PCM Flash to Grand-AM Cup PCM

[Rusty]

I need advise on tuning a PCM.
We have replaced a 400CI with LS6 Heads (550HP / 510 torque) to a 383CI stroked with LS6 Heads. The new engine has lower compression (around 11 or higher) while the old was slightly higher. Same intake, injector, headers. Cam is close to the last Engine.

I read the PCM and Changed the CI parameter and had to perform a full flash because the Grand-Am racing PCM did not have a serial number. We have not proceeded with the crank sensor learning.

Current results: Engine cranks and idles at 1600 RPM although the desired idle is at 1000 RPM. The MAP and MAF values are within allowable range. I believe that the rational test fails which triggers the MAF P code. We do not have O2 sensors installed for sensor 2 and are disabled in PCM but we do have sensor 1 installed for both bank 1 and 2. The values at idle of the O2 sensors seems real low (.0020 mV). The command AFR is set for 13.5. The fuel learning is enabled but are resulting in 0%.

This is a road race car and we are expecting around 510Hp with 475 foot pounds of torque.

Question: Does low mV readings from O2 sensors reflect lean or rich? What is the best plan to start tuning?

[mr.prick]

play with the desired airflow table {B4307}
and see if that helps your idle.
get a WBO2 for AFR, it`s bad to start a new motor overly rich.
would you mind posting the Grand-AM Cup tune?

[5.7ute]

Under 450mv is leaner than stoich, over 450 richer.
I would do as mr.prick has stated & try lowering your desired airflow tables.
Depending on the cam in your car using a wideband or even the narrowband sensors could be erronous due to overlap. If this is the case just give it what it wants for now.(use the bidi controls)

[joecar]

Low mV from front O2 sensors indicates lean, 450 mV indicates stoichiometric 14.7, high mV indicates rich.

Front O2 sensors mV should look like this:
- closed loop -> cycling from below 300 mV to above 700 mV and back again at least once per second.
- open loop part throttle -> more or less flatlining.
- open loop WOT-> flatlining... at WOT anything below 900 mV may be too lean.

At WOT (which is also open loop), you want to see about 930 mV on the front O2 sensors.

See the AutoVE tutorial: http://download.efilive.com/Tutorials/PDF/AutoVE%20Tuning%20Tutorial.pdf

AutoVE requires a wideband O2 sensor (some, like the LC-1, support serial/digital AFR output (instead of analog) which FlashScan V2 supports).

The idea behind AutoVE is this: the AutoVE procedure corrects the VE B0101 table; when B0101 is correct, then the OL B3605 and PE B3618 tables then produce the correct AFR (i.e. the wideband measured AFR equals the commanded AFR), which means you can set the AFR you require (i.e. something like 12.6 to 13.2 for WOT).

After AutoVE, you can do AutoMAF which is the same procedure but with MAF enabled and B0120 set to zero; this corrects the MAF B5001 table;
when this is done, you set B0120 back to stock.

Before doing AutoVE/AutoMAF you may want to recalculate the IFR B4001 table to match your injectors using the spreadsheet.

You may also want to adjust the MAF/MAP/ETC sanity tables (yes, ETC, even if you have cable thottle).

What PCM do you have (from what donor vehicle)...?

You may even want to run COS5 which has a MAPxRPM OL table versus the stock ECTxRPM OL table.

[joecar]

Post some logs and tune file if possible.

[joecar]

Is your fuel pressure regulator at the fuel rails up front (has a vacuum hose connecting it to the intake manifold) or is it in the fuel tank (no sign of it anywhere near the fuel rails; don't mistake the pulse dampner for the FPR... the pulse dampner has no vacuum hose attached to it).

[Rusty]

I have downloaded the AutoVE Tuning Tutorial. I plan to make revisions to the tune file tomorrow. I am going to start with lowering the desired airflow table values. After several tunes, I will post some data.

Thanks for the advise!

