struggling newbie needs direction

[jeeper]

I've been struggling with trying to tune my LS1b for about a week on and off.

I'm trying to follow the Calc.VET tutorial using the Calc.VET summary notes.

I downloaded several calc.vet.txt files and a couple pid files. I think I finally have the correct calc_pids.txt for my AEM 30-0334 which I am using in analog mode for the LS1b. (I will be using it in CAN mode on my next project.)

What I can't seem to figure out is why I can't find the correct pids as shown in the attached picture.

I have attached pics of my available pids and the ones I am unable to locate anywhere.

[pinstripebob]

Can you attach a copy of your calcpids.txt file? What is the folder that it's in?

Sent from my Pixel 3 using Tapatalk

[jeeper]

My calc_pids.txt is attached.

It is in C:\Users\user\Documents\EFILive\V7.5\User Configuration

I found it on another forum and it looks like it does have the correct info for my wideband.

[jeeper]

Ok maybe I get it now? I combined the one from the tutorial and the one for my wideband found elsewhere. Will this new one work if I rename it and put in in the configuration directory?


edit: it errors out when I try to open scan tool

[jeeper]

It appears the dynamic air temp portion is erring out. I have commented line 60 and 61 out but then when it gets to line 100 it errs again.

[jeeper]

So I started over and deleted all the calc_pids,txt files I had been accumulating. I now have one that doesn't error out. I'll report back if it doesn't work. So far I seem to have found the missing pids but I'm struggling to get rid of the red x's.


Example:

Calculated VE, not corrected {CALC.VEN}

Expression:
% = {SAE.MAF.gps}*{CALC.DAT.K}/{SAE.MAP.kPa}/{SAE.RPM}*3445.2/displacement()
is NOT valid because:
Required parameter {CALC.DAT.K}, at position 27 is not selected.

PID value cannot be determined because
the following error would occur:
Expression not valid: Required parameter {CALC.DAT.K}, at position 27 is not selected.

Calculated VE, not corrected {CALC.VEN}

Expression:
g*K/kPa = {SAE.MAF.gps}*{CALC.DAT.K}/{SAE.MAP.kPa}/{SAE.RPM}*15
is NOT valid because:

PID value cannot be determined because
the following error would occur:
Expression not compiled

[pinstripebob]

It looks like the two errors in your last post which are as follows:
1. You do not have CALC.DAT selected, one of your PIDs.
2. There is an issue with your CALC.VEN PID calculation.

To try to make things easier, I've attached my calc_pids.txt file which should have all the calculated PIDs you need, plus a few extras. I've also included my PID selection file.

Go ahead and put this calc_pids.txt file in your User Configuration folder. Put the PID selection file in Documents\EFI Live\V7.5\PIDs.

Once you do that, make sure you have all your scan tool windows closed, then reopen V7.5 scan tool and choose GB-LS1-CALCVET when you go to the PIDs selection tab and open a PID selection file.

You may see some red x's if you load this PID file without being connected to your vehicle. Make sure you connect to your vehicle and see how that goes.

23335
23336

[jeeper]

Thank you!

I don't even have to have it connected to the V2 or ECM for it to show in the scan tool software now. No red x's.

My next problem is I am using an analog input from my AEM 30-0334 wideband, not serial. I am using analog because dr.mike (the guy who modified the AEM firmware to work CAN with EFILIVE) suggested it would be better. I made a special cable to connect to the gauge and it doesn't include a serial connection. Should I just use serial instead of analog?

From another thread:


The analog output from the 30-0334 is preferred over the serial. In either case, you would need to add the associated pins to the connector.

The serial output of the 30-0334, as shipped, is compatible with the old 30-4100. So, it is slow, and low resolution. If you send me the gauge for re-FLASH, I can upgrade the serial to be 5x faster with a full extra digit of resolution.

I would use the analog output. It has more resolution and better response times than even the upgraded serial connection. Unlike most widebands, it does not suffer from the dreaded ground voltage offset errors, due to its built-in differential compensation circuitry. Because of the, remember to connect the WB analog- pin to the scanner's ground.

