PE filters, etc...

hey guys, i have searched high and low for what ppl are using for a pe filter string and everytime i find a file it wont let me open it:rant:, so if its all the same to you guys what are you guys using to filter out pe when tuning the VE table, ill be done breaking in my new clutch soon and when im done ill be traveling higher in the rpm band and don't want any contamination from the PE. if there is any other filter string you guys use to help sift through the false data it is greatly appreciated.

thx guys,
Tom

There is really no need to filter out PE unless you are tuning VE using the LT/ST trims. PE merely increases the enrichment but your combined VE (variable being tuned) and commanded enrichment (fixed) still must show a 0.00% error if VE is tuned properly. But to tune VE (or MAF for that matter) properly you will need a wideband.

The filter is your commanded AFR/Lambda PID >14.0 or .99 respectively. [PID.6001]>14.0 The histo will only populate with data when the comanded AFR/Lambda is at stoich. Remember to add the Commanded PID (example is the Hi Res version) to your chart display table. You need to be looging it for it to work.

Ed M

Or you can use this and make corrections all throughout with PE still being used.

(([AUX.20121]/[PID.6001])-1)*100

You need to plug in your wideband where the AUX.20121 is. Just go through the pid list and find and copy yours. Then you have to datalog commanded Hi Res AF, which should be the PID.6001.

This is what I use, because you want PE on boosted cars. Especially when tuning ;) This will also give you a plus/minus value identical to fuel trim corrections which helps knock out the tuning quicker.

Then for fine tuning in closed loop using fuel trims, you can use this for an all out average, which also helps knock it out quicker... Just make sure your logging stft's and ltft's for both banks.

((([PID.6]+[PID.8])/2)+([PID.7]+[PID.9])/2)))

Not to pick hairs but PID 6001 is the AFR hi res, not the EQ Ratio. If using the EQ Ratio, just also make sure you check the "polarity" of the EQ Ratio...I have seen them called this but really turn out to be Lambda and vice versa....check the scan that the pid values transition in the right direction lambda 1.0 to say .85 and EQ Ratio 1.0 to 1.15). To Make EQ Ratio equal to lambda just take 1/EQ Ratio. Obviously your commanded PID and Wideband reports have to be of the same format.....can't mix Lambda and EQ Ratio......

Ed M

Quote:

---

Originally Posted by mowton

Not to pick hairs but PID 6001 is the AFR hi res, not the EQ Ratio. If using the EQ Ratio, just also make sure you check the "polarity" of the EQ Ratio...I have seen them called this but really turn out to be Lambda and vice versa....check the scan that the pid values transition in the right direction lambda 1.0 to say .85 and EQ Ratio 1.0 to 1.15). To Make EQ Ratio equal to lambda just take 1/EQ Ratio. Obviously your commanded PID and Wideband reports have to be of the same format.....can't mix Lambda and EQ Ratio......

Ed M

---

On the gen 3 and the E40 EQ is the inverse of lambda. On the other gen 4 (at least the E38) it is commanded lambda.

Quote:

---

Originally Posted by mowton

Not to pick hairs but PID 6001 is the AFR hi res, not the EQ Ratio. If using the EQ Ratio, just also make sure you check the "polarity" of the EQ Ratio...I have seen them called this but really turn out to be Lambda and vice versa....check the scan that the pid values transition in the right direction lambda 1.0 to say .85 and EQ Ratio 1.0 to 1.15). To Make EQ Ratio equal to lambda just take 1/EQ Ratio. Obviously your commanded PID and Wideband reports have to be of the same format.....can't mix Lambda and EQ Ratio......

Ed M

---

Your both right. Sorry for the miswording. Post corrected.

Quote:

---

Originally Posted by Eaglegoat

On the gen 3 and the E40 EQ is the inverse of lambda. On the other gen 4 (at least the E38) it is commanded lambda.

---

E67 is this way too.

Should note...the E38 and E67 use EQ ratio in the tune, but the data in the scanner is lambda. So when you want to hit 0.8 lambda, you need to command 1.25 eq.

Thats why I developed 2 different sets of PIDs (one for commanded and 1 for error) to handle each of the derivations to stay in Lambda on my Wideband.

Ed M