Subaru Outback Forums banner

1 - 20 of 35 Posts

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #1
Having issue with a flashing check engine light on my Suby (description in the signature box). Codes are P0302, P0304, P0306.

Need some help, please.

Initially appeared as a steady CEL with those codes, but it ran normally. Cleared the codes. Then a couple drive cycles later, it became a flashing CEL. Took to my trusted mechanic. Confirmed the codes i had read. Recommended to change A/F ratio sensor, so I bought and had him install new Subaru sensors both sides. CEL stayed off for one day. (Sensors were previously changed in 2010 at 183,xxx miles. Rear sensor not changed this time.) Inspections during the installation revealed nothing out of the ordinary.

The next day, when I started the car, within a minute or two the light was back on and flashing. Again 2-4-6 misfires. Sometimes it will briefly switch to steady CEL, but then back to flashing.

Current mileage 357,xxx.

Headgaskets replaced at 300,xxx miles and the re-build included pretty much anything that you could think of to replace, including timing chains/guides, water pump, completely rebuilding the heads, new plugs (factory spec plugs), plug boots, et al. Exhaust replaced with cat back aftermarket system at 329,xxx miles.

Recent 500 mile trip about two weeks ago - it ran flawlessly. Monitored with Scan-Gauge the whole trip. Temps, batt voltage, etc., very stable and as expected. Mileage was good/normal for highway on the trip. Didn't use any oil or coolant at all.

Nothing done to the car since then except gas and drive. If it matters, I've been driving much less than normal (lost job, no longer commuting. but I need this fixed as I start a new job in a couple weeks....). During the limited driving time, short trips and all, the mileage is down a few MPG, but nothing out of the ordinary range for winter.

Knock sensors never replaced. I run Mobil 1 High Mileage oil, factory recommended weight. Subaru coolant and ATF. Synthetics in the diffs.

I checked/cleaned/refreshed all the grounds I could find several months ago.

When the CEL is flashing, the engine runs rough. Exhaust smells rich. And after a minute or two of idling if warmed up, it smells very hot outside the car.

Last 5 tanks of gas have been mid-grade with no ethanol. Car has used and liked this quite well for years, so this is not a change. Same gas stations even.

Takeda air bag replaced under recall about 2000 miles ago. Cabin air filter done at the same time.

I have all my maintenance taken care of on schedule, with Subaru parts/supplies for anything significant.

Mechanic is currently stumped, so I'm looking for suggestions to help pinpoint the issue.

Any ideas or suggestions? All inputs will be appreciated!

Thanks and regards,

wilsonhp
 

·
On the Super Mod Squad
02 Pair: 3.0 VDC Wag & 2.5 Limited Sedan
Joined
·
26,215 Posts
PCV last changed, and if changing now make it a OEM,

_______


EGR is also asymmetrical like the pcv.

__________

engine grounds. = something on that side of the engine.

http://www.subaruoutback.org/forums...13210-engine-ground-location.html#post4662186

________

did anyone look at fuel trims from side to side. ?

did anyone pull the plugs looking for oil, like from all the orings or a valve cover? ...worse would be coolant.

what are you running for spark plugs. (hopefully NGK double plats). and the 6 were installed at the same time,
and not 3 and the other 3 forgotten about. (such things have happened on this board = 3 old NGK, 3 new champion turds...that engine was tossed in the scrapper eventually).
 

·
Registered
01 Outback LL Bean
Joined
·
2,428 Posts
Given the quick onset of the issues I would not suspect plugs.

A stuck EGR isnt a bad idea. Although that isn't a typical problem with these cars, this engine has a lot of miles.
 

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #6
Thanks for all the quick replies.

Answers:
PCV changed @271K miles (2014), OEM
Fuel pressure not measured. can we measure both sides independently? I think I have a spare fuel filter I can put on, just in case. Is the FPR the concern here?
Fuel trims not measured, or at least not recorded. Mechanic used Snap-On scan tool to read codes. Does the engine need to be fully warmed to read these at idle? Worried about killing cats idling.
Plugs NGK PLFR6A-11. All replaced at same time when head gaskets done. (I supplied them and got the old ones back for inspection.) Not pulled/inspected as part of this effort.
Passenger side airbag was recall related replacement.
Everything is dry underneath. No obvious/significant oil or coolant leaks. Mechanic said he checked/snugged up all the hose clamps underneath while he was there. All the various gaskets were replaced in the head gasket effort.
How do we test for a stuck EGR? Where physically is this located?