The connectors and pins are available from digikey.com

connector: https://www.digikey.com/product-detail/en/jst-sales-america-inc/PUDP-10V-S/455-2469-ND/1989458

pins: https://www.digikey.com/product-detail/en/jst-sales-america-inc/PUDP-10V-S/455-2469-ND/1989458

I would order a few extra of each, as they are usually cheaper than the shipping costs. And, it prevents the need to hack up a perfectly good connector/harness that comes with the WB :)

I did order extra and could even make a new cable just for serial but if the analog has more resolution it sounds like it may give better results so I'd rather try that first.

Looking at the files you provided you are using slots that my AEM was using and I am not really sure how to properly assign the "CLC-00-xxx" placeholders. Is there a link that describes what those are and how to assign them so as not to interfere with those already existing in the file you posted?

[pinstripebob]

I have not used an AEM wideband before, I've always used an LC2 communicating via serial. I don't post a lot on forums, but I've seen dr.mike on a couple other forums and he seems to have a pretty good understanding of tuning EFI. I would go with his recommendation for using the analog output of your AEM.

This should be a pretty easy thing to do once you figure out how to add or modify the wideband PIDs in the calc_pids.txt file, it sounds like the hang up is assigning numbers to the PIDs. You can have just about as many PIDs as you like in this text file, so I would only recommend deleting PIDs if they are obsolete. I should have done that with the CALC.VE PIDs so I (or now you) don't use the wrong PID when trying to do CALC.VET tuning, I guess it's easy to hoard when storage is essentially free!

If you haven't done so, head here: https://www.efilive.com/documentation and download the V7.5 scan tool manual. Open it up and go to page 90, your PID identifier numbers can be 000-899. So if you want to add or change PID identifiers, pick a number between 000 and 899 and use it. If you want to try grouping your PIDs together, maybe pick 500 or 600 and start from there, where your PIDs are 500, 501, 502, etc. That will help you troubleshoot any issues that may come up. Each calculated PID is listed twice in the text file, so you will want to see each identifier twice, once for the calculation of the PID, and once for the units and description.

Give that a try, do some experimenting, and if you get hung up let me know where you got stuck.

[jeeper]

Thanks for the help.

I'm not sure yet if this is going to work but here is my calc_pids.txt and my pid file.

Now I have to go try and make a log file.

[jeeper]

Can someone confirm the log I took has the info I need? I have a short log made and I need to learn how to proceed. So far most of my effort has been in trying to get this far I haven't started reading what to do next. I'm just trying to verify that where I'm at is where I should be so far.

[jeeper]

I decided I'd better make another log. This time I used an old tablet computer. Is this better?

[pinstripebob]

Your last log is close. If you open it and click on the data tab, you can see your minimum, maximum, and average values for each PID. There should be data for the CALC.VET, CALC.LTFTBEN, and CALC.SELBEN PIDs but they stay at 1, 0, or N/A for the whole log.

I clicked on the PIDs tab, it looks like CALC.SELBEN has a red x. Right clicking and choosing More Info shows that you need to log the CALC.WO2BEN PID as well. Make sure you have that PID selected, do a short log, then stop the log and take a look at the data tab to make sure you're getting data for each PID you have selected.

[jeeper]

I'm not really sure why but on my computer screen I see two lines for CALC.VET and one line is populated. Same for CALC.VEN. Just curious if you used my calc_pids.txt file to view the log or doesn't that matter?

Here is what my log looks like on my screen.
23342
Curious why mine shows data?

Also I went ahead and completed the steps in the tutorial and came upe with this. I don't plan to try running it. It doesn't look right to me.
23343

Is it normal to have the scan software need to replot the log file everytime it opens?

Edit:
I see why there are two lines each for those. One is percent and one is g*K/kPa.

[pinstripebob]

I might have seen incorrect data if I'm missing your wideband PIDs.

It looks like you're really close. Make sure the units from your logged correction map are the same as your tune's VE table. I highly recommend keeping everything in metric units and using the VE units, not percent.

Your first VE update looks like you copied percent values into a table with VE units. Notice that the maximum allowed value in your VE table is 10.0, and the peaks are at that maximum.

