Help please, PCM in dead poll will not take flash, cos 5 upgrade

[johnmaster]

Trying to flash just the calibration from my base file into the pcm after I loaded the initial cos 5 file and it won't let me flash the new calibration. It says that the os's have different numbers.

99 chevy c1500 5.3. I built a file from an 02 file downloaded from tunefiledepot flashed the cal and os and got it running just fine. sucked out the new file and saved it. shut it off, flashed the cos 5 base file 02020003 and it went well. Now it is stuck without a calibration because I am trying to follow step 4 which says to flash just the calibration from the file I read after it was running good and it gives an error that the os have different numbers and that it won't allow it to calibrate.

"Open the *.tun file that you saved in step 1. It contains the PCM’s calibrations from before the EFILive Custom Operating System was installed. Perform a calibration only reflash of that *.tun file.
At this stage, the PCM is capable of starting and running your engine. However, don't get too excited just yet, there are a few more steps that must be completed."

What am I doing wrong?

[johnmaster]

Just realized I installed 02020003 wich I thought was a version of cos 5, looks like the 03 on the end makes it cos 3. What should I do, will cos 5 overwrite the cos 3 without a calibrating it first? Hoping to not turn this pcm into a doorstop.

[ScarabEpic22]

Full flash COS5, then do the cal only reflash. Hope that sets you straight, Ive had a PCM or 2 in limbo before and its nerve racking!

[johnmaster]

Thank you, I had 2 problems, wrong cos then wrong file saved as my calibration read from vehicle file...had the old 99 file saved somehow and that wouldnt work with either cos. I reflashed the new 02 donor file and read it out and saved it properly, then flashed the right cos 02020005 and flashed the saved 02 cal over it and read it back out so I can fix the out of range cells. Should have it going in short order if things keep working out.

[johnmaster]

Fixed oor cells, found I had to change them and change them back to get them to stick as changed after flashing and reading it back out. Also found the 99 to 02 conversion problem of dtc P1638 regarding the alternator terminal wire not used on 99's. Changed it to not reported and flashed it, drove it to the store with no problems, working great, Next step 2 bar map sensor.

[joecar]

For alternator wire look at G1206 and G1207.

[joecar]

You got it going on COS5...

Did you read it out and save to file...? (this is your starting file for any future edits).

Did you fix all the OOR's...?

Did you set B3647 to sane values...?

[johnmaster]

Yep, running great, have a couple copies saved of this base file now. Out of Range cells changed, saved, changed back and saved again to get them to not read out of range anymore. Changed B3647 to the way it showed in the tutorial and flashed it. Vehicle has worked flawlessly as far as I can tell from a few short drives.

PM sent, but also looking for help from anyone who can lend it. I want to run SD (using stock maf only for IAT, eventually replacing it with a separate IAT sensor) and use a 2 bar gm map. I want to operate in Open loop and closed loop now (I think), but eventually I will be operating in open loop only. I am lost in some of the "How to steps" found in the tutorials, notes and threads, looking for some "why" so I have a background on what I am doing and why. I want to make sure I do everything in the right order for the right reasons. 99 c1500 5.3 completely stock so far, preparing the pcm for all of the modifications I plan on doing to the motor, want to get this working before I make the other changes, (one thing at a time with an insanity check in between each).

I have a good grasp of the Boost VE table, when the new map sensor reads 105kpa or higher it will reference the appropriate values, 105 kpa copied from the end of the standard ve table and then scaled to 300% to roughly estimate the increasing ve at the rest of the cells under boost. Eventually refined but rough for now to get started.

I understand you have to trick the pcm into thinking the maf has failed so that it reverts to using other inputs like rpm, map, tps, baro, VE to calculate the cylinder mass.

Maybe the best way to ask what I all need to do yet is to cut copy paste the tutorial and notes the way I think I should do it and see if anyone can verify/clarify the rest of this process for me:

1. Get the vehicle running properly on COS5 according to the pdf tutorial directions (here) (http://download.efilive.com/Tutorials/PDF/Custom 20OS 20Upgrade 20Tutorial.pdf) with the original vehicle calibration flashed over the COS, oor cells fixed and verified, basically steps 1 through 6 which ends on page 12. Done. Except I believe I omitted this step on pg 10:
[To control the fuelling exclusively via this calibration, you must also configure:
Engine->Fuel->Trim.
• {B4205} Closed Loop Temp Enable: Set to max setting
Engine->Fuel->O2->Parameters.
• {B4108} STFT Idle Enable: Disable]

I didnt change these settings yet, I don't understand what this is doing. Looks like it is killing closed loop and disabling stft's at idle. Should I have changed these settings, why?