What does this lead us to?

Thanks again....
 

·
Registered
01 Outback LL Bean
Joined
·
2,428 Posts
EGR is behind driver's side head. Maybe disconnect from intake and plug intake.

Is there anyway you can get a scanner on it to measure fuel trims?

Hard to say what it's doing without any scanner data.
 

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #8
The scanner i have available is a ScanGaugeII.

I connected it, started the car, confirmed again the P0302/4/6 codes. When i first started the car, the CEL was off (!) but it came on flashing within a minute or two.

I then accessed the FRZ data available when the codes are set, via the SGII. Data is captured per PID, with some labelled as to the parameter. Most PIDs queried had no data recorded/displayed.

PID definitions looked up at: obdcon.sourceforge.net

Here's what I recorded:
PID02 = 0304 (Look up gives "FREEZE DTC". I think this means that the PID data to follow relates to the P0304 failure code. Presumeably there is a similar set of data below for 0302 and 0306, but I didn't want to keep the motor running any longer than I had to...)
PID03 = 0202 (don't know how to interpret the 0202, but the label is "OPEN LOOP". makes sense as the engine wasn't warmed up.)
PID04 = 19 (Engine Load, idling in park, RPM approx 1400)
PID06 = 83 (Look up gives STFT Bank 1)
PID07 = 80 (Look up gives LTFT Bank 1)
PID08 = 66 (Look up gives STFT Bank 2) and this is my problem bank
PID09 = 80 (Look up gives LTFT Bank 2)
PID0B = 29 (Label says MAP, value converted to 5.9 by SGII)

nothing more had data stored up to PID20, and no PIDs that i could access higher than that-probably because I didn't know how to go there...

From the lookup source listed above, it looks like the recorded values are more toward the lean side rather than the rich side. Maybe i missed the "-" sign before the values when reading the SGII?

Also, during the 5-10 minutes the car ran while I played with the scanner, it ran progressively rougher, and the smell got progressively worse. Looking under the car, no obvious overheating of any of the 3 cats. It was sounding progressively hotter, but the coolant temperature wasn't increasing abnormally, per the gauge on the dash.

Trying to judge under-hood temperatures by feel and smell, I think the drivers side was warmer. The foul smell was all at the back of the car by the exhaust. The exhaust sound became more of a "putt putt putt" tone the longer it ran.

At one point, I shut the motor off and a few seconds later attempted to re-start. It took much longer to start (more cranking time). MAYBE felt like it was a little flooded the way it responded.

If I'm correctly interpreting PID06 and PID08 as the short term fuel trims for banks 1 and 2, is the difference between 83 and 66 significant? I assume that large of a difference would be. Are these "normal" values? The question then is "what is causing the difference?"

Any more thoughts? Where do I go from here, based on this data?

Again, many thanks and regards,
 

·
On the Super Mod Squad
02 Pair: 3.0 VDC Wag & 2.5 Limited Sedan
Joined
·
26,215 Posts
if the scan gauge 2 data is not enough:

maybe get a $10 Vag cable if you got a windows laptop handy, and put free rom-raider on it and get some 1 minute logs

this thread has logs and discussion using rom-raider as related to a problems in a 2004 US spec H6

http://www.subaruoutback.org/forums...e/462009-2004-h6-ez30-stalled-stop-light.html

rom-raider is cool as it makes recorded logs of a wider bunch of parameters then one person can think of. so more then one person on here can look at them and maybe see anomalies that are not easily apparent.
 

·
Registered
01 Outback LL Bean
Joined
·
2,428 Posts
WilsonHP. Can you show live short and long term fuel trim with the engine running. The freeze data does not appear to show fuel trims. Possible because it was in open loop when the code set.

Wait until it is in closed loop then provide fuel trim data.
 

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #11
Sorry. No laptop, Windows or otherwise... No way for me to get live data. Don't know if mechanic's scanner can output live data.

Were PIDs 06-09 not fuel trims?

