Set up Closed Loop O2 Feedback

[MattSpyderVR4]

What about O2 FB Over Clear?  Does this set the O2 FB Value to 0 when the Maximum Load or Maximum RPM values are exceeded?  

And, a related question if the above is true:  If I have O2 FB Maximum Load set to 0 PSI, and O2 FB Over Clear is off, I know the O2 FB value will not be updated while I'm boosting.  But will the O2 FB value still be applied to the fueling under boost, even though it's not being updated?  I'm seeing some behavior on my car in boost that leads me to believe this is the case.

My car appears to act as yours does also.  I ended up making the whole O2 feedback table have sane values and have O2 feedback on all the time to compensate for it.  Would be nice to have a new help file or something to document all the new checkboxes/options for v1.11.  

[FWombat]

What about O2 FB Over Clear?  Does this set the O2 FB Value to 0 when the Maximum Load or Maximum RPM values are exceeded?  

And, a related question if the above is true:  If I have O2 FB Maximum Load set to 0 PSI, and O2 FB Over Clear is off, I know the O2 FB value will not be updated while I'm boosting.  But will the O2 FB value still be applied to the fueling under boost, even though it's not being updated?  I'm seeing some behavior on my car in boost that leads me to believe this is the case.

My car appears to act as yours does also.  I ended up making the whole O2 feedback table have sane values and have O2 feedback on all the time to compensate for it.  Would be nice to have a new help file or something to document all the new checkboxes/options for v1.11.  

I confirmed that O2 FB Over Clear does what I thought it did; it sets the O2 FB value to 0 when you exceed your max load or max RPM for O2 FB.  My boost fueling is consistent again!  I'm so happy.    Before, my O2 levels in boost were always a straight line, but varied quite a bit from pull to pull, because it was applying whatever the last O2 FB value was from cruising.  Took me a while to figure out what was going on.  

I go back and forth about O2 FB while boosting; I'd hate for a loose O2 sensor connection or something make me go too lean while boosting!  But on the other hand, it would be nice...to have some correction.  I'll probably end up trying it one of these days.  Has it worked well for you?

Yeah, updated docs would be nice. Actually, I'm still running 1.03,  but even that old version has the new undocumented options.  I'll get around to upgrading one of these days...

- Brian

[BLKMGK]

Hrm, O2 FB Over Clear is OFF on  my mapping 8O Appreciate you pointing this out! The latest AEM DL has updated docs, have you checked them? I've not loaded the new version so I don't know if the .HLP file is updated but I've gotten a copy of the new docs - just not gotten them printed and read yet

[Flr Power]

You can always limit the level of corection ta a safe -5% if you worry about getting too lean. I boost up to 14 psi with O2 Feddback enabled and it works so nice. Even if the sensor connection go bad it will only correct to the min max you have choosen. I have actually tuned my car on the rich side forcing it to trim fuel in the negative value. I am using -8% and 8% for O2 FB rich and lean limit. Try it

[barrydklein]

Proportional feedback only (O2 FB integral table = 0)
O2 FB proportional table = 0.5 (Gu)

Proportional and Integral feedback
O2 FB proportional table = 0.45 (Gu)
O2 FB integral table = (Pu)/1.2

These are the same equations for P+I control given in the Boost Control sticky. For boost control it seems that you just plug these numbers in on the options screen.

For O2 feedback, there are tables/graphs for proportional and integral control. I'm not sure how to enter the "0.45 Gu" and "Pu/1.2" values graphically.

What am I missing?

Thanks,
Barry

[asphalte]

the learning mode will adjust the fuel MAP using AFR target map isn't it?

[mx5]

the learning mode will adjust the fuel MAP using AFR target map isn't it?

not the closed loop target map. there is another map for the autotune targets - just check your software

[JasonB]

Proportional feedback only (O2 FB integral table = 0)
O2 FB proportional table = 0.5 (Gu)

Proportional and Integral feedback
O2 FB proportional table = 0.45 (Gu)
O2 FB integral table = (Pu)/1.2

These are the same equations for P+I control given in the Boost Control sticky. For boost control it seems that you just plug these numbers in on the options screen.

