Jump to content
LegacyGT.com

AFRs off and small backfiring


Recommended Posts

The other night while driving home I heard the fuel pump whining way more than normal, pulled a Learning View and lots of fuel was getting added.

 

0-<5.60 - 15

5.60 - <10.00 - 5

10 - <40.00 - 15

40.00+ - 2

 

I got home and pressure tested the intake just in case and found very little. Since my pump was a pre-e85 compatible Aeromotive that I've been using with e85, and it was whining, I figured it was the pump and replaced it with an AEM 320, getting a new unmodified basket in the process. I didn't replace the two green o-rings as I didn't have any and they didn't look swollen or rolled. (Note: I'm going to replace them.)

 

Drove around for about 10min and Learning View seemed to look better, but I was still skeptical so I put in a wideband and had the intake smoke tested. The smoke test showed no leaks.

 

In 3rd the wideband is showing about 12.4 at redline, when it should be 11.4 - 11.8. Learning View is wacky again, almost the exact numbers as my initial problem.

 

0-<5.60 - 14.70

5.60 - <10.00 - 5.70

10 - <40.00 - 15

40.00+ - 2.10

 

I'm also getting a small backfire as I press the throttle if I granny shift in super low RPMs. If I set the cruise at 40mph in 3rd it almost feels like it's missing as the o2s jump wildly between 13.9 and 15.0. I've never had a wideband before so I don't know if that's normal behavior? At a static speed and RPM I would assume it would hold 14.7 easier and more consistently?

 

Idle is smooth as butter and WOT power is smooth and strong. You wouldn't know there was a problem if you didn't look at Learning View or the wideband.

 

What should I be looking for? Next steps in my mind are to replace the two green o-rings and check fuel pressure. The low rpm backfire confuses me though. An ignition problem would show up as extra richness vs leanness? If the FPR was going bad I assume it would also pour in too much fuel? The behavior feels like an intake leak, but I've checked everything and the smoke test was clear. I'm stumped.

Link to comment
Share on other sites

How does the stock O2 look (A/F Sensor #1 Ratio or similar)? That should line up with the wideband until you hit the mid-11 range of AFRs. If your wideband is reading 12.x and the front O2 reads 11.5, then there's an issue with one of your sensors.

 

 

Since WOT power is good, this sounds like a broken sensor more than broken fueling, especially if you're not getting knock under heavy loads/high RPM. Since it seems the car's running fine open-loop, it sounds like a sensor error during closed loop. How many miles on the stock front O2? Those are known to go bad without actually throwing codes. Also potentially worth cleaning the MAF sensor.

Link to comment
Share on other sites

Thanks for the suggestions. Unfortunately this is the 3rd tank of e85 with the problem, from different places.

 

I had the thought that the stock o2 sensor could be bad and that would make sense in a lot of ways. I should pull up RomRaider and compare the stock to the wideband. If it were the front o2 sensor I'd assume, maybe incorrectly, I'd run rich. The front o2 sensor has less than 20k on it though. I'm e85 so I don't think we'll see knock as readily.

 

I considered the MAF sensor was dirty and not reading air correctly, cleaned that.

 

Checked the vacuum lines to the FPR when I smoke tested.

 

I also ordered the pump holding portion of the fuel pump canister that Subaru sites list as a filter (42072AG16A), as well as the o-rings (42025AG09A x3, 42025AG10A x1). I didn't realize it was considered a filter now. When I did the pump I grabbed an unmodified one from a Pull-and-Pay that was dirty as heck, and my original had 151k miles on it. I put a new sock on, but not the 42072AG16A.

Link to comment
Share on other sites

Do the WBO2 AFRs under WOT look okay? If you're running fine under WOT (i.e. in open-loop), I think that would indicate that fueling is probably fine. Remember that learning view will only give you data on running in closed-loop (idle and light-throttle cruising).

 

 

I'd suggest some more diagnosis by logging the front O2 before throwing more parts at the issue.

Link to comment
Share on other sites

They don't, they are about a full point lean (12.4 vs 11.4 target) and sometimes more. I've stopped stepping on it because I saw 13.3 last time, way too lean. It will shoot down to 11.4 when I step on it and quickly go lean from there.

 

I agree, don't want to throw parts at it, that filter and the o-rings were probably due though. Having a fuel pressure gauge would help too.

Link to comment
Share on other sites

The new filter section and o-rings didn't do it.... the old one from Pull-n-Pay was disgusting and there was more crap in the surge tank than in 151k miles of my old one, but still, no go.

 

It initially started up nice, idled nice and I thought, awesome, it's good. Then about a minute in it stumbled, went lean and barely caught idle.

 

WOT confirmed it, scary 13.5, I shut down quick. Learning View is only at +3 so far, but it's still struggling to idle when I let off the gas, so it hasn't learned all the way up yet.