Since i can't produce more data for you to review, I'd like to at least summarize what I've learned so far, with your guidance, for inputs to my mechanic to start acting on on Monday.

1-check long and short term fuel trims in closed loop mode. Use to determine:____________________________
2-Check for plugged/stuck EGR.
3-Check for plugged cat.
4-Check fuel pressure/FPR for possible leaks. Check both key-on/engine-on, and key-on/engine-off.
5-Check for engine grounds on the frame rails areas.

And change my fuel filter.

Anything else to add?

Regards,
 

·
Registered
'08 3.0R Limited Sedan; '09 Outback 3.0R
Joined
·
152 Posts
Here's my journey from when I had 3 misfires all on one side of the engine (just like you): http://www.subaruoutback.org/forums/showthread.php?t=421889

Misfires on one side of an engine usually point to something on that side: cam timing (variable valve oil control solenoids freezing) or timing chain.

In my case, one of the two timing chains had jumped off by a tooth due to either low oil pressure or a failing chain tensioner. You can debug this by running a compression test on all 6 cylinders and comparing the relative results. Unfortunately, your misfires are on the driver side cylinders which are harder to get to.
 

·
Registered
01 Outback LL Bean
Joined
·
2,428 Posts
Interested to check fuel trims to identify if you are dealing with a rich or lean condition. If it is a rich conditon I would suspect spark or compressing issue. If it is lean I would suspect fuel delivery.
I am suggesting getting this data becuase it is pretty easy to obtain. I suspect next steps will lead to a compression test. I suspect, similar to the previous post, a timing issue. Compression diagnosis is a sort of a pain on this motor so do the easy diagnosis first,fuel trim and fuel pressure then moving to spark test then to compression test.
 

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #14
Started the car briefly yesterday to move it within the garage. Check engine light was steady, not flashing, and the car ran perfectly smoothly. Not rough/misfiring. Didn't smell bad for the minute it was running.

I have no doubt had I kept it running longer, the light would have started flashing and the rough running would have returned.

Does the fact that it ran smoothly, with the steady light, indicate that the timing is still intact? I.E. that the chain hasn't jumped a tooth?

And then, if it does indicate timing is OK, what could be coming into play as the car warms up to begin misfiring?

I still plan to change the fuel filter.

It's going back in to my mechanic this afternoon for deeper diagnosis... I'll print out this message thread for him as reference.

Thanks again for all the learned assistance.

Regards,
 

·
Registered
03 H6 OBW & 06 WRX Sportwagon
Joined
·
17,625 Posts
wilson, if you own a smartphone, I 'think' the Torque app with an ELM327 BT adapter (cheap) will let you monitor live data.
 

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #16
Some diagnostic information from the mechanic today.

No live data available for short term fuel trim from the SnapOn scanner. But he did say he saw:
Bank 2 stft -18 to -25
Bank 1 stft +5 to +7
while watching it run closed loop.

is the correct interpretation that the computer is trying to pull fuel from bank 2 to correct a rich condition?

The feeling now is that it is related to the coils. He pulled all the coils and plugs from bank two.

in a couple of the coil connectors he found some apparent oil (4 and 6). Coil 6 was covered in heavily gritty oil externally as well (see pic, and this is after wiping it off a bit before removing).
The attached pic shows inside of the coil connector, showing the apparent oil.
I'm assuming there is no oil internal to the coil that could be leaking into the connector from the coil it self. (anyone sectioned a coil? are they fully potted with epoxy, as they appear to be? see pic.) These appear to be weatherpack type connectors so its curious as to how the oil got in there....
It also appeared that plug 6's insulator was slightly darker (sootier?) near the center electrode than those of the other two plugs, indicating recent misfiring. Note-the sticky wetness on the external plug end is dielectric grease, there was no oil in/on the boots as they were all replaced last year during the headgasket rebuild.

Could this oil in the connector result in the same type of misfiring as oil in the plug boot can cause? Could this be my problem?

Next step, tomorrow, is to remove the coils from the other side, clean them all up completely (including inside both the coil and harness connectors) and swap the coils side to side. The theory being if it is related to the oil, there should be no more misfiring anywhere. And if there are bad coils, then the problem should move to bank 1. If they are bad, replacement is in order.

