Megasquirt and EDIS troubleshooting

This is my Fiat. For being the cheapest car you could buy in america 41 years ago it is awesome. It is so awesome that it single handedly chased off all the fleshy headed mutants from sector 16B of the forbidden zone. It is so gnarly that girls get sterile just looking at.......insert more 80's movie references as you see fit.

But when it is not awesome it is frustrating.

It is running MS2 on a Ver.3 board. Firmware 2.88 and using an EDIS4 setup from a mid 90's escort. TBI injection from a Yamaha Raptor 700 quad with an extra injector stuffed in for good measure.

The problem is that it completely randomly decides to misfire. 7 starts out of 10 it is so utterly awesome that it intimidates the other larger cars on the road. It builds boost quickly and linearly, but every now and then it fires up and misses. It has a low load miss. say cruising rpm drawing a fair amount of vacuum it sounds like an STI with a header. Boost builds slowly and if I put my foot into it the miss will go away, boost will come up to about 12psi and just when you think things might be going well it will pop the BOV and buck.

Now here is the funny bit. If I turn the car off and right back on again while sitting at a traffic light the problem will often go away. This issue will never appear if the car is running well and not restarted. It is only on startup. I can drive for hours with no issue, stop for gas and it will run like poo when I leave. It isn't a hot start cold start issue either. Sometimes it acts up first thing in the morning at 35 degrees, other times when it is hot and restarted.

I have run new grounds, run new power wires and even replaced the relay that powers the EDIS brain. Everything that should be twisted or shielded is.

I beseech thee oh helpful GRM'ers. What say you?

It's trying to tell you that it wants to be sold to me at a challenge friendly price.

I've had this issue with the Wreck Racing miata (running EDIS 8). There were some poorly connected ground wires that were causing the problem, as I recall. You probably know MS is very sensitive to the grounds.

We've also had a fair number of EDIS modules go bad or have intermittent problems at mid to high load and rpm, but that doesn't sound like your issue.

Please tell me there's a build log for this on here somewhere That would be beauty, eh!

Readers ride is here

http://grassrootsmotorsports.com/reader-rides/820/

Full build thread is here http://retrorides.proboards.com/index.cgi?board=readersrides&action=display&thread=70073

I have gone through the grounds a million and one times. The MS ground is a 12ga wire soldered to the relay board and is grounded via a ring terminal to a valve cover stud. The EDIS ground is exactly the same, just grounded to a different valve cover stud.

Looking at this diagram

I see there is a 25uf cap grounded from the coil plug pin 2. I was told I didn't need that when I did the install so I left it out. Problem? The Escort I removed it from did not have it either.

What's coil plug pin 2 do, is that 12VDC+? Don't have a diagram here, can you explain?

Dont know why that diagram isn't showing up

Or

This one says radio suppression on the cap.

http://www.seattlerobotics.org/encoder/jun97/basics.html

Seems like adding that cap could be useful in keeping the EDIS module happy, or at least increase chances of it staying happy. Really nothing to lose by adding it.

I have it in place on my EDIS setup. I figured since ford put it there I should have it. It is supposed to be for radio interference and I have been told that it is not needed. I could see it playing a part in grounding the EDIS unit. Put it in it can not hurt.

The cap is between the power feed and a power ground; co-locate the cap as near as possible to the module with a good clean ground; its function is to drain AC ripple off of the DC power wire feeding 12V+ into the module. Integrated circuits like this sort of device work off of DC power, which is sort of easily available in a car, just with noise attached. Alternating current noise will ride the DC line and cause problems for the circuitry and this cap's job is to pull that noise off and shunt it to ground, leaving cleaner DC power behind.

Check out the link I posted, should help show what I'm talking about.

This post edited for clarity.

Mine does the same thing. Out of the blue, runs like it lost a cylinder or two. Quick restart usually fixes it, but no apparent "cause." Same code and EDIS. I'm very curious what the solution is. I've checked all my grounds too, AND I'm still using the cap.

It is nice to know others share my frustration.

