High idle during cold start

[Julien]

Hi Adam,

During cold start if I get out of the idle control for high RPM lockout I get stuck at 2k rpm.

It's weird to me because when the idle control reengages the idle position goes to my idle base setting and rise it meaning it's too low.

Also the idle position is 2.5% higher than the base setting and I don't know where this correction comes from (I'm out of the startup offset at this stage)

Do you have any guess? Thanks

 

[Adamw]

The 2.5% offset is from the dashpot, I dont think this is your issue though.  The rev hang after the first blip is hard to see the cause, but before that you have very poor vacuum, (80Kpa is like 3/4 throttle on most cars), then when you blip it the vacuum suddenly improves and the lambda goes very rich - like a load has been removed?  The battery voltage doesnt really change but maybe the voltage reg has just dropped the alternator load back when you gave it a blip?  Does it have power steering or air-con or anything like that? 

There is some other quite unusual stuff going on in the log too, although I dont know if you were dragging the clutch or anything but I will assume not.  Why does the RPM come down at 1:32 when the throttle is open 10%?  Why does it only need 57% idle position at 1:34 to hit the target idle speed, then 6 seconds later it needs 65% to stay on target?  I can see after the 2nd blip the purge valve opens so that possibly is causing some variation sometimes but it didnt seem to upset much this particular time.  Is it normal for your EGT to be 650°C at idle?  I assume the B1 sensor isn't working?

[Julien]

Thanks for your reply but we are Sunday, you shouldn’t be here!! We can all wait until Monday!

No electrical load.. A/C is gone, and if I plane to install an electro hydraulic steering it’s not done yet. The car hasn’t run during the last week and this log is on the first startup so the battery may have been low. So the blip could have increased the current from the alternator even if the tension was stable. Giving the impression a load has been removed.. 

Could this lambda drop be because the timing goes from 9 to 27° producing a better combustion?

I agree for the 2,5 not being the cause and ok, it’s active during “idle lockout” so now I understand why I got this

 

I’ll make a fume test on the inlet to make sure I’ve no issue here.

I’ll also unplug the purge solenoid and disconnect it’s hose to breath the tank.

 

I took the car out of the garage then I blipped to get this behavior and then I drove away at 1:30 and stopped at 1:34 that’s where the rpm dropped and the idle control get back in action. I blip again but rpm dropped enough for the idle control to catch this time.. and at this point I stoped the log to keep it small.

I’ve no clue why it behaves like this.. maybe a leak in the intake but not constant 

 

Correct for B1, the thermocouple is plugged but I need to have a bung welded on the manifold to install it.. so atm it’s only laying in the engine compartment..

The B2 EGT is installed before the turbo that may be why you find the temp high.

Once warm the idle is stable..  but MAP is still at 67kPa..

[Julien]

Hi Adam, I tried to unplug the purge solenoid but the no improvement in the vacuum value...

I let the car warm up a little longer than usual and I couldn’t reproduce the 2K rpm issue. I'll give it another try later.

I also set the lambda target back to 1 were my hesitation are and they are still here so it doesn't seems to be related to the purge actuation..

I enclosed a longer log since cold start. I used the GP rotary switch in order to flag were the hésitations happen..

https://www.dropbox.com/s/4dj7t71tggd9q0x/PC Datalog - 2023-03-19 10%3B38%3B48 am.rar?dl=0

Thanks for your help !

[Julien]

Did another star this afternoon and same behavior at 2k..

The carbon canister purge can be eliminated. 

On this log at 1:35 I lowered the rev from 2K by slipping the clutch (the car being steady) and the vacuum dropped but the idle get back to normal.

Could it be too stiff springs in the dump valves?

 

cs5.llgx

[Adamw]

I think the idle valve is possibly working backwards.  Can change idle control to open loop mode, then with it idling, see if increasing the base position increases RPM or decreases it.

[Julien]

Hi Adam,

This was tested during the initial startup. I used open loop to define the base table before going to closed loop.

But I tested as you suggested and I can confirm that the valve works as intended. Increasing value increase the idle speed.

I don’t know how much pressure can make the turbos at idle. But I wonder if my dumps valves could be the problem. They are fitted the ‘wrong’ way from factory, I mean the high pressure is entering the side of the valve and not the piston. I replaced them long ago with Forge ones when I found one leaking.. I need to check but I think they are fitted with a stiffer spring to cope with 1.0 to 1.6b were I may should have fitted the sotfter one with this arrangement... Do you think that taking one out to see if I get more vacuum could be a good try? They are not really easy to access..

[Adamw]

The dump valves only bleed air from turbo side of the throttle dont they?  They could only infuence vacuum if they were connected to the engine side of the throttle body.  

The low manifold vacuum may just be normal for this car.  It seems a bit odd the it is different at idle some times than others but it may just be sensative to alternator load or something.

To me the the idle valve still has the most odd behaviour.  It seems it needs the valve open by a different amount everytime you come down to idle to achieve the same target.  Sometimes it can vary by 15% without a change in RPM.  I have seen the bosch barrel ones get sticky before - have you pulled it off and had a look inside it recently?  With them off, if you look through the side port you should see just a small window, it wont be fully closed - they spring return to a partially open position.  You shoud be able to rotate that barrel very easily in either direction with a screwdriver or similar and it should spring back to the partially open position by itself.   

Can you try the RPM lockout a bit higher to see if allowing idle ign to kick in earlier can help prevent the rev hang.  This may create other drivability consequences but it is worth a try.  Try it at around 800RPM. 

 

 

 

 

[Julien]

I did clean it a while ago and it was free at that time, I’ll check it again.

I enclosed a log of open loop the trial if you notice something wrong.

Concerning the dump valves, yes they are before the TB but they are supposed to be open by the vacuum when you lift the throttle.. and at idle if the spring is too strong they may not open, then the air from the compressors will be forced through the TB gap and the OSCV I suppose. Depending of the very low pressure they’ll make at idle..

idleopenloop.llgx

[Julien]

I tried with the hose from the intercooler to the dump valve out and it's the same

I'll increase the idle lockout and see how the car drives like this..

Thanks Adam

CS7.llgx