2G Weird low RPM behavior.

[A Sharp]

Hey,

My 2g Talon TSI has been having issues at low RPM ever since I bought it. It idles at around 1,200 RPM and hesitates really badly right around 2,300 RPM under load. These problems are even more pronounced when the car is cold. When I first start the car on a cold day, it will idle as high as 2,700 RPM. The hesitation is much worse when it is cold as well. I even remember one occasion where the car was cold and it misfired like crazy under load at 2,300 RPM.

However, the engine runs just fine past 3,000 RPM.

Things I have tried:

A full throttle body rebuild, including a new BISS.
Tightening all IC pipe connections.
Installing properly gapped NGK copper spark plugs.
Some boost leak testing with a bike pump (ghetto as hell, but I don't have a compressor).

At this point I feel like there aren't any boost leaks big enough to cause problems of this magnitude. I suspect that maybe the MAS is dirty. Also the car's throttle response is very jerky when cruising around in first gear. When I initially step on the gas it jerks, kinda like it instantaneously goes from 0% throttle to about 5% throttle. This makes me suspect that I have a problem with the TPS.

The car is pretty much stock as far as the engine goes. I replaced the T25 with a 14b from a 1g (still running stock boost, tune, and IC piping) and the guy that had it before me did an EGR delete, but that's about it.

Any suggestions would be great! Sorry that I don't have a data logger or wideband to provide additional info, but this is my daily driver and I'm just looking to make it a solid runner.

Thanks!

 

[Colt4G63 also]

I'm not sure how you do a BLT with a bike pump, because the air that normally leaks past the rings and turbo seals would practically exceed the capacity of the pump.

The excessive RPM indicates too much air getting into the engine, so either via the throttle body, or some external vacuum leak. I suspect you need to go back to the drawing board regarding the BLT.

The surging can certainly be caused by a faulty TPS signal. Ordinarily the ECU will cut fuel if it senses a closed throttle while the RPM is above around 1200. Closed throttle is approximately .63 volts. WOT is approximately 4.5-4.8 volts.

 

[A Sharp]

I hooked the bike pump up directly to the IC piping, bypassing the turbo. I was able to use soapy water to find the major boost leaks.

I did a bit of reading and decided to try disconnecting the TPS. When I took it for a drive, the hesitation was much improved and the throttle surging was gone. However, the idle was still just as high as before. Would this indicate a bad TPS or could the orientation of the TPS be off?

Thanks for the info!

 

[Colt4G63 also]

I can't confirm or deny the orientation of the TPS, but I gave you some voltage values to test for.

As for the high RPM, I stand by my previous statement. Air is entering the engine; that's the only way the RPM can rise.

 

[A Sharp]

*BUMP*

Alright, so I've done a few things since I started this thread. The extremely high cold start idle was caused by the FIAV. It was loose and backed out almost all the way. I remedied this by tightening it all the way down. However, the car still idles at about 1,200 RPM when warm.

As for the IC piping, I have installed a FMIC and full metal piping with a 1g BOV (which is recirculated). I boost leak tested this setup, again with the bike pump , by pressurizing the piping immediately after the turbo. With the bike pump, I was able to pressurize the system to 10 PSI, which is what I'm running right now. I would pump like a maniac and then jump into the driver's seat to look at the boost gauge. The pressure in the system slowly releases over the course of about 30 seconds, and then stops at about 3 PSI. Correct me if I'm wrong, but I think that this rate of leakage wouldn't have any large impact engine performance. I'm pretty sure that the only remaining leaks are at the injector seals, since they are the only spots that bubble when sprayed with soapy water. I will be replacing the seals soon.

I have also removed and cleaned the MAF with MAF cleaner.

Unfortunately, the low RPM problems have not gone away. The car still hesitates, especially around 2,300 RPM, regardless of throttle position and boost. Though it hesitates worse if I put my foot down and/or if it's cold.

I'm just not sure where to look next.

Thanks!

 

[Talonman86096]

Do you have link? Sounds similar to symptoms I get sometimes when my timing is get pulled, also when cold mainly. And also only till about 3,000 rpms and then it goes away. With link though I have it set to flash the check engine light anytime a certain degree knock is detected. Could be pulling timing at lower rpms causing the hesitation, sounds similar to mine which mine is phantom knock that I need to track down

 

[A Sharp]

Unfortunately, I do not have link, a wideband setup, or any kind of data logger. However, what you described does sound very similar to my car. It also starts behaving normally at about 3,000 RPM.

I haven't heard of phantom knock before, but now that I've read into it a bit it makes sense. My lifters can be a bit ticky and the knock sensor itself has probably never been replaced. But from what I understand, this problem sucks ass to diagnose and fix because it could be anything in contact with the engine that rattles or makes noise. From what I have read, most people just use link to tune around it. Since I don't have link, it might just end up staying the way it is .

