Have you cleaned your throttle body and Idle Air Control Valve? Not sure if this could cause uphill surging or not but it would definitely effect idle speed.
(I believe all '01s had?!) Some other details would help: miles, turbo or N/A (trim), 4EAT or 5MT.

O2 sensors generally go about 100,000 miles before problems. My past 2 '00s had theirs replaced under warranty before I got them and I didn't quite get to the point to replace on the second. I did have to replace on my '04 at about 90,000 b/c the heater circuit went out and threw code.

The do have TPS. I'm not sure how to test as mine have not yet gone bad. I would say it is a definite possibility but then the cars are pretty good at throwing codes (such as the knock sensor you mentioned, crank and cam sensors, misfires for various reasons, etc.).

Have you run it a lot with less than 1/4 tank of fuel? The fuel pump could be on the fritz as well. I got the surge/buck a couple of times after almost running out of fuel but didn't have to replace the fuel pump. Considering the cost and others' experiences, I'll probably do mine about 150,000 mi as preventive maintenance on my '04.

 

The throttle body has been cleaned (at least the best I could do without taking off, should I remove it completely to clean it?) but not the IACV. It's a 2001 2.5L N/A, 5 speed about 140,000 miles. Normally I run the car to about 1/4 tank and the fill it but I never let it get below 1/8 even at the worst of times (I have never seen a low fuel light) today I was driving it and had about a 1/2 a tank of gas and the things was acting horrible. I got a cheap code reader and a Haynes manual for this thing tonight at Autozone and I will check to see if any codes are still stuck in the ECU and I guess I will yank the O2 sensor and test it tomorrow to see what it does.

 

Just checked the codes and its giving me a p301 ,302, 303. Don't know if this is accurate though as I forgot to connected one of my plug wires when I was changing the plugs and started it up and it gave me an error then so it might be the same code. I am going to clear the codes and check my coil pack and the plug wires for a break.

 

Throttle body is OK to clean on car because it still has cable and is easy to open throttle to clean backside and inside with car off (versus drive-by-wire throttle control with electronic module - haven't done one of those on a subaru yet). I cleaned mine last weekend. You clean inside some with the car off and crank it and spray some in with it running too. Be sure it is safe for electronics and not just "carburetor cleaner". I did try to avoid spraying some on the shaft joint at the TPS. It's easy to remove the IACV and clean it and the seat area off too. It's on top of the throttle body with blue sticker, electrical connector and 2 small screws mounting it. Be careful not to drop the screws or damage the o-ring seal if you decide to clean it too.

Sounds like coil and/or wires is a good next step. Be sure and use a light coating of dielectric grease on the boots and connections. Moisture as well as heat often times make some noticeable difference with coil problem symptoms.

Don't forget fuel pumps can cause symptoms with no codes.

If you're still on the original O2 sensor I would plan on changing it when you can afford it rather than waiting for it to go completely bad.

To be clear the performance issue was there both before and after the dealer changed the timing belt and other parts and symptoms didn't change noticeably?

 

The symptons were there to a point before the timing belt change but seemed to have gotten worse over the past month or so (been about 2 months since the TB was changed) it didnt like hills before the change but the stuttering is a fairly new development.

 

Just checked the resistance on the coil pack the secondary looks ok but the primary isn't showing up at all does that mean its dead?

 

When you test the primary, does the meter show infinite resistance i.e. it's open or no resistance at all i.e you have continuity? I would expect the primary to be almost a dead short.

Since the car runs, albeit somewhat badly, I would suspect a fuel system problem, since a misfire or spark problem would likely throw a code.........

 

When I test it I get 0L ohms on all the meters I have tried 2 cheapy pocket multi-meters and 1 pretty good one that I use for everything at work, all of them at giving me 0L. But when I go to the secondaries I get about 13kohms which is in spec according to the Hayens manual. I tried doing a continuity test and I get nothing. But if I test between pins 2&3 I get a resistance of 1.3Mohms don't know if that is supposed to happen or not.

As far as the codes it actually did throw the coded for misfire on 1,2,3 but im not getting a CEL just the codes from the reader so they might be old.

I guess I can get a fuel pressure gauge to test the fuel pump but I didn't want to buy more tools if I didn't have to.

