Diagnosing a non-working IAC Valve in a 92' Galant

[sebas-1024]

Hello, I posted a few days ago about a reliable brand for an engine coolant temperature sensor and received a lot of help, so I'm back with more problems that I hope you can help me with.

The vehicle is a 92' Galant with a 6-bolt 4g63 and an automatic transmission. As far as I understand it the AIC Valve is not working, the car right now idles at about 900 RPMs in Drive or Reverse and when placed in Park or Neutral at about 1500 RPMs, the AC doesn't work but I'm guessing if it did the car would probably stall when idling. I've been chasing this problem for a while now and this is what I have so far:

1. The original IAC Valve has an internal electrical issue, according to the manual the coil resistance should range from 28 to 33 Ohms at 20 ºC in the specified terminals (Figure 1), the values are Ok for terminals 4-5 and 5-6, but terminals 1-2 show an OL (Out of Limit) reading and 2-3 show values of 62 Ohms which is almost double, this leads me to believe that there is a short of some kind within the coils of the IAC Valve and it needs replacing.

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

Figure 1 - Checking Coil Resistance

2. I bought a new IAC valve on Ebay, after installing I noticed that the car idles even higher, when checking the coil resistance again, it showed values of about 39 Ohms for all terminals, seems a little high but I guess it should still work to an extent, so I thought it could be a harness or ECU problem, which leads me to the next step.

3. I followed steps 1 and 2 of the Harness Inspection process shown in Figure 2, step 1 is Ok because the readings for terminals 2-Ground and 5-Ground show 12 V or so, but step 2 is where I'm a bit lost, when checking the connections I found the same result for 58, 59, 67 and 68, they're all in continuity with terminals 2 and 5 on the other side of the harness, and they have no continuity to ground or to terminals 1, 3, 4 and 6. If I'm reading the circuit diagram correctly (Figure 3), there should be continuity between terminals 58-1, 59-3, 67-6, and 68-4. So this would mean that somehow terminals 58, 59, 67 and 68 are connected to terminals 2 and 5, and not to terminals 1, 3, 4 and 6, which would explain why the AIC Valve isn't working at all.

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

Figure 2 - Harness Inspection

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

Figure 3 - Circuit Diagram

I have 2 questions here:
1. Is the AIC valve with the 39 Ohms coil resistance able to work properly with this vehicle once the harness issue is sorted out, or do I need one with the proper resistance values, and if so, can someone recommend one?
2. Is my understanding of the harness issue correct and maybe the cables are crossed or spliced at some point? I'm not an electrician or anything of the sort, so feel free to correct me. I'd just like to be somewhat sure of the problem before I start messing with the IAC harness.

Sorry for the long post, but this issue has been bugging me and I want to make sure I'm on the right track. Thanks in advance for any help!

 

[chrysler kid]

Im not sure if its a useful tip, but on my automatic 2g that was having idle problems, the only way I was able to verify the idle air control motor was working was to leave the old idle air motor installed on the car, unplug its harness, and take the new part and plug the harness into it. Basically youre going to watch the idle control valve to verify its clicking to 4 different positions as commanded by the ecu

I took a video of the new idle air motor when i started the car and saw the new motor was doing 4 different lengths for the stepper motor to verify the new part was doing exactly what the ecu commanded it to.

Also with the part installed in the car and the harness unplugged the car should idle flat at the exact idle it should, because unplugged its has fully closed the opening in the throttle body that lets air bypass the throttle body to idle higher(at warm operating temp)

Since these cars are so prone to ecu caps leaking, sometimes the ecu failing will prevent the stepper motor from doing its fully cycle of positions.

 

[steve]

The old style metal cased ISC/IAC are know to be a problem due to the coils shorting and causing the driver IC's in the ECU to burn up. Your measurement showed that the one coil had gone open and the second one had a high resistance. It's still possible that the old ISC damaged the ECU driver before the coil went open.

I think the second test is just checking the wiring between the ISC and the ECU for opens or shorts. You would want both the ECU and the ISC to be disconnected for that test.

