Trigger problem data log.

[qwkswede]

I am running a 36-1 Ford steel wheel with a Hamlin Electronics hall sensor reading the teeth.

Here is some info on the sensor I am using.
http://www.hamlin.com/product-detail.cfm?productid=102

The car runs nicely for about 5-10 minutes, then suddenly it starts dropping the trigger signal. My thoughts are that the sensor might be faulty?? Its strange that it begins so suddenly. I managed to capture a datalog last night. You can see the smooth rpm line, followed suddenly by a very erratic signal. Can someone help diagnose this?

[qwkswede]

Here is the complete Datalog.
http://www.denverspeed.com/Files/VEMS/v3.3_n002295-2010.04.16-20.42.45.vemslog

I just realized you can look at these trigger error logs. I don't know exactly what they mean, but it can't be good.

[mattias]

You are running it far too rich. Don't just go driving looking for trigger errors when your tune is far off.

Lambda <0.80  means mis-fire and can cause bursts of a lean lambda reading. Your ignition system can create serious issues with the electrical system in the car depending on how some things are wired, including making your trigger signal confused.

Your idle sites seem close to the truth. The rest need to be lowered and smoothed out. No rocky mountains going on down there. You're running a stock cammed 16V 4 cyl? The VE table should be pretty flat and boring..

This advice might not solve your problem, but it doesn't hurt to fix that. Check your EGO correction gauge reading and make corrections as you go, increasing load and rpm slowly.

[qwkswede]

I adjusted the sensor closer to the crank wheel and that seems to help the trigger problem. My starting map p was smooth but way richer than expected. I borrowed it from someone on the wiki.. I have only run the engine about 5 minutes so far. And started pushing the idle bins around to get my Afr closer.. But it hasn't even gone long enough to warm up the coolant yet. The VE numbers are alot different than I am used to in megasquirt. The bottom 2 rows were always 20-30 range there. I can only guess there must be some different fuel calculations. I am learning. 

I think I can go for another drive now. Maybe farther than wAlking distance this time. The trigger problem stranded me yesterday. . I'll reduce the fuel map and smooth first.

[tonci]

Did you solve your problem with bringing sensor closer to the wheel?
I have almost same problem. But it in my case it only happens on idle. Every 60-70 sec. rpm signal goes to 0 for a millisecond and than it goes back to normal. It usually resets warmup and lambda reading. I have tried different firmwares, configs and even vems boxes. I have connected new shielded cable from sensor to vems connector but it didn't help. Now I believe the problem is in vr sensor. Since the engine has trigger wheel on crank inside the block I can't bring sensor closer, so my only option is to try to replace the sensor. Do you guys know how can I recognize faulty VR sensor without oscilloscope?     

[qwkswede]

Yes, closing the gap helped clear up the rpm signal. I get a nice clean signal now, but the sensor is very close to the teeth. Maybe .5mm clearance. I had it close enough to rub on one try, so I backed it up just a tiny bit. It seems too close for safe running, but its working. I think my hall sensor is maybe a bit finicky with my teeth design.

On my past megasquirt setup, I ran a Cherry Hall sensor with an aluminum body, and it would run perfect with 2-3mm clearance.