Sent from my Pixel 3 using Tapatalk

[pinstripebob]

Also yes, when you open a log everything will need to update. Same when you apply a filter.

Sent from my Pixel 3 using Tapatalk

[jeeper]

How do I verify what my logged units are?

[pinstripebob]

Edit your map in the scan tool. The first tab for data should let you choose a PID and it's units. Check there, make sure it is g*K/kPa.

Sent from my Pixel 3 using Tapatalk

[jeeper]

It looks like they match but I'm probably missing something?

23344

23345

[jeeper]

I'm curious if maybe my results are indicating I have a vacuum leak somewhere? I've already checked for one because I suspected one earlier in the process but I didn't fine one. I'm going to test for a vacuum leak again and this time try to be more deliberate.

I just ran through the process again of calc.vet but got the same results. If I continue to try the same thing over again expecting different results I think there's a word for that...

The thing is, the engine is running the best it's ran since I started this whole process and I haven't even made a proper tune yet.

With LTFT and closed loop disabled the Jeep is very drivable. It is a tad rich at idle but at cruise speed it is sometimes staying right around 14 AFR on my wideband.

[statesman]

jeeper, your fuel trims will work better if you connect your bank 1 O2 sensors to the exhaust on the left side of the engine (drivers side). Bank 2 is the right side of the engine.

[jeeper]

Thanks statesman. I remember looking up what side they were supposed to be on after I posted in that other thread and I have not changed it yet. I will start there for sure.

[jeeper]

So I did another quick run after swapping the O2 sensor inputs around. I will say the AFR did improve slightly on my warm up ride, but the overall results are similar to before. This leads me to start wondering if maybe the formula is incorrect from my WB. The calc_pids.txt I copied from was actually from another WB that isn't exactly like mine but I thought may work. I'm guessing the lambda equation may be wrong though. I'm going to try and modify it.

[jeeper]

No change after modifying my calc_pids.txt. What am I missing?

[pinstripebob]

Can you attach a copy of your calcpids file?

Sent from my Pixel 3 using Tapatalk

[jeeper]

This is my current calc_pids.txt. I am renaming it and attaching it here so I don't get confused by it.

I'm wondering if I don't have the proper understanding of how this works. My 30,000 foot assessment is the calc_pids.txt tells the software what hardware to look for and how to use the data from that hardware once (or during) the data is collected? And if this assumption is correct which part of the calc_pids.txt is getting used to calculate the new VE table data?

Edit: after further consideration it seems likely it's the calc.selben portion? If so it would make sense (assuming I understand the theory correctly) that in my calc_pids file the calc.selben is listed in % and it is first in the file. Maybe that needs to be either removed or relocated?

[pinstripebob]

I made a few small changes to your text file as follows:

1. Moved your definition of your wideband to chronological order (30 before 32).
2. Added AFR as a unit.
3. Removed CALC.AEM.LAM from two of the expressions and used the explicit calculation for the lambda of your AEM calculation.
4. Updated the calculation for PID 110. It was using the external WO2 PID, not your AEM calculation.

See if that helps.

23353

[jeeper]

Thanks Bob. I was just in process of reading how to create the pids using the Auto VE tutorial. It spells it out pretty good so far.

One thing I noticed when I put your file in place was an error "duplicate AFR unit". I commented out the AFR under *UNITS and can open the software now but I didn't see a duplicate entry for AFR in the file. I couldn't understand why it was giving me an error for it. I even did a search for it using "Ctrl + F" and let the computer find all instances of "AFR". It didn't list any duplicates...Then I read the explanation tab of the error. I assume it will work without the added AFR unit. I will try it tomorrow and report back.

Thank you for the help. It is really appreciated.

[jeeper]

I'm guessing I still don't have a proper formula.

[pinstripebob]

Can you make a map that shows your wideband's AFR or EQR plotted against RPM and MAP? This should show if you're seeing reasonable values from your wideband.

Also, are you applying the CALC.VET filter mentioned in the tutorial and replotting your map?