 

The behavior makes me think the FPR is bad as a vacuum leak wouldn't be intermittent. I guess I'll get a fuel pressure gauge to test that theory next.

 

A post MAF, pre turbo leak is still on my mind, but I sure don't see anything and smoke test was clear too.

 

I guess if the fuel pressure turns out ok then I'll start tearing stuff apart until I find a leak.

Edited by SSpeed
Link to comment
Share on other sites

How much pressure dose the smoke test put into the system ? May be try a short blast of air from a compressor. may like 15psi.

305,600miles 5/2012 ej257 short block, 8/2011 installed VF52 turbo, @20.8psi, 280whp, 300ftlbs. (SOLD).  CHECK your oil, these cars use it.

 

Engine Build - Click Here

Link to comment
Share on other sites

I think that idea is sound. The smoke test was only 3-5psi. It didn't build as much pressure as I thought it should. I attributed that to going past the valves since I didn't see smoke, but I don't recall ever having that problem in the past.

 

If I grab a couple of these and put them on the EVAP purge valve line and the PCV (AOS) line to block them off, I should be able to pressurize the intake to whatever amount I want right?

 

https://www.handsontools.com/OTC-4506-2-Piece-Fuel-Line-Clamp-Set_p_6691.html?gclid=CjwKCAjwr-PYBRB8EiwALtjbz8jYPIdntwVJzg6GmkTbFLi5s68H5KlWikjS8ivtG1pfFuFsce3s-xoC-SwQAvD_BwE

 

Can I block the EVAP line over at the FPR or do I need to do it close to the intake? I see it T's off at the purge valve under the intake manifold and it's not clear to me where it goes, can't tell in the service manual either.

 

The BOV line and boost control reference line shouldn' t matter. Maybe pop the oil cap off just in case?

Link to comment
Share on other sites

I'm starting to think this is an ignition issue. The mileage on the last tank was horrible. Perhaps a plug problem is causing leanness on an appropriate mixture due to misfire and the ECU is adding fuel back. The weird thing is the ECU is totally happy, no misfires registered, no roughness, but I can feel the misfire occasionally at cruise. At idle once it recovers it's nice and smooth. Edited by SSpeed
Link to comment
Share on other sites

  • 1 month later...

Slowly getting around to this. Took video tonight of the boost, fuel pressure and AFR gauge. It's definitely not fuel pressure causing the leanness.

 

 

Next up, plugs.

Link to comment
Share on other sites

Plugs go in Tuesday.

 

Another thing I just noticed. Airflow g/s at peak are only 229 g/s. Normally they are up around 255 g/s. That seems like a large disparity especially given the car is running as fast as it ever has. I saw the MAF in Romraider at 0.66V with engine off and key on, Pin #1 is 12V. Nothing changes when I wiggle the connector. The middle pin is a rust color compared to the others, not sure if that is supposed to be that way. I've cleaned the MAF and the connector.

 

I wonder if I should borrow another MAF too just for fun.

Link to comment
Share on other sites

Sorry I didn't check back in sooner. In the back of the service manual there are charted that show where all the lines go.

 

Something like this from the sticky above,

 

https://legacygt.com/forums/showthread.php/2005-2006-lgt-colored-vacuum-routing-diagram-143225.html

Edited by Max Capacity

305,600miles 5/2012 ej257 short block, 8/2011 installed VF52 turbo, @20.8psi, 280whp, 300ftlbs. (SOLD).  CHECK your oil, these cars use it.

 

Engine Build - Click Here

Link to comment
Share on other sites

Thanks for that diagram.

 

New plugs are in, leanness and stumbling/misfiring at low load persist. One step colder plugs were whitish, so leanness is occuring. Not hot enough for melting or electrode damage though.

 

I ordered a new MAF today. The fact that at WOT I'm 30g/s below what my logs have shown forever, with the car still running great at WOT, seem to point to the MAF or the harness.

 

The only other real possibility is 30g/s of unmetered air getting in at WOT. I've smoke tested the intake from the MAF in and it tested solid. The unknown to me is both the PCV/AOS and evap purge seem like they would be unmetered air? Or am I thinking wrong? If I'm getting 30g/s of unmetered air through the AOS that's a big problem. :)

Link to comment
Share on other sites

Sigh.... The new MAF didn't fix it. I am super confused.... Still getting a bit of a backfire at low loads. New plugs are in. Idles smooth as butter...

 

Pulled my map off the ECU and made sure it was the e85 map, it was. Pulled the airbox apart and made sure that was sealed up correctly.

 

On a hunch I blocked the AOS port that goes into the intake tube and vented the AOS to atmosphere. Could feel the AOS pulsating, but it did nothing for my leanness at WOT.

