Is there any way to determine that the ECU is going bad prior to it dying completely? Car's been shutting off randomly while driving - pretty much just like how it was a couple weeks/months prior to last ECU dying. Replaced in 2004 with a used one off a member on here.

I guess it wouldn't hurt to just buy a spare ECU so I don't get stranded anywhere, but it'd be nice if there was a way to diagnose a dying ECU.

Is it completely shutting the car down, or just for a fraction of a second while driving? If its idleing and it happens it will probably stall. I was having the same problem, never actually figured out what was wrong, just went to MS.

Originally posted by MCubed45
Is there any way to determine that the ECU is going bad prior to it dying completely? Car's been shutting off randomly while driving - pretty much just like how it was a couple weeks/months prior to last ECU dying. Replaced in 2004 with a used one off a member on here. I guess it wouldn't hurt to just buy a spare ECU so I don't get stranded anywhere, but it'd be nice if there was a way to diagnose a dying ECU.

If it's resetting, shutting down while driving, starting hard while hot or not starting at all and displaying no CEL for any of the symptoms, chances are it's going bad.

when it just resets while driving, the CEL disappears for a couple minutes (mine is normally always on b/c they didn't reflash when they did the secondary air recall thingy). sometimes it completely shuts down while driving too tho. pretty much the same things it was doing when the last ecu died. blah.

Thats weird, I usually had better luck with one that wasnt reflashed compared to one they did.

Heres the real irony. The only ECU I ever had to never show any signs of failing on me, was for an auto car. I could only rev to 6400 before it would rev limit, but it never failed.

Originally posted by 740 turbo brick
Thats weird, I usually had better luck with one that wasnt reflashed compared to one they did. Heres the real irony. The only ECU I ever had to never show any signs of failing on me, was for an auto car. I could only rev to 6400 before it would rev limit, but it never failed.

lol wow. i think i'll still to MTX ecu's nonetheless. 9 yrs on the original and 6yrs on a used ECU isn't too bad. kinda worried that most of the ecu's people have for sale now are gonna be that much closer to dying tho.. :(

I do remember seeing rebuilt ECU's available recently. IIRC they were $150+core

That's how my last 95 MT ECU died. It would stop providing the 9V power for the crank and cam sensor. You could use an adjustable voltage regulator to provide this power to the sensors. I did this and use weather-pack connectors to choose between the ECU or my circuit providing the sensor power.

Corbin
'95 ESI-T

Now with more power and fewer leaks



Humble servant of the Wiki

Originally posted by Corbin
That's how my last 95 MT ECU died. It would stop providing the 9V power for the crank and cam sensor. You could use an adjustable voltage regulator to provide this power to the sensors. I did this and use weather-pack connectors to choose between the ECU or my circuit providing the sensor power.

I'd like to see that. It sounds absolutely genius

Originally posted by RoninEclipse2G
I'd like to see that. It sounds absolutely genius

There isn't much to see. I just added a daughter board to my MS case and put a couple voltage regulators on it, heat sunk into the upper case. The adjustable regulator puts out 9V for the cam and crank sensors. The fixed regulator puts out 5V for the MAP, TPS, etc. I actually have connectors on the fuel injectors, coil pack, and IAC so I can switch back to stock ECU control for emissions testing.

As a side note, this failure mode is only one of several that could occur. My first 95 AT ECU lost its upstream O2 input. That took me forever to figure out why the car kept fouling out and running on 2 cylinders.

Corbin
'95 ESI-T

Now with more power and fewer leaks



Humble servant of the Wiki

question:

i've been trying to find a used mtx 95 ecu and one of the sellers offered me one that is 95-96 compatible. i thought none of the ecu's are interchangeable? he said late 95 and 96's are the the same. is this correct? i'm pretty sure mine is an early 95 anyways, but just wanted to verify if i need to specify early vs late 95 mtx when looking for an ecu... thanks

96 ecu's have the two harness plugs and won't work on a 95. You need to find out if the ECU has one harness plug or two

Originally posted by RoninEclipse2G
96 ecu's have the two harness plugs and won't work on a 95. You need to find out if the ECU has one harness plug or two

ok that's what i thought...

do you happen to know if the plug layout changed with the model year or if it was one of those odd mid-year changes?

tia

IIRC it was a model year change.

Originally posted by RoninEclipse2G
IIRC it was a model year change.

hmm ok well mreigle is really confusing me. guess i'll wait to see pics or for him to tell me how many plugs his ecu has. he says it's a late 95 that's compatible with 96's...