Testing of the EGR via the scan tool showed it responding as expected.
Testing of the FPR by removing the vacuum hose from the FPR and pressurizing the fuel system by cycling the key to the run position multiple times resulted in no fuel coming out of the FPR nipple. Hopefully indicating no leaks in the diaphram (it would have come out of the nipple under the pressurization). Actual fuel pressure was not measured.

Further thoughts?

Thanks and regards,
 

Attachments

·
Registered
'03 Outback H6-3.0 Black Granite Pearl, base model with cold weather package and cloth seats.
Joined
·
1,061 Posts
Discussion Starter #18
I'm not the one doing the actual diagnostics and work. My mechanic is. And he has satisfied himself that the timing is intact. He is also the one who did the head gasket replacement/rebuild, so I trust his judgement. You all have provides some great insights and ideas to evaluate, and I very much appreciate that, but he's responsible for getting this fixed and reliable again.

My goal here is two-fold. One-to understand in my engineering mind what could be causing the issue and be able to discuss it intelligently with him, and Two-to help feed him insights from a larger group of experts.

He doesn't plan to remove any more plugs. The plug/coil removals were part of his diagnostic process. And I'm completely OK with that. The next phase of diagnostics is detailed above, and it is one he and I jointly agreed on. Does that plan make sense?

I'd really like someone to comment on the oil in the connectors as a possible cause of the misfires, the stft results, and hear any comments on the possible source of oil in the connector being from the coil itself.

I've read here over the 14 years that I've been a member, that the coils are super reliable and rarely fail, but when they do fail-what does the failure look like? What are the chances that it is actual coil failure on multiple coils on the one side? Doesn't seem that likely to me, but I'm no expert here so I'd like to validate that idea. I can understand #6 failing, given the environment it lives in (heat over a cat, oil/dirt coated, 357k miles.....) but multiple coils at the same time? Or is it an age thing? They are all the same age so is it reasonable that they will all start failing at approximately the same time? Without understanding the failure mode(s), I don't have enough information to be certain. And I've never seen mention of oil in the connectors, whether or not it caused an issue.

How do I factor in the fact that on cold startup it runs fine, but as soon as it starts to heat up, the misfires start. Is it the transition from open to closed loop and the computer kicks in and starts trying to change things? And after a hot soak, re-starting has a lot of misfiring and is difficult-because it is already closed loop?

Your comments continue to be appreciated!

Regards,
 

·
Registered
03 H6 OBW & 06 WRX Sportwagon
Joined
·
17,625 Posts
here's one idea, after it is warm enough to exhibit the problem, do a battery disconnect 'reset'.

fire it back up - if it is running fine, that does kinda point to a sensor that is feeding the ECU an out-of-threshold signal. If it runs just as poorly, could be something more - uh - 'mechanical'/physical.

it's worth noting that each side's fuel rails have their own 'damper' or pulse regulator gadget (? I think ???)- maybe something weird with that?

the odds of 3 coils being bad is vanishingly tiny. easy to swap around to another position with the other bank.
 

·
Registered
01 Outback LL Bean
Joined
·
2,428 Posts
When the car is cold it uses extra fuel and lower spark timing. As it warms up it cuts fuel and advances timing. Cold start up is just masking your issue.

It is not possible for all of the coils to fail at one time. It is possible for there to be some sort of wiring issue causing all coils on that bank to not fire but I doubt that is your issue. Swap coils if you want from side to side.

The amount of oil residue on the coils is typical for a car of that age and is not your issue. Oil on the coils is due to leaking valve cover gaskets. Did you say the HG were replaced by this mechanic? They should have replaced the valve cover gaskets as well.

You have a rich fuel condition which is the result of incomplete combustion. You have the condition with no load on the engine, ie, while idling. I would not suspect an ignition problem unless the coils were completely dead. The most likly cause of incomplete combustion is a compression issue and since it is affecting the entire bank I would suspect timing, or excess fuel.

Swap the coils 1 with 2 and 3 with 4 and see if anthing changes. If not, next do a compression check on cylinder 2. If not then do a proper fuel pressure test and an injector flow test on the engine.
 
1 - 20 of 35 Posts
Top