 

This still feels like a massive airleak pre-turbo. I may re-try the smoke test. I'd almost suspect a burned valve if it didn't idle so nice.

 

Looking around the pulsating air from the AOS is normal. I see no smoke from the oil filler tube revving it and the oil cap test reveals more vacuum than pressure. I had just done the same thing on my 3.6 and felt nothing from the oil filler tube so I got a bit confused. :)

Edited by SSpeed
Link to comment
Share on other sites

Thinking out loud....

 

Every plug when replaced was a uniform white. Not bad, but about what you see with stock plugs at a higher hp. That makes me think it's not an individual injector or individual coil, but probably something more global as every plug looks a bit lean? Does that makes sense or am I overlooking something? I'd expect one plug to be quite different if it were a problem with an individual cylinder? Kinda glad I did plugs so I can rule out individual cylinders.

 

I also logged the cylinder roughness to see if something showed up an individual cylinder and there is absolutely nothing showing up. Makes me wonder how bad it has to be before the ECU pegs it. I know when I had an injector go bad it showed cylinder roughness. This seems like it's across the board and a problem before it gets to the cylinder. Honestly, without the wideband or Learning View you wouldn't know there is a problem, it idles smooth as butter and drives nice, even WOT feels ok until you see it creeping up over 13 on the AFR, the backfire/stumble is just while granny shifting and feels like an incorrect mixture to me or a burnt exhaust valve. I can't see a burnt exhaust valve showing lean on 4 cylinders though and I would think that would idle poorly. And if it's not showing up as cylinder roughness that has to mean it's excess fuel igniting in the exhaust vs misfiring in the cylinder?

 

It still puzzles me that the logs show reading 30g/s lower than normal on the MAF at WOT than what I've seen since the beginning of time. Maybe it's sending me down a rabbit hole of condemning the airflow metering, but it's almost got to be that or a huge pre-turbo leak that I'm somehow missing with the smoke test. You don't just lose 30g/s of airflow and run lean at the same time.

 

Here's my next plan of attack.

 

1) Check continuity on all wires from ECU to MAF (this will be a huge pain and not sure how to do it)

2) Smoke test again

3) Tear the intake manifold off and replace all hoses and look for anything that I might be missing

4) Check the compressor housing

5) Pull the injectors and send them in for cleaning

6) Sell the car

7) Park it in Aurora and leave the keys in it

Edited by SSpeed
Link to comment
Share on other sites

Zoinks, it appears it may be the e85 "black goo" problem. Running a can of 44k through now and the AFRs are dropping, but still not perfect. I filled up at a new station twice just after dynoing with great AFRs. I may have to pull the injectors and have them cleaned.
Link to comment
Share on other sites

if you have access to 44K, do you have access to the BG Fuel Injection Service? with the tool that hooks up to the fuel line? It'll save you the hassle of taking the injectors out and sending them away.
Link to comment
Share on other sites

I don't know that, sorry. I work for a dealership and have access to the tool. There is a new recall on the Toyota Camry so every dealership has at least one of the tool.
Link to comment
Share on other sites

44k goes into the fuel tank. Anyone can purchase a fuel injection cleaning kit, Amazon probably has a ton of them. You will need to hook up air so a compressor that's capable of maintaining a pressure higher than idle fuel psi is required.
Link to comment
Share on other sites

  • 1 month later...

Ugh, this is annoying.

 

Injectors were cleaned, they had a 1% increase in flow after cleaning.

 

I tore the intake tract down to the cylinder head and inspected everything.

 

Fuel pressure is good, so it's not voltage or the fuel pump controller.

 

MAF is brand new, wires look good, voltage shows up within spec.

 

Cleaned the K&N filter

 

I mean, what really is left? I don't know what could be wrong? Could it just be the summer blend of fuel?

Link to comment
Share on other sites

What does your E85 test out at? I've seen +/- 10% differences. Our E85 in Iowa was closer to E90/E95 in the summer. Unless you tuned on winter fuel and are running on summer blend, I don't think you would see that severe of a shift in your fuel trims.

 

Are you hitting boost targets? I had a problem on my car with a similar problem and found that EWG line was leaking. I saw overboosting, weird fuel trims, high AFR, back fires, etc. It took me forever to find this leak and I took everything apart (similar to what you are doing now). Ultimately, I had to soap test every single joint to find it. It was buried far enough I couldn't hear it and it would bleed slow enough that I couldn't see it on my gauge, but it had a serious effect on how the car performed.

 

As I'm sure you are away, it's not surprising to see the high numbers in the A and C categories. You rarely drive your car in the D area (unless you are full throttle everywhere). Similar to your B area, it's more just off idle at low speeds. Typically your A and C are the first to show issues.

 

It really seems like you have an intake leak or vacuum leak somewhere you just haven't found yet.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...

Important Information

Terms of Use