[Rusty]

It seems that tuning is not may main problem. The engine is only running on 4 cylinders. After increasing RPM to over 3000 then some of the remaining four pick up. All have spark. I believe that my O2 sensors are reading so lean due to the dead cylinders.
I have a problem with being able to Full Flash / Read / flash Cal. I always have to perform a Full Flash.

I am attaching 3 files (2 tune files and 1 log file). My first PCM read tun file has VIN number but after I perform a Full flash the VIN number is not in the PCM.

Again, the PCM has a Grand-Am Cup racing sticker on it but the tun file has a VIN number that matches a GM-Pontiac-Canada-1999-Trans Am Formula-LS1

How can I get to only perform a Cal Flash so I can use other tun files that I have for a similar engine?
:wtf1::wtf1:

[joecar]

Use the scantool to set the VIN into the PCM: Bidirectional->Change VIN.

You mean that a cal-only flash does not work...? Can you copy the log text and paste it here.

[Blacky]

How can I get to only perform a Cal Flash so I can use other tun files that I have for a similar engine?
:wtf1::wtf1:

Download a completely stock *.tun file that matches your existing operating system 9381344 and transmission type from here: http://www.holdencrazy.com/EFILive/stockTunes.php

Full flash that stock file into your PCM.

Once that is complete, select the menu option: Flash->Show PCM Info... and click on the [Get Info] button. Hopefully you will see a VIN displayed, you can change it here, to your vehicle's VIN.

If you do see a VIN, you can use the menu option Flash->PCM Security... window to change the VIN to the correct value for your vehicle.

If you don't see a valid VIN, then please read this web page:
http://www.efilive.com/index.php?option=com_content&view=article&id=111&Itemid=121
and send the relevant *.blx file(s) to me at [email protected]

Regards
Paul

[Rusty]

Use the scantool to set the VIN into the PCM: Bidirectional->Change VIN.

You mean that a cal-only flash does not work...? Can you copy the log text and paste it here.

I will try this tomorrow. I also will need to do the crank learn but concerned about the dead cylinders. Thanks for the help. If I can not just flash a tun then I will post log tomorrow.

[joecar]

Rusty,

First do what Blacky/Paul said in post #10...

if you can't get cal-only flash working, then there may be other PCM functions that are not working (it might explain 4 cylinders not firing).

[Rusty]

Blacky,
Can I use a newer operating system with my PCM? I am concerned that my PCM is not normal. From my posted tun file is that possible. On the outside it reads "Grand Am Cup". It also reads something like "No spark, Must have separate controller"

If the operating system and hardware is just normal, then I really would like to program to US as the Metric defaults from Canada are a little hard to convert on the track. I currently just try to hit the dash button at 150mph, which can be a little distracting.

Thanks for your help.

[Rusty]

The reason why I can not perform a tune flash and am required to perform a full flash is because the PCM serial number is all zeros. How can I get a Serial Number into the PCM?
Here is some data from EFI LIVE:
PCM
VIN 2G2FV22G2X2226801
Hardware Number N/A
Calibration ID 4294967295
Serial Number 000000000000
Operating System 09381344
Engine Calibration 16253254
Engine Diagnostics 16263887
Transmission Calibration 16264134
Transmission Diagnostics 16253323
Fuel System 16264551
System 16264738
Speedometer 9359157
Operating System BCC N/A

[mr.prick]

4407

[Blacky]

Blacky,
Can I use a newer operating system with my PCM? I am concerned that my PCM is not normal. From my posted tun file is that possible. On the outside it reads "Grand Am Cup". It also reads something like "No spark, Must have separate controller"

If the operating system and hardware is just normal, then I really would like to program to US as the Metric defaults from Canada are a little hard to convert on the track. I currently just try to hit the dash button at 150mph, which can be a little distracting.

Thanks for your help.

Yes, you can reflash it with an updated OS file. Make sure the OS file is for a 512KB PCM like yours and not a 1MB PCM. However, that won't help with the initial problem of "no serial number".

