Pages: [1]
Author Topic: 1.8T AGU high idle  (Read 277 times)
doktor
Jr. Member
**

Karma: +0/-0
Offline Offline

Posts: 48


« on: March 20, 2025, 11:19:19 AM »

I'm really clueless with this. It's not a big deal but it drives me nuts nonetheless.

This is on a M3.8.3 cable throttle system.

After a cold start, for a minute or so I get a high idle at around 1500-1600 rpm. The weird thing is that it seems to drop to normal (~1100) when coming to a complete stop, but if the car is moving even so slowly, the idle increases again (does not matter if with clutch disengaged or in neutral).
When warmed up, the idle is sometimes spot on, but other times it is slightly higher. But it is always stable, without misfires/hunting etc.
I also get inconsistent engine braking (overrun) behavior, sometimes it transitions instantly, sometimes after a while and sometimes never.
Except than this, the car works perfectly fine. There are no DTCs.

What I tried:
-intake manifold gasket, TB gasket
-cleaning and adapting the TB
-throttle cable does not stick
-MAF
-checked all vacuum lines and also PCV plumbing, replaced suspicious ones but there were no obvious cracks
-tried blocking vacuum lines to charcoal canister and to PCV without effect
-tried another ECU
-coolant temperature sensor works properly


The only thing I can think of which I did not test are vacuum check valves and PCV valve on the intake, but this should be ruled out by blocking the PCV vacuum line.

Could the idle regulation mechanism in the TB be shot? I removed the cover and everything looked nice and clean inside, no scratches on the resistive tracks.

To me this looks like metered air which is taking a different route to the manifold than through the TB, but I cannot figure it out. Or bad TB itself...


Thanks for any help  Wink
« Last Edit: March 20, 2025, 11:29:12 AM by doktor » Logged
grayjay
Newbie
*

Karma: +3/-1
Offline Offline

Posts: 23


« Reply #1 on: March 20, 2025, 05:39:40 PM »

Idle stabilization valve can get gummed up, try spraying out with penetrating oil and hook up 12V to it to see if it moves freely.
Logged
JeanAwt
Sr. Member
****

Karma: +11/-12
Offline Offline

Posts: 260


« Reply #2 on: Yesterday at 07:01:57 AM »

"The only thing I can think of which I did not test are vacuum check valves and PCV valve on the intake, but this should be ruled out by blocking the PCV vacuum line."

The small air vent valve on the lower engine located on the T-hose may be broken and allow air to enter the intake. Also check for leaks in the downpipe joints as this will distort the lambda reading...
Logged

A4 B5 avant 1.8T 2000 NO QUATTRO
doktor
Jr. Member
**

Karma: +0/-0
Offline Offline

Posts: 48


« Reply #3 on: Yesterday at 03:39:51 PM »

grayjay: This does not have an IACV, it is a part of the TB. I already dismantled it and everything is clean and nice, the resistive tracks are not worn, I even hooked up a multimeter and there are no dropouts.

JeanAwt: I will replace the valve as I am really not 100% sure it is good, but anyway, completely blocking this vacuum line did nothing so I am sure this will not help with my idle.

I have no other ideas than to replace random parts now. The throttle body looks clean and nice, moves smoothly, but I will try replacing it anyway as I am out of ideas...  Huh Huh I cannot see any other source of this problem.
Logged
JeanAwt
Sr. Member
****

Karma: +11/-12
Offline Offline

Posts: 260


« Reply #4 on: Yesterday at 11:57:33 PM »

The small check valve works like this: when the throttle is closed, so when you idle/decelerate, you suck in some of the low engine gases, and in positive gear it is closed. The entire PCV system must be in good condition, this amount of air is calculated. But not enough to make it go up to 1500 rpm...there must be something else, leaks.


Check with vcds the lambda values...rpm...temperatures...idle speed...
« Last Edit: Today at 12:02:55 AM by JeanAwt » Logged

A4 B5 avant 1.8T 2000 NO QUATTRO
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.03 seconds with 18 queries. (Pretty URLs adds 0s, 0q)