Troubleshooting ECU drivers and the ISC

[GSTwithPSI]

So, driving to the track yesterday, and started to smell something funny. I thought something outside was to blame and didn't give much thought to it. Ran the car at the track all day and made an hour drive home without issue. I get home, park the turd in the garage and it sits overnight.

I go out this morning to dick with the tune, and jump in to start it up. As soon as I got in, I noticed the same smell. Now I'm thinking, WTF. It must be something in the car. I smell around and narrow it down to what seems to be the passenger floorboard. ECU, no sh*t right? Before I do anything else, I start to car just to see what happens. It starts fine, idles fine, and acts completely normal. Now, I'm really like WTF?

I pull the ECU just to see what I can see. I smell the case before opening it up and immediately know something's up. I pop off the cover and here's what I found:

Now, my question is, why? I've recently swapped in a revised (black) ISC, and have been running it for a few hundred miles now, so I don't think that's it.

The one factor I'm wondering about is my wideband. This seems to have happened less than 12 hours after I hooked it up. But, the wideband is still functioning as it should. ECMlink still works, the wideband is reading correctly and everything. I hooked it up using the EGR temp circuit as an input, all according to instructions. I'm confident it is wired correctly, because the damn thing works. Could this be coincidence?

Thoughts, input, suggestions would be awesome.

 

[turbowop]

Coolant leak from the lower section of the throttle body shorting out the ISC and blowing those drivers...possibly.

 

[turbowop]

Oh wait, you have a SMIM, which means you probably don't have coolant at the TB anymore. Weird.

 

[GSTwithPSI]

I deleted the FIAV and coolant lines a while ago, using a blockoff plate with an ISC bypass. There wasn't any coolant running to the TB yesterday, if that means anything. Thanks for the reply.

 

[tektic]

Well if its still working take out the melted parts clean it up and replace them with new ones. It shouldn't be that bad but I would check all the 5v stuff.

 

[GSTwithPSI]

Eh, I emailed the guys at ECMtuning, to see what they have to say. I'm planning to mail them the ECU regardless.

 

[89Patches]

That's good that you caught that before anything else happen!! My Ecu took a sh*t a couple of days ago causing two injectors to stay open. Ended up hurting my motor /ubbthreads/images/graemlins/frown.gif

 

[GSTwithPSI]

Ok, so as a testament to the amazing customer service at ECMtuning, Thomas already sent me back a response. Who answers whiny customers on a Sunday night?

Anyway, it looks like IC105 is blown. According to this link: click, the ISC is to blame. I'm pissed, because the tan ISC that was in the car was working perfectly. Ironically, the only reason I swapped it out with a black revised ISC was to avoid blowing up ISC drivers, as it's rumored the revised ISCs are far less likely to cause such a failure. Well, guess that's what I get for being proactive. I'll test the ISC tomorrow to confirm it's bad, but suffice it to say something bad happened and completely fried the ISC driver.

The ECU seemed fine up until recently, when I installed the WB. I mentioned this to Thomas, who seemed to think the WB install couldn't have caused the issue. That was a relief. I just immediately thought of the WB possibly causing an issue, since that's the only change I've recently made to the car. Actually, I suppose "fine" is subjective, seeing as the car somehow still runs and logs like nothing is wrong. I'm sure I wired the WB correctly since ECMlink was reading it just fine, but wanted to make sure this issue with the ECU and the WB install weren't connected since there are 12 hours separating the two events. Just mere coincidence I guess.

1G PARTS ('90-'94 ECUs)
IC104 controls ECU pin 14 ('90) or 6 ('91+), MAF Reset signal. The other half of the chip is unused on US DSMs.
IC105 controls ECU pins 67 and 68, Idle Speed Control (ISC) Servo (coil pair B).
IC106 controls ECU pin 57, Fuel Pressure Solenoid (FPS), and ECU pin 105, Boost Control Solenoid (BCS).
IC107 controls ECU pins 58 and 59, ISC Servo (coil pair A).
IC108 controls ECU pin 53, EGR Solenoid, and ECU pin 62, Evap Purge Solenoid.
IC109 controls ECU pin 56, Fuel Pump Relay, and ECU pin 65 Air Conditioner Compressor Clutch Relay.

 

[dsmless]

I had the same problem on 96/1000, it fired the same IC. I just ran it with the ISC unplugged for a bit then replaced the ecu and the ISC. Every now and then I test the ISCfor proper resistance, Im still a bit weary about it since my first fried ecu, but so far its holding up fine.

 

[ghostinthevr4]

That sucks has the car been idle surging? That's the end product of a car that has idle surged, send it too Tom and Dave they'll square you away. Check the ohms of your ISC Position sensor on a black one should be 38-40 ohms if it's a bit higher that's okay.

When you get everything back together do a boost leak test also, boost leaks cause idle charge and idle tuning problems even if no surge is present.

 

[GSTwithPSI]

No, that's the weird thing. The car ran normally up until I pulled the ECU. I'm going to pull the ISC and test it as soon as I have time.

 

[EfiniX]

You know, there's two types of ECU's in these cars. Ones that have failed and ones that are going to fail. Look on the brightside: once this is all buttoned up, you'll have a nice, shiny, probably-not-going-to-burn-up-for-a-long-time ECU.

I've been told by many people that the first "mod" you do when you get a VR4 (or any 1g for that matter) is pull the ECU and send it off to ECM Tuning. Sucks, but ultimately you'll end up with a better car.

 

[coyotes]

The ecu went bad in my car before I bought it, I should probably get the ISC checked out, although I don't know if that's what made the original ECU go bad.

 

[GSTwithPSI]

I finally got around to pulling the ISC today, and it looks like I found my issue. Resistance on ISC coil pair B between pins 5 and 6 was 0.6 Ohms, which caused ECU driver IC105 to get fried.

I'd like to explain a little bit about how to correlate which ECU driver is responsible for running which ISC coil. I had to cross reference a wiring schematic because this is not clearly spelled out in any one place, so hopefully this saves someone the trouble in the future.

The ISC has 2 coils. They are referred to as coil pair A, and coil pair B.

Coil Pair A
-When looking into the ISC, coil pair A is pins 1, 2 and 3.
-ISC Pin 2 is ground for coil pair A
-ISC Pin 1 is a coil and goes to ECU pin 58
-ISC Pin 3 is a coil and goes to ECU pin 59
-IC107 is the ECU driver that will blow if coil pair A is bad.

Coil Pair B
-When looking into the ISC, coil pair B is pins 4, 5 and 6.
-ISC Pin 5 is ground for coil pair A
-ISC Pin 4 is a coil and goes to ECU pin 68
-ISC Pin 6 is a coil and goes to ECU pin 67
-IC105 is the ECU driver that will blow if coil pair B is bad.

ECU Injector driver information page

 

[ktmrider]

^^ Good info here, mods should consider a Sticky/Bookmark.

 

[prove_it]

Second adding that post to the how-to section. Good testing info there.

 

[GSTwithPSI]

Got the ECU back from ECMtuning. The guys did a great job patching this thing up:

Before:

After:

Before:

After:

Before:

After:

Edit:
Here are 2 other helpful threads regarding exploding ECUs:

Would you like some fried ECU with your order, Sir?

Cooked ECU - deep fried - with pics.