GVR-4 Cyl 2 and 3 misfire, may be ecu?

[Dante923]

Hey, recently got my car tuned not too long ago and it was running perfectly after and the next day i started having minor fuel cut problems. They started to get worse so i took action and changed my spark plugs and fuel filter... didnt help. The problem then got to the point where it lost all power stalled at idle. A CLE came on for ignition coil so i took it to a shop because i didnt have time to look at it and they said it was bad and the power transistor unit was also bad. They were going to change it and before they did they checked the voltage going to the it and they said it was lower than it should be and that cyl 2 and 3 were misfiring. They suggested that it could be the ecu but they didnt want to get further into it because there was "so much aftermarket". I figured they would say this so now im back to where i started with a little more info. Also, they said that the fuel pressure was fine which suprised me. I read on the forums that somebody had a similar problem where the car would run fine on cold starts (mine does too) and then when its warmed up, it ran like crap like mine. He said the problem was the fuel pump. I have an exhaust gas temp gauge and it is located right off of the second cyl and when the car is running like shit it reads 200 deg but if i hold the rpms steady, it climbs to 1000-1400 deg( where it is when it runs fine) and the car runs fine again until higher rpms when it has fuel cut problems again. The tachometer bounces all over the place when it runs bad but it levels out when i keep the throttle in the same spot. When i hold the rpms and the EGT guage goes up and i floor it, it quickly drop down again. Not sure what it could be but i think im getting close. it seems like it all points to the ecu but i cant be sure. Any ideas?

 

[Pupdeck]

There is a possibility it could be the ecu. I would certainly do a compression test on all cylinders, and check the spark plug wires. Could potentially be something with the cam angle sensor, a lot of times the wires fray coming out of thE sensor or just a bad sensor. But that is a few things it could be.

 

[Firegst]

If they tested the coils and said they were bad, why not just replace the coils then go from there.

 

[Rene1998gsx]

new coils would be good idea u will need new ones reardless right?

 

[Dante923]

he was going to replace the coils but he said the voltage to the coil was low and it would just ruin the new one too

 

[steve]

The voltage to the coils is fixed by the battery/alternator voltage and condition of the wiring.

The ground for the coils is generated by the power transistor, and the quality of it's ground. Assuming the Power Transistor is receiving a strong enough signal from the ECU to drive the PT into saturation, it should be able to pull the coil primary very close to 0v to charge the coil and then quickly switch off to fire the coil.

1G ECUs are well known for capacitor leakage causing damage. If your ECU hasn't had the original capacitors changed and any damage repaired by someone who knows what they are doing and cares about making them last, it could be the source.

The CAS also has a role here since it is the source of the engine timing information to the ECU. The 91+ style CAS isn't a likely as the 90 style to have old broken wires coming out of it but by now the oil seals are likely hard and cracked and could be leaking internally.

Start by checking the ECU.
You can check the voltage to the coils yourself and measure the primary and secondary resistances. You should do this before replacing them.
You can check the output of the CAS. Read up on how to remove it from the head and turn it by hand while measuring the output signals.

I can't read posts without some spacing anymore so there may be much more to look at.

 

[boost97gst]

I have this intermittent failure. Very weird. Car will go lean as hell, cel comes on for cyl 3 misfire. If i cycle power it works fine. So its def related to a fuel injector. You may want to look into this as well. if im not mistaken the actual injector drivers are located in the ecu. Im not too happy cause this started happening immediately after i sent it off to ecmtuning for a socket conversion. Could be wiring though, never rule that out.

 

[Dante923]

Steve, so ## saying it could start working again because when i hold the throttle steady it charges the coil enough to fire it, making it seem like its firing? I have seen posts about bad ecus and it doesnt smell or click but it does seem like it is related to the ecu, maybe the capacitor recently broke and now im starting to see symptoms. Guess im gonna have to pull the ecu and check for corrosion. Ill try doin that tomorrow thx.

 

[1996Eclipse420a]

I can be driving my car, and randomly it loses power for like half a second, but instantly regains it. I've tried data logging it, but when the "jerk" happens, it shuts off the data logger. The radio and the other electronics don't shut off though, so I don't think the ECU is the problem for me.... I'm about to start a build and will have pistons, and rods in the next couple of days... so hopefully I can find out what the issue is during that time. I kind of think I have a bad ground wire somewhere, or a wire is grounding out somewhere it shouldn't be... It doesn't happen in a sequence, completely random. It's driving me crazy.

 

[Dante923]

Took out the ecu and checked it out. Nothing stood out but heres the pics. The capacitors appear to be replaced because of the color of the solder. Im no ecu guru but it looks ok to me.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[steve]

You're correct, the two capacitors that usually leak have been replaced and it looks like before they did any damage. The ECU has also been socketed, in this case so that the E943 GVR4 EPROM could be installed.

