Using STFT to tune for now

[pavetim]

I have the tutorial for AutoVE but isn't that to tune OLSD with a wideband? I'll be doing that soon but I need a quick tune while i'm waiting for my LC-1. I want to tune his 02 S-10 real quick to keep it from being very lean. How do I tune without the MAF and using the STFT? I followed the tutorial but that shuts off the STFT and forces CL. Now I need to be in Closed loop, MAF less with LTFT turned off but STFT on correct? Thanks.

[joecar]

You're MAF-less, make sure it throws a MAF DTC (this indicates the PCM has failed over to the VE table).

Leave CL and LTFT's enabled.

Use the attached calc_pids.txt file, place it in My Documents\EFILive\V7.5\User Configuration.

Log the LTFT's and the pids mentioned in Shawn's Calc VE tutorial (no MAF of course).

Use one of the B0101 VE BEN maps, on the Data tab change the pid to LTFTBEN, and save the map with a new filename (B0101 LTFTBEN).

On the BEN map, apply the filter that excludes throttle transitions.

Make sure the button x-bar is clicked.

Copy-with-labels this map, and paste-with-label into B0101 VE table... the table should move up a small amount (eyeball before/after, make sure it makes sense), you might have to fill in any holes and lop off any spikes, and handraulically extrapolate to high MAP/RPM;

save tune with a new filename; do a cal only flash, log again, observe the LTFT's.

[pavetim]

Think I got that calc.pids txt working but I tried to add average STFT pids but it gave me an out of bounds error, did I do it wrong?
LAM None LAM "Lambda"


# ================================================== ============================
# Add slot definitions here
# --------------------------------
# See sae_generic.txt for more information on "SLOT" formats
#
#Units Low High Fmt Expression
#------------ ------------- ------------- ---- --------------------------------------------------------------

*CLC-00-010
AFR 9.63 19.63 0.4 "14.63 / {GM.EQIVRATIO}"
EQR 0.5 1.5 0.4 "14.63 / {GM.AFR}"

*CLC-00-020
AFR 9.63 19.63 0.4 "14.12 / {GM.EQIVRATIO}"
EQR 0.5 1.5 0.4 "14.12 / {GM.AFR}"

*CLC-00-051
EQR 0.5 1.5 0.4 "{GM.EQIVRATIO} / {EXT.WO2EQR1}"
LAM 0.5 1.5 0.4 "{GM.EQIVRATIO} * {EXT.WO2LAM1}"

*CLC-00-052
EQR 0.5 1.5 0.4 "{GM.EQIVRATIO} / {EXT.WO2EQR2}"
LAM 0.5 1.5 0.4 "{GM.EQIVRATIO} * {EXT.WO2LAM2}"

*CLC-00-200
factor 0.5 1.5 .4 "({SAE.LONGFT1}+{SAE.LONGFT2})/200+1"

*CLC-00-300
% 0.0 100.0 .2 "{SAE.MAF.gps}*({GM.DYNAIRTMP_DMA.C}+273.15)*3445.2/(5.669*{SAE.RPM}*{SAE.MAP.kPa})*{CALC.LTFTBEN}"
VE 0.0 3.0 .4 "{SAE.MAF.gps}*({GM.DYNAIRTMP_DMA.C}+273.15)*15/({SAE.RPM}*{SAE.MAP.kPa})*{CALC.LTFTBEN}"

*CLC-00-001
% -25.0 25.0 .2 "({SAE.LONGFT1}+{SAE.LONGFT2})/2"

*CLC-00-003 -10.0 10.0 .2 "({SAE.SHRTFT1}+{SAE.SHRTFT2})/2"
%
# ================================================== ============================
*PRN - Parameter Reference Numbers
# --------------------------------
# See sae_generic.txt for more information on the *PRN section
#
#Code PRN SLOT Units System Description
#------------------------- ---- ------------ ---------------- ---------------- ------------------------------------------

CALC.AFR_E00 F011 CLC-00-010 AFR Fueling "Commanded AFR (14.63)"
CALC.EQR_E00 F012 CLC-00-010 EQR Fueling "Commanded EQR (14.63)"

CALC.AFR_E10 F021 CLC-00-020 AFR Fueling "Commanded AFR (14.12)"
CALC.EQR_E10 F022 CLC-00-020 EQR Fueling "Commanded EQR (14.12)"

CALC.BENS1 F051 CLC-00-051 "EQR,LAM" WO2-Serial "BEN Serial Wideband 1 (based on EQR or LAM)"
CALC.BENS2 F052 CLC-00-052 "EQR,LAM" WO2-Serial "BEN Serial Wideband 2 (based on EQR or LAM)"

CALC.LTFTBEN F200 CLC-00-200 factor Tuning "LTFT BEN"
CALC.VE_Table F300 CLC-00-300 "%,VE" Tuning "CALC VE Table"

CALC.LTFT F001 CLC-00-001 % Tuning "LTFT Average"

CALC.STFT F003 CLC-00-003 % Tuning "SFTF Average"

[joecar]

Post your calc_pids.txt file here.

