Filtering help needed

[bk2life]

Thanks to bindsquirrel and sirius for introducing me into filtering in the scanner.

Below is a scan, and three sections in the 'graph' area that are all the same, but error 1 has one type of filter, error 2 has a compbined filter, and error 3 has no filter.

I'd like to "Add a filter to remove all the data that's not EQ = 1 (e.g. open loop warmup/PE).."

And also used the picture attached below.

My quesiton is, how do i know what each filter does? And yes, i have read the hptuners instructions, and i see the filtering section, but, is there anymore indepth instructions on the topic? There is very little info here in the forums, and some of that info is well over 10 years old, and the instructions or pictures are outdated and dont work.

Seems to me filtering woudl be a big help, I'd just like more info to fully understand the topic. For example, what does EQ =1 mean?
And if I am way over analyzing this, sorry, i like to know every detail. I look at it as, hptuners is a tool, if i dont fully know the tool and how to use it, then it amkes me look dumb in the end.

thank you.

here is my current scanTRIP 3 stft ltft modified ve.hplwb eq filter.png

[gtstorey]

First step and hard to grasp sometimes is that although we call it a filter, what we put in the "filter" is what is getting passed on, not what is removed. If the expression is true, it passes on the data, if false it excludes it.

(EQ=1) as a filter means that the only data passed will be when EQ=1.

((EQ<1) or (EQ>1)) will pass data when EQ does not equal 1.

[SiriusC1024]

EQ is equivalence ratio. It's normalized AFR for the stoichiometric ratio of the fuel being used. EQ units are lambda.

When you see commanded EQ=1 that means that stoichiometric AFR is commanded (14.7/14.7 = 1).

When commanded EQ = 0.8 that means 0.8*stoichiometric AFR. 0.8 lambda = 11.76 AFR.

[gtstorey]

Is this a question of filtering or definition? Rereading OP, I could see it being either way.

[bk2life]

Is this a question of filtering or definition?.

yes to both your questions.

that is why i included the scan, did i write the filter correclty?

EQ is equivalence ratio. .

I understand that, for some reason i was putting fuel cells and EQ as one. For example, Fuel cell x is the same as EQ = 1

not sure how i got them mixed, but i was going to ask for a EQ/Fuel cell spredasheet to know what is what....lol

[SiriusC1024]

For what it's worth I don't use filters for NA. Fuel trims are automatically filtered. Might catch some STFT warmup, but it gets averaged out pretty well. Where you need to start paying attention to the wideband, PE, it's pretty obvious.
wb PE.png

You'll get the feel for it the more you do. I get a really solid VE table mapped after about 6 iterations.

I'll use a filter for OL, like a turbo application, that filters for stable MAP.
https://forum.hptuners.com/showthrea...l=1#post748278

ABS([50030.91]-[50030.91.shift(-10)]) < ([50030.91]/[50030.91.shift(-10)])*0.1

Speaking of. If you're interested in trying filters, put this into one of your filtered fuel trim readouts. Should make it more accurate because transients are removed (unstable MAP):
CL SD (filter for UMAP):

ABS([2311.71]-[2311.71.shift(-10)]) < ([2311.71]/[2311.71.shift(-10)])*0.1

[SiriusC1024]

I understand that, for some reason i was putting fuel cells and EQ as one. For example, Fuel cell x is the same as EQ = 1

not sure how i got them mixed, but i was going to ask for a EQ/Fuel cell spredasheet to know what is what....lol

What do you mean by fuel cell? The fuel trims from STFT or LTFT?

[bk2life]

What do you mean by fuel cell? The fuel trims from STFT or LTFT?

the following is from thread: https://forum.hptuners.com/showthrea...-HP-Tuners-3-0


Add Fuel Trim Cell ([6310]) to your channel and chart and depending on your setup, and data you want to address generate a filter for a specifc value or a 2 part (AND) for a range. Use + < > to setup the equation.

Ed M

Fuel Trims LS1 V8
0 - 15 defined by the LTFT RPM and MAP boundaries
16 - Idle PN (AC on)
17 - Idle PN (AC off)
18 - Idle InGear (AC on)
19 - Idle InGear (AC off)
20 - Idle (EVAP closed)
21 - Coastdown (EVAP closed)
22 - Non-Idle (EVAP Closed)

Usually EVAP closes at WOT hence cell #22 is mostly used at WOT, but you will see it used during warmup (before evap purge begins). At WOT you will also see cell #15 if the EVAP purge opens (ie. vacuum drops enough for the purge to begin).


Fuel Trims LS2/LS7/L76, Cobalt SS 2.0L s/c and other later model vehicles that use RPM/MAP boundaries
The LS2/7 and others have more cells but follow a similar pattern to the LS1:

EVAP Open
0 - 15 defined by the LTFT RPM and MAP boundaries
16 - Idle PN (AC on)
17 - Idle PN (AC off)
18 - Idle InGear (AC on)
19 - Idle InGear (AC off)
20 - Decel

EVAP Closed
21 - 36 defined by the LTFT RPM and MAP boundaries
37 - Idle
38 - Decel


Fuel Trims later LS3 and other later model vehicles that use Airflow Mode boundaries

EVAP Open
0 - 5 defined by the LTFT Airflow Mode Purge On boundaries
6 - Idle
7 - Decel

EVAP Closed
8 - 13 defined by the LTFT Airflow Mode Purge Off boundaries
14 - Idle
15 - Decel


Fuel Trims V6 vehicles
0 = Idle
1 = Decel
2 = Light Load / Engine Not Running
3 = High Load
4 = Med Load

[SiriusC1024]

Oh that fuel trim cell stuff. Don't worry about it.