If you looked at the pins from the ECU side with both connected you would typically see 12v on the 4 ECU pins unless the ECU was actively trying to step or hold the ISC in position and then one of the pins might be close to 0v.

The new black plastic ISC's typically measure about 39 ohms. That's fine, it just means less current will flow through the coil as the ECU steps the pintle.

These tests don't prove that everything is fine. Watching the ISC step (not just buzz) as it rehomes on powering up the ECU is as about as good a check as you can do without special equipment.

Also as mentioned, the ECUs of this time all have a problem with the capacitors leaking and eating up the circuit traces. The need to be inspected, caps changed and repaired as needed. If you have blown a driver IC it's usually visible when the ECU is checked over.

 

[mitsubishikid]

Would be smart for the op to remove the ECU from the car, open the case up and inspect all of the black ISC drivers, especially ic105 and ic107, both drivers run the ISC coils, doing a visual inspection of the circuit board just to see if they have any obvious signs of failure like chips, cracks, cracked, burn marks or burned out sections around or on the drivers, both the top and bottom of the circuit board.

 

[sebas-1024]

Basically youre going to watch the idle control valve to verify its clicking to 4 different positions as commanded by the ecu

I tried doing this at one point and saw no movement with either of the AIC valves, the new one should be working so there is a problem in the harness or ECU, I'm leaning towards harness first because the inspection seems to indicate a faulty connection or a short-circuit somewhere along the wires, but they're both possible causes.

The old style metal cased ISC/IAC are know to be a problem due to the coils shorting and causing the driver IC's in the ECU to burn up. Your measurement showed that the one coil had gone open and the second one had a high resistance. It's still possible that the old ISC damaged the ECU driver before the coil went open.

I think the second test is just checking the wiring between the ISC and the ECU for opens or shorts. You would want both the ECU and the ISC to be disconnected for that test.

If you looked at the pins from the ECU side with both connected you would typically see 12v on the 4 ECU pins unless the ECU was actively trying to step or hold the ISC in position and then one of the pins might be close to 0v.

The new black plastic ISC's typically measure about 39 ohms. That's fine, it just means less current will flow through the coil as the ECU steps the pintle.

These tests don't prove that everything is fine. Watching the ISC step (not just buzz) as it rehomes on powering up the ECU is as about as good a check as you can do without special equipment.

Also as mentioned, the ECUs of this time all have a problem with the capacitors leaking and eating up the circuit traces. The need to be inspected, caps changed and repaired as needed. If you have blown a driver IC it's usually visible when the ECU is checked over.

Definitely possible, the old AIC does seem to be shorted, what does ISC stand for by the way?

I think the same about the second test, the harness was disconnected from both ends when I checked for ground or short circuits, that's why I think there should be continuity between the corresponding wires and nothing else, there aren't any other connections between those two ends.

Good to know about the 39 Ohms, will install that one eventually then.

What are "caps" and "IC driver" in the ECU? Have never touched the ECU except for disconnecting the harness for this test so I have no idea really.

Would be smart for the op to remove the ECU from the car, open the case up and inspect all of the black ISC drivers, especially ic105 and ic107, both drivers run the ISC coils, doing a visual inspection of the circuit board just to see if they have any obvious signs of failure like chips, cracks, cracked, burn marks or burned out sections around or on the drivers, both the top and bottom of the circuit board.

How safe is it to take apart the ECU, just a few screws and don't drop water on it? Wouldn't want to cause more headaches than I already have.

 

[chrysler kid]

How safe is it to take apart the ECU, just a few screws and don't drop water on it? Wouldn't want to cause more headaches than I already have.

No worries of damage unless you drop it. There are videos on how to remove it from the metal case on your tube. Its Been a while since I cracked one open and all I can remember are the hondas ive done

 

[steve]

What does ISC stand for by the way?

What are "caps" and "IC driver" in the ECU?

