EDIS fails under load - loss of PIP and CTO

[sberry]

I'm 99% there on getting my Volvo B20 running with MS2 + EDIS.

I'm running a preassembled V3.57 MSII from DIYAutoTune that works perfectly in all respects. The trigger wheel and adapter are homebrew pieces I made myself and had trued up by a machinist - the trigger wheel portion is made from the thin universal wheels sold on DIYAutoTune. Photo of triggerwheel and sensor

The car drives great and ignition behaves well under almost all conditions. It will rev cleanly all the way up to 6500 rpm. I can cruise at 5000rpm and the ignition is rock solid. Unfortunately I am experiencing momentary loss of ignition under heavy loads. This is an inconsistent problem: I can drive the car very hard for a while before experiencing this. The problem happens often enough that I cannot track or autox the car and I've had one near-issue with the car dying at the end of an on-ramp acceleration that was a near disaster.

Symptoms:
- Happens while accelerating at or near WOT in 3rd or 4th gear
- Usually happens when the car is warm - especially after a heat soak in a parking lot.
- My Autometer tach (driven by EDIS CTO) will drop to zero and the engine will die.
- After a varying period of 2-5 seconds the engine will come back to life and the tach will resume working as if nothing had happened
- If I feather the pedal the engine will continue running but if I resume WOT it will fail again - it seems very sensitive for a period after the ignition cut-out
- In the datalog everything else appears to be normal - the ignition TACH input from the EDIS PIP signal just disappears as if the car has stalled (datalog and MSQ attached).
- The fact that both the TACH/PIP signal to MS /and/ the CTO signal driving my tachometer drop simultaneously leads me to believe the problem is in the EDIS system and not an issue with megasquirt
- I cannot reproduce this problem by wiggling any portion of the EDIS or MS harness while the car is idling.
- So far I've replaced VR sensor, EDIS unit, EDIS coils with good used Ford parts and this has had no effect on the problem.

In the attached datalog you can see the issue pretty clearly between the 808 and 816 second mark. The ignition TACH goes away, then it suddenly comes back and quickly dies again.

The fact that the engine runs reliably at all rpm except under heavy loads is confusing to me.

Any of these symptoms ring a bell for anyone?

Thanks in advance,
Steve Berry
Austin, TX

[trakkies]

Have you also checked the basics - is the EDIS unit still getting 12v and a good ground when the engine cuts?

[sberry]

The only logging ability I have is what I can see from MS - BATT voltage seems to be normal for what that is worth.

EDIS unit is grounded to the engine. Coil is grounded to the chassis. I'll run a secondary ground for both coil and edis just to totally eliminate them for my next test.

Power I am pulling from a switched 12V source. I'm not sure what else is on that circuit - I'll get a better grip on that.

Tomorrow I can isolate EDIS power feed and make sure it's on a switched 15 amp fuse all by itself. Should I instead just wire up a relay or a fused line direct from the battery?

Any other advice for ruling out power supply or grounding as an issue?

Thanks for responding. This is a tough problem to diagnose. Every time I make a change it takes about 20-30 min. of driving to reproduce.

[trakkies]

EDIS, of course, should work totally independent of MS, although with fixed timing, and your diagnosis that it is failing does seem right. Does the EDIS unit get very hot when it fails? I dunno if this would cause it to shut down, though. If you're certain all the EDIS parts are good, the only other option is a wiring fault of some sort. As regards the supply to it, you could rig up a test lamp or LED across the feed and see if that remains on when the fault occurs, rather than changing anything on spec.
How is it wired? It's common to have used connectors hacked off the original loom, so lots of splices to your loom. Are you certain these are perfect? Are the connectors good - do the actual terminals in them look undamaged? Intermittent faults are the hardest to find.

[sberry]

I moved the EDIS power to share the same circuit as megasquirt. My thinking was that if I was datalogging and EDIS lost power then megasquirt would as well and it would show up in the log - proving that the EDIS issue involved loss of 12V power.

I drove it as hard as I could up and down a fast, hilly road near my house for 30 minutes. 80 degrees outside. Over and over full throttle pulls from 1st - 4th, shifting between 5k and 6k. I shut it off an let it heat soak for 5 minutes and then resumed. I could not duplicate the cut-out condition.

I'm hoping it's that simple. I'll be able to put some miles on it this weekend and try it out at an autocross.

Thanks for the advice. I'll post back to this thread if I run into any more issues.

[trakkies]

You may be ok using the same supply as MS for the ignition - but it's generally safer to use a separate one to keep noise down. But this should show on a log if it causes a problem.