Have you tried full-flashing a totally stock OS file with a matching OS number like I suggested? If that does not fix your serial number then the PCM's memory is most likely damaged.

Also make sure you are using the latest version of our V7.5.5 software. You can download the latest version here: http://www.efilive.com/index.php?option=com_content&view=article&id=48&Itemid=124

Once you have installed the latest version (or if you already have it installed), select the start menu option: Start->All Programs->EFILive->V7.5->Tools->Check for New Update.

The note on the outside that says "no spark" is a concern. Is that just written on by hand or is it a factory sticker/warning?

As for the metric units, do you mean the dashboard in your car or the dashboard in the EFILive software?

Regards
Paul

[joecar]

Post pics of your PCM.

[dc_justin]

Yes, you can reflash it with an updated OS file. Make sure the OS file is for a 512KB PCM like yours and not a 1MB PCM. However, that won't help with the initial problem of "no serial number".

Have you tried full-flashing a totally stock OS file with a matching OS number like I suggested? If that does not fix your serial number then the PCM's memory is most likely damaged.


I've run into quite a few situations lately where, after a TIS flash, the PCM will report a blank serial number, EFILive will update it. A subsequent "PCM Info" report will show a valid serial number, but any attempts to reflash the tune will show a blanked out serial. I can repeat this on just about any 99-00 OS.

[Blacky]

I've run into quite a few situations lately where, after a TIS flash, the PCM will report a blank serial number, EFILive will update it. A subsequent "PCM Info" report will show a valid serial number, but any attempts to reflash the tune will show a blanked out serial. I can repeat this on just about any 99-00 OS.

Interesting, the plot thickens...

Do you happen to have any *.blx trace files from when you reflashed a PCM with a blank serial and EFILive added the serial back in, followed by a *.blx file where EFILive refuses to see the serial number?

Regards
Paul

[dc_justin]

I'll try and remember to grab one the next time I encounter this problem. I do what I can to avoid it (upgrade a customer to an 01-02 OS instead), but I'm sure I'll encounter it again.

[Rusty]

I performed a full flash with a stock operating system. I added my VIN number and then read the PCM. I changed the VE table and flashed cal changes. EFI required me to purchase another license which I did. That is another conversation.

This is where I am now. We replaced the cracked FAST intake due to GM valley cover bolts pushing on bottom of the aftermarket FAST intake. After correcting intake, car is idling correct and seems to be running better but needs a lot of adjusting to open loop.

After 15 seconds after start, I get flashing check engine light. Only P1689 and P1541 codes that are for traction control. I imagine that it is flashing MIL due to values from O2 sensors 1 bank 1 and 2. O2 sensors 2 bank 1 and 2 are removed.

Engine Information:
Largely ported and polished heads (243),
Cam has lift and dur. greater than 2006 Z06 specs.
hooks headers
Z06 injectors
383 CI
3" exhaust, no cats, no mufflers

I am attaching my tune file and logged data. I appreciate any advise on changing settings to correct.
Thanks!

[Rusty]

Post pics of your PCM.

I have attached a couple of pictures of PCM. Do you think I should go back to older OS?

[joecar]

Rusty,

Is that a PCM that the Grand Am rules require you to race with...?

Is that the purpose of your vehicle, GA racing...?

The flashing MIL indicates misfires, unless the GA PCM gives it special meaning.

If the rotating assembly (from pulley to flywheel) and/or the crank sensor, then a CASE (crank angle sensed error) relearn may be needed, as this steers the misfire algorithm.

[joecar]

Your log shows LTFT's pegging at 25% at idle... your NBO2's may not be working right which would throw some DTC's which would light the MIL steady (but misfire overrides and flashes the MIL).

Hmmm.... that PCM is set for 31* advance regardless of the spark tables and spark modifier tables...? Am I interpreting correctly...?

