Forester situation

[2002 FORESTER L]

Hello all,

 

I am having a Forester problem. Engine related. I was getting cel on and codes p0130 and p0420. I changed both forward and rear O2's. Then the fun started. I replaced the front O2 with a Bosch. I read these don't work too good with the CPU. I started getting terrible idle problems, engine surging, new codes (p0065, p0171 and p0130). I then changed the Bosch O2 with a Denso. No change in problem. Brought car to dealer, they wanted to change the O2 and then the CPU to the tune of 1300.00. I did not do this. I did change the front O2 again. I installed a Subie O2 and returned the Denso. No change.

I then purchased a used cpu and installed. No change.

I have disconnected the front O2 while running and it did not change the idle problem. Also disconnected the MAP, AAIS while running and no change in idle problem. I also swapped the ICU with another and checked it for operation and cleaned out the throttle body and ICU area. The throttle body was very clean, which was nice to see. I checked the AAIS lines for clogs, they are fine. Have a friend who is a Subie tech bring it to work at a different dealer and did not see anything wrong with exhaust or engine lines, stated it is a weird odd problem.

I just changed the coolant temp sensor, pcv and fuel filter. No change. I also re-installed the original CPU.

 

These are the real time scan readings:

RPM: 700

 

MAF: .20 (even thought it does not have one?)

 

MAP: (HG")  5.1-5.4

 

COOLANT: 185

 

IAT: 136 (actually 50 degrees outside)

 

STFT: 37.5 (?!?)

 

LTFT: 14.8 (?!)

 

O2S12 (V): 0.015 (I assume this is the rear O2)

 

O2S11 (V): 2.251 (I assume this is  the front O2)

 

OBD2 STAT: CA

 

FUEL SYS 1: CLSD

 

So those fuel trims look way messed up.

 

At approx 2500 RPM:

 

STFT: 0 (ADJUSTING BETWEEN - AND POSITIVE VALUES WHILE LTFT ADJUSTS)

 

LTFT: 27

 

The higher idle fuel trims look like a vacuum leak?

 

To sum it up, car ran fine before O2 change. Now it has a surging idle and stalls after re-starting after driving. It seems that all these things are happening when it goes into closed loop (warmed up). When I first start it cold start, it quickly goes to cold idle RMP then drops down for a second and comes back up until it drops to hot rpm while warming up. The cel light seems to come on at around 2200 rpm after driving for approx. 20 miles. I am now getting p0130 and p0171. Was going to perform a smoke test to look for vacuum leaks next.

 

Any thoughts...

 

Thanks for reading,

Mike

 

[idosubaru]

Vacuum leak.  Check all air hoses around intake and recirculation/PVC hoses

"ran fine before the O2 change?" - but still had a check engine code for oxygen sensor so it wasn't entirely symptom free. 
did you buy this thing from a wrecked and drive it 500 yards before doing this work or have you owned it since new?  what are the lengths of time it was "asymptomatic" and "driven with an O2 code showing"? 

[2002 FORESTER L]

idosubaru,

I am the second owner, my folks the first so no it was never in an accident. Car has 125k. The cel was on about 2 months before changing the O2's. I changed the O2's in August, it has not been driven much since. I have changed the timing belt, tensioner, water pump, belts, cam and crank seals and t-stat. The rear O2 was changed as well. This was all done at approx. 85k. I have owned it since 2012. I will try my home made smoke test to look for vacuum leaks. Just weird it ran fine before changing the O2's. Wish I just kept clearing the code and left it alone. Although I suppose whatever is causing this issue would eventually start to make it run like it is. It is an automatic tranny as well. Was looking into fuel pressure regulator as a cause. Thoughts on that? Engine is definitely hunting at idle, like a leak.

Thanks for the reply,

Mike

[GeneralDisorder]

MAP sensor readings are whacked. Being the MAP is an absolute pressure sensor, it should read 29.92 in/hg with the engine off. Subtract the usual idle vacuum of a healthy engine (18 to 22 in/hg) and the sensor should read ~10 in/hg at closed throttle idle. 

IAT is pretty far off but if it was idling for a long time the under hood temps might have got it there. Still - with an 86 degree delta over ambient I would question it. Checking it in free air would be wise. 

Check the MAP sensor with the engine off against local barometric pressure. And then with a shop vacuum gauge. 

Dealers and even most shops are unable to handle this level of troubleshooting for the most part. You may need to ask around at who can actually do real diagnostics and understands EFI live data. Usually best to find someone that is well versed in performance tuning. And yes this will cost - their time is valuable. 

[2002 FORESTER L]

GeneralDisorder,

 

Would a vacuum leak give a out of limits MAP reading? I will see what I get for a MAP reading with the engine off. I was seeing 6.79 on another scan tool, that does not live stream.

 

My understanding is that if fuel trim increases with RPM then usually vacuum leak, if opposite usually fuel delivery. I am thinking vacuum leak as others have pointed out.

