1G Bad Intermittent misfire when warm, and warm starts

[1slw3sgte]

My 91 tsi randomly when warm gets a bad misfire as if it’s running on 2 cylinders, basically not drivable. When it happens also my wideband reads super rich (at 10.0).

After waiting for like 10-15 minutes for the car to cool off it will turn on and run fine until it wants to misfire again. I’ve replaced coil pack, spark plug wires, and power transistor but still have the issue. I’ve also took my injectors to get cleaned and tested and their good. No check engine light.

This issue has to be electrical because the car runs solid until that happens then have to wait for it to cool off, no overheating issues.

 

[DSMPT]

Could be a voltage issue due to a bad alternator/regulator. Or a bad fuel pump/fuel pressure regulator. Check the voltage and fuel pressure when the issue is occurring.

 

[1slw3sgte]

Could be a voltage issue due to a bad alternator/regulator. Or a bad fuel pump/fuel pressure regulator. Check the voltage and fuel pressure when the issue is occurring.

I was able to make the car start to misfire when I got home and checked my FPR, while the car was running, while misfiring, and it was good, at 42psi just what I set it at.

I heard iridium’s are not good, I do run e85 so what spark plugs you think I should run next since I’m prob gonna go pickup some tomorrow. Thanks

 

[1slw3sgte]

Any other idea of what my issue can be? Would love to hear your thoughts.

To make it clear, the intermittent misfire is pretty bad like as if the timing is completely messed up, I have to hold the gas to keep it on and sounds like running off 2 rough cylinders. Then works fine after 5-10 minutes.

The ECU is clean and rebuilt by ECMTuning, no leaky caps, no corrosion, nothin, clean ECMLink work. Replaced power transistor, coil packs, wires, and plugs. Maybe the cam sensor on the intake cam?

It’s a 6 bolt so no crank sensor.

 

[We're on Boost]

Any other idea of what my issue can be, would love to hear your thoughts. To make it clear the intermittent misfire is pretty bad like as if the timing is completely messed up, have to hold the gas to keep it on and sounds like running off 2 rough cylinders. Then works fine after 5-10 minutes. The ecu is clean and rebuilt by ecmlink, no leaky caps no corrosion nothin, clean ecmlink work. Replaced power transistor, coil packs, wires, plugs. Maybe the cam sensor on the intake cam? It’s a 6 bolt so no crank sensor.

When that happened on my car it turned out to be one bad terminal in the connector that goes onto the power transistor. In that case we were able to verify that there was something wrong with the connector by pushing on it in different directions with the engine running and one guy in the car with foot on the gas a little to keep it running.
It definitely was going back and forth between running on 2 cyls and running on all 4.

 

[T bags]

Engine coolant temp sensor in good working order? Had a simile issue with a non dsm an it was an afermarket unit. Not a big fan of iridium’s since I’m used to running ngk v powers. 2 on connecting long leads or alligator clips to the wiring on the fuel pump.

 

[1990TSIAWDTALON]

I run BR8ES's in my 9:1 and 10:1 E85 motors with high (over 30psi) boost and they are gapped at a whopping .018". You could try necking down the gap, but I don't think you are supposed to mess with iridium plugs. Are they the ones with the very tiny electrode? I go NGK normal copper plugs in all of my DSM's. If you are stock you should be in the 6 heat range not 8 like me.

 

[1slw3sgte]

I did swap from a 90 transistor to a 91-94 transistor since I couldn’t find a 90 transistor for sale. Had to do the rewire, car was doing the same thing before the transistor swap but will recheck my work.

I did swap my temp sensor a while back with a Autozone one, I’ll go steal I mean buy another one and see how that works. I swapped my iridium’s with bpr7es kept the same gap I had before at .20

 

[We're on Boost]

I did swap from a 90 transistor to a 91-94 transistor since I couldn’t find a 90 transistor for sale. Had to do the rewire, car was doing the same thing before the transistor swap but will recheck my work.

Do you mean the car had a 1990 power transistor on it when you bought it? Your car is a 1991 isn't it?

The 90 and 91 transistors don't even have the same number of terminals in the connector. So they need different connectors there.
Probably you should show the details of the rewire you did for the transistor. Then somebody like @steve in here could take a look for problems with that.

Along with that we should verify whether your ECU and coil pack are '90 or '91. The connector on a '90 coil pack has 4 wires. The '91-'94 has 3 wires.
It might be that the PO did some wrong parts mixing.

 

[Tyeler18]

Pull plug wires and see what cylinders are going down when it acts up. If it's 1/4 or 2/3 you can pretty much rule out anything that isn't spark related. ECU, wiring, I have had the cam sensor fail hot and take out only 2 cylinders.

 

[1slw3sgte]

Sorry for the late reply post, took time to zen and recollect my mind to tackle the issue on the DSM .

So now I met a new friend who had a GST part out. Got his green top CAS and the fuel injector transistor that is on the firewall as well, gonna try these parts in the next day or two.

Will be back with results hope it work out

 

[steve]

Got his green top CAS and the fuel injector transistor that is on the firewall as well, gonna try these parts in the next day or two.

I assume you mean Injector Resistor Pack not fuel injector transistor, cause I've never heard of that.

Is this still an issue that only happens when the car gets hot?

in DSMLink you can disable cylinders one at a time to see which ones make no difference when it's misfiring. Let us know which and please post a log of it runing ok and when it's not.

 

[waltah]

You really need to test/measure stuff for example by shutting off cylinders one at a time as Steve suggested. There's just too many things that can be heat related to reliably find a problem by replacing parts. For example a slight head gasket leak, an air leak or a sensor issue that only shows up closed loop.

