Lambda tuning with AEM OBDII WB

[Ghostnotes]

Been tuning in AFR and my trims are good but I want to be able to tune in lambda.

Problem is I'm not 100% sure how to set it up. With the old versions you would set your offset along with commanded/14.7 against your WB. Here it's a little different.The properties are still the same but again....the how has me confused. Since this WB is already reporting lambda do I just plot it against the EQ ratio commanded or similar?

Would it be a matter of:
[6001]/14.xx]/[36.238] OR [36.238]/[68.238]

or something similar?

Thanks

[MakesBadDecisions]

I just set all my channels and graphs to report lambda for everything pertaining to CMD and Actual. I then use a custom math to AVG both WBO2s and apply a small buffer, (100 * ((([36.238]+[37.238])/2) - [50118.238]) / [50118.238])*.85 . It took me a long time to convert my brain to Lambda, still have to use my cheat sheet conversion to change GM EQ to AFR then to Lambda when my brain stops working and for sanity.

Hope this helps some. I can share my layout and channels list with tune with current log so you can see how I look at it, might help some.

[Higgs Boson]

if you're using the AEM just right click the logged channel and Units -- Lambda

EQ Ratio displays as lambda in the scanner so yes, you would log both of them and use the preset Math in your layout for EQ Ratio error. You don't have to build anything anymore.

[Ghostnotes]

I just set all my channels and graphs to report lambda for everything pertaining to CMD and Actual. I then use a custom math to AVG both WBO2s and apply a small buffer, (100 * ((([36.238]+[37.238])/2) - [50118.238]) / [50118.238])*.85 . It took me a long time to convert my brain to Lambda, still have to use my cheat sheet conversion to change GM EQ to AFR then to Lambda when my brain stops working and for sanity.

Hope this helps some. I can share my layout and channels list with tune with current log so you can see how I look at it, might help some.

Thanks I appreciate that I assume the buffer is for the narrowband swing?

if you're using the AEM just right click the logged channel and Units -- Lambda

EQ Ratio displays as lambda in the scanner so yes, you would log both of them and use the preset Math in your layout for EQ Ratio error. You don't have to build anything anymore.

Been meaning to post back but have been out of town and then dealing with a major disaster her in Houston...

I kind of thought that but it's been forever since I tuned in lambda. The preset knows to look at the WB?

When I did the [36.238]/[68.238] and tried to correct it, it seemed not to work like I remember. For instance if it's reporting 1.10, I add that to the table but it seemed to add even more. So I set the value back and just to see, I went the other way just in case it was calculating backwards, so I multiplied by .90. Trims were still off so I'm assuming that equation was incorrect.

[Higgs Boson]

yes, just use the math, it's simpler. you can look at the equation in the math if you want to see the right one.

[8850]

Thanks I appreciate that I assume the buffer is for the narrowband swing?




Been meaning to post back but have been out of town and then dealing with a major disaster her in Houston...

I kind of thought that but it's been forever since I tuned in lambda. The preset knows to look at the WB?

When I did the [36.238]/[68.238] and tried to correct it, it seemed not to work like I remember. For instance if it's reporting 1.10, I add that to the table but it seemed to add even more. So I set the value back and just to see, I went the other way just in case it was calculating backwards, so I multiplied by .90. Trims were still off so I'm assuming that equation was incorrect.

I used this spread sheet to correct my offset.

[MakesBadDecisions]

Ive found that the "buffer" allows me to hammer the changes with paste and a "full" multiply to rough things in, then go to paste and multiply by 50% to polish it off. The "buffer" helped with overshoot and has made the process a little faster, in my opinion of course. This is all in open loop MAF only. I filter my trims by trim cell using high cell counts and "buffer" that math when closed loop "polishing". My process takes a little longer than most as I build my VE table while doing my MAF. By the time I get to the closed loop "polishing" my VE table has taken shape. When it is time to kill the MAF and dial the VE table it takes little time to go over and check it. Most of time it is so damn close it takes very little time.

These are only my observations and opinions.