[Rusty]

joecar,
Thanks for the help on the flashing MIL without codes. We have lower than expected exhaust temps from cylinder 8. We will resolve those issues today. Assuming the mechanical and crank sensor are resolved. And assume that I relearn the CASE after removing the misfires.
What do you suggest to get open loop running closer to stioch?
What PIDs should I add to the log list help?
Should we get a wide band O2 to work with EFI? If so, please recommend one. I see the list which works with EFI but not sure which is best for us. Need no more technical problems and it to connect easily to EFI live and FAST systems. Also, if it can be a hand held sensor, it would help with other open loop cars.

[Rusty]

Rusty,

Is that a PCM that the Grand Am rules require you to race with...?

Is that the purpose of your vehicle, GA racing...?

The flashing MIL indicates misfires, unless the GA PCM gives it special meaning.

If the rotating assembly (from pulley to flywheel) and/or the crank sensor, then a CASE (crank angle sensed error) relearn may be needed, as this steers the misfire algorithm.
I can change the PCM with other systems per our rules. We have remained with this GM PCM because it was working great with the last engine couple of engines. All other people involved are wanting to switch over the a FAST and removing the complexity of the GM PCM.

[joecar]

Yes, fix the physical misfires first before attempting a CASE relearn.

To get open loop running closer to stoich you could do one of:
- raise either of VE B0101 and/or CFOL B3605 a little (trial and error);
- do AutoVE using a wideband, this will then let you specify the AFR you want to see (can you see the time-saving advantage of this);

By doing the second option, then the AFRs specified in B3605|B3618 will show up on the wideband (when PE is enabled, the richer of B3605|B3618).

Do you have FlashScan V2...? If so then a wideband like the LC-1 is cheap and simple, it connects serially to V2 using a RJ-12 to stereo 2.5mm cable as sold by forum member TAQuickness; see this: showthread.php?t=8115 (http://forum.efilive.com/showthread.php?t=8115)

By placing a connector on the end of the LC-1 you can make it portable.

For racing, you could probably run with an OLSD tune... avoiding the MAF and NBO2 has several advantages including improved transient response time (in my $0.01 opinion). How much proportion of the time do you spend at WOT vs Part Throttle...?

To have even more control over OL AFR, you could run COS5 (altho this requires a little more work copying all the tables over... and you have to first see if OS 12212156 is compatible with your PCM... if your PCM is a "0411" PCM then it will run.

I suppose you should consider/experiement with these during off-season.

Hmmm... you should also carry a spare PCM tuned with your latest good tune.

How did they disable/fix timing advance to 31*...?

[joecar]

Include these pids:
VSS
RPM
TP
MAF
MAP
IAT
ECT
AFR
SPARKADV
KR
IBPW1
IBPW2
DYNAIR
DYNCYLAIR_DMA
LONGFT1
LONGFT2

Maybe include these:
HO2S11 (not HO2S12 or HO2S22)
HO2S21 (not HO2S12 or HO2S22)

For AutoVE or AutoMAF, keep the channel count to 24 or less (giving 10 updates/second); see bottom of PIDs tab in scantool.

[Rusty]

I will adjust the PIDs as you listed. I use flashscan v1.2. Do you have an opinion of a wide band that will work with my hardware / software? I see that FAST sells a stand alone that logs data but I do not know if compatible with EFI Live. Also it is costly ($400+). I would rather pay for quality and a device that is somewhat easy to use and common.
I appreciate your advise on the tuning. BTW: cylinder 8 had intake value open due to stock valve cover and rocker arm clearance. I think we got that fixed. Ready to tune!

[joecar]

Any basic wideband (LC-1, PLX, etc...) will do... you don't really need to have a gauge, you just feed the wideband analog AFR signal into FlashScan V1... the requirement is that it outputs a 0V-5V linear analog signal representing AFR (e.g. LC-1 outputs a linear 0V=7.35AFR, 5V=22.35AFR).

If you use one of the "supported" widebands, you can simply use the pre-made calc pids (for AFR and BEN) for that wideband... or we can make custom calc pids... (I reprogrammed my LC-1 to give better resolution (e.g. 0V=11AFR, 5V=16AFR) so I made my own calc pids (see attached... oh, I log EQIVRATIO instead of AFR, where EQ = stoich/AFR... I taught myself to think in terms of EQ rather than AFR).

