Surging Idle and P0638 CEL Issue Solved

gandm99 · 08-25-2010, 09:55 PM

For the past few months I’ve had this bizarre surging idle problem that would manifest itself whenever I would go into boost and then slow down for a stoplight by holding the clutch in or shifting into neutral such that the engine rpms would freely drop from the higher rpm range. The rpm would freely drop from, say, 4000-5000 until they got past maybe 1500 and then it would catch itself and begin to surge up and down between maybe 2500 and 1000 rpm. And it would actually throw a Throttle related CEL code if I let it (P0638). I learned that I could prevent the code by goosing the throttle which seemed to snap the car out of the surging do-loop.

I spent all kinds of time looking for vacuum leaks; I swapped out both the ECU and the TB with used replacements, took the accelerator pedal assembly apart and cleaned the contacts, and also swapped out the front and rear O2 sensors. None of it made a lick of difference and I was pretty much at wits end and on the verge of bringing it to a trained professional.

So, with all that said, today I solved the problem. Turns out it was all due to a stuck open check valve on my catch can system (vendor is sending me a free replacement valve asap). This check valve is positioned between the intake manifold and a Tee for both the Turbo Inlet and the crankcase, and was there to prevent the intake manifold air from ever flowing into either of those places. It basically acts in similar function to the OEM PCV valve, so this failure would also be equivalent to an OEM PCV valve stuck in the open position.

I’m trying to visualize what may have been physically taking place but it’s not totally obvious. I’m pretty sure that it did create some sort of weird airflow loop around the turbo under boost since much of the boost pressure was also blowing out into the intake pre-turbo. And it was also likely pressurizing the crankcase. Maybe even creating some sort of hydraulic accumulator effect such that when I shifted and the intake manifold pressure went negative, that accumulated air in the crankcase would suddenly get blown back into the intake manifold… maybe that explains the surging.

I also had been smelling methanol after spraying (nozzle located in TMIC pre-TB). As it turns out, the methanol was actually getting blown all the way out through the air intake filter when I shifted and the intake suddenly/briefly pressurized. I had erroneously thought that I had developed a leak in the HSF-5 system, but now, after fixing the check valve problem, I no longer smell it.

Hopefully my troubles and final solution may help someone having similar issues. I know there are numerous posts involving throttle related CEL/Issues and Idle problems.

G

CN’s: Stuck open check valve on catch can system causes surging idle and a P0638 CEL code.

* Registered users of the site do not see these ads.

gandm99 · 08-30-2010, 03:04 PM

turned out it was engine bay heat that killed the check valve (krank vent). i put a brand new one in over the weekend and within 30 minutes of driving, it too had stuck open. Fortunately i was able to get it working perfectly once it cooled down.

i ended up re-routing my catch can tubing to avoid the high heat in the valve, and also added a heat reflective sleeve over it. so far it's all good.

gandm99 · 08-30-2010, 03:14 PM

one small clarification about my original post... I had stated that the Meth was blowing out the intake due to the bad valve. That is no longer confirmed and factual. I sprayed meth briefly over the weekend and did still notice the smell when the windows were down.

Really not sure where the meth is leaking from at this point. The nozzle is located in the TMIC just before the TB, and i've checked for leaks from the nozzle to the HSV and haven't found any (also re-trimmed the plastic hose ends and re-did the connections everywhere). I also know it's not leaking between the HSV and the pump since that line holds pressure and I don't smell meth unless the windows are down. I'm starting to think that my Intake manifold/engine mate may have a very small leak that only leaks under higher boost pressure... maybe upwards of 18 psi or so since I'm able to build boost and hold my 18 psi target without issue.

looks like another round of intake pressure testing is in my future...

wackbards · 01-10-2011, 05:12 PM

duuuuuude....
I think this may.... just... have... save... my... @$$...