[Alvin]

Its very important to filter by fuel trim cells. The decel fuel trims dfco etc.. can skew your data. The fuel trims in those cells do not matter and do not carry over to on throttle.

[bk2life]

Oh that fuel trim cell stuff. Don't worry about it.

Its very important to filter by fuel trim cells.

Well, that is confuzing

[SiriusC1024]

I've never needed them. Gen III. A definition of 0-15 would be cool, but look you're filtering basically for idle or not. I'm not sure how decel will throw off CL either. You turn off DFCO while tuning to map those points specifically. If it went DFCO that means no CL, and that means no fuel trims.

0 - 15 defined by the LTFT RPM and MAP boundaries
16 - Idle PN (AC on)
17 - Idle PN (AC off)
18 - Idle InGear (AC on)
19 - Idle InGear (AC off)
20 - Idle (EVAP closed)
21 - Coastdown (EVAP closed)
22 - Non-Idle (EVAP Closed)

Fuel Trim Cell filters for Dummies.....

Gen III Closed Loop only -- [6310]<16 (only let 0-15 through) --OR-- Gen III Open Loop -- only cells that will work is idle (EVAP CLOSED) 20, Coastdown 21 and non-idle 22 which is everything else; Use [6310]=22

Ed M

[6310]<16 (filter for Closed Loop)
Woohooo. We're filtering CL for CL. That'll help sort things out.

[TheMechanic]

For what it's worth I don't use filters for NA. Fuel trims are automatically filtered. Might catch some STFT warmup, but it gets averaged out pretty well. Where you need to start paying attention to the wideband, PE, it's pretty obvious.
wb PE.png

You'll get the feel for it the more you do. I get a really solid VE table mapped after about 6 iterations.

I'll use a filter for OL, like a turbo application, that filters for stable MAP.
https://forum.hptuners.com/showthrea...l=1#post748278

ABS([50030.91]-[50030.91.shift(-10)]) < ([50030.91]/[50030.91.shift(-10)])*0.1

Speaking of. If you're interested in trying filters, put this into one of your filtered fuel trim readouts. Should make it more accurate because transients are removed (unstable MAP):
CL SD (filter for UMAP):

ABS([2311.71]-[2311.71.shift(-10)]) < ([2311.71]/[2311.71.shift(-10)])*0.1

I've always disabled DFCO during tuning and then turn it back on after VE or VVE and MAF is done. Am I doing something unnecessary?

[SiriusC1024]

No that's what I do, too.

[bk2life]

wb PE.png

All good info, thank you from all.

For what it's worth, I just took a run to the lcoal burger joint and here is the results.
A few more runs where i can do some more WOT and I should be good to go.

pic.jpg

[SiriusC1024]

Right right, but those zones are where you need to use CL.

[8aHemi]

Would you really have to filter that out? isn't eq error just comparing the commanded and "actual" as reported by the WB? forgive me for my ignorance, just got my WB to log finally and am setting it up myself.

[gametech]

Would you really have to filter that out? isn't eq error just comparing the commanded and "actual" as reported by the WB? forgive me for my ignorance, just got my WB to log finally and am setting it up myself.

There are many ways to end up at the same place tuning. However, I have found it to be most accurate and efficient to tune areas where PE=1 using the fuel trims, and tune areas where PE is not =1 with the wideband by way of filtering. This way I can use closed loop and one datalog to copy/paste corrections from idle to WOT, but have the accuracy of the narrowbands when PE=1. Since the computer is using narrowband fuel trims for these areas anyway, if there is any discrepancy between your wideband and narrowband readings you will be tuning to the same reading your PCM is going to use.

[Alvin]

Its very important to filter by fuel trim cells. The decel fuel trims dfco etc.. can skew your data. The fuel trims in those cells do not matter and do not carry over to on throttle.

I'm going to say it again.

You'll get where you going faster if you filter by fuel trim cell. Things that happen in decel are a good example. Fuel trims can go nuts there but aren't stored and are not important. If they get averaged into your good data it will take you longer.

[gtstorey]

There are many ways to end up at the same place tuning. However, I have found it to be most accurate and efficient to tune areas where PE=1 using the fuel trims, and tune areas where PE is not =1 with the wideband by way of filtering. This way I can use closed loop and one datalog to copy/paste corrections from idle to WOT, but have the accuracy of the narrowbands when PE=1. Since the computer is using narrowband fuel trims for these areas anyway, if there is any discrepancy between your wideband and narrowband readings you will be tuning to the same reading your PCM is going to use.

My LNF/E69 uses a wideband instead of a narrow band stock, and it stoworks a lot better to use fuel trims than commanded vs actual AFR.