[jwick]

Hello All, I am a new member to 3si and recently bought a 1991 dodge stealth TT with 90k miles. It is in very good shape, but has been having some intermittent engine issues that are becoming more consistent. I am going to try to keep this post concise, but it will be a bit of a novel as a want to present you with all the information. Any help or advice is greatly appreciated!

The day I got the car, it ran great, I drove it probably close to 300 miles with absolutely 0 issues. The previous owner said it had some intermittent issues with misfiring and a jumpy tach but it was taken to a mechanic and apparently sorted out. In the coming days of driving it, I saw the issues resurface, and they have become more and more consistent. over the course of the last week, the car has gone from sometimes having the issues to having them just about every time I turn it on.

The issue:
When I turn the car on, it will fire up fine, but after a bit of idling and when the rpms drop, the tachometer will begin to get jumpy, flickering back and forth between a correct rpm count and 2/3 of what the rpm should be. The rpm changes in the engine are also heard and felt. The engine will then turn on the check engine, and disable the 3rd coil pack (Cylinders 3-6) and run in a limp home mode with only 4 cylinders firing and the tach reading 2/3 of what it should be. The check engine code given is code 53, which corresponds to a lack of communication in the 3-6 coil pack, the coil that the engine turns off when it goes into limp mode. The car will continue to get me from a-b if necessary on four cylinders, but nonetheless it is a 6 cylinder engine running without cylinders 3 and 6. Before the ECU shuts off communications with the 3-6 coil, the car will drive with the tach jumping, and I will feel heavy misfires as the tach jumps, which I believe corresponds to the ECU not having proper communication/intermittent communication with the coil pack. From what I understand, the ECU has to be in communication with the coil pack to get a tach speed, as each coil provides 1/3 of the tach speed to the ECU (since there are 3 separate coil packs).

What I have done:

• I have replaced the PTU with a new OEM replacement as I have heard these can be the cause of this type of issue. No change in the problem after the fact
• I have replaced the old battery with a high quality battery as the old one was not performing well. The overall performance of the car is better but it did not resolve the issues with the misfiring/code 53
• I have checked the front 3 plugs and wires. They are seemingly in great shape.
• I replaced the coil pack associated with the check engine light with a new OEM replacement and have moved the coil packs around. The issue has not been resolved and the code 53 does not follow any specific coil pack
• I cleaned all harness connections with OBD connection cleaner between the ECU, PTU, and ignition coils.
• I have heard that faulty IACs can cause some issues like this, so I took out the old IAC, disassembled it and cleaned it with carb cleaner (it moves much more freely now), but there has been no change to performance
• I pulled the ECU and visually inspected it and cleaned it with some Isopropyl alcohol. There were signs of a bit of leakage underneath the capacitors, but nothing extreme
• I replaced the fuel filter as a shot in the dark and because I suspected it could not hurt, the car had been sitting for awhile
• I had seen some issues on the forums with a short/ground happening between the ptu/coil pack 3-6 blue/yellow wire connection that has the jumper for engine speed diagnostics in the engine bay. I suspected there may have been an intermittent ground interrupting the tach signal, so I simplified the wiring and cut out the diagnostic connection, running the 3-6 ptu line straight to the coil pack connector. The issue still persists afterwards, but is seemingly somewhat better, the car will run longer before going into limp mode. I checked the continuity of all 3 ptu-coil connections before splicing the wires, and they all had good coninuity but I went through with the splice as I did not think it could hurt.

After all of this I am still having the persistent code 53, and an engine that will eventually work its way into limp home mode, seemingly cutting off communications with the 3-6 coil, and having the tachometer display 2/3 of the engine rpm. I have the following thoughts of how I can continue to approach this and would like some opinions.