Do you know of anyone who was able to fix this problem without using link?

Thanks!

 

[A Sharp]

So I just got an OBD II cable so I can do some basic data logging and code reading with my laptop. After hooking it up, it revealed that my car does actually idle at about 780 RPM when warm. This means that my tachometer reads about 500 RPM too high. After revving the engine a bit, I discovered that the tach does this at all engine speeds. Is there any way to adjust the tach? Could I perhaps put a resistor in series with the tach so that the tach sees less voltage?

Also, since the OBD II interface doesn't provide knock sum or knock retard, is there any way to see if timing is being retarded? I'm still trying to figure out what is causing my slight hesitation in lower RPMs. My logger does however show timing advance for the #1 cylinder, so could I use this to check for timing retard?

BTW I changed my lower injector seals and my IC piping now holds boost indefinitely.

Thanks!

 

[A Sharp]

UPDATE:

One thing that I noticed about the hesitation is that it only happens at partial throttle under 3k RPM. This means that it is only happening in closed loop. I also noticed that the car seems to generally run rich. There is a huge soot stain on the body by the tail pipe and the exhaust smells rich. These factors made me think that the O2 sensor was playing a role in the problems. I replaced it, but it didn't help.

A while later I was at a buddy's house and we decided to try seafoaming it just for the hell of it. After sucking some seafoam through the brake booster line, it was clear that there were TERRIBLE exhaust leaks all around the area in front of the engine. There was practically as much smoke coming from the engine bay as there was from the tail pipe. Upon further inspection, the gaskets on both ends of the exhaust manifold and the O2 housing gasket were leaking.

A few weeks later I pulled the turbo off of the car to fix the leaks, and this is what I found in addition to the bad gaskets: cracks on the exhaust manifold and on the 14b's exhaust housing. Take a look at the pics.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

ht

You must be logged in to view this image or video.

You must be logged in to view this image or video.

I can even see light through the crack on the EM. Anyhow, I'm assuming that both of the EM and hotside housing will need to be replaced.

All things considered, could these exhaust leaks cause the O2 sensor to make bad readings, causing the car to run rich and hesitate?

 

[97 eclenger]

these manifolds are bad for cracking. itd read a little off but i don't think itd be that bad. try cleaning the maf, replacing the o2 sensors and see if you can look at the knock sensor. usually with a bad knock sensor the car won't have any wot power.

 

[lasershark]

The hotside of the turbo is fairly common, does it go all the way through?
The em can/will cause the readings to be off and also the karman type mafs don't need to be cleaned.
http://www.dsmtuners.com/threads/how-to-clean-your-mas.345616/
http://www.dsmtuners.com/threads/bad-or-good-maf.403921/

 

[A Sharp]

Thanks for responding, guys.

these manifolds are bad for cracking. itd read a little off but i don't think itd be that bad. try cleaning the maf, replacing the o2 sensors and see if you can look at the knock sensor. usually with a bad knock sensor the car won't have any wot power.

As I mentioned earlier, I did replace the upstream O2 sensor, though it didn't help. Also, I agree that my car may very well have some knock sensor issues. However, the hesitation is only noticeable at partial throttle. It is the worst around 70-80% throttle. If I start from 2k RPM and stomp the gas to the floor, the engine won't hesitate. This is why I'm thinking that it is an issue with the closed loop system, in particular the mixture the O2 sensor is reading. At WOT the ECU completely disregards the O2 sensor and only pays attention to airflow metered by the MAF, correct?

By the way, when I say hesitate, I mean that at around 2,300 RPM the engine seems to miss a bit and the car jerks slightly.

The hotside of the turbo is fairly common, does it go all the way through?
The em can/will cause the readings to be off and also the karman type mafs don't need to be cleaned.
http://www.dsmtuners.com/threads/how-to-clean-your-mas.345616/
http://www.dsmtuners.com/threads/bad-or-good-maf.403921/

The crack in the hotside does not go all the way to the exterior, so it is not externally leaking exhaust. I'm just thinking that a crack right near the wastegate would slow spoolup some, though I haven't noticed any slow spoolup. Lastly, I have cleaned my MAF sensor recently with some CRC cleaner. Thanks to you, I now know that is a dumb thing to do because it's a karman . I didn't mention it earlier because cleaning it had no impact on the problem.

 

[A Sharp]

I guess I could throw some JB weld on the exhaust manifold. I don't know if it will actually help though. Has anyone tried this? By the way, the 3rd and 4th pics I posted show the same crack on the EM. The 3rd pic is a view from inside the runner, while the 4th is, obviously, a view from the bottom of the EM.