[Stealth97]

I can change the PCM with other systems per our rules. We have remained with this GM PCM because it was working great with the last engine couple of engines. All other people involved are wanting to switch over the a FAST and removing the complexity of the GM PCM.

Why not put that GM ECM on the shelf, buy yourself a 12200411 and pick an OS that works for you and upgrades to COS3 or COS5 or something that fits your needs.

If you aren't limited to PCM's, you can still stick with the GM PCM and remove the unknowns from yours.

[joecar]

What car do your engine/trans/PCM and wire harness come from...?

If you say 1999+ F-body or 1999-2003 Y-Body (can someone please check the year ranges),
then a 0411 PCM would plug in with no harness modification...
A 1997-1998 would need PCM connectors to be re-pinned.

If you do buy a 0411 PCM like Stealth97 suggests, get the seller (any of the forum sponsors are good) to flash it with OS 12212156 from a 2002 F-body with manual trans (as you have a MT)... should be under $100 shipped for a used one.

12212156 offers various options:
- run 12212156 with your tables;
- jump to COS5 with your tables;

And will give you control over timing advance.

(I don't know if your GA PCM fixes advance at 31* via physical means or via OS code patch means).

[joecar]

I'm thinking that a known used PCM will fix your flashing problems (i.e. the GA PCM has been "tampered" with).

[Rusty]

We fixed all mechanical issues that have been causing the strange results. All cylinders are running correct. (No flashing MIL, it was the dead cylinder 8). I have a FAST wide band that should be here tomorrow afternoon. I plan to use the advise that Joecar gave and tune.

As for the PCM. I currently have 2 GM PCM that are labeled Grand-Am and 2 other stock GM PCM (2001 and 2004).
The currently installed GA GM PCM had operating system 09381344 (200 Camaro LS1) which I had problems flashing. I downloaded a stock from http://www.holdencrazy.com/EFILive/stockTunes.php and performed a full reflash operating system 12221588 (2002 Z06). I meant to get the 2001 Z06 but must have clicked wrong.

I currently have no problems flashing or reading. I do wish to have the best performing OS for racing. I plan to start tuning soon. Should I use a different OS? I hope to remove the MAF sensor from closed loop after tuning, if better performance is a results.

The car will run 95% at WOT and the rest of the time between 50% to 100% throttle. Of course, this excludes the full breaking periods.

Attached is a log file with Joecar suggested PIDs. I relearned the CASE and the zeroed out fuel learn tables. Car was not on Dyno. Just parked (1000, 3000, and 4000 RPM), Its amazing that the trim goes to 25% so quickly. I am finally getting O2 readings that make sense!

Thanks for all the help and I appreciate any more advise about operating system selection or tuning! You guys have help a lot!
:good:

[joecar]

Stick with whatever works...:cheers:

[joecar]

Rusty,

You don't use boost or nitrous, so I don't think COS5 will help you...

The other features of COS5 may not help you either:
- open loop AFR table is indexed by MAP vs RPM (instead of ECT vs RPM);
- includes Alpha-N VE table indexed by TP vs RPM (in addition to the normal VE table);
- allows semi-open loop (SOL) which uses STFT to trim to stoich.

I really don't think any of those will help you...

so I suggest to keep developing 12221588 which is working for you.

Cheers
Joe

[joecar]

When you go MAF-less, show us some more logs...

Include these pids:
VSS
RPM
TP
MAP
IAT
ECT
AFR
SPARKADV
KR
IBPW1
IBPW2
DYNAIR
DYNCYLAIR_DMA
LONGFT1 (exclude for OLSD)
LONGFT2 (exclude for OLSD)
HO2S11 (not HO2S12 or HO2S22)
HO2S21 (not HO2S12 or HO2S22)

Keep your channel count to 24 or less for fastest sample rate (important during AutoVE).