Jump to content
LegacyGT.com

Beyond frustrated with unstable idle and surging in neutral


Recommended Posts

Here's the background story. Car was tuned at IAG Performance in December 2014 after adding ETS manifold with Tial 38mm wastegates. Car ran great for about two months then out of the blue the idle was all over the place primarily on cold start. RPMs would drop normally then rise as I shifted into neutral or pushed in the clutch to coast to a stop light, etc.

 

Cleo tuned it in December and has sent me a handful of maps with tweaks here and there. Nothing seems to resolve the issue. The idle is more stable now but the RPMs still drop then surge inexplicably when shifting into neutral or pushing in the clutch when coasting. The colder the car is the greater the surge in RPMs. After emailing data logs and new maps back and forth with Cleo I took the car to IAG about two weeks ago. Prior to scheduling an appt at IAG the car was smoke tested locally and no vacuum leaks were found.

 

After spending most of the day with the car Cleo concluded that it was likely a bad air/fuel sensor. I replaced the sensor last Wednesday, flashed back to Cleo's tune from December and no change. Idle was horrible again upon initial start up. It seems clear that this rules out the air/fuel sensor being bad. I sent Cleo the logs upon replacing the sensor and got a new map from him. Again idle is more stable put same surging when coasting and not in gear.

 

I hope this post doesn't come off as anything negative towards IAG. I have had nothing but pleasant experiences and have received great customer service from them over the past two years. I'm just reaching out to other avenues and hoping to get some other possible opinions. I would love to hear from some of the tuners and experts on here. That's you Mike @ Infamous, McDowell, and fahr_side.

 

Thanks to any and all for your input. I'm sick of this car running like shit for months now.

Link to comment
Share on other sites

I'm not sure it's your tune. I'm having this same issue (as I did on my 09 N/A) and it might be a Subaru ECU thing. When I come to a stop and I press the clutch in, the RPMs drop way down and then bump back up pretty high, then coming to stop around normal idle. I have a video of this if you wanna see it and compare.

 

Edit: If you google "subaru rpm idle drop" there are a ton of posts about it. Here's one excerpt:

 

"In my opinion my 6MT idles way too low, often shuddering for a second or two as if it wants to stall. Even a mere 100 rpm boost in idle speed sure would help when starting out from a stop, too, as that can be tricky with the goofy first gear. I've only got a few hundred miles on it so far... I thought I would give the car's computers a chance to "teach" itself a proper idle before I asked the dealer about it."

Link to comment
Share on other sites

ebo03, I feel your pain...trust me, I feel your pain.

 

'Nuff said.

"It's within spec" - SOA :rolleyes:

"Depth is only shallowness viewed from the side." - Fredism

"So, how much did it cost for your car to be undriveable :lol:." - Stephen (very close friend)

"You have done so much it would be stupid to go back." - Sunny of Guru Electronics

 

2018Q50RS | 2015WrxThread | Shrek

Link to comment
Share on other sites

How high is the "surge" in RPMs?

 

Are you speed density or running a MAF?

 

Speed density.

 

Vacuum leak possibly?

 

This, vacuum leaks are a common cause of surging.

 

Reread initial post. Car was smoke tested with no leaks found.

Link to comment
Share on other sites

I had a similar problem. It turned out to be a blown vacuum hose under the tmic. I forget exactly which hose it was. It was near the pcv. How much vac is your car pulling? Mine is normally around 11.6 but with the blown hose it was less than 10.
Link to comment
Share on other sites

Speed density.

 

Reread initial post. Car was smoke tested with no leaks found.

 

Ah, I had missed it as it was at the end of the tuning comments. Smoke tests aren't fool proof and the fact that nothing else changed and adjustments to the tune aren't helping, I would take a few minutes and manually recheck all of the vacuum lines and couplers. Perhaps take a closer look at the vacuum lines going to the EWGs as I assume there's now a T there and hose that wasn't there stock (or you're not running one of the DWG manifolds and I'm just talking out of my ass).

Link to comment
Share on other sites

I had a similar problem. It turned out to be a blown vacuum hose under the tmic. I forget exactly which hose it was. It was near the pcv. How much vac is your car pulling? Mine is normally around 11.6 but with the blown hose it was less than 10.

 

My AP shows about 8.2 at idle, only see over 10 if I downshift hard.

Link to comment
Share on other sites

Definitely sounds like a vacuum leak somewhere.

I'd check for exhaust leaks too while @ it.

