Jump to content
Ultimate Subaru Message Board

Recommended Posts

Yes there is all kinds of conflicting info out there on this one.  Rock Auto has sent me the wrong part more than once.

 

here's what AIP Electronics says:    1999 Subaru Legacy 2.2 Liters; w/OEM #22680-AA29A

 

They have a note that says this supercedes 22680-AA29 which makes sense since there was a recall on this part and SOA replaced them with another part.  So far their listing seems the most knowledgeable - perhaps more so than SOA

 

http://www.ebay.com/itm/NEW-MASS-AIR-FLOW-SENSOR-METER-MAF-FITS-2-0L-2-2L-3-0L-/400315740058?fits=Year%3A1999|Make%3ASubaru|Model%3ALegacy&hash=item5d34ad6f9a&vxp=mtr

Link to comment
Share on other sites

ok so the duct in my car that the MAF sensor fits into actually has a part # of 22680-AA29A so AIP has the right part for my car and

SOA gave me the wrong part # that's why it didn't work in my car.

 

However I am now pretty sure the MAF sensor is NOT the problem, see next post.

Link to comment
Share on other sites

Drive test, live scanner data, should have done this long ago.

 

Following, there will be three readings/numbers for each parameter; first, coasting w/no acceleration attempted; second, acceleration attempted below 2500 RPM and not successful; third, successful acceleration when finally get above 2500 RPM.

 

spark advance: 15, 0 to -4, 25

MAP: 8, 26

MAF: 0.8, up to 4.5, up to 6.3

STFT: negative to -23 (slight acceleration), small #s, small #s

LTFT: negative to -27 (slight acceleration), small #s, small #s

 

So MAF during the drive test looks fine to me.

 

Spark advance, 0 to -4 when malfunctioning during attempted acceleration, this looks completely wrong? recovers to 25 or so when normal acceleration resumes at 2500 RPM.

 

STFT and LTFT during low RPM slight acceleration, these numbers look bad, due to spark advance wrong?

 

So it looks like the ECU is screwing up the spark advance during acceleration below 2500 RPM this could be due to knock sensor (I have replaced this twice with the cheap China one), ECU screwing up, or is there any other possibility?

 

I'm certainly no expert, does this sound right? sometimes I feel like I'm talking to myself here on this forum.

 

Comments?

Link to comment
Share on other sites

Don't worry, we all talk to ourselves... ( I think)...

 

Try the fake knock sensor and see what you get.

 

I don't know much about the MAF sensor fitment because there were several different styles. Typically there are 4 pin sensors and there are 5 pin sensors. When ordering a replacement you have to make sure you get the one with the correct number of pins in the plug.

Link to comment
Share on other sites

So I installed the 564k ohm fake knock sensor last night. It was finally rainy today so I took that as an excuse to drive around.

 

In four different drives it accelerated flawlessly. I kept stomping on the accelerator at every chance and it accelerated smoothly from low RPM through 2500 with no change at 2500 RPM. Previously when I thought it was running pretty well, you could still notice a slight surge as it passed 2500 RPM.

 

With all the accelerating, I did hear it knock a little but only once.

 

I need to keep testing but I think this is it, a hypersensitive knock detection and response system. Since I tried two new ($15) Aisin sensors and they did not correct the problem, probably not the knock sensor. ECU must be hypersensitive to the knock sensor.

 

I did see a reference to this problem, a recall I think for 1998 Legacy where the ECU responded to the knock sensor incorrectly. Mine is a '99 but was probably one of the first ones, manufactured in June of 1998, so those Hoosiers might have thrown some left over 1998 parts (ECU?)at it.

 

I guess I could try a different ECU but not sure I want to get involved in that.

Link to comment
Share on other sites

Well the car was probably originally phase I since June '98 must have been one of the first '99s made, but the engine (installed Nov. 2014, all I know is used, ca. 135k, from a '99 Legacy) is definitely phase II.

 

Hmm, maybe that's the problem?  Phase I ECU with a Phase II engine?  Maybe I should see if SOA will give me ECU #s?

Link to comment
Share on other sites

There's no easy way to make that swap work.

 

Phase 2 has ignition control built into the ECU, has a 4 wire plug on the coil pack, and has totally different Idle control system built into the throttle body.

Phase 1 has a separate ignition module (igniter) mounted on the firewall with 3 wire plug on the coil pack, and IAC is stuck on the side of the intake manifold behind the #3 intake runner.

Which does your car have?

 

The manifolds are different so you can not bolt a phase 2 manifold to phase 1 heads. (and vice versa)

 

You can have a phase 2 short block with phase 1 heads and intake manifold. (Or vice versa)That's the only way that swap works.

But the ECU doesn't know/care which block is in there. All it knows is the electrical connections, which are different enough between phases that they don't interchange.

 

June 98 could make your car a 99 model year which would make it a phase 2 from the start.

Edited by Fairtax4me
Link to comment
Share on other sites

Yes my car is a 1999 model year (manufactured in June 1998), and I assumed it now has the phase II engine because as I understood it the phase I engines had the spark plugs exposed up on top whereas the phase II engines have them buried down below (thus the long tubes), mine is like this.  The IAC is on top/front of the throttle body to the right of the TPS and I have not seen anything that might be an igniter on the firewall, it has the ignition module at the top front of the engine, where ignition wires run to the plugs.

Link to comment
Share on other sites

yeah, likely to spew codes if a ground were really bad. But, some codes stay 'pending' unless they occur 2 or more times within X drive cycles w'ever.