For O2 feedback, there are tables/graphs for proportional and integral control. I'm not sure how to enter the "0.45 Gu" and "Pu/1.2" values graphically.

What am I missing?

Thanks,
Barry

I am curious how you turn this into a graph as well.

[barrydklein]

I spent some time tonight playing with O2 feedback (at idle) after doing some reading about P+I feedback on the internet.

Observations:

Proportional gain is fairly insensitive. I turned integral gain off and tried to get a persistant oscillation by increasing proportional gain. You have to max out the settings (-120) and increase global gain to 2 to achieve this. Increasing proportional gain increases the speed that you return to your target setting.

Integral gain is very sensitive. If you make it slightly more negative (-10 or lower) your oscillation increases. A smaller negative number (-1) decreases oscillation, but it takes longer to reach the target AFR. Integral gain is what determines how accurately the feedback tracks your target setting.

Conclusion:

The settings in the startup calibration seem very good as is. Any of my adjustments only seemed to make things worse. It also helps to have your fuel map as close to your target AFR as possible in order to minimize target error.

I am pretty low on the learning curve on this.

Regards,
Barry

[Mk3ukr]

After tuning my off-boost area of the fuel map I enabled O2 feedback. All options are the same as in base calibration. For some reason my parameter O2 #1 FB always stays +7.8% and not oscillating at all    
Any ideas how to help it?
Thanks

[mehlosTT]

After tuning my off-boost area of the fuel map I enabled O2 feedback. All options are the same as in base calibration. For some reason my parameter O2 #1 FB always stays +7.8% and not oscillating at all    
Any ideas how to help it?
Thanks

It appears that your fuel map in the off boost areas ( zero psi row and downward to 12 ish psi, needs about 7-8% more fuel in each cell. The O2 #1 FB showing a + 7.8% means that the ecu is trying to get the dialed in AFR value , as listed in the O2FB Table,  however_ the Fuel Map cells values in those Off Boost rows are still too small to accomplish that _without allowing the O2FB to do it's thing_ as in adding in  7.8% more fuel to reach the desired AFR.

[twoqik4u2]

or the ecu is trying to add more than +7.8% fuel to achieve correct afr and the limit is set to 7.8%.  Try increasing your limit to +-15%.

[Mk3ukr]

Thank you guys for trying to help me .I wish everything was so simple. I need to explain all details of my setup.
O2 feedback enabled only in off-boost area (less than 100kPa)
O2 feedback target table set to 14.5 all  cells except idle
O2 feedback rich/lean limits set to 7.8%
UEGO sensor installed about 3 feet after the turbo
I tuned my off-boost area cell by cell to 13.5-14.5 AFR, so target O2 value is actually leaner almost everywhere to compare with my real AFR

When I enable O2 feedback my parameter O2 feedback value goes straight to +7.8% and stays there all the time. I have flat line of  O2 FB on my datalog. If I set O2 FB rich limit to 15% parameter goes to +15% and stays there. Even under boost when O2 FB is supposed to be off it is still there. So, I run very rich everywhere. I turned off O2 FB, don’t know what else I can do. I tried already to play with options and P + I Gain tables to no avail
I downloaded 1.19 version of software, maybe will have more luck with it. Though I am 1.5 month away from home and my car. Won’t be there soon to play with O2 FB again.    
I feel I am making some mistake in basic things.
It's too bad I can't enclose my calibration file here

[redevo7]

Did you have set the right O2 sensor in the Options-Injectors-Menu?

If you using the wideband as O2 Sensor #2, you also have to enable the feedback for the Injectors from O2 #2

i had also done this mistake when starting with a fresh map ( there is O2 #1 as default, but running the wideband as O2 #2)

[Mk3ukr]

Yes, I use O2 #1 and feedback for injectors enabled from the same #1 sensor. My laptop is not with me right now, but I remember I took care about this. Will check again my calibration tomorrow
Thanks