[mr.prick]

Average STFT & STFT_BEN PIDs


#Units Low High Fmt Expression
#------------ ---------- ----------- -------- --------------------------------------------------------------

*CLC-00-013
% -100 99.22 .2 "({SAE.SHRTFT1}+{SAE.SHRTFT2})/2"

*CLC-00-014
factor 0.00 2.00 .2 "(({SAE.SHRTFT1}+{SAE.SHRTFT2})/200)+1"

#Code PRN SLOT Units System Description
#--------------------------------- ----------- ------------------- --------------- -------------------- -----------------------------------

CALC.STFT F013 CLC-00-013 % Fuel "STFT Average"

CALC.STFTBEN F014 CLC-00-014 factor Fuel "STFT BEN"

[pavetim]

What are BEN pids anyways?

[joecar]

A "BEN" is a correction factor... you multiply a BEN map into the VE or MAF tables to correct them.

Since lambda = AFR / stoich AFR = 1 / eqr, BEN has several forms (all of them equal):

BEN
= wideband AFR / commanded AFR
= wideband lambda / commanded lambda
= commanded eqr / wideband eqr
= commanded eqr * wideband lambda

It was named after someone.

[pavetim]

Well I created pids for average LTFT and average STFT. I did some VE changes to his truck tonight and then logged av ltft and av stft they were way off getting alot of -7 and -10. I copied them and pasted-add them to the VE table, now my LTFT are showing alot of zeros and my STFT are close to zero and some small positives so i'm guessing in did the initail VE table right, right? Ok now before all my trims were negative and I pasted-add them, so if I have positive ones now do I still paste-add or something else? And I didn't do a BEN pid for STFT do I really need one, it seems what i've done so far is working.

[mr.prick]

You added LTFT/STFT as a percentage or just plain added them? :confused:

BEN (Base Efficiency Numerator) is a percentage multiplier in this case.
To add 1% (1.01*X.XX)
To subtract 1% (0.99*X.XX)

Adding LTFT or STFT directly to the VE or MAF table can cause some problems. :bad:

[pavetim]

I just copied my LTFT numbers and pasted, added them, then i went and did the same to the STFT numbers and pasted, added them to the main VE table. thats what i've read to do, now when I had problems is I read to paste add them, and then on the second round past then multiply by 50% to the VE table, the truck wouldn't even run so i went back and just added them. Most of my Trims are coming in-line now and the truck runs alot better. I'm guessing if I did it wrong the trims would be further off. I can email you the logs if you like so you can see. And i'm pretty sure the LTFT and STFT averages are percentages the way i'm logging.

[WeathermanShawn]

So your previous hunch that the truck was lean does not appear to be the case. ..:confused:..According to your Trims the fueling was actually very rich.

Did you disable the MAF? If you could post a tune & log it might be easier to comment on.

Thanks..

[pavetim]

I did disable the MAF and since I don't have my WB O2 yet I used both STFT and LTFT. So negative fuel trims means it's rich? Cause thats saying it's taking away fuel right? Thats wierd then cause for the most part when i've heard of headers being put on it usually leans it out. Although most ones i've done was long tubes and full exhaust so I dunno. Ok I feel like a dummy, see if I posted the logs correct. And they are numerical so 3 is the latest adjustment.

[pavetim]

Ok I added a BEN for STFT. I see how it goes, so basically if I use the BEN pids as an average I would copy the numbers, the paste-mulitply those?

[minytrker]

You normally paste and multiply when doing your fuel table.

[WeathermanShawn]

I am a little curious as to the vehicle's ECT. Does it ever get warmer than 145-150F?

You see you have B3802 set to 70C and B4108 is set to Disable. So on your last log I see a lot of LTFT's that remain at zero (0). So while you are getting some STFT data, it looks like your LTFT's are not really giving reliable data.

