Subaru Idle Fueling Issue

[Rob W]

This is becoming rather annoying.

Engine is a ej25/20 hybrid with 8.2:1 comp, has ID1000X in Radium fuel rails and BC272 cams degreed to advertised specs.

I am using a Ballenger AFR500 wideband which seems to be slow to come out of lean condition on overun and start up but other than that the guage reads as per the laptop readout.

It is in open loop idle. Closed loop fuel is turned off (it didn't help with this issue) Idle ignition control is on and generally works very well as set up. I have some extra fuel in IAT correction at higher ECT/IAT start up 0%TPS but is not affecting for this log. I did do some work with the charge temp table to try and work around this but it wasn't really helping so I put it out of the equation.

I've been spending time on the overall fueling and accel enrichments which I am reasonably happy with at this point, getting it ready to put on a dyno and optimise the set up. The cruise and WOT fueling is consistent at higher MAP and RPM.

The issue I'd like another set of eyes to help with is an idle issue that has developed. Once I go to 0%tps it starts to richen up dramaticaly and the revs drop and sometimes stalls out. Pulswidth seems to increase for no apparent reason and just dumps fuel in. Then the MAP starts to jump about. It's not all of the time. I can start it up and have it be fine, although the lambda is not always consistently the same each time under similar conditions.

The Idle Air Control Valve is 20 years old its a 1999 wrx. The idle clamp required at 80*-90*c is 90 points higher than it was a couple of years ago. So clamp was around 200, now its up around 300.

The MAP sensor is relatively new 3 bar GM style.

Not sure if it's a vac leak when warm but I've been over it all a thousand times. Or if the IACV is not good enough to catch it anymore or it's an issue with the MAP sensor. Or if it's tune related. If anyone can have a look over the log and tune file and can see what might be happening with the tune and/or mechanically that would be much appreciated.

You can see from 55:20 approximately where the issue is really bad and I've had to tickle the throttle to catch it. There are a couple of other places in the log there where you can find idle and see I've had to do the same thing

 

 

https://1drv.ms/u/s!AsYB-OUQHJQzzmCWxykRzAR8sPEQ?e=CvaB22

https://1drv.ms/u/s!AsYB-OUQHJQzzl-lRWqKN5kMUnxJ?e=wsXDhV

 

[Ducie54]

Few things i've noticed

While your dead time are correct you don't have any figures in the SPAT.

You have got the idle stepper reset on key off but don't have a ecu hold time.

 

[Rob W]

When i put the ID,s in it I put the figures in the SPAT and it wasn't real happy so i cleared it back out. But i can try those again now. Im not sure about the hold time issue but now you mention it I'll look into thst further. Can you give me an insight into how that should be?

[Ducie54]

In the help file type in Stepper Reset, explains 2 different ways for a reset. 5 sec hold time should be enough i reckon. I would give it a clean first so it give repeatable results.

I would also target a slightly richer AFR at idle until you get the idle sorted .95 is always my start point until i get idle stable.

 

[Adamw]

2 hours ago, Rob W said:

Im not sure about the hold time issue but now you mention it I'll look into thst further. Can you give me an insight into how that should be?

Without this the ECU will turn off as soon as ign is turned off so the stepper cant re-zero itself.  This means next start up it will not be in the position the ECU believes it to be and will mess up idle control a lot.  So this is the first thing that needs to be fixed before we can really diagnose if there is an idle problem.  As per Ducies advice, 5 seconds will do.  Do us another log after changing this. 

 

Edit, added later:

In both places it nearly stalled the fan had just turned on as it came down to idle, Your fan step is set to 70 (crazy number?) but the idle valve only changed by about 3 steps - did you change this setting after the fact?  Your fan is turning on/off a lot so I would set the hysteresis higher so that doesnt happen as often.  You will want to set the fan step properly.

It looks like overrun fuel cut is also playing a part.  I would either turn it off temporarily to see how much it is interfering or bump up the highlighted cells to about 1800RPM.

[Rob W]

Well. You blokes are awesome. No wonder I've been chasing that stepper motor around for the past couple of years. 

I realised that it reset on key off but the ECU Hold Power had not computed. I have set up AO5 as Ecu hold power. The DI was already set up with ignition switch.

At some point in time I was adding fan step to stop it stalling when they came on. Initially it was set to 6. 

This table here seemed to com as part of a firmware upgrade and has never been populated. Any suggestions?

 

I have made all the changes suggested and will get the car out and see how things go and report back with good results or a new log and tune file

[Adamw]

Is this a WRX V5/6 car/ecu?  The ecu hold power needs to be on Inj drive 6 for it to work.  The output was already assigned correctly in your map above but you have the keep alive time set to zero.

[Rob W]

Righto. Thanks. Yes it is v5/6. I set the alive time to five seconds.

[Rob W]

Awesome you blokes. Seriously. Thanks so much for that

I set up the ECU hold time as advised. Went back through the open loop idle procedure. The clamp table needed about 90 points taken out of it at 80* so it's back to where it started from a couple of years ago.  

Rock solid at every intersection. So far so good. Now I might have a half decent chance of setting up the closed loop idle.

[Ducie54]

Glad it works. Always frustrating chasing the little things.

Another thing to look at is the over run fuel cut settings. At 1.2% TP threshold you may find you can reach 1500 rpm with less TP% so when sitting stationary and holding the rpm up it will bounce in and out of overrun fuel cut.