N52 Revs Bouncing on Warm Start

[Mainzer05]

https://streamable.com/lcu3qc?src=player-page-share

3.0si so N52: Is it normal for the revs to bounce around like this then settle and idle normally the whole drive? It was about 34 degrees F ambiently, but the car was already to temp from running errands and I had just run into the store before starting it in this vid.

I’m also getting at best 20 mpg US driving around town responsibly. Had a prolonged crank after the car sat for a week. Battery is brand new. Running rich for some reason?

 

[TOMGREEN413]

Might be Vanos solenoids settling down, mine does this very slightly for a few seconds on cold start up but not as much as yours. I don't think it's anything to worry about. This is quite a common thing but others far more knowledgeable will be along shortly.

 

[r3vmatch]

I suspect this issue is related to the charcoal canister / EVAP / fuel purge solenoid.

Basically I think our cars end up sitting more often than not if they're not a daily driver, and as the car sits any fuel vapor from the gas tank is designed to leave the fuel tank and get stored inside a charcoal canister.

There's some logic in the ECU but basically on a warm start after the car has been driven for >20 mins, on the following start the ECU activates the EVAP solenoid which is located right under the throttle body ; there's a tube runs from the charcoal canister to the "inlet" of the EVAP valve/solenoid and the "outlet" of the valve goes to the intake behind the throttle body.

When the ECU enables the solenoid, vapors are supposed to be pulled from the canister through the intake into the engine.

On my car, while trouble shooting a rough idle, I decide to replace this valve with a new BMW OE part ; the very next time I had a warm the start the revs were bouncing so badly and the car was shaking, I had to turn it off. I put the old valve back and bench tested the new valve which worked fine (I activated it with 12V, current-limited) and just blew air into one port and the air would flow freely.

After doing a deep dive on this what I concluded is the old valve may have been restricted in its flow, the new valve was not, and inside my charcoal canister, since the car sits more than it runs, vapor / fuel has soaked the canister and when the EVAP valve is activated, the engine is pulling down way more fuel than the ECU was expecting.

As usual, no code.

All of this is a theory but it conincides with what've seen / read elsewhere. There's a youtube video of a guy (I think in AUS?) who had a rev bouncing issue on warm start and after BMW did a bunch of worked it turned out the be the EVAP / purge valve which got me thinking down this path way.

 

[r3vmatch]

Also,

Looking closely at your video it feels like an "idle hunting" issue more than a rev bouncing, but the EVAP theory might apply still.

Here's some info on the EVAP / DML system for the curious : https://www.princeton.edu/~sdepasqu/bmw/pdfs/Leak.Diag.Sys.pdf

 

[Mainzer05]

I suspect this issue is related to the charcoal canister / EVAP / fuel purge solenoid.

Basically I think our cars end up sitting more often than not if they're not a daily driver, and as the car sits any fuel vapor from the gas tank is designed to leave the fuel tank and get stored inside a charcoal canister.

There's some logic in the ECU but basically on a warm start after the car has been driven for >20 mins, on the following start the ECU activates the EVAP solenoid which is located right under the throttle body ; there's a tube runs from the charcoal canister to the "inlet" of the EVAP valve/solenoid and the "outlet" of the valve goes to the intake behind the throttle body.

When the ECU enables the solenoid, vapors are supposed to be pulled from the canister through the intake into the engine.

On my car, while trouble shooting a rough idle, I decide to replace this valve with a new BMW OE part ; the very next time I had a warm the start the revs were bouncing so badly and the car was shaking, I had to turn it off. I put the old valve back and bench tested the new valve which worked fine (I activated it with 12V, current-limited) and just blew air into one port and the air would flow freely.

After doing a deep dive on this what I concluded is the old valve may have been restricted in its flow, the new valve was not, and inside my charcoal canister, since the car sits more than it runs, vapor / fuel has soaked the canister and when the EVAP valve is activated, the engine is pulling down way more fuel than the ECU was expecting.

As usual, no code.

All of this is a theory but it conincides with what've seen / read elsewhere. There's a youtube video of a guy (I think in AUS?) who had a rev bouncing issue on warm start and after BMW did a bunch of worked it turned out the be the EVAP / purge valve which got me thinking down this path way.

Thanks for the response man! That makes total sense!

So what are you exactly saying I need to replace because you said you replaced the valve and it made it worse. Ok a start today it almost sounded like the car was backfiring.

 

[r3vmatch]

Hi there,

One thing you could try is to pull the electrical connector to the EVAP valve so when the ECU attempts to actuate it, you can see if it makes a difference on warm starts to see if it's anything to do with the charcoal canister / EVAP system.

Unfortunately the valve is burried at the bottom of the intake so you might have to remove the air/sound resonator cavity to get to it. I think each time I've messed with this, I have to remove the top part of the air filter, and the intake pipes, not hard but definitely tedious and time consuming.

If you try to remove the valve entirely I'm sure it'll be worse because you'll either let unmetered air in behind the throttle body or the emissions tests will catch a leak. Removing the electrical connector the valve should neuter it. I don't know if the ECU's smart enough to measure the current on the purge valve or not.

Before messing with this, you might want to make sure there are no vaccum leaks with a smoke test, and possibly to clean the MAF (100% isopropyl alcohol) & the MAP sensor.