 

I don't know what 0L means. Is that what you see when the probes are not connected to anything? What is displayed when the probes are touched to each other? There may be a diode that blocks the current in one direction or it may be a transistor junction. Try reversing the probe leads.

Regarding the CEL, perhaps it's burnt out or disconnected. Erase the codes from the ECU, take a drive, then do a code reading. That will give you a real view of whats happening.

For a pressure tester, it's possible to assemble one from fittings, fuel line hose, hose clamps and a 0-100 psi water gauge for about $30. How much does your local mechanic charge an hour?

 

Ok just changed the plug wires restarted the car with the OBDII scanner connected and I get all 4 cylinders misfiring and a P0172 error which tells me bank #1 is too rich. Is this because of the O2 or could it be the MAP/MAF which ever this thing has. The idle did level out some and the resistance on the new wires is about 8.5 ohms vs the 18ohms I was getting from the old ones.

 

i would go for the coil pack next.

 

well I think Im going to try and test the O2 sensor on the car just to be safe. hate to buy a new coil pack at 200$ if its actually ok and the O2 is the thing thats bad. O2 is way cheaper

 

The O2 sensors don't actually do much of anything until they have warmed up ~5 minutes then the ECU goes from an 'open loop state' to a 'closed loop state' and starts to regulate the fuel flow closely.

I have a random problem similar to yours but I think yours is related to fuel pressure and the electrical issues with misfires. The Haynes manual details how to do a stimulus/ response test for the fuel flow in detail without a pressure gauge.

If you pull apart the O2 connectors you should see the O2 sensors code fault on the ECU.

 

wow, $200 is pretty pricey. sounds like a good plan.

 

dharmashiv ... sweet thanks for the info I just got my Haynes for it and haven't been completely through it yet. I will try the stimulus/response test next chance I get. I did notice that after clearing the codes and running the car for about 30 miles or so it seems to have gotten a little better, but its still not perfect. Seems to be driveable unless I kick the AC on then it wants to bog down to almost a stall if I'm sitting at idle. I was going to try to just disconnect the O2 and see if it would give me a code but I didn't know if it would run without the O2 plugged in. May have to wait a days or two to test it out between work and the storms we have been getting I haven't had much time to turn wrenches.

On a side note it seems that Autozone and Adavance Auto (the two major parts stores in my area) have a vaccum/pressure gauge that will do fuel for about 20 bucks. But if haynes way of doing things is easy enough I may just go ahead and save the 20 bucks.

 

Just an update. I think I may have figured out the problem (or at least a part of the problem). Oil. The seals around my spark plugs are leaking a bit. I didn't think much of it because I have had a number of cars in the past (not subies) that have had leaky valve pans and never caused any real issues aside from a dirty engine and the smell of burnt oil. I was reading another posting somewhere (maybe here) that talked about the oil getting into the silicone boot of the plug wire and making it conductive. So I have new wires installed (about a week old) and the performance is slowly going back down to how it was before I changed them, so I'm thinking this may have something to do with it. So if work allows for it I hope to be changing the valve cover seals this weekend and apparently getting yet another set of plug wires (gad I got the cheap ones).

 

Let me chime in here ..... I was chasing a neighbors 2003 Outback for a problem that sounds similar to this.

Cold start ... all fine

Few minutes after warmup ...... BAD BAD IDLE ... stalling about 300 rpm

After warmup .... All was fine once more until the next cooldown.

After MANY dollars at a dealer ( front O2 sensor, TPS, Engine Temp Sensor, Plugs, Wires, Throttle body Cleaning ), I asked him to let it with me 1 overnight. In the morning, I put my PC attached OBDII reader on it and monitored: MAP, TPS, Ambient Temp and Short Fuel Trim.

As on schedule it started to act up about 5 minutes after startup. Only thing I noticed on the display was the MAP went from 9.8 to 14 and bounced around there.

So it's either a defective MAP ....... OR the idle solenoid closing down, forcing the manifold pressure to go up. Now they CLEANED the throttle body, so I shut it down and removed the IACV ( idle solenoid ). It looked fine, felt the inside opening and there were no sharp edges or burrs. Pintle looked clean, but a little too dry. So I put a thin coating of silicone oil ( could have used brake fluid as that's a good lubricant for seals ). My thought was it MAY be sticking during warmup. Put it back together and called him.

