Idle goes under target and wants to stall

[Volcom86]

I am having a strange issue that was not present till now.

My idle will drop under my target idle value and it will stall at times. I am noticing  this for example when I am reversing parking when I need to be gentle on the accelerator and just give it lite blips. 

Sometimes when I am pulling up to a set of traffic lights. Does not happen all the time. 

Any idea what is making this? 

Attached log and map

https://www.dropbox.com/s/v46290ewszho9vd/Idle and cruise .llg?dl=0

 

2005 STi with cable throttle body

 

I appreciate any feedback thank you. 

 

[Adamw]

I think it is mostly a fuel issue, everywhere in your log where there is a near stall, including in your screenshot above, the lambda is pegged at max lean for a while before the stall (it looks like 1.25Lambda or 18.4AFR is the leanest your wideband is set up to read) so it is actually even leaner in these areas than the log shows. 

Im not really sure why though, there are other areas where you have the same MAP, same RPM, same PW, same fuel pressure but the lambda is fine.  I suspect possibly the injectors are working in a very non-linear region.  Are they the Bosch CNG injectors?

You could try bumping the min eff PW up to about 0.6 or 0.65 and see if that helps the injector control better.

 

[Volcom86]

2 minutes ago, Adamw said:

I think it is mostly a fuel issue, everywhere in your log where there is a near stall, including in your screenshot above, the lambda is pegged at max lean for a while before the stall (it looks like 1.25Lambda or 18.4AFR is the leanest your wideband is set up to read) so it is actually even leaner in these areas than the log shows. 

Im not really sure why though, there are other areas where you have the same MAP, same RPM, same PW, same fuel pressure but the lambda is fine.  I suspect possibly the injectors are working in a very non-linear region.  Are they the Bosch CNG injectors?

You could try bumping the min eff PW up to about 0.6 or 0.65 and see if that helps the injector control better.

 

Quote

 

Thanks Adamw, I will give that a go...

My Injectors are ID 1,700x 

My tuner has tuned the car with a different closed loop gain this was the original as I had issues with idle AFR at idle hunting however the idle was stable 

 

And these are the new tweaked settings 

 

So I am not sure if this has any affect on the idle issue. 

The other thing I Should point out is this is the first time me really trying this on 98 petrol could this effect anything coming from e85 ??

[Adamw]

Ah yes, now I look closer some of it is caused by closed loop lambda.  Its pulling lots of fuel out in cruise conditions then when you come to idle it cant remove that trim quick enough.

I would change the CLL RPM low lockout to about 1600RPM so it is disabled at idle and when coming into idle.  The min Eff PW that I mentioned above will probably help a lot too.  

[Volcom86]

Thank you @Adamw I will give it a try and report back. As always I appreciate your feedback and help so thank you!  

[Volcom86]

@Adamw

Did some testing this morning and BOOM the idle issues are gone! Now I do have an issues with Idle being at 11AFR not targeted at 14.7AFR 

Adjusted the The min Eff PW and the CLL RPM low lockout at 1,600RPM 

https://www.dropbox.com/s/ojlb6va0vd9hirn/111.llg?dl=0

The CLL seems to be out more then before. One thing is fixed then other issues arrive  

Thank you 

[Mario Schneiderbanger]

following since i have the exact same issue on my car and also running 1700x with e85

[Adamw]

5 hours ago, Mario Schneiderbanger said:

since i have the exact same issue on my car

If you have the same issue then the same fix will apply.

[Mario Schneiderbanger]

adam, by setting the cll lockout to 1600 you are effectively disabling cll at idle always. 
 

can't we set up an activation delay where if there is a change in tps the cll waits say 10 seconds before engaging? that way while backing out of the garage or inching up in traffic, as soon as there's a delta TPS logged it disables cll and waits 10 seconds or whatever you want before it activates again?

 

this way it won't try to reduce fueling while you're puttering around but will engage if you are sitting in a parking lot for a while 

[Adamw]

10 hours ago, Mario Schneiderbanger said:

can't we set up an activation delay where if there is a change in tps the cll waits say 10 seconds before engaging? that way while backing out of the garage or inching up in traffic, as soon as there's a delta TPS logged it disables cll and waits 10 seconds or whatever you want before it activates again?

Yes, this is how it works already.  Anytime it has been locked out, when the lock out conditions are no longer true then there is a re-activation timer that times out to allow Lambda to stabalise before the CLL is re-enabled.  The length of the re-activation timer is hard-coded in the G4+, but from memory is 1-3 seconds depending on conditions.

That would not help the OP's problem as he hadnt even met any lock-out conditions.

 

10 hours ago, Mario Schneiderbanger said:

adam, by setting the cll lockout to 1600 you are effectively disabling cll at idle always. 

That was just the quickest fix to get around the OP's problem which was due to the CLL pulling fuel out in the cruise area so when he pushed in the clutch and suddenly wanted to idle the CLL couldnt add the trim back in quick enough.  You typically dont have this issue but with 1700cc injectors on petrol the PW is so short (he is down to 0.3ms before it nearly stalls) things get very sensitive.

A more correct fix would be to tune the VE slightly lean in the cruise area so the CLL is adding fuel instead of removing - then when it comes to idle it is just slightly rich for a moment until CLL corrects it.