Googling shows a lot of other folks fighting the same problems and llike me, just when they think they have it licked it comes around again......as is the nature with intermittent issues.

There is a set of LS1 coils in my tool box waiting to go on but I really don't want to forfeit this fight.

Of the folks sharing this issue, how many are running the same firmware?

In general for this sort of thing it'd be among the last places I'd look, but the "reboot-and-its-okay" aspect of it has that tang of software finding a corner case and getting stuck...

Occam's razor says it's more likely some condition EDIS doesn't like.

MS is also affected by chassis ground to some extent by sensors, charging system etc... I assume there's a beefy and well-connected ground strap between engine and chassis? How about the alternator case? Is it on rubber bushings?

I ran into this with the SRT4 swapped Neon. I ended up wiping the chip and re-installing the firmware to fix it.

I initially ran B+G 2.88 and it did it.

Last spring I nuked it and ran 2.90 (? I think) and it still does it.

I've checked my grounds thoroughly to no avail.

Someone posted about a similar issue over at www.locostusa.com, and found a fix.

The solution:

"Noise Filtering - EDIS" box under Basic Settings

Reset "next pulse tolerance" normal running up to 70% (from 25%)

SkinnyG wrote: Someone posted about a similar issue over at www.locostusa.com, and found a fix. The solution: "Noise Filtering - EDIS" box under Basic Settings Reset "next pulse tolerance" normal running up to 70% (from 25%)

I WILL TRY THIS ASAP! I still haven't fixed this. With the major upheaval my life took I stopped driving the car for a while and when I do I just cycle the power until the problem goes away.

Could you post a data log of the issue and a copy of your tune file? That could give us a bit more to go on.

Same problem, different system. Ignition switch was going. Gave a shaky connection to power for the system.

We really need a log file and a copy of your MSQ.

Your pip and saw connections are running through shielded cable?

How old / good are the high tension leads? water with a little vinegar in a spray bottle at night will quickly show any "leaks" in the high tension ignition cables . If they are parking under the hood and the pip and saw wires are near them things go bad fast. A wiggle here or there or just the relative humidity can changed things enough to make the problem come and go.

How do you know it is not a fuel issue? If you disconnect the pip and saw to the edis it will default to 10 deg (I think) Can you get it to repeat the problem with the pip and saw disconnected?

Wait what gap are you running on the plugs and are you using resistor plugs?. For that matter what brand plugs are you using?

I am using NGK resistor plugs and have swapped plug wires in the course of troubleshooting. I doubt bad plug wires would clear up after cycling the power though. I don't have the computer that I did when I started this thread so I don't have access to the log file I made. It should still be here though

http://msefi.com/viewtopic.php?f=89&t=40644&p=239668&hilit=fiat+edis#p239668

Not too many 40 year old engines run well at 16:1 AFR...

That should only be from fuel is cut on decel. I did turn that off in troubleshooting and it changed nothing. Where I see the misfiring the AFR is between 12 and 14:1.

Disconnect the SAW wire at the MS end.
Depending on your wiring, could be as easy unplugging the cable from EDIS to Megasquirt, and unpinning the wire from the connector. (note the MS 'plug and socket' suggestion)

The EDIS will fall back to 'limp mode', ie timing will be fixed at 10 degrees BTDC (or whatever is set by the position of the VR sensor relative to the crank timing wheel)

Now drive the car, and see if misfire still occurs.

If it stiil misfires, it has nothing to do with MS, it's on the EDIS side, otherwise, it's in MS or the EDIS-MS wiring.

The SAW (Spark Angle Word) allows the ECU to set the EDIS timing, dwell, and rev limit by the length of control pulses. If noise and/or other signals somehow occur on this line, the EDIS will interpret them as commands, (and then troubleshooting hilarity ensues !).

REV LIMIT, as in deliberate ignition misfire over specified RPM?

Of course, power off/on resets the EDIS back to default.

Edit: Might be better disconnecting just the SAW, since MS will likely use the PIP signal to help control fuel injection.

Carter