Sent from my Pixel 3 using Tapatalk

[jeeper]

I'd be happy to if I knew how. I'm still trying to learn how this all interacts and thus is why I'm struggling with this so much I guess.

Until I figure out how to make the map can I just look at the Data tab and see the values from the AEM?

[jeeper]

So I found a calc_pids.txt in another forum that has a completely different calculation for BEN. I'm wondering if I'm supposed to be dividing instead of multiplying in my equation?



# ================================================== ============================
# File details
# ------------
#
# This section defines various details about the file format.
*FILE
#Parameter Value Description
#---------- ---------------- ---------------------------------------------------
VERSION 7.1.1 File version
DECSEP . Decimal separator used in this file

# ================================================== ============================
# Units
# -------------------
# See sae_generic.txt for more information on the *UNITS section
*UNITS
#Code System Abbr Description
#-------- ---------- -------- -------------------------------------------------------------
MPG Imperial MPG "Miles Per Gallon"
# ================================================== ============================
# Add slot definitions here
# -------------------------------- ({SAE.VSS.mph})/({GM.INJFLOW.lbpm} *{GM.IBPW1}* {SAE.RPM} * 2 / 6250 )"
# See sae_generic.txt for more information on "SLOT" formats
#
#Units Low High Fmt Expression
#------------ ------------- ------------- ---- --------------------------------------------------------------
# MPG
*CLC-00-001
MPG 0.00 99.99 .1 "{SAE.VSS.mph}/({GM.INJFLOW.lbpm} * {GM.IBPW1} * {SAE.RPM} * 4 / 6250 )"
# AEM Ext Input 1
*CLC-00-002
V 0.0 5.0 .1 "{EXT.AD1}"
AFR 10.0 18.5 .1 "{EXT.AD1}*1.7+10"
# BEN AEM 1
*CLC-00-003
factor 0.0 2.0 .1 "{CALC.AFR_AEM_1.AFR}/{GM.AFR}"
# AEM Ext Input 2
*CLC-00-004
V 0.0 5.0 .1 "{EXT.AD2}"
AFR 10.0 18.5 .1 "{EXT.AD2}*1.7+10"
# BEN AEM 2
*CLC-00-005
factor 0.0 2.0 .1 "{CALC.AFR_AEM_2.AFR}/{GM.AFR}"

# ================================================== ============================
*PRN - Parameter Reference Numbers
# --------------------------------
# See sae_generic.txt for more information on the *PRN section
#
#Code PRN SLOT Units System Description
#------------------------- ---- ------------ ---------------- ---------------- ------------------------------------------
CALC.MPG F001 CLC-00-001 MPG Fuel "Miles Per Gallon"
CALC.AFR_AEM_1 F002 CLC-00-002 AFR O2-Wide "AEM Wideband - Input 1"
CALC.BEN_AEM_1 F003 CLC-00-003 factor O2-Wide "Base Efficiency Numerator - AEM 1"
CALC.AFR_AEM_2 F004 CLC-00-004 AFR O2-Wide "AEM Wideband - Input 2"
CALC.BEN_AEM_2 F005 CLC-00-005 factor O2-Wide "Base Efficiency Numerator - AEM 2"

[jeeper]

I think I may have it. I'll report back tomorrow with details. Otherwise I'll report back in frustration...

[pinstripebob]

Good! Keep it up!

Sent from my Pixel 3 using Tapatalk

[jeeper]

So time for another update.

I still don't really "understand" what I'm doing but I hope I'm getting closer to that.

I played around quite a bit yesterday and had many unsuccessful attempts. I left the lines in the file to help me remember what I had tried already. I'd do some reading and come back and make another attempt. Here is a sample of how I modified the calc_pids.txt file:


#BEN 0.5 1.5 .3 "({EXT.AD1}*0.1621+0.4990)*{GM.EQIVRATIO}"
#BEN 0.5 1.5 .3 "({CALC.AEM.EQR}*{GM.EQIVRATIO})"
#BEN 0.5 1.5 .3 "({CALC.AEM.EQR}/{GM.EQIVRATIO})"
BEN 0.0 2.0 .3 "({CALC.AEM.AFR}/{GM.AFR})"
#BEN 0.5 1.5 .3 "({EXT.AD1}*0.1621+0.4990)/{GM.EQIVRATIO}"
#BEN 0.5 1.5 .3 "({GM.EQIVRATIO}/{CALC.AEM.AFR})"


For some reason copy and paste truncates the text but you get the idea.