2. test drive the vehicle to make sure it is working good on maf, no dtc's or mil. save a copy of the working cos5 with maf tune as a base. Done.

3. ignore valet mode instructions pg 13 because I wont use that feature or nitrous features.

4. note the reading from the current map sensor so I can use it to calibrate the 2 bar.

5. Install the 2 bar map sensor.

6. change {C6301} MAP Sensor Scaler: to 190 for a 2 bar map to get me close.

7. Flash the new cal, scan for the new MAP reading and compare.

8. adjust {C6301} up or down a bit to get the value closer, flash, then scan again and repeat until I can get the map reading the same or very very close to the original 1 bar. Any idea how much a change to the scalar changes the MAP reading?

9. Change {C5501} MAP Rationality Test RPM High (RPM) to 2000RPM. (so that it isnt performing the test under boost?)

10. Disable the maf sensor (but leave it in because mine has integrated IAT) by performing the following steps
Set the Engine Diagnostics->Engine Diagnostics->MAF->Parameters:
• {C2901} MAF High Frequency Fail 1: 1Hz
• {C2902} MAF High Frequency Fail 2: 1Hz If calibration {C2902} is not available in the operating system that you are using, then ignore this step.
• {C2903} MAF High Frequency Fail Limit: 1
• {C2907} MAF Test Min Engine Speed: 300 RPM
• {C2908} MAF Test Min Run Time: 0.1 seconds
Set the Engine Diagnostics->Engine DTC processing Enablers:
• P0101 ‘C’ Non Emissions
• P0102 ‘C’ Non Emissions
• P0103 ‘C’ Non Emissions
Set the Engine Diagnostics->Engine DTC MIL Enablers:
• P0101 ‘No MIL’
• P0102 ‘No MIL’
• P0103 ‘No MIL’

11. I would ignore the step on pg 15 regarding setting {C6101} because I have a cable type throttle boty.

12. Then follow these steps for turning off transmission dtc's even though in my current cal they are already set to "no"
Set the Transmission Diagnostics->Transmission DTC Fault Max Pressure Enablers
• P0101 ‘No’
• P0102 ‘No’
• P0103 ‘No’
Set the Transmission Diagnostics->Transmission Failsafe Enablers:
• Set all airflow related enablers to ‘No’ to prevent the transmission from going into failsafe mode due to the MAF being disabled.

13. Reduce the barometric pressure offset learning window by changing the following so that if I drive in different altitudes an offset can be calculated:
Engine Calibration->Barometer->Parameters:
• {B0301} Min RPM for Barometer Update: 1200 RPM
• {B0302} Max RPM for Barometer Update: 2000 RPM (or before high boost levels)

14. Since I plan on eventually running under boost I would follow the directions on pg 18. I don't know why you would wait to observe "significant spark retardation" instead of just changing these parameters right off the bat? I am new to this, I don't understand why you would want torque management.
"If you observe significant spark retardation as you pass through 110-115 kPa MAP, the likely culprit will be torque management. You can confirm that by logging the PID {GM.EST_TRQ_DMA}.
To prevent the torque management subsystem removing spark timing, change the following calibrations:
Engine Calibration->Torque Limiting:
• {B1902} Torque Loss from Spark Retard: Set to all zeros.
Engine Calibration->Torque Limiting->Parameters:
• {B6619} Max Torque, Trans Input Shaft: Maximum value.
• {B6620} Max Torque, Trans Output Shaft: Maximum value."

15. That is all the changes I could come up with from the tutorial, now to dig into Joecar's notes taken from the .txt in the cos5 tutorial I would confirm
The COS allows adaptable spark (i.e. sliding between HO and LO tables depending on
detection of knock) even when the MAF is failed; to do this these parameters must
be set as follows:
- A0000 set to Disabled,
- A0002 set to 100%.
Does this mean I am operating with adaptive spark right now (COS5 with MAF)? This means I should leave my high and low octane tables original?

