Dignostic help Q: 01 Forester rough idle/miss

[RobS]

Looking for help diagnosing an issue on a friend's Forester:

2001MY Forester 2.5 SOHC Auto
(Note: these are speed density, no pesky MAF to point the finger at)

Here's the rundown:

Idles rough, distinct misfire visible (engine shakes/car vibrates) and audible at idle. Clears up when opening the throttle / driving.
Tube seals and rocker cover gaskets just replaced, new ignition cables, coil pack, spark plugs (correctly gapped), new PCV valve.

Problem persisted in the same manner after the replacement of the above.
Had one code for an O2 sensor, replaced the sensor and the code never returned, no difference present in idle/drivability.
Drove the car for several miles, several key cycles, sat idling for a good while and still hasn't thrown a code.

Here's the strange part:
Out of ideas, I decided to pull the (newly installed) plugs to see if that would lend me any clues.
On the passenger side bank (1&3) the plugs looked black (not fouled, rich)
I only had enough time/light/patience to remove the leading plug (cyl. #2) on the driver's side. This plug looked like this cyl was running very lean, it was completely while (no tan appearance at all.)

My Thoughts:
Vaccum leak?
Timing off?
Burnt exhaust valve?
Valves out of adjustment? (They are completely quiet)
^ All seem theoretically possible to me, however the different appearance of the plugs has thrown me for a loop.
Faulty/clogged injector (cyl 2)?



Once again: no check engine lights. Plugs indicate Bank 1 running rich, Bank 2 Lean

Thanks in advance.

 

[Roma]

Sounds like you may have a coolant leak(head gasket)due to that #2 cyl. plug being white. Just guessing

 

[SatinSilver]

Have it professionally diagnosed so you don't waste more money throwing parts at it.

 

[RobS]

Originally Posted By: satinsilver
Have it professionally diagnosed so you don't waste more money throwing parts at it.


These parts the owner already had and the plug wires were already soaked/plugs oil fouled, so that was a necessity regardless. The initial post represents the beginning of the diagnostic process.

I was under the impression a white spark plug indicates a lean condition.
Regarding a headgasket failure, there is no smoking from the exhaust when it's misfiring.
If that plug's appearance was the result of coolant entering the combustion chamber, wouldn't white smoke be visible from the exhaust?

 

[Rand]

early 2.5's were notorious for HG issues.

any coolant consumption or gasses in coolant?

 

[Roma]

Depends on how much is leaking. Cracked head will have white smoke. After warm up remove radiator cap and check for bubbles in the coolant. Is it using any coolant?

 

[The_Eric]

If it only has one o2 sensor then running lean on one back will cause the pcm to add fuel to all cylinders. Do you have access to a scan tool that will do live data?

 

[RobS]

Not readily available, no. I'm sure I could track one down.
I should check Long/Short term fuel trims, correct?
Yes, the car has one primary A/f sensor.


That does make sense,..the plugs were pulled and examined after we replaced the o2 sensor and drove it for a bit.
Possible the o2 sensor that caused the DTC was fouled from running excessively rich to compensate for cyl 2's lean condition?

So, what would cause at lean miss at idle that clears up under load? (and doesn't cause a CEL)

Vac leak? (leaning towards this now, seems to make the most sense. The VC breather lines/pcv lines and other assorted vacuum lines are rock-hard)

Faulty injector? (I've never witnessed a clogged/poorly spraying injector to accuse it as a diagnosis, I have no idea how it would manifest itself)

 

[datech]

Sometimes those plastic air intake parts crack and create an intermitant vacuum leak.

 

[cptbarkey]

air intake cracked, bad or not connected MAF sensor near the air filter, or you forgot to plug back in one of the various vacuum hoses connected to the airbox.

 

[martinq]

This is a common problem/symptom.

Absolutely check/replace the front O2 sensor. Had this exact problem in a 2002 and a new sensor fixed it for years following.

Please report back with an update!

 

[Quest]

I'm with satinsilver on this one.



Q.

 

[The_Eric]

Originally Posted By: cptbarkey
air intake cracked, bad or not connected MAF sensor near the air filter, or you forgot to plug back in one of the various vacuum hoses connected to the airbox.