Told him if this doesn't fix it, either the MAP is defective ...... or the IACV. Neither is cheap. That was last Sunday.

This afternoon he was on the answering machine when I got home ....... told me I was a Subaru god of a mechanic ........ Said it never ran so well.

I think a lot of that may be in his mind ..... because what I did was fix a 5 - 10 minute warmup problem he had.

I searched .... the IACV is 22650AA192 and lists for $345 from Subaru. If I needed one, I might just go to Ebay and buy an entire used throttle body with one on it. I see the entire throttle bodies for under $100 with the TPS and IACV at a minimum on them.

 

I searched .... the IACV is 22650AA192 and lists for $345 from Subaru. If I needed one, I might just go to Ebay and buy an entire used throttle body with one on it. I see the entire throttle bodies for under $100 with the TPS and IACV at a minimum on them.

Before anyone goes out and buys a new IACV.

There is a IACV cleaning procedure on nastysock. Peaty also has the endwrench article on ISC (as they call it) cleaning on scoobymods.

I just got back from a 2 week holiday, my wrx sat the whole time. It had a really bad surging idle and came up with a p1507 cel. I did the iacv cleaning procedure and so far so good.


dm

 

Thanks for the info I will give that a try. I just changed the the valve cover seals along with the spark plug well seals. That seems to have taken some of the high speed missing issues out. Which I still found to be really odd. But I guess I have some more work to do after I get my CV axel changed (ripped boot)

 

So this morning after changing the belt tensioner on my AC belt (stupid me broke it cause I forgot to loosen the nut). Anyway, I got the belt tight on the the AC and after little while of driving, I have noticed that the engine seems to bog down a lot when the AC compressor kicks on. Its intermittent so I can't be sure but was wondering if anyone else has had any issues with the AC compressor going out and causing this sort of issue.

 

Sounds like my problem is more common than I thought. My valve cover gaskets have been leaking for awhile, and I suspected that might have something to do with the ignition stumbling at low rpms on uphills in extreme heat.

 

Oil is not a good conductor of electricity.

A/C on is mechanical load. So yes if you have a problem it will be there more with the A/C on.

 

The 2003 Outback above ... CAME BACK ....... and NOW I'm CERTAIN it's fixed. Here's a thread I just posted ..... and since BOTH you Dave and the OP have manual transmissions, you might look into this:

http://www.subaruforester.org/vbulletin/f88/neutral-switch-78655/#post870591

 

Update: new plug wires cured my problem. 600 miles into a 900-mile road trip (all highway), the sputtering was getting worse in 5th gear when hitting a rise in the road. SO, I stopped and wiped the oil off the front spark plug boots; I couldn't get to the rear two without moving the air intake on one side and the washer fluid reservoir on the other. But, while there was some oil from the leaking valve cover gaskets, it wasn't nearly as much as I had expected, and definitely not enough to saturate the rubber boots.

A little further down the road, I found an AutoZone, and decided to just try putting new wires in (after seeing how easily it could be done on the 2.5 N/A motor). One set of Duralast wires for under $30, and 10 minutes of work, and I was back in business. It runs like a champ now, even 400 miles later.

One note (and maybe the source of the original problem): One of the wires came apart when I took it off. The inner connector separated from the boot and the wire at the distribution block. Luckily, it was up top and easy to pull the metal connector off. I guess it's possible that it was already coming apart, and the connection was iffy.

Anyway, pull the wires at the top and make sure they're not coming apart. On a side note, does anyone know for sure if the motor needs to be pulled to replace the valve cover gaskets? That's the only major issue with my Foz now.

 

I recently had an issue of stuttering and shaking at idle on my 2003 RS that kept getting worse. Went to replace the plugs and found that two of the plugs had oil all over the top and there appeared to be no electrode left (crappy bosh plugs the issue here).
Replaced the valve cover gaskets (easy as changing the plugs except you have to take the battery out to get more room) and replaced the plugs with NGK and my idle is nice and smooth now.

 

I just purchased an 01 Forester / manual / 115k and it recently developed the same problem. It had a new catalytic converter (not sure about O2 sensor) 1k miles prior to my purchase. Problems since fuel filter (not OEM Subaru part) a few hundred miles ago.