It does look like you successfully failed the MAF and you seem to have gotten some valid data. I am curious as to whether you are applying any filters (transient, ECT, etc), especially when the engine is that cold. Normally you get better results with the engine at normal ECT.

I personally would be hesitant in disabling B3616 (PE Mode) as you have. You don't activate it all all. Since your ECT's are so cold, you never hit the fuel enrichment you have in your OLFA Table B3605. Worst, it looks like you went the wrong way on your WOT EQ's. You are commanding an extremely lean AFR for some of the Throttle Openings/MAP's you hit. Looks like 16.6 AFR at WOT!:confused:

Sorry if I sound overly critical, but you posted the question and I am just giving you my observations. I would wait for the wideband and follow the AUTOVE to the letter. A wideband would have indicated how lean you were. The AUTOVE Tutorial also indicates the Commanded Fuel you should have and examples of the filters.

Good luck.

[pavetim]

I don't mind criticism im.learning and what im here for. I get my wb today just wanted to.get him close, had some issues I told him he needs to fix told him to get new tstat I think its sticking, and he has a missing issue think some plug wires are not all the way on. Well the commanded pe is stock and I disabled it to make sure I didn't hit it but as you can see I kept it about 2500rpm max ill re enable it when I get my wb so I can tune higher rpm too. Ok so you said multiply that's if you use either% average or ben?

[joecar]

Make sure the tunetool is set to AFR units or EQR units (Edit->Properties->Display->Commanded Fuel Values).

When viewing a table sanity check yourself by looking at the units on the 3 axis's.

[joecar]

Your ECT is too low to allow CL, which means trims can't be used for anything... is there a reason why ECT isn't coming warm enough...?

I would avoid STFT and use LTFT (when CL is achieved) since the LTFT is an integrated/averaged STFT (i.e. the PCM has already done some of the work for you).

I wouldn't add the LTFT to the VE table (even when displayed in %), this is the whole point of the BEN.

It might be good to set all the tables back to stock and start again...

Is there any reason to not run the MAF... the simplest thing to do right now is to run the MAF and follow Shawn's Calc.VE tutorial... and then clean it up by following with Auto VE tutorial using a wideband.

I only had a quick look at your files, I'll look closer tomorrow, as I'm sure Shawn will also. :)

Criticism... none of us are criticizing, we're just eager to see you succeeding... :cheers:

[pavetim]

Ok sorry i'm going by a tutorial i seen from someone that said take the LTFT then add it to the VE and then take the STFT and add it to the VE again. I'll start over once he gets the problems fixed, plus i'll have my wideband so won't have to mess with NB sensors anymore. I'm running with maf disabled for now to get the VE inline since I didn't have a WB but can run the CalcVE setup since I have a WB now. Ok so when I run the BEN for LTFT I get my numbers, then copy the LTFT table after I run my filter, then paste and multiply to the VE? I didnt' mean criticizing as a bad thing, a better word would of been critiquing LOL.

[pavetim]

Oh one more thing in the CalcVE tutorial when you select your pids the page that says your final pid selection should look like this. There are 30 pids selected, but can have 24 max, whats up with that?

[WeathermanShawn]

Use the PID Example on Page 3 of the CALC VE Tutorial. Ignore the example of page 4..(sorry it has extra Pids..corrected on next Edit..). That should keep you at 24 or less.

As helpful as other Tutorials and Tuning Guides from different sources are, mixing and matching might get you confused. Its like 'baking a cake'. You have to follow the recipe exactly. Don't worry we have all been new to tuning..you are probably taking on more than I did when I first started.

The CALC.VE Table Tuning method actually utilizes the narrowbands. Having a wideband just expands your ability to track the AFR when you open the throttle and enter Open-Loop. Narrowbands are your friend. They can be extremely accurate at stoich. Use them.

Same as MAF. There is nothing inherently wrong with utilizing a MAF. Yes, it is somewhat of a engineering 'contraption', but working against it takes some skill. When you begin tuning it is probably best to keep the narrowbands, keep the MAF. Tune for LTFT's, and just calculate the VE Table. Later you can do an open-loop AUTOVE or go to a CLSD Tune.

You have to fix that ECT problem. Try installing a new thermostat if that is the problem. You need normal ECT temperatures to tune correctly.

