Can someone test my MegaSquirt II 3.57?

[jangell]

I've run out of ideas about why my 3.0L MegaSquirt conversion won't run. Would someone here be able to test my ECU? The unit is a MegaSquirt II 3.57. I figured I'd mail it out to someone who has a working conversion, and they could flash it with their tune and see if it works in their car, then mail it back to me. I'd pay shipping both ways, of course. Although if anyone is in the souther Massachusetts area, I'd be happy to meet them for a test.

I just can't figure out what is wrong. I've documented what I've tried in this thread (http://dmctalk.org/showthread.php?16657-3-0L-Megasquirt-w-EDIS-Headers-running-at-different-temps/page13) (tunes attached to posts in the thread). It was actually idling properly. I then re-routed the harness a bit (which required lengthening a couple of wires), replaced a coolant hose and fixed two swapped coils, and now it won't start. It cranks, occasionally fires a cylinder, occasionally backfires, but won't actually start.

Of note:
- Fuel pressure is good (there's a gauge reading ~40 PSI in the engine bay). Fuel is fresh (a few weeks old).
- All plugs and coils are firing (I pulled each plug/coil pair and tested them separately while cranking).
- All fuel injectors are good (pulled them and put them in bottles, ran MS in test mode to check spray, replaced the one bad one I found).
- All coils are connected to the correct plug banks.
- VR sensor is correctly positioned 6 teeth before the missing tooth, and ~1 mm from the gear.
- Oil pressure is good.
- MAP reading is good.
- RPM reading from EDIS to MegaSquirt via TunerStudio looks good.
- TPS and temp sensors read correctly.
- Ignition timing to EDIS appears to be correct, other than the fact that the car doesn't seem to want to start.
- Removing the EDIS SAW signal from MegaSquirt has no effect.
- Changed batteries for good measure.

Most recently, I replaced the near 40 year old bulkhead connectors with a single modern 31 pin weatherpack bulkhead connector. I traced and tested every wire as I went, making sure they were good from the MegaSquirt DB37 connector, EDIS connector, fuses and other original connections in the car to their ends in the engine bay, and both crimped and soldered each wire before inserting its pin into the connector. Everything looks good and tests positive.

The most important note here is that the car was idling before. I just don't know what has changed to break it. Since I've tested everything else, I'm wondering if something has gone wrong with the MegaSquirt unit itself.

Thanks!

-- Joe

[dn010]

Did you try the jimstim or other simulator boards with your ECU since?

[jangell]

I do have a Stimulator. I got it so I could do tests with a computer-based instrument cluster without having to work inside a running car all the time. I played with it a bit a few weeks ago when I completely failed to get CAN Bus to work. In fact, I seem to get no useful signal from MS at all for CAN Bus, even with the right settings in TunerStudio, although my oscilloscope is older than the DeLorean and is a bit hard to read (really need to get a new one that doesn't weigh as much as a car battery...)

...of course, I brought MS home, but I left the serial to USB cable at the garage, so I can't actually look at anything in TunerStudio. I'll have to run down tomorrow morning and grab it. All I can do right now is watch the fuel pump light turn on and fuel injector lights flash as I change the RPM dial.

-- Joe

[opethmike]

I no longer have an EFI PRV in my car, but I do still have my 3.57 MSII ECU. I could lend you that if you'd like.

[jangell]

That wold be great! I really wanted to try that, but I didn’t want to deprive anyone of a working car while they mailed me their ECU. I’ll PM you my address, and if you want to PM me a PayPal address I can send you the shipping payment.

Thanks!

— Joe

[dn010]

Any updates?

[jangell]

Nothing useful, unfortunately. Mike's MegaSquirt behaved exactly like mine does, so that's not it.

I'm back to thinking it's an ignition issue, which seems to be the consensus anyway. I'm going to try a coil pack in place of the coil-on-plug setup. I've tried two different EDIS6 modules already, so it's probably not that, but since this all started when I was pulling on wires, it might be something related to the coil wiring on the engine side of the firewall.

Luckily, 3.0L plug wires are about $15 on eBay. I also picked up a coil pack and wiring pigtail, but the coil terminals are the wrong size for the 3.0L plugs, so I'm going to have to crimp new ones on (or get another coil pack).