What you keep calling the AIC is called either a ISC (Idle Speed Control Servo) or IAC (idle Air Control Motor) depending on if the car is a 1G or a 2G. Same part (except on a 90), same function, likely different engineer writing the manuals.

Cap is short for capacitor. A electrical device that stores charge.
IC is short for Integrated Circuit. Another electrical device that contains many other transistors, resistors, capacitors, and diodes etched into a piece of semiconducting material.
Driver in this case refers to a special type of transistor that can handle larger currents than your garden versions so it can control motors and valves.

 

[sebas-1024]

Ok, checked my diagnostic process and I was checking the wrong wires, turns out those are fine and there is continuity from both ends of the IAC motor harness, so I don't have any reason to suspect that anymore. Based on this I started checking again to see what could be the cause.

Im not sure if its a useful tip, but on my automatic 2g that was having idle problems, the only way I was able to verify the idle air control motor was working was to leave the old idle air motor installed on the car, unplug its harness, and take the new part and plug the harness into it. Basically youre going to watch the idle control valve to verify its clicking to 4 different positions as commanded by the ecu

I took a video of the new idle air motor when i started the car and saw the new motor was doing 4 different lengths for the stepper motor to verify the new part was doing exactly what the ecu commanded it to.

Also with the part installed in the car and the harness unplugged the car should idle flat at the exact idle it should, because unplugged its has fully closed the opening in the throttle body that lets air bypass the throttle body to idle higher(at warm operating temp)

Since these cars are so prone to ecu caps leaking, sometimes the ecu failing will prevent the stepper motor from doing its fully cycle of positions.

Retook this test just to be sure and the result is the same, no movement or vibration at all in the new IAC motor when connected to the harness while the old one is installed in the throttle body, so unless I bought a non-working IAC, it is not receiving a signal.

No worries of damage unless you drop it. There are videos on how to remove it from the metal case on your tube. Its Been a while since I cracked one open and all I can remember are the hondas ive done

I removed the ECU from the vehicle to give it a look, turns out it was supposedly repaired in 2015 due to a faulty IAC (Figure 1), I guess they didn't do a good a very good job. There's a burn or explosion mark to the right of the inside of the case, so something definitely got damaged in there (Figure 1).

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

Figure 1 - Inside of the ECU case with burn mark

Would be smart for the op to remove the ECU from the car, open the case up and inspect all of the black ISC drivers, especially ic105 and ic107, both drivers run the ISC coils, doing a visual inspection of the circuit board just to see if they have any obvious signs of failure like chips, cracks, cracked, burn marks or burned out sections around or on the drivers, both the top and bottom of the circuit board.

It seems like this is exactly what happened, while checking the ECU I noticed some parts that look like they're damaged, maybe they were repaired but didn't last long, or they were never repaired, I'm attaching some pictures (Figures 2-5)

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

Figure 2 - Line of pins that look a bit burnt

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

Figure 3 - the pins in IC105 looks a bit burnt

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

Figure 4 - IC107 has seen better days as well

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

Figure 5 - Some collateral damage I suppose, these don't look damaged just stained from whatever got burnt

So the bottom line would be that the ECU is not working properly, is there an easy fix for this or should I just look for a used one? I'm guessing that fixing this would require some basic soldering skills at the very least (which I don't have), so maybe this is as far as this ECU goes.

Thanks everybody so far, this has been a headache but I was able to find the culprit with your help.

 

[steve]

Not only do you need some average soldering skills you need the pinout of the IC, a source of spare parts and know where the various pins go. All the capacitors should be changed at the same time.

Back when I was repairing them I tracked down New Old Stock (NOS parts) and also salvaged them from NA and otherwise dead ECU's.

I can't tell from your pictures if that is an EPROM ECU, which would make it more valuable, or a less common version like a 90 or a GVR4 ECU. In any case I'd send that one to the guys at ECMTuning for repairs and also keep an eye out for a spare EPROM version.