 

Also consider switching from a pure SD tune or any emulated-MAF transfer function tune. Instead of a 100% SD tune - many tuners can now do:

 

a. hybrid (MAF+MAP-based tune that transitions from MAF to MAP and back as needed to take advantage of the strength of both sensors and methods (and limit the weaknesses of each) - in fact the FA20DIT OEM tune is like this - MAF driven under certain conditions, MAP driven under others)

 

b. RPM and Boost-driven fueling maps (removes 'Load' out of the equation all together, allows the engine to hit the ECU target lambda much better under OL and CL). Perfect for MT Scoobies.

Link to comment
Share on other sites

Vacuum on idle when the car is warm is around 9.5. No clue what it was prior to that so I don't really have a baseline for that figure.

 

A vacuum leak makes sense based on how the car is behaving. But after having a local mechanic (he's also a good friend) smoke test it, then IAG looking at it for 5.5 hours don't you think a vacuum leak would have been detected by now?

 

Here are the main points of all of this:

 

1) The car ran great for two months after the last round of mods and tuning. All of a sudden the idle when cold was surging up to 2500 RPM then it would drop off below 1000 RPM and repeat this cycle until the car was warmer and the idle leveled off. Sometimes it would nearly stall when dropping from 2500. Idle fluctuations were worse during the colder months when ambient temps were down. Idle is still uneven when running the map from December. Idle has improved with each of Cleo's revised maps.

 

2) The surge in neutral varies based on operating temp. When cold and I'm coasting down my hill first thing in the morning it will cycle between 1200-1800 RPMs a few times before leveling off. Once the car is warm, it will surge around 300-400 RPM. I never recall seeing surging like this before. Is this considered normal? Cleo certainly didn't think it was normal.

 

It seems pretty clear to me that there is a mechanical issue that started all of this. If it were an issue with the tuning this would have showed up from the get go.

Link to comment
Share on other sites

Thanks for the detail ebo03. Good point, but you never know; -9.5psi for an EJ LGT is likely decent.

 

A vacuum leak would have the biggest effect at idle or low RPM and under low load conditions. How does the car run @ speed, under boost, etc? If the RPMs are stable then or if you raise them to say 2-3K RPM when warmed up in neutral, then I'd still think its some type of vacuum leak somewhere (or a few small ones in a number of locations).

 

Check that your intake is well secured too - all clamps leading to the turbo, up-pipe, intercooler inlet/outlet, etc. And maybe re-seat the MAF sensor to make sure its snug and that no false air is getting in that way or downstream of the MAF.

 

One other thing to check (as-in have the tuner check) on the tuning side is the OL-to-CL transition.

 

If you can log a cold start and let her idle for say 3+ minutes, you should capture it.

 

What you're looking for is the moment when the ECU switches from OL to CL operation and to see if the worst RPM surge/drop swing happens at that exact point.

 

Then you can repeat the same log for warm start (constant CL) and compare the two, see how unstable the RPMs are then in comparison.

 

You'd want to log:

 

RPM

Fuel Trims for Bank 1 (STFT, LTFT or Fuel Correct, Fuel Learn - depending on whether you have access to OBDII PIDs or SSM EPIDs)

MAF output (g/sec, cfms or raw voltage)

AFR for Bank 1 (or better yet Lambda if you can)

Boost/Vacuum

Coolant Temp

Oil Temp (if you can)

Catalyst 1 Bank 1 Temp (if you can)

CL Flag (if you can)

 

What you should see is when the ECU 'thinks' the cat has reached a high enough temperature (its actually estimating this, doesn't know the real cat temp, the OBDII output is just a guesstimate), it will switch from cold-start OL open-loop to CL closed-loop operation.

 

This transition is where the worst surge likely occurs, as the car will initially go extra rich, then extra lean, and level off resuming 'normal' CL operation. Its also the place where you'd see the biggest impact from a vacuum leak and yes, it would be more pronounced when cold out (cat takes longer to reach operating temp, intake temp typically lower longer; engine bay heat-soaks much later).

Link to comment
Share on other sites

Thanks for the detail ebo03. Good point, but you never know; -9.5psi for an EJ LGT is likely decent.

 

A vacuum leak would have the biggest effect at idle or low RPM and under low load conditions. How does the car run @ speed, under boost, etc? If the RPMs are stable then or if you raise them to say 2-3K RPM when warmed up in neutral, then I'd still think its some type of vacuum leak somewhere (or a few small ones in a number of locations).