Car runs fine, 30mpg, then will go 'whoomp', 'pop' stumble and finally smooth out again and mileage drops to 15mpg. Idles at 500/700 running good or bad.

Would like to take to dealer for 120,000 mile service to be checked and for records, but would like to have the 'surprises' known/repaired first.

Fuel issues? Get OEM fuel filter? Is there another fuel filter in tank or somewhere that could effect this?

Hope you get your vehicle worked out, I love mine.

 

OP's problem sounds like a TPS failure to me... a lot like what my car was doing when mine was bad. Shifting erratically, and acting odd off throttle and at idle.

Also, 'erratic' problems can often be caused by a loose grounding wire. I found that out the hard way... so check all the grounds on and under the intake mani to make sure nothing is loose. :shrug:

 

Hey Guys, I hope you can help me with this.

Now, I have a 2004 Sti, but, I do have "the surge". It occured anywhere from 1800 to 3500 rpm, worse at 2400 rpm, BUT, only at VERY small throttle or on the over-run.

I first noticed this 18 months ago, but only when starting from cold, and it didnt seem important, but it has steadily got worse, and now occurs when warm.

I took scooby to a respected Subaru specialist, from whom I originally bought the car, a 400 mile round trip, and they said nothing showed up, and suggested changing the air flow meter. That done, the car is smoother overall, but the fault is getting worse. The surging now occurs from 1800 to 3800 revs, still on almost closed throttle only as before.

Now this seems to me to be a throttle position related fault, can anyone point me in the right direction to cure it ??.

Please !!.

VBR Chris Adams.

ps: the car is a 2004 Forester Sti. ( 2.5L Turbo, Japan and New Zealand model ).

 

I will update this thread as my diagnoses progresses.

After fixing many a Subaru for idle problems ... now I have had a phantom and not everyday one for a month or so. At first I thought I had bad gas. Car always starts, but from a cold start, after a few miles, if I stop, idle was very rough, I sometimes had to step on the gas to keep the engine running and the exhaust smelled REAL BAD. If I let the car set a few days, it seems to take a longer warmup time before I have a normal idle.

With an OBD reader/monitor I would only see the O2 sensor go nuts ..... that's why I started there. Was it the Cause or Result? Turns out it was the result. Also I had PENDING misfires on all 4 cylinders. P0301, P0302, P0303 and P0304. Never went over the criteria for a misfire, but always pending.

I tried a new front O2 sensor, Engine temp sensor, swapped AIS, TPS, Air Temp Sensor, Map... All no help. Neutral switch was BAD ... always open telling ECU I was never in gear. Again no help there.

Over this weekend, I though about it. I had removed the EGR and cleaned it when I did my engine rebuild last year. ( http://www.subaruforester.org/vbulletin/f88/every-2-5l-interference-engine-64410/ ). Being an ELECTRONIC EGR with a stepper motor, there is no way to see or test if it is open or closed. And at almost $200, I wanted to be certain. So I took an aluminum can, cut it and laid the EGR gasket on top of it. Then I cut the outline and used a hole punch to ONLY punch the bolt holes. Put it in and started driving.

Since Sunday, my bad idle has not returned. I DO get an expected CEL P0400. Happens during the EGR test. About 5 mins into a start, while decelerating ( fuel cutoff time ) the ECU opens the EGR and expects a DROP in the MAP value. With my aluminum plate in place, this obviously does not happen.

So later today or tomorrow, I will bite the bullet and order an EGR. After further searching the internet, Mitsubishi's use a similar valve and their boards are full of bad idle problems caused by it. The Mitsubishi versions however is more than $400 .....

Something to think about if you have any Subaru with this type EGR and an intermittent bad idle problem.

Item #11 in the following Illustration:
http://opposedforces.com/parts/fore...11/type_3/intake_and_supply_system_turbocharger/intake_manifold/illustration_1/

Pictures attached ... 1st is the EGR on the engine
2nd is the valve, gasket and cutout blockout aluminum from a can
3rd is the valve with the gasket then the blockout and bolts in place ready to be put back on..... and smooth my idle ......but give me the P0400 code.

 

:
Well, everything was going great until i go to put the egr back in and pulling out of my garage i get p1494 and p1496