check my calculated pid

[ddnspider]

I was digging back through some of my calculated pids and realized I may have an issue. See the calculated PID below


*CLC-00-110
factor 0.5 1.5 .4 "{GM.EQIVRATIO}*{EXT.WO2LAM1}"

CALC.WO2BENSERIAL F110 CLC-00-110 factor Tuning "BEN from Serial Wideband"

From what I gather, GM.equivratio is (stoich/actual) and lambda is (actual/stoich), so isn't multiplying them together going to square the BEN? I feel like I am missing something or maybe I don't understand what GM.equivratio actually is. Something set me straight :)

[darcy]

No, lambda is lambda. Equiv.ratio is 1/lambda and AFR is stoich * lambda. AFR is a function of lambda, not the other way around.
Also, you're not making a distinction between commanded and actual.

That BEN calculation is the same as GM.EQUIVRATIO / ACTUAL.EQUIVRATIO where a bigger number indicates more fuel is being commanded than is achieved and thus MAF/Ve should be increased.

[ddnspider]

No, lambda is lambda. Equiv.ratio is 1/lambda and AFR is stoich * lambda. AFR is a function of lambda, not the other way around.
Also, you're not making a distinction between commanded and actual.

That BEN calculation is the same as GM.EQUIVRATIO / ACTUAL.EQUIVRATIO where a bigger number indicates more fuel is being commanded than is achieved and thus MAF/Ve should be increased.

Ah. I see my issue. Lambda is actual/commanded....not actual/stoich.

SO (1/lambda)*lambda=my BEN

[darcy]

Ah. I see my issue. Lambda is actual/commanded....not actual/stoich.


Not quite. COMMANDED/ACTUAL is BEN. This can be measured in AFR, Lambda, Equiv.Ratio or any combination of.

Lambda is a measure of stoichiometry.
There is Commanded Lambda (from ECM) and Actual Lambda (from WB02 as {EXT.WO2LAM1}).
Just as there is Commanded Equiv.Ratio ({GM.EQIVRATIO}) and Actual Equiv.Ratio (from WB02)

[joecar]

{GM.EQIVRATIO} --> commanded EQR

{EXT.WO2LAM1} --> measured LAM --> 1/(measured EQR)

[joecar]

From what I gather, GM.equivratio is (stoich/actual) and lambda is (actual/stoich)...

GM.EQIVRATIO = commanded EQR = 1/(commanded Lambda)

EXT.WO2LAM1 = measured Lambda = 1/(measured EQR)


GM.EQIVRATIO comes from the following:
- in CL it is 1.0000,
- in OL it is from B3605 (non-COS) or B3647 (COS),
- in PE it is from B3618.


EXT.WO2LAM1 comes from the wideband.

[joecar]

Lambda = 1/EQR (and vice-versa) independently for each of commanded and measured.

[ddnspider]

{GM.EQIVRATIO} --> commanded EQR

{EXT.WO2LAM1} --> measured LAM --> 1/(measured EQR)

So my BEN from Serial Wideband was correct, intentionally or unintentionally, because I was taking (1/commanded lambda)*(actual lambda from wideband)=BEN :)

[ddnspider]

Stupid question I know....I just finally had an afternoon to go play with my tune and I looked at my old logs and realized I hadn't messed with the tune in the car in 6 months, I guess life just gets in the way. Now if I could only resolve:

a) OLMAF tune car wants to go lean after driving for awhile (I played with the charge temp blending table with some decent results).

b) this stupid bucking at light throttle (10%) low speed (10-20mph) between 1500-2k rpms. I increased my throttle follower substantially based on the idle tutorial, but maybe I should add some to the throttle cracker. I tried logging my idle trims and they never move the entire log.

[joecar]

So my BEN from Serial Wideband was correct, intentionally or unintentionally, because I was taking (1/commanded lambda)*(actual lambda from wideband)=BEN :)Yes, correct.