losing EDIS spark (and PIP signal) intermittently

[tonkatoy]

I'm going to apologize in advance for the long post, but hopefully someone will recognize these symptoms and be able to offer some guidance.

I've searched the forums all day today to try and find someone who has had the same symptoms I'm experiencing, but nothing seems to match up well.

I am running a MSII board version 3.01 with firmware version 2.36. It was built per the manual for EDIS wasted spark. The EDIS system was constructed with the module, VR sensor, coil pack and wiring from a 4.0 liter V-6 Ford Explorer. I used all the factory wiring but altered the lengths to fit my installation by disassembling the VR sensor plug and the coil pack plug, shortening and re-soldering the wires at that point. The module plug was unaltered and spark plug wires were made by installing the Ford coil pack ends onto 8 mm Chev plug wires. I machined the crank trigger wheel from scratch. It is 1/4" thick mild steel with a square tooth profile measuring .283" wide for the tooth, .283" wide for the space between teeth and .250" depth of the space between teeth.

The problem I seem to be experiencing is a loss of coil triggering by the EDIS module, but ONLY on restarts after the initial engine start. I can start the engine, but it dies quickly which I kind of expect since I haven't done anything except take a calculated guess at the tables. Then, when I attempt to restart the engine right after it dies, I get only the initial 2 second energizing of the fuel pump relay and priming pulse. The injector LED on the MegaSquirt does NOT blink, the fuel pump does not run and the injectors do not squirt. One time the injector LED started flashing after a long crank (>15 seconds), the injectors started squirting and the engine fired up--and then, of course, died. I can't remember if there was a key on/off cycle between the engine stalling and that attempt or not. It seems that key-on/off cycling does not change things, but waiting for a minute or two before attempting a start does.

I have checked for high tension discharge from the coil by hooking a timing light to the spark plug wires and it correlates to the injector LED (e.g. no LED flash, no timing light strobe), and have swapped EDIS modules (no change) My suspicion lies with the VR sensor or trigger wheel. I plan to swap to a different VR sensor and adjust the gap from .030" to .045" and .060" to see if there is any change. I don't have (and wouldn't know how to use) a scope so I really don't know what the signal to the module looks like. I do know that I used the factory shielded VR wiring, only shortened as I said earlier and it runs isolated to the module except for the MAT and coolant sensor leads for half the distance.

After I had machined the trigger wheel and mounted I believe I read on the forums that ideally, the tooth tip should be the same width as the diameter of the pole piece in the VR sensor--but now I can't seem to find the thread that talked about it. Could the fact that my tooth width is nearly 4 times the diameter of my VR pole piece have anything to do with my problem? Anything else anyone else can think of that I should check? ANY suggestions or ideas are welcome and appreciated!

[PSIG]

I think your wheel dimensions will be okay. You need to determine where the failure is - MS or EDIS or the wiring. I would disconnect the SAW wire to let EDIS run on its own default timing and try. While trying and if you get no activity from MS, probe the PIP wire to see if EDIS is sending a signal, to verify whether EDIS is funked or MS is deaf. Once you know what end it's on, the diagnosis will get simpler. BTW - I use large straight pins or T-pins (hobby and craft stores) to probe between the wire and connector weather seal to get readings. It 's a simple way to test factory sensors, connections and other components without damage to insulation or weather seals.

If you haven't seen it, this page gives more MS related EDIS info that may be educational - but the settings and such are for different code.

David

[Heribert]

Hi there,
there is a fair amount of diagnostics built into most if not all EDIS units.
One feature is monitoring spark voltage by monitoring the primary pulse , somehow.
I do not know HOW that functions , only that if the resistance in the HT part ( 2 plugs and two HT wires)
drops to below some kiloohms, there may be problems. Some EDIS 6 has the primary coil charging current
available for scoping on pin 4, and this is what I monitored.
So use resistor spark plugs and/or resistor leads , and keep each plug/wire sum > 5 kiloohms.
Also, make sure that you use the filter cap directly on the coil +supply lead, 1 to 20µF works well. The ground lead needs
to be short, very short.

Good luck, to me EDIS has always been totally trouble free.

It may be a good idea to have a 50/50 rate on tooth/gap, but it works otherwise too.
On my bench test I have a wheel with 280mm diameter where the tooth to gap is 20/80 % . The wheel is made
in 3 mm mild steel, Never a problem.
PM your email, and I can send pics of the test wheel.


Heribert

[tonkatoy]

Thanks for the input, guys...much appreciated!
I'll hold back on re-profiling the crank trigger until I try the suggestions you have offered. I'll report back with success, (hopefully) or more data to troubleshoot. I'm jealous of those who are working through tuning issues--that's what I want to be thinking about!
Since I'm so early in the process, I plan to download the newest version of the firmware to MS, even though I doubt it would have any effect on my current issue. Might as well be up to date with the latest and greatest...

[tonkatoy]

Thanks to those who offered help on this problem. Ultimately, I did re-profile the crank trigger but it didn't seem to change things. I have moved forward and installed the engine in the truck, merging the wiring harnesses and permanently mounting the MS-II. For reasons I can't explain, the problem has not manifested itself while in the vehicle. I can't think of anything I did differently when wiring it in the truck, except that maybe some wierd ground bias existed in the test stand harness setup, or possibly some kind of voltage issue to the EDIS module due to wiring, or possibly the keyswitch I was using. I wish I could explain it, but I'm just happy it is working when it really counts!