• A lot of places I have seen point to this being an ECU issue. I have continued to suspect this too, but I would like to explore my options/confirm this is the issue before I spend a lot of money on a new ECU. I was thinking of pulling the ECU again, testing the capacitance of the capacitors with a multimeter to see how close they are to what they should be, and then possibly replacing them on the board. Even though the leakage was slight on them, all the work I have done leading up to this makes me think the capacitors are causing this and the slight leakage confirms this. I could also see the intermittency of this issue being caused by a capacitor sometimes having the correct charge and sometimes not.
• I believe there is a chance there is a short in the line from the ECU to the PTU for the 3-6 coil pack signal. I was thinking of jumping a line from the ECU out harness right to the PTU in to test this
• I have also heard faulty IAC's can cause ECU issues as well. I am unsure that this is my issue, but I believe it is something I should check off. I cleaned it physically, but bad electrical internals could still persist. I can always replace it, and probably should nonetheless.
• I will eventually get around to checking/replacing the rear 3 plugs and wires as well, I just have not had the time to tackle taking off the intake manifold to get there yet. The condition of the front 3 plugs and wires makes me doubt this is my issue though.

Any thoughts on how I should continue to troubleshoot this would be greatly appreciated! Thank you for reading my novel!

 

[beegeezy]

Welcome! It sounds like you are very adept at diagnosing the common issues and have checked everything I can think of off the top of my head.

I agree with your plan…run a jumper wire from the ECU to the coil pack. While a short is possible, a 1G ECU with leaking original capacitors is much more likely the culprit. You can have your ECU rebuilt for <$200, so it’s not bad. If you want to mod the car, now would be a good time to invest in a new tuneable ECU though.

Sent from my iPhone using Tapatalk

 

[BlackStealth]

I had someone local with the same issues, he changes all the same parts like you did, but it ended up being a bad ECU. I would strongly suggest replacing those electrolytic capacitors, there's only 3 of them. If you do the work yourself, be sure not to pull on the capacitors and rip the copper traces underneath. Also after the capacitors are removed, take a high power flash light and verify the copper traces under the green solder mask is still there, and not eaten away by the leaky capacitors.

 

[BlackStealth]

Also I will add that leaky capacitors make it hard to solder again, so you will need to scrape and clean the solder pads. Also I would suggest mounting the capacitors up higher so you can solder both sides of circuit board.

 

[BlackStealth]

I gave this some more thought, there's a good chance those leaky capacitors may have done more damage than you realized. I would scrutinize all the circuit traces nearby the leaks and look for damaged copper traces that may have "opened up" (opposite of a short).

 

[jwick]

@BlackStealth @beegeezy Thank you both for the information. I am going to go ahead and replace and inspect the capacitors sometime in the next week probably. My bank account is currently recovering from buying the car, so the thought of getting a chrome ECU is enticing but if fixing the capacitors gets my stealth running right for cheap, I would rather try that first and leave the chrome ECU as a backup option/future plan. I should be good as long as I order capacitors that have the same capacitance and voltage rating as stock, correct? I believe this link to 3swiki lists the right parts for my 91 tt stealth, they mention 47µF @ 50V, 22µF @ 50V, and 100µF @ 25V capacitors ECU: Replace Capacitors (1991-1993) - 3000GT/Stealth Wiki . If I take my time and have care with the soldering work, I believe I can get past whatever damage may have been done by the capacitor corrosion.

 

[HOSS]

iirc the ecs module can also cause ecu problems

 

[mothsmoths]

Also I will add that leaky capacitors make it hard to solder again, so you will need to scrape and clean the solder pads. Also I would suggest mounting the capacitors up higher so you can solder both sides of circuit board.