In the end what I think may have been happening was the same numbers were being used in two equations and I was getting the sum of both equations? I'm not really sure because the light hasn't come on yet for me and I'm making guesses still. I find it very difficult to troubleshoot when I don't really know what to look for...

I finally decided I would "start over" and replace my calc_pids.txt with this one (https://forum.efilive.com/attachment.php?attachmentid=20472&d=1480124326).

It was as close as I could find and I was fairly sure the parameters for that AEM matched mine so I went for it.

I ended up making some changes to my pid selections in the scan tool and viola. The numbers coming back to the map looked more reasonable.

I then had to modify my filter because it was (apparently) based on a pid I was no longer using. I applied the transient filter and low count filter and did a copy and fill with labels. I opened the v7.5 tune tool and did a paste and multiply with labels. I did this with both maps.

I smoothed the tables and renamed the tune and saved as a new tune. I also turned closed loop back on. I did a cal flash and took it for a drive. It ran okay but not really any better than before. In fact it was much more lean than before.

I took it out again this morning and did a log. I reset the LTFT before I started. I drove my dog to the park for a swim and took the long way. It ran fine on my way there. Maybe 15 minutes worth of drive on my way there. I looked at the map and was a little disappointed I wasn't seeing zeros. I saved the log and shut down everything..

I let my dog run off some energy and take a swim in the river for a while. Maybe 20 minutes later I got back in and fired up my computer and started the engine. It wouldn't idle. If I didn't apply throttle it would stall out. This is the first time since I started trying to tune where the engine would just stop running altogether. I decided I'd just flash it back to a pre-tune and drive home. It runs great on that tune.

Sorry for the long read but that's where I'm at this morning.

[statesman]

Get the vehicle speed sensor hooked up to the PCM.

[jeeper]

Time for another update and some more questions.

So I will start out by admitting I still don't know much about EFI but I have learned some things that have helped. First off I was not aware that I wasn't ready to even try tuning when I did. I wasn't even aware that my engine was running extremely lean. I knew it was lean but didn't realize it was dangerously lean. It would be around 18+ AFR pretty consistently during acceleration and even WOT.

Because I was so green I didn't know there was so much I didn't know. I'm still green but I have advanced slightly I think.

What I discovered so far is my lean condition was likely self induced. I think (but still need to verify) that when I changed the OS to 12212156 I didn't change all the needed parameters within that OS to allow proper fueling. Nothing I did seemed to allow the engine to run rich. I ended up reflashing the PCM with the stock OS 09381344 and removing VATS and changing some torque management settings. It is running really well now. More importantly it doesn't run lean. It idles great and seems to have good power under load.

So on to the questions.

I keep being told I need a speed sensor but I don't understand why at this point. It seems the Jeep is running and driving just fine without one. It doesn't appear to need to know that I am moving. I can't imagine what benefit I'll get from adding it. So I need to ask, why do I need one?

Also I keep reading I need a manual transmission specific OS. I've spent several hours driving this Jeep at this point with the stock PCM that was originally controlling an automatic transmission truck. All I have done is disable torque management in several places and it seems to be running just fine. So why do I need a manual transmission OS? What will it do better that I am missing?

[Highlander]

When you are running in ve mode, vacuum leaks don't really affect your resulting VE table as by nature it accounts for those leaks.

The purpose of the speed sensor is so that the ECM can run correctly between idle mode and non idle mode and other situations, but I agree that it may not be needed if you simplify your calibration.

[jeeper]

I just wanted to say thanks to everyone who helped me get this going. It has been running beautifully for a while now. In fact I had an offer to buy the vehicle so I ended up selling it. Time to start on my next project I guess.

[joecar]

I just saw this thread and I'm reading thru it... those calc pids could probably be simplified/reduced down.