 

might be nice to see what the fuel trims are if you can get frezeframe or live data. Some smartphones apps can do that with a $20 elm327 BT adapter i think. My old Innova scanner can get FF data.

 

related to other posts - if the car runs better in open-loop or immediately after a hard reset, there's likely bad data from an A:F sensor. or related sensor. O2, MAF, Knock, maybe engine temp, etc.

You need a rooted android with the "torque" app. On fancier cars you can even mess with tuning I think.

Link to comment
Share on other sites

  • 4 months later...

I wanted to share an interesting and surprising discovery with a timing problem that exhibited all your symptoms and thought it worth sharing as it was something that you wouldn't have thought of.

I purchased an Impreza that has an EJ204 with AVCS timing and would intermittently suffer terrible acceleration, rough idle but if you pressed the accelerator slowly, it would gradually accelerate as opposed to virtually nothing if you tried to accelerate quickly. After swapping out many sensors (MAF, IAC, Oxygen, TPS, knock) without success I too was running out of options. The isolation of the knock sensor or an ECU reset improved the situation for a period of time, then go back to the intermittent poor accelarion. I thought it may be fuel pressure or timing.

 

One thing I did notice was at cold start, a funny sounding exhaust note and always a slightly harsh vibration of the engine at idle, warm or cold even without this other problem.

 

I hooked up a software based scan tool and like every one else, no fault codes whatsoever and no real clues. Timing was however a little erratic and sometimes would jump around with the knock detection. Isolating the knock sensor made a difference but I had replaced this also. The timing indicated 15 degrees at warm temperature but when put in drive with the brakes on to load the engine, it would bounce around a bit and certainly give some backfires and was retarding the ignition. The only thing I did notice was the advance /retard timing of the right hand bank of cams was very slightly different to the left hand bank (this side has the cam pulley sensor) After some deep thinking, I decided to go right back to basics and check the cambelt auto tensioner and timing.

After removing the covers and crank shaft pulley and checking the timing, to my surprise, the right hand bank of cams was one tooth out on the timing! On top of that the hydraulic auto tensioner was not exerting enough force which compounded the problem of erratic timing. I am certain it had not jumped a tooth.

According to a label under the bonnet, the Cambelt had been replaced about 50,000k ago so looks like it was not timed correctly. The interesting thing to note is that the timing is derived from the left hand bank of cams (has the cam shaft pulley sensor) and so the ECU would not pick this problem up. In addition, because the engine has the AVCS, it was trying to compensate for this and if I ran lower octane fuel (91 instead of 95) the probelm would show up worse but most of the time would run ok.

 

Well, after, setting the timing correctly, replacing the cambelt auto tensioner, the engine engine idle is super smooth and the intermittent poor acceleration problem has been eliminated. Also the funny sounding ehaust note has disappeared. I was able to put the knock sensor back to original as with all the other sensors! The scanner also showed that the AVCS timing between the two banks of cams is now very similar.

 

I had spent many hours on this problem and rather than keep going round in circles, I thought ouside the square and decided to go back to basics. Glad I did as otherwise this problem would never have been found. It had been to two different mechanics and they could not find the cause. The previous owner did 40,000k and I had done 10,000k before it began to show up as a real problem. This experience may help someone else with one of these intermittent poor acceleartion issues.

 

 

So I installed the 564k ohm fake knock sensor last night. It was finally rainy today so I took that as an excuse to drive around.

In four different drives it accelerated flawlessly. I kept stomping on the accelerator at every chance and it accelerated smoothly from low RPM through 2500 with no change at 2500 RPM. Previously when I thought it was running pretty well, you could still notice a slight surge as it passed 2500 RPM.

With all the accelerating, I did hear it knock a little but only once.

I need to keep testing but I think this is it, a hypersensitive knock detection and response system. Since I tried two new ($15) Aisin sensors and they did not correct the problem, probably not the knock sensor. ECU must be hypersensitive to the knock sensor.

I did see a reference to this problem, a recall I think for 1998 Legacy where the ECU responded to the knock sensor incorrectly. Mine is a '99 but was probably one of the first ones, manufactured in June of 1998, so those Hoosiers might have thrown some left over 1998 parts (ECU?)at it.

I guess I could try a different ECU but not sure I want to get involved in that.

Link to comment
Share on other sites

Well that's really interesting. I don't know what AVCS timing is. I think I understand your solution but have no experience with adjusting timing belts/tensioners etc. with these newer cars. (I worked on 1960s/70s era cars as a teenager - much simpler)

 

I definitely noticed that it would not malfunction with very gentle acceleration, but really acted up with harder acceleration. I had been thinking that the root of the problem might be with the timing belt/tensioner etc.

 

With the current setup where there is no timing adjustment for engine knock, it does knock a little but only at slower speeds up steep hills - which I figure I can live with.

 

Maybe once I get to the end of the list of dozens of things to fix (3-4 things left including brake replacement, cv joint cover etc.) I will expand my horizons even further and look into it

Link to comment
Share on other sites

  • 6 years later...
49 minutes ago, Julian Rodriguez said:

I've got a 98 Subaru Legacy with a 2.2 L engine. I'm also having this issue at the moment. Was there any solution to this at all? I know it's been quite some time since anyone has commented but I could really use some help.

Probably better to start your own new thread with all the information you can, rather than revive such an old thread.

Link to comment
Share on other sites

11 hours ago, Julian Rodriguez said:

Sorry new here 

It's no problem. You can do it this way...but it's just easier to start your own, especially since this one was never resolved and could be a completely different cause.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...