I agree with BlackStealth and beegeezy that since you already did a good job changing and rotating the coils and replacing the PTU with no return of the missing 3,6 signal, this is very likely an ECU malfunction. Seeing evidence of leakage from the electrolytic capacitors confirms this especially in a 1st gen ECU (?MD159966) from the C104 capacitor. Electrolytic capacitors, unlike most other capacitor types is polarized meaning it has a + charged end and a - charged end. The property of an electrolytic capacitor that defines how good or bad it is functionally is not the measurement of its capacitance, it is the measurement of its ESR (Equivalent Series Resistance). It evaluates the leads, internal plates, insulator material, electrolyte, etc. all at once. That cannot be done with a multimeter, it requires an ESR meter which does the measurement while the capacitor is still on the board and is compared to an "expected value table" standard. Capacitors with faults leading to high ESR often overheat and thereafter bulge and leak as the electrolyte chemicals decompose into gases, making them somewhat easy to identify visually; however, capacitors that appear visually perfect may still have high ESR, detectable only by ESR measurement. But, if it leaked it's bad.

You should read through this currently active post and look at his pictures and my posts and others. 92 R/T TT - engine starts fine every time but stalls.... The leaked electrolyte runs down and onto the pins of the vertical daughter board labeled IC105 which I believe is responsible for generating and timing the ground signal output which fires the 3,6 coil. The ground signal for the other 2 coils is primarily from elsewhere in the ECU. You could post a couple good close-up pics of your ECU board if you are gonna give self fixing a go as experienced others can point out issues.

For clarification purposes...this 1991 OBD1 ECU was pretty amazing when it came out in 1991, but is very primitive comp with 1996 OBD2 and up. There was no serial communication data feedback to ECU and it would not be for another good decade before ECUs would be sophisticated enough to shut-down ignition or injectors to non-functioning cylinders or as part of a "limp mode". In your car the ECU limp mode involved solely keeping the car A/F, timing, and injection functions fixed based on preprogrammed tables in the ECU. ie. runs in "open loop" even after up to desired temp.

 

[beegeezy]

Lots of good info!


Sent from my iPhone using Tapatalk

 

[BlackStealth]

Here's my wiki page, although someone had edited it. I used primitive soldering tools back in the day (10+ years ago), today I'd recommend a temperature controlled soldering tip @350 degrees, with removable tips. Using a flat soldering tip allows the solder wick tape to work much better and open up the solder pads for inserting new cap. Use solder that contains tin/lead it flows better than lead free stuff. Don't over work a pad for too long or the heat will make the pad come off the circuit board.

Electrical: Replace ECU, ECS and Climate Control Capacitors - 3000GT/Stealth Wiki

www.3swiki.org

 

[AndrewRoyal91]

Also agree with the capacitors. I’ve had to do this one 5 out of 6 of my 3000gts

 

[jwick]

I agree with BlackStealth and beegeezy that since you already did a good job changing and rotating the coils and replacing the PTU with no return of the missing 3,6 signal, this is very likely an ECU malfunction. Seeing evidence of leakage from the electrolytic capacitors confirms this especially in a 1st gen ECU (?MD159966) from the C104 capacitor. Electrolytic capacitors, unlike most other capacitor types is polarized meaning it has a + charged end and a - charged end. The property of an electrolytic capacitor that defines how good or bad it is functionally is not the measurement of its capacitance, it is the measurement of its ESR (Equivalent Series Resistance). It evaluates the leads, internal plates, insulator material, electrolyte, etc. all at once. That cannot be done with a multimeter, it requires an ESR meter which does the measurement while the capacitor is still on the board and is compared to an "expected value table" standard. Capacitors with faults leading to high ESR often overheat and thereafter bulge and leak as the electrolyte chemicals decompose into gases, making them somewhat easy to identify visually; however, capacitors that appear visually perfect may still have high ESR, detectable only by ESR measurement. But, if it leaked it's bad.

You should read through this currently active post and look at his pictures and my posts and others. 92 R/T TT - engine starts fine every time but stalls.... The leaked electrolyte runs down and onto the pins of the vertical daughter board labeled IC105 which I believe is responsible for generating and timing the ground signal output which fires the 3,6 coil. The ground signal for the other 2 coils is primarily from elsewhere in the ECU. You could post a couple good close-up pics of your ECU board if you are gonna give self fixing a go as experienced others can point out issues.