Post any logs or tunes if you run into problems..:)

[pavetim]

If the CalcVE tutorial does the VE table and MAF at the same time, why would you need the AutoVe anyways? And so basically the WB is just to use other than stoich (like in PE mode) cause the NB's are more accurate right? When I use the WB's I just use the tables exactly the same way as the CalcVE just will be using wide band and not LTFT right? what pids would I need to change to use the WB? Thanks for all the help guys i'm actually learning quite a bit, I never knew how to use the NB sensors to tune until you guys showed me.

[pavetim]

And what confuses me is I thought to tune your VE table accurately you had to disable MAF so it doesn't provide feedback or something like that. But your saying if I follow the CalcVe tutorial I can tune the MAF, and the VE table as if I were in SD mode?

[WeathermanShawn]

pavetim:

Technically if your utilize the MAF and stay closed-loop you do not need to utilize the AUTOVE. The AUTOVE is somewhat of the 'gold standard' of tuning an automobile, but it is primarily for those running MAF-less and whose needs usually include an open-loop Tune.

CALCVE Table is using mathematics to convert your existing MAF Airflow along with narrowband feedback to simultaneously calculate your VE Table. So you will be using your WB, NB, and LTFTS. The beauty is that the formula does most of the 'heavy lifting' for you, so in many cases you do not have to 'paste and multiply' the results to your new VE Table..you can simply paste the results directly into your tune.

So you do not ever need to disable the MAF in order to accurately construct a VE Table. The magic of math eliminates that procedure.

I would just start over from near stock on your tune (including the VE Table). Hook up your wideband. I am sure someone can walk you through that if necessary. You might start a new thread if you need help on that. Same with Pid selection, etc.

The actual VE Table Tune is pretty easy. We have probably 'walked' through a half-dozen or more on this board. Rarely have I seen it take more than (3) 30 minute logs to accomplish. Just print out the CALC VE Tutorial and follow it to the letter.

Good luck..

[joecar]

Calc.VE = calculates corrected VE and corrects MAF, based on NBO2 and MAF (requires them to be in good condition).

AutoVE = corrects VE, based on wideband, requires OL and MAF-less (i.e. isolate source of airmass to VE and eliminate trimming).

AutoMAF = corrects MAF, based on wideband, requires OL and VE-less (i.e. isolate source of airmass to MAF and eliminate trimming).

The limitation of Calc.VE is that is can't calculate/correct VE/MAF when you go WOT/OL:
closed loop is exited so now there is no trimming of any kind,
if the previous LTFT was positive, the PCM adds it to WOT fueling, otherwise not.

[pavetim]

Ok makes more sense now, i'll just try the calcve tutorial by the letter and see what happens. Now just saying the way i've done it before, and this was 5 yrs ago it technically isn't wrong just takes longer right. I would put it in olsd and tune the VE table with my wide band and then get my PE/WOT down to the AFR I want, then after that is good, i'd go and calibrate/tune my MAF sensor with CLMAF. That is pretty much what the CalcVE tutorial does but pretty much in one step right? And by looking at my logs when would I know i'm ready to start my PE/WOT tune, when the LTFTBEN are close to 1.0 thru 4000rpm?

[WeathermanShawn]

Agree.

You were never doing anything 'wrong' with any of the other methods. Calc.VE combines a lot of the other methods. As Joecar stated, when you hit PE/WOT you are back into open-loop..and essentially doing an 'AUTOVE' type AFRBEN to determine your airflow/fueling.

You can always log the FUELSYS Pid (under Fuel Pid) that will indicate when you go from closed-loop to open-loop. Easiest way is to just watch your Commanded Fuel AFR/EQ change. It will go from stoich to enrichment (AFR). Also your LTFTBENS will 'lock'..hopefully at 1.00.

[pavetim]

Yeah it makes perfect sense now, I don't know if i'm just stuck in old school since I haven't tuned in years, I just thought you couldn't tune VE and MAF at same time, thought it had to be seperate. But I guess with your guys knowledge and R&D you can tune both now.

[joecar]

Calc.VE does this:
- calculates VE using MAF, MAP, RPM, DAT and the ideal gas law (P.V = n.R.T),
- corrects the calculated VE using LTFTBEN,
- corrects the MAF using LTFTBEN.

[ LTFTBEN is simply the multiplier version of the LTFT percentage ]

[joecar]

When you've finished doing AutoVE, you can now do AutoMAF (using the wideband) by doing this:

enable the MAF, set B0120 to zero (disables VE), repeat the AutoVE procedure on the MAF (i.e. make the BEN map look like the MAF table).