Another detail I noticed is that the spark plugs fit much more snuggly in these new wires than they do in the coil-on-plug ones I'm using now. I'm not sure why the CoP ones are so bad, since they were made from 3.0L spark plug wires.

Anyway, I'm hoping to test this either this weekend or Thanksgiving weekend, depending on how much time I want to spend in the cold. :) My wife also just got a '72 Charger that we need to do some work on (new rear leaf springs/shocks, new belts, and the rear main seal seems to be leaking), which has distracted me a bit.

Thanks!

-- Joe

[Josh]

Be 100% certain the sheilded wire from the CPS to the EDIS is in good condition.

I made a disconnect in the wire, it was not a great idea. Whenever it rained i had to unplug it and plug it back in. That is why I cut it off when I shipped you the harness.

I would re-run that whole wire.

This has to be something minor.

[jangell]

Ah... I installed a nice quality headphone jack/plug pair on the wire again when I set it up. I never did find a great way to run it, but I intended to put it behind the bulkhead connector plate to keep it out of the weather. It has been low on my list of problems to worry about just yet. :)

I did a test a while back where I pulled the plugs (so there would be no compression) and cranked the engine with the fuel pump off, and I got a stable 200 RPM with my high-torque starter. This implies that the signal is solid, at least. While doing normal cranking the RPMs seem appropriately variable due to compression. The ignition errors seem consistent, with one cylinder in particular seeming to backfire, an occasional proper firing, but most of them don't want to fire at all, although the spark plugs themselves are sparking.

I am tempted to replace the entire run of shielded wire just to be sure, but I'll give the coil pack a go first and see what happens. I was considering redoing all the wiring to the coil-on-plugs as well. Again, all of this stuff worked back in August, so I'm really not sure what has changed that is causing all these problems.

Thanks!

-- Joe

[FABombjoy]

headphone jack/plug pair
:shock:

[jangell]

It's a shielded headphone jack/plug pair... :)

...I'll hardwire it.

-- Joe

[dn010]

https://www.diyautotune.com/product/wire-2-core-shielded-22-gauge-10-39/

run it from the sensor -> EDIS. Solder it. Cover it. Be done with it.

I did not make any sort of disconnect for mine.

[jangell]

I didn't get around to removing the disconnect on the CPS cable, but I did install the coil pack. I realized that for this test I didn't have to modify the wires it came with, as the plug ends are longer and thinner than the stock 3.0L ends -- I just snapped them right onto the plugs. I cut the four wires from EDIS to the coil-on-plugs and installed a 4-pin weather pack connector so I can swap pack setups more easily.

Longs story short, this is what I got:


http://www.youtube.com/watch?v=V0bNZTq2XtQ

It's idling at 1500 RPM, spits out lots of white smoke from the exhaust, and seems to take more cranks than the coil-on-plug setup did (around 10 vs. 2-3 with CoP), but it runs! And it runs smoothly (no swapped spark plug wires this time).

At this point I'm thinking that I pulled a wire in the coil-on-plug setup a bit too hard, but not enough to completely disconnect it. It may be that it's sending a lot of current down the few strands that are still holding it together, and this is confusing EDIS and keeping the CoP from working correctly. Googling suggested that EDIS is sensitive to such things, so it might well be the issue.


I'm deciding how to proceed from here. My options are to use this coil pack, get a different coil pack, or redo the CoP wiring to remove any breaks.

This coil pack's issues are figuring out how to mount the original 3.0L boots on to the 3000GT wires, or figuring out whatever terminals are on the other end of the 3000GT wires so I can put them on the 3.0L wires. I also have to figure out how to mount the coil pack. Getting another coil pack has similar issues with sourcing terminal ends and mounting.

Reworking CoP has the risk that it still won't work, but it's cheap (since I have everything already), and doesn't have any mounting issues. I would also butcher one of the new $15 sets 3.0L wires I got and install those boots on the coils, since they seem to hold spark plugs far more securely than the ones that originally came with the engine.

That's something for after Thanksgiving though.

Thanks again for the help everyone!

-- Joe