel runout or something.  Can you reduce air gap to more like 0.5mm? I have shaved material off the plastic sensor mounting flange before to reduce air gap.