 

[1slw3sgte]

I’m back after some tests. So I replaced CAS and injector resistor pack and no change. I did disconnect the #1 injector plug when it was misfiring and it didn’t change, it actually leaned a bit to around 13 instead of the 10 it was reading on the AFR, do keep in mind when it randomly misfires it smells like straight e85 in the air. ran a bit better so I guess my injector is getting stuck open randomly. Any ideas, the ecu was sent to link for their treatment like 2 years ago and I opened it up and it was super clean. Might just haft to upgrade to the 1200cc

Also I did take my injectors to get clean recently to a well known guy here in SoCal, got them back and he stated that they where not that dirty, about 5% difference in peramiters between the 4 injectors. I do remember I have a noid light set at my fathers shop and will use them while the car decides to act up again

 

[waltah]

This still smells like an ignition system failure. Plugs, wires, a coil, one transistor, or the ECU.
It's just about impossible to come up with anything else that would produce that 'sounds like it's running on two cylinders' when hot.

The running rich (when on two cylinders) is because the other cylinders aren't burning their fuel and it shows up unburned in the exhaust.

What about a shop with an ignition analyzer?

 

[1slw3sgte]

It started doing it on cold starts recently and hot starts

 

[pauleyman]

I’m back after some tests. So I replaced CAS and injector resistor pack and no change. I did disconnect the #1 injector plug when it was misfiring and it didn’t change, it actually leaned a bit to around 13 instead of the 10 it was reading on the AFR, do keep in mind when it randomly misfires it smells like straight e85 in the air. ran a bit better so I guess my injector is getting stuck open randomly. Any ideas, the ecu was sent to link for their treatment like 2 years ago and I opened it up and it was super clean. Might just haft to upgrade to the 1200cc

Also I did take my injectors to get clean recently to a well known guy here in SoCal, got them back and he stated that they where not that dirty, about 5% difference in peramiters between the 4 injectors. I do remember I have a noid light set at my fathers shop and will use them while the car decides to act up again

pulling an injector wire and it goes lean indicated you were injecting raw fuel and getting no spark. It makes sense that it was rich then you removed the fuel. It still sounds like a spark/ignition issue.

 

[MyTalonTSi1991]

I'm having a similar issue with mine! I've done everything you have done as well including a new fuel pump. Replacing the coil next and perhaps look into changing out a couple vacuum lines.

 

[T bags]

I'm having a similar issue with mine! I've done everything you have done as well including a new fuel pump. Replacing the coil next and perhaps look into changing out a couple vacuum lines.

Swapping coils on my buddies was the key. For some reason the car did not like aftermarket coils when warm. Get a smoke tester for the vacuum lines or a homemade mineral oil smoker.

 

[pauleyman]

You guys need to stop throwing parts at the car. Especially if you have ecmlink. Log the car and diagnose. You can artificially heat parts off the car with a hairdryer and test to see changes. It's way less expensive to do this.

 

[1slw3sgte]

So did the noid light on all injectors specifically #1 cylinder injector which was the one I suspect gets stuck open and all 4 flash so the ecu is good, I already bought a set of injectors a couple days ago so I will be replacing those

 

[pauleyman]

So did the noid light on all injectors specifically #1 cylinder injector which was the one I suspect gets stuck open and all 4 flash so the ecu is good, I already bought a set of injectors a couple days ago so I will be replacing those

I don't agree. You pulled the injector plug and the misfire didn't change. However you did say the fuel ratio changed because you just stopped that injector from firing. So it seems all injectors are working as they should and that cylinder and maybe others aren't firing. Ecu could still be suspect and so could wiring and items withing the ignition side. New injectors won't hurt it but I would say you've already ruled that out.

 

[XC92]

Same thing has been happening on my 2010 Kia Rio LX, which of course is a totally different kind of car, SOHC, NT, lower compression, etc. It's also a lot newer so it has OBD2 and more/better sensors and a more sophisticated ECU.

I haven't had time to diagnose and fix it as it's not my DD and use it mainly as a backup car when I'm working on my '92 Talon TSi, and I intend to sell it later this after I fix this issue and several others.

But the point is that it's probably something pretty basic like bad coils, alt, wires or voltage regulator, or a clogged fuel filter/line or failing fuel pump.

And I agree with Pauleyman that the wrong way to fix this is to just replace part after part. That's expensive and improper diagnosis. Figure out what's causing this and fix/adjust/replace it, and also what may have caused it to fail as sometimes issues cascade and it's not just one cause.

 

[waltah]

You guys need to stop throwing parts at the car. Especially if you have ecmlink. Log the car and diagnose. You can artificially heat parts off the car with a hairdryer and test to see changes. It's way less expensive to do this.

What he said. STOP throwing parts and diagnose the thing. No time now but maybe later I'll take a swing at the step-by-step.

 

[1slw3sgte]

managed to get a log of the car doing its misfire
-the first log i started it up and it was misfiring, few seconds later it goes back to normal (24 sec) .

-the second log i continuously try to turn it on but its always misfiring.

-the third log it turned on fine no misfire, until the end of the log it shuts off(659 sec).

-the fourth log it turns on good, o2 sensor pegged at 0.8 for a bit then goes to normal fluctuation until (93 sec) mark it misfires.

I feel like my o2 sensor might be going out bc when i log it sometimes stays pegged at 0.8 volts and does not fluctuate. when it misfires i can see the the airflowperrev jumps from lets say 0.45 to like 1.45 which i dont think its suppost to do. a while before the new injectors i put in i managed to get the fuel trims hovering around 0 which it drove pretty good.

so since the start of this issue i replaced
-PTU
-coil pack
-CAS
-Injector resistor pack
-injectors
Atleast i have spare parts now so im not bothered.
it is on speed density on a omni 4 bar

I should have posted logs from the start but its happens randomly.