EXACT SAME symptoms.

guess what? I removed my OEM PCV check valve when I put on a air/oil separator.

gonna stop by the parts store on the way home...

wackbards · 01-10-2011, 05:28 PM

wow. I'm still in a bit of shock.

my setup sounds similar to yours. I run a HFS-6 on a stock turbo. High boost (23lbs). I have a Grimmspeed AOS on it. I didn't follow the recommended directions, and deleted the PCV...

I'll update if this works...

gandm99 · 01-11-2011, 12:09 PM

i forgot about this thread. my meth/exhaust smell was entirely due to a small exhaust leak near the turbo. no big deal. just annoying.

and the P0638 was finally fixed by replacing the "krank vent" valve with the factory PCV. No problems since. For whatever reason, my car literally eats those supposedly indestructible Krank Vent things.

wackbards · 01-12-2011, 01:30 AM

Well, I had completely removed the PCV and capped the line running from the PCV to the manifold (as per crawford's AOS instructions). I'm running the Grimmspeed AOS, and they recommend leaving the PCV in tact. I just don't know... I wouldn't be experiencing problems for the reason you outlined because manifold pressure couldn't travel out any more... we'll see I guess. Should find time in the next couple days to button things up and take it for a spin.

08-25-2010, 09:55 PM	#1
gandm99 Scooby Newbie Member#: 58709 Join Date: Apr 2004 Chapter/Region: SCIC Location: Los Angeles Vehicle: 2004 WRX STi WRB	Surging Idle and P0638 CEL Issue Solved For the past few months I’ve had this bizarre surging idle problem that would manifest itself whenever I would go into boost and then slow down for a stoplight by holding the clutch in or shifting into neutral such that the engine rpms would freely drop from the higher rpm range. The rpm would freely drop from, say, 4000-5000 until they got past maybe 1500 and then it would catch itself and begin to surge up and down between maybe 2500 and 1000 rpm. And it would actually throw a Throttle related CEL code if I let it (P0638). I learned that I could prevent the code by goosing the throttle which seemed to snap the car out of the surging do-loop. I spent all kinds of time looking for vacuum leaks; I swapped out both the ECU and the TB with used replacements, took the accelerator pedal assembly apart and cleaned the contacts, and also swapped out the front and rear O2 sensors. None of it made a lick of difference and I was pretty much at wits end and on the verge of bringing it to a trained professional. So, with all that said, today I solved the problem. Turns out it was all due to a stuck open check valve on my catch can system (vendor is sending me a free replacement valve asap). This check valve is positioned between the intake manifold and a Tee for both the Turbo Inlet and the crankcase, and was there to prevent the intake manifold air from ever flowing into either of those places. It basically acts in similar function to the OEM PCV valve, so this failure would also be equivalent to an OEM PCV valve stuck in the open position. I’m trying to visualize what may have been physically taking place but it’s not totally obvious. I’m pretty sure that it did create some sort of weird airflow loop around the turbo under boost since much of the boost pressure was also blowing out into the intake pre-turbo. And it was also likely pressurizing the crankcase. Maybe even creating some sort of hydraulic accumulator effect such that when I shifted and the intake manifold pressure went negative, that accumulated air in the crankcase would suddenly get blown back into the intake manifold… maybe that explains the surging. I also had been smelling methanol after spraying (nozzle located in TMIC pre-TB). As it turns out, the methanol was actually getting blown all the way out through the air intake filter when I shifted and the intake suddenly/briefly pressurized. I had erroneously thought that I had developed a leak in the HSF-5 system, but now, after fixing the check valve problem, I no longer smell it. Hopefully my troubles and final solution may help someone having similar issues. I know there are numerous posts involving throttle related CEL/Issues and Idle problems. G CN’s: Stuck open check valve on catch can system causes surging idle and a P0638 CEL code. * Registered users of the site do not see these ads.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Rough idle and throwing CEL	marcus712	Factory 2.5L Turbo Powertrain	4	11-30-2009 01:25 PM
High Idle and P1507 cel - Need new IACV?	Asterix	Factory 2.0L Turbo Powertrain	38	01-27-2009 11:47 PM
Rough idle issue solved via TSB sticky!!!	Gregmatic	Baja Forum	0	03-07-2006 11:49 PM
Erratic idle and surge	shikataganai	Factory 2.0L Turbo Powertrain	15	02-13-2006 09:55 PM
idle and no cel	cowlypso	Normally Aspirated Powertrain	1	05-17-2004 01:13 PM