16. I also dont understand when you would want a TP VE table as opposed to a MAP VE table? Map when throttle is steady, TP when making quick transitions between throttle positions?

17. After all this I think I would save the new file, flash it, try to start it, check dtc's, try to test drive it and then start into the tutorial on VE tuning.

What do you think, how far off am I, what am I missing or not understanding?

If you can read this you don't need glasses.

[johnmaster]

Latest update, installed the 2 bar and changed the scalar. was 14.4psi baro and 14.3psi MAP with the 1 bar. 190 scalar value got me 14.0 baro and 13.9 map. 191 made no change, 200 put it up to 14.6. 195 got me the same initial readings that my 1 bar gave me so that was what I stuck with. Dabbled at trying to change the units to KPA for finer resolution but gave up quickly and pressed on as all of this was undo-able if need be. Need to figure out what units to be working in and how to change them easily.

Then I made all the changes above including what Joe PM'd me which was:
Make sure:
- B3616 PE Enable allows easy enable of PE (65% below 3200, 35% above 3200),
- I would run with B3618 PE set to the same value as the last column of B3647,

Fired it up and drove it to get some gas and a unibit for drilling the hole in my exhaust for the wideband.
It was totally driveable. Not necessarily refined by any means but light throttle and cruise you would never know it was a custom operating system and SD untuned. Shifts and power under heavy throttle were different, definitely need to work on this and learn more but so far the thing works at least. I am going to install my serial wideband to this vehicle and hopefully tonight or tomorrow I can start logging some drives.

[joecar]

In scantool: on PIDs tab, on MAP pid do rightclick->Metric.

In tunetool: go Edit->Configure Display Units and search for any data/row/col that uses MAP and do rightclick->Metric.

[ hint: by clicking on the column heading the list of tables gets sorted on that column ]

[joecar]

When in SD make sure VE is correct or trans may slip in-gear... bump VE up a little when in doubt.

[johnmaster]

Awesome, I think that will help. I do have MAF and all nb02's still on the vehicle to log and get the MAF and VE tables into shape. Which cal/os would I use to drive the vehicle for this tuning? Bone stock GM, COS5 after first set of changes and 1 bar map, Cos5 with 2 bar and maf failed, or one of these with some slight changes?

I think the picture will get clearer for me after the ve tutorial and some experimentation.

[joecar]

Use COS5 with the 2-bar MAP if you have it, otherwise with the 1-bar MAP (you won't be able to tune boost).

[johnmaster]

It's still bone stock naturally aspirated so no need to tune for boost yet, just trying to get it running as well in SD as it was on the MAF before I tear it down and change everything around. Trying to learn on it now so I know what to do when the time comes. So same plan, cos 5 with 2 bar?

[joecar]

Ah, in that case the easiest thing to do is run COS5 with the 1-bar, leave MAF and CL/trims both enabled, do the Calc.VET tutorial, and then do the Calc.MAFT tutorial and compare the results from these two.

[johnmaster]

Awesome Joe, this guidance is what is keeping me from going insane, my head is spinning from information overload:shock:

[joecar]

Take one step at a time, and question everything.

[johnmaster]

Ok, I went until I got stuck, which wasnt very far:) I can't get most of the calc PID's to come up basically CALC.VE_Table and CALC.LTFT. I unchecked "supported" to show them all and they custom ones won't come in. I have restarted the scan tool and I swapped the calc_pids file in user config with the one that was posted on the "A new twist on Calc. VE.." thread. restarted the scan tool a couple times. I'll attach a copy.

Any ideas what I am doing wrong? I haveV2, V7.5 loaded from spring of this year. I saved the pid list with all the ones I could bring up selected, just need to get the others to show and I'll be on to the next step.

[joecar]

Look at the bottom of that calc_pids.txt file... (you want the pids mentioned in there)

CALC.VE_Table has been replaced by CALC.VET

CALC.LTFT is named CACL.LTFTBEN

[johnmaster]

Awesome, Thank you!
I have 24 pids selected now, I chased down the ones that weren't supported and selected the pids they required by right clicking info and seeing what it needed to work. Now I have 24 pid's, says 25 channels, and on Data (F9) tab it shows 26 different lines. I know I am only supposed to use 24 channels for highest sampling, which one is unneeded? I'll attach the pid file in a second from my laptop if it might help.