OP already stated that there was no MAF, it is a speed density system.

Originally Posted By: martinq
This is a common problem/symptom.

Absolutely check/replace the front O2 sensor. Had this exact problem in a 2002 and a new sensor fixed it for years following.

Please report back with an update!


OP stated that there was a code for the O2 sensor, that is was replaced and drivability hasn't changed, nor has the code returned.

Originally Posted By: Quest
I'm with satinsilver on this one.



Q.


Originally Posted By: OP
These parts the owner already had and the plug wires were already soaked/plugs oil fouled, so that was a necessity regardless. The initial post represents the beginning of the diagnostic process.




Come on guys... Why are we making the OP cover the same ground multiple times?

 

[The_Eric]

Originally Posted By: RobS
Not readily available, no. I'm sure I could track one down.
I should check Long/Short term fuel trims, correct?
Yes, the car has one primary A/f sensor.


That does make sense,..the plugs were pulled and examined after we replaced the o2 sensor and drove it for a bit.
Possible the o2 sensor that caused the DTC was fouled from running excessively rich to compensate for cyl 2's lean condition?

So, what would cause at lean miss at idle that clears up under load? (and doesn't cause a CEL)

Vac leak? (leaning towards this now, seems to make the most sense. The VC breather lines/pcv lines and other assorted vacuum lines are rock-hard)

Faulty injector? (I've never witnessed a clogged/poorly spraying injector to accuse it as a diagnosis, I have no idea how it would manifest itself)


Yes, do track one down. In the meantime get a mechanics stethoscope or an ratchet extension and listen to all injectors. If the injector on the dead hole sounds different of doesn't make any noise, then you know where to investigate. You can also take a meter and measure the resistance of all injectors and see how they compare.

Like another poster said, spray for vacuum leaks around the intake.

For head gasket issues, just because you don't see white smoke doesn't mean that you don't have a leak. It takes very little coolant to make a bad misfire. Although I think there would be wetness on the plug and I don't think that the PCM would fatten up (I'm guessing here) the fuel trim to the point where the others would be black.

 

[Challenger71]

No. As Eric just posted recently, a HG related misfire does not automatically mean white smoke/steam from exhaust and this occurs only in the most severe of HG losses. In many situations with late 90's / early 2000's Subaru 2.5's is that the HG will slow leak, mainly externally via oil seepage on the block rear. Could be either side - look at block just above CV boots for external oil and/or coolant leak and use a leak dye if vision is a problem. Check the overflow bottle for discolored coolant, primarily the presence of exhaust gases will change the color of green coolant to brown/rust red. Also, check oil levels daily and naturally you want to check for presence of coolant contamination which changes oil hue to mocha coffee. In my particular case I had no evidence of coolant into oil so it was an external leak out the block, into the exhaust and into the coolant (compression and pressures coming from combustion cycle are much higher) and this caused a significant misfire at idle and randomly under load. $1700.00 later and she is like new with the revised HG. If you do see evidence of coolant contamination in the oil, deadline the vehicle immediately as you do not want to ruin the rod/crank bearings and the rest of the engine for that matter.

Originally Posted By: RobS
Regarding a headgasket failure, there is no smoking from the exhaust when it's misfiring. If that plug's appearance was the result of coolant entering the combustion chamber, wouldn't white smoke be visible from the exhaust?

 

[meep]

I HAD THIS EXACT MYSTERY ISSUE ON A 97 LEGACY.

Engine was WELL maintained and had zero issues, leaks, etc..

The timing belt stretched. It would generate random misfire codes at idle but otherwise ran fine. I had done the TB at 60k, and went with a Gates belt. it was significantly less robust than the oem. I replaced it with oem at 106k along with the WP and ALL of the stumbles went away.

I had a shop try to diagnose it at one point-- they said misfire on cyl #4, but the tech said he was skeptical too and it didn't all add up. they were going to try injector, then wiring harness, but said it was a by-the-book guess that didn't completely add up in this case.

For our 97, it was the TB.