Closed Loop Lambda control and Mixture Map

[Tim D]

Hi, I have been trying to refine the fuelling on my Subaru Impreza STI 2.5. 

Please can I have some feedback on a couple of questions?...

1. Using Mixture Map to tune it, my fuel table looks quite spiky, despite using lots of samples.  I've seen other posts where the table looks much smoother, should mine look smoother?

2. During cruise, Lambda is steady, but on a 'pull', I can't get it much better than shown, is this good enough or should it be better?

Thanks in advance.

Tune and log attached...

Log 2019-02-3 3;42;13 pm (611 wickware).llg

611. Smoothed.pclr

[ClintBHP]

The problem you have is that the mixture map is using the AFR Lambda target table to work out what fuel ratio is required and where you are getting the spikes you have no direct relational axis so it has to guess, once you have a base tune use the mixture map and get the axis of the AFR table in sync where you need to tune it better and don't correct the sites you have already done.

Also turn off Lambda correction and overrun fuel cut while tuning or you will get no-where.

[Tim D]

Thanks Clint,

Are you saying I need to change my target lambda table from MAP to MGP and dimension it like the fuel table?

Thanks in advance.

Tim

[ClintBHP]

More getting the axis correct in revs and pressure, weather you use MAP or MGP or PSI makes no difference as both are only a math calculation of the raw ADC.

[Tim D]

I see, thanks, I will try that.

Cheers

Tim

[Tim D]

Clint, I'm confused... 

My fuel table uses MGP (to compensate for altitude effects) ? 

My Lambda target table uses MAP.

I'm not sure what you are suggesting when you say:

More getting the axis correct in revs and pressure, weather you use MAP or MGP or PSI makes no difference as both are only a math calculation of the raw ADC.

My target table does not have many cells (it's not a complex map). 

It was my assumption that the ECU would interpolate the X axis as required anyway.

Please could you expand on what you are suggesting?

Cheers

Tim

[ClintBHP]

No worries buddy; it's easier to work when both axis are the same so either work in absolute or gauge pressure and not both, so choose one and it will be easier to tune for the brain.

You engine will not want 14.7 AFR from -100 to -40 kpa MGP at all revs, your also using super unleaded which is 14.5 stoic anyway, and you will want to be somewhere on the rich side of this while your engine ROC is positive but you have no control over that with your current AFR target table as its not fine enough at the bottom.

Your overrun fuel cutoff being on is stopping you from mapping the low end of the table because it is sending the Oxygen sensor into a massive lean condition, so turn this off while you are tuning.

Also your Closed loop lambda is making the mixture map almost impossible, just look at the 2750 spike and all the corrections going on.

 

[Tim D]

Clint, you're a star, thanks for your help.

The more I learn, the more I realise I don't know!

With regards the overrun fuel cut, I use the filters on Mixture map to filter out any samples when this was active.  I guess you don't approve of that method from what you say?

Do you think I should keep 'IAT fuel trim' active during tuning (but all other corrections off)?  As I see it, I need to compensate for inlet temps otherwise the tune will be wrong when the weather warms up?

I always use V-Power or Momentum, so should I change my stoic value to 14.5, or keep it at 14.7 to cater for any fuel?

Sorry for all the questions!

Cheers

Tim