For clarification purposes...this 1991 OBD1 ECU was pretty amazing when it came out in 1991, but is very primitive comp with 1996 OBD2 and up. There was no serial communication data feedback to ECU and it would not be for another good decade before ECUs would be sophisticated enough to shut-down ignition or injectors to non-functioning cylinders or as part of a "limp mode". In your car the ECU limp mode involved solely keeping the car A/F, timing, and injection functions fixed based on preprogrammed tables in the ECU. ie. runs in "open loop" even after up to desired temp.

@mothsmoths is there anything that I can do to clean the IC105 sister board? the fact that it is separate from the ignition commands for the other two coil packs makes me suspect that the capacitor electrolyte may be shorting parts of that connection. I believe I know what board you are talking about. I could see possibly de soldering it and re soldering but I am unsure if it is an option, and it has many pins and I would be reluctant to attempt it.

One piece at a time though, I am going to tackle the first step of replacing the capacitors and see if that solves my issues.

 

[jwick]

@AndrewRoyal91 Were you having similar issues to me on any of those 3000gts? And did replacing the capacitors fix it?

 

[BlackStealth]

I would leave IC105 in place, and just clean the connection and then re-apply some solder.

 

[Marshja3]

I’m thinking about trying to rebuild an extra ECU I have from my parts/spare vr-4 car. I’m sure the capacitors are most likely bad or close to being bad so I want to rebuild it and use it in the future when I get my daily vr-4 running. Where do we go about getting the esr meter?

 

[BlackStealth]

You can find a bunch of "ESR meters" on ebay. Keep this in mind, if the ESR meter device puts out more than 5 volts during it's test, it may damage the microcontroller. Better to remove the capacitors during the test. But if you have gone this far already, might as well replace the capacitors. Electrolytic capacitors have a limited life cycle, and after about 10 years of use they need replacing. Capacitors are the number one reason why electronics fail in everything, such as TV, radios, and car electronics.

 

[Marshja3]

Yeah it looks like they are relatively cheap as well. 3sx sells 4 replacement caps for what looks like 12 dollars or so. I’m just kinda needing something to do indoors. I live in Iowa and it’s much to cold outside to be doing anything with the car itself. It is in an old machine shed that is full of rotting holes of wood so all kinds of elements make it inside of the building. I will have to wait until spring to get back in the car. That’s kinda why I was thinking about getting into all of this. I can bring my ecus indoors and tinker and learn while being warm 😎

 

[mothsmoths]

Yeah it looks like they are relatively cheap as well. 3sx sells 4 replacement caps for what looks like 12 dollars or so. I’m just kinda needing something to do indoors. I live in Iowa and it’s much to cold outside to be doing anything with the car itself. It is in an old machine shed that is full of rotting holes of wood so all kinds of elements make it inside of the building. I will have to wait until spring to get back in the car. That’s kinda why I was thinking about getting into all of this. I can bring my ecus indoors and tinker and learn while being warm 😎

I have used my Peak Atlas ESR70 meter for over 15 yrs. It tests electrolytic capacitors in circuit. For ESR measurement it discharges capacitor then pulses it on/off 100,000X/sec so esr can be measured without allowing the capacitor to charge or activating other in circuit components that would interfere with reading or get damaged. Max current it uses is 20mA and 40mV and takes about 2 seconds. In most circuits it will also report capacitance as well if able...will if out of circuit. But its the ESR that in 2 secs tells you good or bad.

 

[Marshja3]

I would leave IC105 in place, and just clean the connection and then re-apply some solder.

jwick did you by any chance have any luck with you ecu?

 

[BlackStealth]

jwick did you by any chance have any luck with you ecu?

It wasn't my engine ECU, it was someone local to me. They purchased a second engine ECU before I had a chance.