08-30-2010, 03:04 PM	#2
gandm99 Scooby Newbie Member#: 58709 Join Date: Apr 2004 Chapter/Region: SCIC Location: Los Angeles Vehicle: 2004 WRX STi WRB	turned out it was engine bay heat that killed the check valve (krank vent). i put a brand new one in over the weekend and within 30 minutes of driving, it too had stuck open. Fortunately i was able to get it working perfectly once it cooled down. i ended up re-routing my catch can tubing to avoid the high heat in the valve, and also added a heat reflective sleeve over it. so far it's all good.

08-30-2010, 03:14 PM	#3
gandm99 Scooby Newbie Member#: 58709 Join Date: Apr 2004 Chapter/Region: SCIC Location: Los Angeles Vehicle: 2004 WRX STi WRB	one small clarification about my original post... I had stated that the Meth was blowing out the intake due to the bad valve. That is no longer confirmed and factual. I sprayed meth briefly over the weekend and did still notice the smell when the windows were down. Really not sure where the meth is leaking from at this point. The nozzle is located in the TMIC just before the TB, and i've checked for leaks from the nozzle to the HSV and haven't found any (also re-trimmed the plastic hose ends and re-did the connections everywhere). I also know it's not leaking between the HSV and the pump since that line holds pressure and I don't smell meth unless the windows are down. I'm starting to think that my Intake manifold/engine mate may have a very small leak that only leaks under higher boost pressure... maybe upwards of 18 psi or so since I'm able to build boost and hold my 18 psi target without issue. looks like another round of intake pressure testing is in my future...

01-10-2011, 05:12 PM	#4
wackbards Scooby Newbie Member#: 219625 Join Date: Aug 2009 Chapter/Region: SCIC Location: socal Vehicle: 2004 wrx sti psm	duuuuuude.... I think this may.... just... have... save... my... @$$... EXACT SAME symptoms. guess what? I removed my OEM PCV check valve when I put on a air/oil separator. gonna stop by the parts store on the way home...

01-10-2011, 05:28 PM	#5
wackbards Scooby Newbie Member#: 219625 Join Date: Aug 2009 Chapter/Region: SCIC Location: socal Vehicle: 2004 wrx sti psm	wow. I'm still in a bit of shock. my setup sounds similar to yours. I run a HFS-6 on a stock turbo. High boost (23lbs). I have a Grimmspeed AOS on it. I didn't follow the recommended directions, and deleted the PCV... I'll update if this works...

01-11-2011, 12:09 PM	#6
gandm99 Scooby Newbie Member#: 58709 Join Date: Apr 2004 Chapter/Region: SCIC Location: Los Angeles Vehicle: 2004 WRX STi WRB	i forgot about this thread. my meth/exhaust smell was entirely due to a small exhaust leak near the turbo. no big deal. just annoying. and the P0638 was finally fixed by replacing the "krank vent" valve with the factory PCV. No problems since. For whatever reason, my car literally eats those supposedly indestructible Krank Vent things.

01-12-2011, 01:30 AM	#7
wackbards Scooby Newbie Member#: 219625 Join Date: Aug 2009 Chapter/Region: SCIC Location: socal Vehicle: 2004 wrx sti psm	Well, I had completely removed the PCV and capped the line running from the PCV to the manifold (as per crawford's AOS instructions). I'm running the Grimmspeed AOS, and they recommend leaving the PCV in tact. I just don't know... I wouldn't be experiencing problems for the reason you outlined because manifold pressure couldn't travel out any more... we'll see I guess. Should find time in the next couple days to button things up and take it for a spin.