[johnmaster]

Here is my pid file with 25 channels. Is there an extra one not needed?

Edit, actually I added the wrong file, the larger one is the one I was having trouble with. I think I got it though, I am using the minimum and suggested Pid list from the "a new twist..." thread and the calc_pid file farther down on that thread, the first calc pid is missing CalcCL on my copy.

[joecar]

You can remove GM.AFR, we are diligently avoiding AFR

:)

[johnmaster]

Excellent, I got it down to 24. Gonna flash my cal with the changes for doing this log, and if all goes well I will be able to drive and log in a few minutes.

[johnmaster]

Took a nice long drive and tortured my truck, logged a lot of great data and was able to cover a ton of of the map with cell counts over 10. It was fun, kind of like a video game "spread out the cell counts". A good long hill is your friend for this I found. Truck was running lean on my wideband but didn't overheat and ran well overall. Not sure if I didn't do what it took to make sure it was going to be richer than it needed. I followed the adjustments on the summary notes thread part III but maybe I missed something.
The weird thing is that I left it on cell counts the entire time and now I go to check the data and map a is all zeros. I closed and opened another scan and then reopened this one and luckily it is populated with real data. I read the note about changing the engine size in the thread to 5328 (5.3L) in Edit>>log file information, not sure if that had something to do with it, that part was confusing. Either way I have the data now so I am wondering what is next? I see different ways to filter and was wondering how to set up and apply a filter properly to get rid of the junk data. "a new twist..." step 3 differs from the pdf tutorial, it says apply the calc vet filter but doesnt exactly say what to do or how. What should I do?

Thank you!

[joecar]

If you're using % units for the map, then yes you have to enter the engine size (and then save the log to retain the engine size).

You set up the filter logic, and then you click the filter apply button.

You want to exclude data where the throttle was changing (i.e. the airflow/airmass would be changing), and and data where the ECT is too low or too high, and a few other things.

[johnmaster]

Okay, so I got that to work on accident then:) Couldn't figure out why it wouldn't let me set that up ahead of time. So after I logged my trip I changed it to 5328 (5.3L) in the list, saved it, closed it and reopened it and viola the data was there in the map. Just got lucky I guess. Was worried I tortured my poor little truck for nothing (it has no idea what I have in store for it yet!)

I just figured out how to set up the filter, didn't realize that you put all 5 parameters with "or" conditions on them into the same filter. I made the TPS % change 10% though because with 5% it was really spikey. Figured it would get rid of some of my faster throttle moves.

It makes a pretty big dent in my stock VE table and a homely lump on my maf curve. Can you check my tune and make sure I didn't have something related to fueling improperly configured? The wideband was reading lean more often than I would have expected it to. Two tunes pasted, the starting tune is the one the truck is running on in my log, the other is with the ve and maf adjustments pasted with no smoothing.

I threw my log in there too...
Many thanks,
John

[joecar]

Hi John,

I'm looking over your files...

Hmmm, MAF doesn't seem to get up with RPM as high as I expected...

Do you have a log file taken running with CALCVET After VE and MAF correction.tun...? Do the LTFT's come down...?

[johnmaster]

I haven't yet, I saw the dent that the new VE made in my stock ve table so I started to look for a culprit and found the dead cat (leaves and air filter that needed changing bad). So now I am ready to make a new log but wondering if I have fueling parameters set up properly as the last run was running lean very often. Looking at my tune did I make the necessary adjustments to keep this thing rich enough for logging this run? Not completely familiar with the changes made to prevent lean burndown for this purpose so relying on your advice.

[johnmaster]

It ran well, dont get me wrong, but 02 was rarely below stoich and usually closer to off the chart...

[joecar]

Hi John,

I looked at your B3647, you have it set incorrectly... the last 2 columns are heading lean and leaner... same with B3618;

go Edit->Properties and check the fueling units, if it says Lambda then change it to EQR and re-edit B3647 and B3618;


11737


hint: when viewing/editing a table, look at the units shown in top/right of the tunetool window.

