interesting AFR behavior @ idle

Just checked for boost leak using the method described on this post and did not find any.

 

Here is my situation: car is in gear, rolling say ~30mph and then I put it in neutral. AFR then goes to ~20 (which is to be expected), as the car slows down and engine is idling.

Then as the car gets to a stop and engine is still idling, AFR drops to 13.5-13.7 instead of 14.7. It then tries hard to reach the target AFR but even after 15 sec of idling on a standstill, it only reaches about 14.1-14.3.

 

It also appears that this behavior happens sometime while cruising, but I need to confirm this via logging.

 

I read over on NASIOC that this could be the AVCS slightly out of sync.

 

Any thoughts?

that's all pretty normal behavior. the slightly rich idle can be improved in the tune by removing the rear o2 sensor's bias.

Can you expand a bit on the rear 02 sensor's bias?

Here is the romraider post about it: link

It doesn't seem to be a commonly available value to change in the tune. For me, I had to add it to my ecu_defs file in RR, then I was able to zero out the 2 values.

 

Basically the rear o2 sensor does actually influence the a/f ratio. The goal of the above tweak is to remove that influence. For me, I did see my idle get closer to 14.7:1 rather than 14.2 ish. Another nice side effect of this tweak is that it reduced a slight off-idle hesitation that was present once the engine was warm.

Here is the romraider post about it: link

It doesn't seem to be a commonly available value to change in the tune. For me, I had to add it to my ecu_defs file in RR, then I was able to zero out the 2 values.

 

^I'll have a look.

 

In the meantime, interesting read.

Basically the rear o2 sensor does actually influence the a/f ratio. The goal of the above tweak is to remove that influence. For me, I did see my idle get closer to 14.7:1 rather than 14.2 ish. Another nice side effect of this tweak is that it reduced a slight off-idle hesitation that was present once the engine was warm.

 

After reading the LGT thread link I posted above and re-reading your comments, it looks like it is exactly what is going with my car.

 

Can the rear o2 sensor negatively impact (read rich condition) the AFR while cruising too then?

i suppose it could affect cruise, but it mainly affected idle afr on mine.

 

although cruise afr seems to bounce around less now.

Thanks for your input heiche.

Interesting...my car has always irritated me because it won't idle as close to stoich as I would like. Even cruising in closed loop puts my car around 14.1-14.2. I have zero boost/vacuum leaks and my maf scaling is all within 1%.

 

I would love to have a fix for it as well. Thanks to both of you. Now to learn more about how to remove the rear 02's bias.

The idle air/fuel ratio is normal to be 14.1-14.3 vs 14.7 on your model. You can change it in the Closed Loop Load Compensation Table.

So when someone goes stage 2 with a non catted d/p and flash the ecu with an ap,does the stage 2 map disable the rear o2 or do you still have to do what was posted above?

The idle air/fuel ratio is normal to be 14.1-14.3 vs 14.7 on your model. You can change it in the Closed Loop Load Compensation Table.

 

This is it guys, just confirmed by doing this in my tune, and it definitely brings closed loop right to stoich if you want it that way. In my case, i took all the slightly negative values (which subtract from the AFR, making it richer) and just zero'd them out.

 

I would say that if the above doesn't work for you, also investigate disabling the rear o2 bias as well, since that does seem to affect things too.

So when someone goes stage 2 with a non catted d/p and flash the ecu with an ap,does the stage 2 map disable the rear o2 or do you still have to do what was posted above?

 

I don't know for sure, but my guess would be yes, that would already be part of a tune designed for a catless dp.

I don't know for sure, but my guess would be yes, that would already be part of a tune designed for a catless dp.

 

It will get rid of the check engine light for using an aftermarket downpipe, but I don't think it will disable the rear 02 sensor. I could be wrong, but they design all of the Cobb maps to be run with catted DP's.

This is it guys, just confirmed by doing this in my tune, and it definitely brings closed loop right to stoich if you want it that way. In my case, i took all the slightly negative values (which subtract from the AFR, making it richer) and just zero'd them out.

 

I would say that if the above doesn't work for you, also investigate disabling the rear o2 bias as well, since that does seem to affect things too.

 

Which ones do you zero out that are "slightly" negative? Could you post a screen shot of your load compensation table? Thanks.

Which ones do you zero out that are "slightly" negative? Could you post a screen shot of your load compensation table? Thanks.

 

In the top left corner, there are 2 values for -0.441, these are the 2 responsible for the richer values at idle.

In the top left corner, there are 2 values for -0.441, these are the 2 responsible for the richer values at idle.

 

There was a discussion with some screenshots here.

There was a discussion with some screenshots here.

 

Thanks. There isn't actually a screenshot of an updated CL Compensation table via RR but there is at least a nicely outlined way to correct the table when comparing it to your open loop fueling table.

After reading the threads posted, it seems that just zeroing out the values in the load compensation table will just force the car into open loop in the cells you zero out. This works great at idle because most open loop tables will have 14.7 afr at idle. However, during cruise I don't think I want to be in open loop all of the time.

maybe then try setting the 2 cells in the upper left that are -0.441 to 0.

maybe then try setting the 2 cells in the upper left that are -0.441 to 0.

 

Yeah, that is fine and is what I will do.

So I changed my CL Load Compensation Table. The two idle cells were changed to zero and the cruise cells (up to 1.00 and 3200) were changed to -.001. Idle is at stoich and cruising is much nicer as well. After logging, it also appears that I stay in closed loop during cruise too.

So I changed my CL Load Compensation Table. The two idle cells were changed to zero and the cruise cells (up to 1.00 and 3200) were changed to -.001. Idle is at stoich and cruising is much nicer as well. After logging, it also appears that I stay in closed loop during cruise too.

 

Nice. I need to figure out how to do this.

If you have a way to opensource tune it is easy. Actually, it would be just as easy, if not easier with accesstuner.

If you have a way to opensource tune it is easy. Actually, it would be just as easy, if not easier with accesstuner.

 

My tune is opensource and I am using romraider to log and ecuFlash to read/write to the ECU. My tuner is still in the process of refining my tune though. I'll point him to this thread.