 

Check that your intake is well secured too - all clamps leading to the turbo, up-pipe, intercooler inlet/outlet, etc. And maybe re-seat the MAF sensor to make sure its snug and that no false air is getting in that way or downstream of the MAF.

 

One other thing to check (as-in have the tuner check) on the tuning side is the OL-to-CL transition.

 

If you can log a cold start and let her idle for say 3+ minutes, you should capture it.

 

What you're looking for is the moment when the ECU switches from OL to CL operation and to see if the worst RPM surge/drop swing happens at that exact point.

 

Then you can repeat the same log for warm start (constant CL) and compare the two, see how unstable the RPMs are then in comparison.

 

You'd want to log:

 

RPM

Fuel Trims for Bank 1 (STFT, LTFT or Fuel Correct, Fuel Learn - depending on whether you have access to OBDII PIDs or SSM EPIDs)

MAF output (g/sec, cfms or raw voltage)

AFR for Bank 1 (or better yet Lambda if you can)

Boost/Vacuum

Coolant Temp

Oil Temp (if you can)

Catalyst 1 Bank 1 Temp (if you can)

CL Flag (if you can)

 

What you should see is when the ECU 'thinks' the cat has reached a high enough temperature (its actually estimating this, doesn't know the real cat temp, the OBDII output is just a guesstimate), it will switch from cold-start OL open-loop to CL closed-loop operation.

 

This transition is where the worst surge likely occurs, as the car will initially go extra rich, then extra lean, and level off resuming 'normal' CL operation. Its also the place where you'd see the biggest impact from a vacuum leak and yes, it would be more pronounced when cold out (cat takes longer to reach operating temp, intake temp typically lower longer; engine bay heat-soaks much later).

 

Car runs fine under throttle. All issues are off throttle.

 

Full speed density so no MAF sensor.

 

Cleo played around with closed loop and open loop for quite a while. That is why he was pretty certain it was the air/fuel sensor. To my best recollection he said the surging issues went away when switching to closed loop so that is why he thought the air/fuel sensor must be bad.

Link to comment
Share on other sites

My AP shows about 8.2 at idle, only see over 10 if I downshift hard.

 

Idle vac goes down from the normal 22 in/hg or 11 psi by 1 in/hg (or .5 psi) for every 1000 feet of elevation you are above 2000 feet. Guessing you are at about 7000 ft.

 

ebo03, I'm putting my money on a vac leak as well. Aren't you running an FMIC? Have you had it pressure tested? Maybe a rock damaged it.

Link to comment
Share on other sites

Idle vac goes down from the normal 22 in/hg or 11 psi by 1 in/hg (or .5 psi) for every 1000 feet of elevation you are above 2000 feet. Guessing you are at about 7000 ft.

 

ebo03, I'm putting my money on a vac leak as well. Aren't you running an FMIC? Have you had it pressure tested? Maybe a rock damaged it.

 

I can't tell you the exact elevation but would guess it's around 1000 feet. Look up elevation for Roanoke, VA. It's nowhere near 7000.

Link to comment
Share on other sites

IAG attempted to pressure test the FMIC when I first installed it in October 2013. I think they said their tester runs 14psi. They attempted to pressurize it from the intake and the OEM inlet leaked like crazy so there wasn't really a good way to pressure test the entire system. Based on how it performed during tuning, there was no indication of any leaks.
Link to comment
Share on other sites

I can't tell you the exact elevation but would guess it's around 1000 feet. Look up elevation for Roanoke, VA. It's nowhere near 7000.

 

Right. I was replying to Unspoken.

 

If you are at 1000 feet, your idle vac should be between 21-23 in/hg or about 11 psi. Any less and you have a vac leak or a sick engine.

Link to comment
Share on other sites

Based on how it performed during tuning, there was no indication of any leaks.

 

An errant rock bouncing down the road and blasting into the FMIC at speed can damage the aluminum enough for it to leak, and that could have happened any time after the tune. Inspect it for damage.

Link to comment
Share on other sites

Right. I was replying to Unspoken.

 

If you are at 1000 feet, your idle vac should be between 21-23 in/hg or about 11 psi. Any less and you have a vac leak or a sick engine.

 

Oh ok, that makes more sense.

 

An errant rock bouncing down the road and blasting into the FMIC at speed can damage the aluminum enough for it to leak, and that could have happened any time after the tune. Inspect it for damage.

 

I haven't looked at it very closely but there is aluminum mesh over the core. I haven't noticed any damage to the mesh. The core is sprayed black and I haven't noticed any new or sizeable rock chips.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.



×
×
  • Create New...

Important Information

Terms of Use