You didn't include a picture of the white connector but I expect it doesn't have 7378E or 7377E printed on it. What part number is printed on the edge connector and top case?

The configuration resistors are incorrect for a GVR4, it looks like this started life as a DSM Federal AWD ECU. The car should have a MD165810 ECU. I haven't looked at the GRV4 code to see that it does when misconfigured. It may not make a difference but you should have 1k resistors in both R129 and R130 for a MD165810.

Overall the ECU very clean but not all electronic problems are plainly visible so it's not 100% ruled out just not clearly bad. If you have a Friend with a good running 91+ 1G trying your ECU in it to see if it misfires with your ECU would tell for sure.

Lacking that assume it works and continue investigating.

I struggled through your first post and it raises the question of what is the electrical system voltage when it starts acting up. Try to measure it at the battery and at some other place like the lighter.

The tach acting up at the same time is what you would expect if the ignition wasn't firing. It also sounds like this is heat related. so either the coil or the PTM could be failing when hot.

 

[Dante923]

the top case says md165810/e2t37378/0810/e
the white connector says 0878e

The problem has gotten worse so im assuming the coil and ptm gave up. It now has the same problems all the time. When i took the ecu out and put it back in i must have messed with some wire because my safc wont turn on and the car wont start, although when it stalled at idle, it had trouble starting back up and i would have to give it gas to keep it on so i probly didnt do that. I looked up some info on ecus and i assumed the 7378e came from the e2t37378 and i found a e2t60878 ea01(because 0878e being printed on the connector) which is for a 7/92-5/93 federal, thats all the info it gives for that number.

i contacted the person i got the car from and he confirmed that it was socketed and rebuilt

 

[steve]

the top case says md165810/e2t37378/0810/e
the white connector says 0878e

When i took the ecu out and put it back in i must have messed with some wire because my safc wont turn on and the car wont start, although when it stalled at idle, it had trouble starting back up and i would have to give it gas to keep it on so i probly didnt do that.

You confirmed that this is a 93 M.Y. DSM MD193299 Federal AWD ECU swapped into a GVR4 case with the GVR4 EPROM but not strapped correctly.

Now it looks like the ECU isn't able to turn the MPI relay on. Do you a CEL for five seconds when you first turn the ignition on? Does the GVR4 boost LEDs turn on to the 0 psi light?

The SAFC is usually connected to the same wire that powers the ECU up, so I don't expect either of the above to be happening. Make sure the ECU is firmly plugged in and check the MPI fuse.

 

[Dante923]

Got the ecu power and started it, when it was cold, it idled on its own but if i revved it it shut off. I started it again and held the rpms up for a few secondsand then it idled on its own again. It idled for about 5 mins and then it shut off. After that it wouldnt idle on its own anymore. I took it around the block and i couldnt get the coils to start firing like i used to. When i got back i noticed my blitz twin sbc wasnt on and it wouldnt turn on. I checked the wires and they all looked fine. While i was checking them, i noticed that the line going to my boost controller had what appeared to be oil in it, or some dark fluid. Not sure what it was but ill post a pic

You must be logged in to view this image or video.

also, i checked the voltage at the battery while running and at the clock on my dash, the battery was at 13.62-3, clock was at 13.50-1

 

[Dante923]

I contacted the shop and they said that they replaced the ignition coil and ptu and the problem persisted, so to protect the parts,they put the old ones back in and thats where he stopped.

Also i got a reciept of parts/services that were done by the original owner (i am the third) but it is only part of reciept so i dont know who actually did it. Anyway, it said "existing ecu was socketed" and i know that the motor was rebuilt and now has about 60k miles on it and the body has 147k

I always thought "why did he rebuild it at 80-90k miles?" but now it occured to me that his caps must have leaked and destroyed his ecu and it must have messed the motor up so he got the current ecu and rebuilt it.The turbo has about 15k miles on it and was installed when the ecu was socketed along with most of the other mods

Since this problem happened after it was tuned i thought about what they did. My original plan was to get a map ecu2 installed, so they set that up but it wouldnt connect to the laptop. It would connect for a second and then boot them off. I read somewhere that day that 99% of all connection problem are caused by usb to serial connections, which they were using. They didn t have a serial port so they couldnt try that. They ended up reinstalling the safc, but first they tried tuning it with just the maf translator. They messed with that a bit and said that it couldnt be tuned on just that for some reason so they said they would "zero it out" and just tune it on the safc. They did that and after it ran fine until this started happening

 

[Dante923]

I found out that the groundwires were grounded properly and the wires were a little messed up. I replaced them and properly grounded them and replaced the ignition coil and this fixed the misfire problem. Although the main issue is fixed i found out in the process that my maft isnt working right so i decided to ditch the old apexi safc and maft and get dsmlinkv3. Should have that set up in the next week and the old galant vr4 should be running good like it should. Thanks for all your help guys.