 

Going to try a smoke test. Do you know if there is a check valve in the line from break booster to intake? I disconnected it and it was holding vacuum as I could hear it release when I disconnected it. This is where I am planning on connecting the smoke delivery tube.

 

I am going to search the area for a shop that does have the skills needed for this problem. I am still amazed at how this manifested itself from an O2 change.

 

Thanks for replying,

Mike

 

 

[GeneralDisorder]

If fuel trim decreases (from high positive trim) with RPM, it's a vacuum leak. If it increases it's a fuel delivery problem.

Those are the rules for MAF cars. They only apply loosely to SD cars like your Forester. It could have a massive below-throttle vacuum leak I suppose but it would be obvious. More likely the MAP sensor has failed. 

GD

[2002 FORESTER L]

GD,

 

I did disconnect the MAP and there was no change in what is happening. Just wondering if this would rule it out.

 

Mike

[GeneralDisorder]

You would have to disconnect it, then clear all the learning from the computer to see if it makes a difference. What's the MAP live data read with KOEO? And then what does it read with the sensor disconnected while running?

GD

[2002 FORESTER L]

KOEO - Reads good, same as current barometric pressure (30.00 on the current weather and was reading 29.8 on reader). Have to check tonight with it disconnected with engine running.

Thanks,

Mike

[2002 FORESTER L]

MAP at idle disconnected, 13.8....connected at idle the same. KOEO is reading 29.8.

 

Going to picknpull and getting a MAP and an AAIS...see what happens.

[GeneralDisorder]

13.8 at idle means you have about 16 in/hg of idle vacuum. Acceptable idle vacuum range is typically 18 to 22 in/hg and a non-turbo Subaru EJ engine will usually be around 20. Given that the KOEO sensor reading coincides with barometric pressure I would be looking for why the engine vacuum is low. Obviously there can be many reasons for this - possibly a vacuum leak or retarded valve timing comes to mind at the top of that list.

GD 

[Ferret54]

Does it have an EGR? I have seen a sticky EGR cause this. 

Wrote an entry here many years ago when MY04 Forester had similar symptoms. Also had pending misfires on all 4 cylinders. 

[2002 FORESTER L]

GD,

 

Seems that when I did my scan test last week the MAP was at 5.5, now it is 13.8...just wondering what that could mean.

Retarded valve timing...would this be a timing belt issue? I can put a timing light on to see where the mark is. It runs fine though at speed.

 

Ferret54,

Just looked into the EGR...it does have one, going to see about the line form the EGR to the head and take it off to check for build up.

 

Going to try spraying starting fluid around for vacuum leaks.

[GeneralDisorder]

5.5 seems too low to even be possible - I've never seen an engine pull that much vacuum. Remember that the absolute sensor reads ~29.92 at sea level and the most the engine can pull would be about 22-23 if it were absolutely perfect and new and idled smooth. So that's about a theoretical minimum sending unit reading of around 7. I would be suspecting that sensor or the wiring/connector with that much variation in the readings. 

Use water to check for vac leaks. Take a water bottle, poke a hole in the lid, and sprinkle it around. 

GD

Edited October 23, 2018 by GeneralDisorder

[Ferret54]

Here’s a link to my EGR write up on the Forester board years ago....

http://www.subaruforester.org/vbulletin/f88/low-idle-stuttering-erratic-jerking-various-rpms-76450/index3.html#post887065

[2002 FORESTER L]

GD, Ferret54 and idosubaru,

 

Welp, its fixed. I went to the boneyard and pulled a MAP and a AAIS from a '01. Only one there and nobody had been in the engine...phew..total cost...6 dollars! I put them in and fired her up...low and behold idled right up and no RPM drop or funny idle. Test drive went good, seems to have more power. No starting problems after it is hot. I cleared the cel and it did not come back on. All symptoms cured! So 6 bucks and about 20 minutes total time to take off parts and put in. Thanks for your input and following along with this situation. My next issue is burping the coolant system since I put in a new coolant temp sensor. It is not overheating, but the fans are coming on. Probably not overheating because it is cool out. I am going to buy one of these:

https://www.amazon.com/Lisle-24610-Spill-Free-Funnel/dp/B001A4EAV0/ref=as_li_ss_tl?ie=UTF8&linkCode=sl1&tag=brucey-20&linkId=a3d02ca1616b44c9f2a4d474cdbfb3d6

Probably drain radiator and then refill with this? I see so many 'methods' about doing this, so any advice would be appreciated. I did not have a problem with this when I did the timing belt and water pump...just a different spot the air is in?

My wife is happy again, so thanks again for helping with this. She was getting tired of driving my '91 740T.

 

Mike

 

 

 

 

[GeneralDisorder]

Yeah - I figured a bad MAP sensor was probably what you would find. The AAIS is really inconsequential as the engine will run fine even with that removed. Glad you found the problem.

GD