 

[mitsubishikid]

^ I Echo what Steve suggested, but do upload some more pictures of the top of the circuit board you might need capacitors too, but for sure contact ecmtuning http://www.ecmtuning.com/product_info.php?products_id=33 they are among the few I would trust to service mitsu/dsm ecu or tcu computers, speaking of which if you are running the stock TCU it's not a bad idea to also pop that open and inspect the capacitors inside, I mean you could also send that to them as well to replace the capacitors on while your at it they are known to just fail out of nowhere on these cars.

 

[chrysler kid]

That board has been tampered with during the previous repair. They have traced wires from connections to a good parts of the circuit on the board

If you look at the burnt aqua looking thing you can see a copper connection point where the soldier failed, and the trace wire coming off of it is hanging freely

I would reccomend buying a new ecu. I would expect even ecm tuning would say the board is damaged beyond what is considered safely savable, and that any future repairs are compromised by previous repairs

 

[steve]

I would reccomend buying a new ecu. I would expect even ecm tuning would say the board is damaged beyond what is considered safely savable, and that any future repairs are compromised by previous repairs

All I can say here is I've fixed much worse. Yes, there are fried traces that will have to be replaced and the rest will have to be inspected end to end to see what else is fried but it's far from trash just because of that. As far as I know those ECU's are still on the road and it's been over 10 years since I last left my bench.

 

[chrysler kid]

All I can say here is I've fixed much worse. Yes, there are fried traces that will have to be replaced and the rest will have to be inspected end to end to see what else is fried but it's far from trash just because of that. As far as I know those ECU's are still on the road and it's been over 10 years since I last left my bench.

Good to know

 

[sebas-1024]

Not only do you need some average soldering skills you need the pinout of the IC, a source of spare parts and know where the various pins go. All the capacitors should be changed at the same time.

Back when I was repairing them I tracked down New Old Stock (NOS parts) and also salvaged them from NA and otherwise dead ECU's.

I can't tell from your pictures if that is an EPROM ECU, which would make it more valuable, or a less common version like a 90 or a GVR4 ECU. In any case I'd send that one to the guys at ECMTuning for repairs and also keep an eye out for a spare EPROM version.

No sure what an EPROM ECU is, but considering that this is a plain Galant without any mods, I would asssume that it has a regular ECU. I saw the ECM Tuning site but I'm not from the US so it would be well outside my budget to send it to them, tried looking around here for an used one in good shape but no luck so far, any idea as to where I could buy one that is in good shape and then send it over to my country? I looked around on Ebay but they're around $150-$200 and that's much more than I can see myself spending right now for this car.

^ I Echo what Steve suggested, but do upload some more pictures of the top of the circuit board you might need capacitors too, but for sure contact ecmtuning http://www.ecmtuning.com/product_info.php?products_id=33 they are among the few I would trust to service mitsu/dsm ecu or tcu computers, speaking of which if you are running the stock TCU it's not a bad idea to also pop that open and inspect the capacitors inside, I mean you could also send that to them as well to replace the capacitors on while your at it they are known to just fail out of nowhere on these cars.

I'm uploading some additional pictures of the ECU, maybe you guys can identify some other things in there. I see what you're saying about the TCU but I'm not from the US, and I don't know of anybody who repairs ECUs here in Venezuela, which is where I'm from.

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.

Sorry for another long post, but tried finding an ECU to replace this one with and haven't been able to find one so far, please let me know if there's a site that sells them or something. Thanks for your replies!

 

[chrysler kid]

So you may not have ecu repair near you, however you should be able to find some one who does TV and electronics repair. This person would repair televisions and cell phones.

Circuit boards use pathing under the green covers. In a very simple terms instead of seeing wires connecting everything on that board the wires are hidden in the green board itself. An electronics repair shop can use a volt meter to follow the path of connections to verify which connection has failed on the board. They can then inspect to see if the board is damaged

Im not able to visually inspect for all possible damage. We may be able to spot a few issues, but some one with a volt meter will need to verify connections and solder joints

In relative terms the ecu is fairly simple compared to a modern gaming system. Ps5 and xbox repairs require micro soldiering because the connections are so small