K he said it's a single plug, so I should be good?

This is the PNC he gave:
#05269730

But i'm not sure how to use it to verify it'll work for me..

Originally posted by Corbin
That's how my last 95 MT ECU died. It would stop providing the 9V power for the crank and cam sensor. You could use an adjustable voltage regulator to provide this power to the sensors. I did this and use weather-pack connectors to choose between the ECU or my circuit providing the sensor power.

seriously? is that the only thing that goes wrong when the ECU dies?

Just to add some more data points to this thread. I have had two more '95 ECUs lose their front O2 input. I found this out when I swapped control back to the stock ECU to pass emissions. The ECU would pull the output of the front O2 sensor down to 0.135-0.155V. The ECU interprets this as a lean condition and will add fuel using the short term and long term fuel trim. The car won't run so well 50% rich. The ECU I was using and the used replacement both suffered from the same failure mode. I first looked at what the ECU thought the O2 output was by using a datalogger (0.135-0.155V). I then stripped the insulation from the O2 sensor output wire and put a voltmeter on it (0.135-0.155V). I then cut the O2 sensor output wire. Put a voltmeter on the sensor side and I get a cycling output from 0V to 1V. Yay. Put a voltmeter on the ECU side and I get 0.135V. Just to make sure it was the ECU, I unplugged the ECU connector and checked the resistance from the O2 sensor output pin at the sensor connector to the matching pin on the ECU connection. It was significantly less than 1 ohm, so I called that good. I put in a Chrysler remanufactured ECU and everything works fine.

Corbin
'95 ESI-T

Now with more power and fewer leaks



Humble servant of the Wiki

My 95 ecu was doing the same thing. Just randomly it would shut off while idling, or start to sputter whenever the hell it felt like it. eventually it got to the point where it didn't even start. Luckily it happened at the shop where I work. I found that the asd relay and the fuel pump relays were not closing and i replaced them less than a year ago. I checked both relays and they work. The ecu was not sending the ground to close the relays. I jumped the relays to see if it would start but it wasn't pulsing the ground to the coil. The ground control citcuit inside crapped out. Got a refurbished one for $200 plus the old core with a year warranty. Plugged it in and it started up like a champ. Also it was not setting any codes to help me diagnose it.

Sorry about bringing up an old thread, but I just remembered that Malachi was having some of the same issues that are mentioned here right before the accident. Basically the engine stuttering around 2500-3500 rpms and occasionally the random "shutdown" while driving where the engine will buck for a split second, and the CEL will kick off for a moment. No codes for a misfire or anything other than my missing EGR.

I'm expecting to need to buy a new ECU, but I have some concerns over buying an original used one of course. Has anyone used the remanufactured ECUs with decent luck? I've seen used Mopar remans floating around between $100 and $200 and aftermarket remans floating around for $200-300.

Any suggestions? :shrug

Edit: This is the one I'm looking at pretty seriously
CARDONE 79-3319
I have checked the part numbers for my production date so it should be correct.

I have a reman cardone ecu that I originally picked up at AutoZone probably 4-5 years ago. Still works. Currently piggy-backed with MS.

Cool. Took me a few tries to find a place that had one, but I got one on order. Should be here before I get the car back from the body shop.

FUCK!

The place I ordered it from just emailed me saying there was an issue and they couldn't get another one.

At least they refunded my money right away.

EDIT: Has anyone dealt with this company?

Triple post!

That company I linked in the last post has a TON of negative feedback once you do a little searching online. I did some research, and the Mitsu factory reman ECU isn't that expensive, and it's still available. My cost through a friend at the local dealer is only $230.

Originally posted by RoninEclipse2G
Triple post! That company I linked in the last post has a TON of negative feedback once you do a little searching online. I did some research, and the Mitsu factory reman ECU isn't that expensive, and it's still available. My cost through a friend at the local dealer is only $230.

did you try the buy/sell forum already? i got both of my previous ones from there for reasonable prices and both sellers had multiple unit available. course there's not really any warranty then and it's kind've a gamble with how much time the ecu's got left.

Originally posted by MCubed45
there's not really any warranty then and it's kind've a gamble with how much time the ecu's got left.

That's why I'm looking at new. My ECU has lasted longer than most at this point.

I'm still sitting on a spare 95 ECU because I bought a reman from an electronics place on ebay two years ago thinking that was my problem. Turned out to be my crank angle sensor...again.