[johnmaster]

Thank you Joe! I will make the changes and get a new good log this afternoon.

[johnmaster]

:doh2:I've been bouncing between the calc vet thread, the summary notes and the old version of the tutorial that is in with my software...No wonder I have been so lost. The new version at the bottom of the post explains it all much clearer and matches the rest of the tutorial. I bet if I updated my software like they say is the first thing you are supposed to do, I might have avoided all these headaches.

[joecar]

Yes, please use the current (most latest) form of the tutorial and summary notes.

[johnmaster]

It is a huge help now, going back over it to make sure I didn't assume something. I logged a new run but I am doubting my new data because it still makes such a jagged dent in my VE table. In my head I am picturing a wrinklier version of GM's for me to just smooth out.

***edit I used the SELBEN to correct my Maf freq, it was way off! That seems like it was a big reason why my VE data was so all over the place, I also am learning how to drive to get better data, mix of low gear on side roads to get high rpm readings and highway in 2nd gear and D to get the lower vacuum cells. Now my corrections are in the .95 to 1.05 ish range and getting closer to 1.0. They were as high as 1.25.

It feels good to be getting closer to understanding this, I changed all my units over to g*K/Kpa and EQ ratio. Until this afternoon I wasnt exactly sure what EQ ratio was:)

Thank you!

[johnmaster]

iT'S aLIVE! Lots if little bugs to work out but it runs and drives! thank you for all your help, now i just need to start tuning it up:)

[johnmaster]

First thing I need to fix is the shifting so I don't destroy the transmission putting any power to it. I went through great pains to make sure the transmission was running and shifting in speed density before I disassembled Johnny 5 to build it up with the turbo and carburetor. It ran great but the maf was plugged in even though it was failed. I should have cut the wires and drove it because I get the feeling it was still influenced by the maf but I can't be absolutely sure. It is not right, I get a clunk like an upshift when I let off, it labors out first pretty noticeably it kind of slides into and out of gears, low throttle shifts are just all around not crisp or timed properly.
How do I log or determine what is being calculated to determine that it's commanding proper shift pressure? A problem that might be influencing this is that the tps is for a completely different application, I set it to zero and it goes to 100 almost exactly at wot, but I am getting a tps low voltage p0122 code so perhaps I need to twist it open a bit and reset the truck with an unplug-on-off-replug to get the tps up to voltage? I will do that in the morning but otherwise any ideas?

[joecar]

Hi John,

It runs :cheers: good job.



Have a read thru these:

Transmission Tuning:
showthread.php?t=1723 (http://forum.efilive.com/showthread.php?t=1723)
showthread.php?t=5876 (http://forum.efilive.com/showthread.php?t=5876)
showthread.php?t=10773 (http://forum.efilive.com/showthread.php?t=10773) in relation to SD (MAF-less)
showthread.php?14493-D2801-TCC-lock-during-upshift (http://forum.efilive.com/showthread.php?14493-D2801-TCC-lock-during-upshift)

Kickdown:
showthread.php?t=3336 (http://forum.efilive.com/showthread.php?t=3336)

...

[johnmaster]

Thanks man! Couldn't have done it without you! lots of sleepless nights learning how to tune with your help. Once I get my software working right I should be able to see what the transmission is seeing and hopefully figure out what it needs to shift right. I will start in on these threads as soon as I can. Then it's spark, and fuel and boost, good fun. I found a nice Mustang eddy current dyno and a skilled tuner who is going to help me fine tune this thing into shape, just gotta get it relatively close before i drag it in there. I have quite a few supporting mods to do but gotta take it one step at a time.

[johnmaster]

A year later and here we are! From what I can tell, The Worlds first propane turbo LS powered Vehicle!

http://www.performancetrucks.net/forums/forced-induction-159/propane-turbo-5-3-first-track-day-video-507389/

This could not have happened without EFIlive forums, especially Joecars dedication to sorting out the lost bolts!

[BLK02WS6]

Joe is the man when it comes to sorting out lost bolts! Couple times I almost lost my nuts, but didn't thanks to him... and I know I have a few screws loose - maybe no help for that though :laugh:

[joecar]

Hey John, good job :cheers:

Thanks for the kind words.