PDA

View Full Version : LBZ Allison simple question



Boost
May 25th, 2008, 08:24 AM
I read my TCM controller today. I know that my truck has a 3.73 final drive. The ECU file says 3.73.

Why does the Allison file say 4.10 final drive? Thanks.

GMPX
May 25th, 2008, 09:51 AM
I know it's a little odd, the diff ratio could probably be deemed as 'not important' on the Allison, I don't know the answer and I'm pretty sure Allison would not answer an Email asking why :angel_innocent:

Cheers,
Ross

Boost
May 25th, 2008, 01:07 PM
Agreed. :hihi:

Maybe when they were fine tuning the speedo they had to alter some values... :grin:

Boost
May 28th, 2008, 03:15 PM
I did not want to open a new thread but I have another question about the Allison contoller.

Why the 30 second countdown? (As opposed to 15 seconds for the ECU)

Is it because it communicates slower? I noticed that it's "not as smart". GM generally recommends a 30 second countdown with the ignition off after programming to allow reset and prevent possible controller lockup. Is this because depending on which controller is flashed, it may be "slow" and need 30 seconds to reset? (such as the TCM).

Just curious, I want to learn as much as possible since I'm hooked on tuning for life now. Thanks.

Blacky
May 29th, 2008, 08:46 AM
As far as I know this is why...

When EFILive (or GM, or any other tuning program) reads or re-flashes a controller, it has to first send a program (called a boot loader) to the controller to handle the read/re-flash duties. That is because the controllers are not capable of doing those two tasks by themselves. (Except the LMM and LBZ but that's a whole other story).

The only place where the boot loader can be stored in the controller is in the controller's RAM. The same RAM that is used to run the controller when it is operating normally. The boot loader "stomps" all over the data that the PCM was using in RAM. When the boot loader is finished, the PCM must clear the boot loader from RAM and reset the RAM to a know state.

Part of resetting the RAM includes resetting some non-volatile memory segments. Those segments are stored in the flash memory and hold data such as the VIN, seed/key, oil life etc. To write to flash memory the PCM must not be "running". i.e. it must suspend its normal operation (i.e. acquiesce), write to flash memory and then restart itself. That usually takes less than a second to do.

The 15-30 seconds is the time that you need to wait after you have switched off the ignition, until the PCM "decides" that it is safe to suspend normal execution and perform the flash write.
Some controllers take longer to make the decision to acquiesce, some take a short amount of time. Sometimes the length of time is determined by other modules on the comms bus. If other modules keep pestering the PCM for information then the PCM will take longer to acquiesce. The 15-30 seconds is just a guide.

Hope that is not too confusing...

Regards
Paul

Blacky
May 29th, 2008, 08:48 AM
Agreed. :hihi:

Maybe when they were fine tuning the speedo they had to alter some values... :grin:

You're probably more right than you realize :secret:

Boost
May 29th, 2008, 09:12 AM
Thank you for taking the time to explain that. It may not seem interesting or important, but it is. That's why GM put out some new bulletins about programming procedures to reduce the amount of fried warranty controllers.

JoshH
May 29th, 2008, 12:54 PM
Part of resetting the RAM includes resetting some non-volatile memory segments. Those segments are stored in the flash memory and hold data such as the VIN, seed/key, oil life etc. To write to flash memory the PCM must not be "running". i.e. it must suspend its normal operation (i.e. acquiesce), write to flash memory and then restart itself. That usually takes less than a second to do.Paul, is the writing something EFI does, or does the ECM/TCM do it itself? I ask because recently my engine hours have started resetting themself. I'm not 100% sure when it started, but it was sometime around the time I upgraded to BB logging. I don't want to blame it on that, as the last two times I did, it turned out to be something else. Just trying to figure out why it went 40k+ miles without resetting, and now it has reset itself at least 2 times.

Blacky
May 29th, 2008, 01:40 PM
The non-volatile memory is updated by the controller under its own control. EFILive has nothing to do with that, except....

When EFILive full flashes a controller it erases the non-volatile memory area. The EFILive software attempts to save and then restore as much information as possible from the non-volatile area. I guess its possible the engine hours are not being saved/restored by EFILive. But if that was happening it would ONLY ever be during a full flash.

The problem with saving and restoring the non-volatile memory area during a full flash is that different operating systems use different formats for the data in that memory area. It is not possible to just save and restore the entire block of memory because the new OS being flashed expects the data in a different format. EFILive saves/restores AND converts the format of the data in the non-volatile memory. To do that the software has to figure out what each byte in the non-volatile memory holds, its format and location in the old OS and its format and location in the new OS. It is far from a trivial task. (Which is why no other tuning product attempts to do it - not even GMs.) If the software can't figure out the existing format for any data in the non-volatile memory then it is not converted and is simply reset.

Even then it would not be a problem on LS2 vehicles as the non-volatile memory is not erased during an E40/E38/E67 full flash.
LS1's could be affected.
Some LB7 duramax trucks have a problems with the non-volatile memory - mostly the oil life value.

Regards
Paul

Cougar281
May 29th, 2008, 02:16 PM
Just FYI, the hours are stored in the cluster, so ECM flashing wouldn't cause the hourmeter to reset.

Blacky
May 29th, 2008, 02:36 PM
Just FYI, the hours are stored in the cluster, so ECM flashing wouldn't cause the hourmeter to reset.

Thanks Cougar :) I learn something every day. Sometimes (like now) I even learn useful info.
Paul

JoshH
May 29th, 2008, 04:30 PM
Thanks for the explanation Paul.

Just FYI, the hours are stored in the cluster, so ECM flashing wouldn't cause the hourmeter to reset.
Do you know what would cause it to reset?

Cougar281
May 30th, 2008, 03:41 AM
Do you know what would cause it to reset?

I've heard of it happening before, but I don't know why. :confused:

Ghosts in the machine? :shock:

modpcm
May 30th, 2008, 05:22 AM
Thanks for the explanation Paul.

Do you know what would cause it to reset?

Hi , it is known by GM for a while , according to some GM engineers the IPC has a battery feed and a ignition feed , the difference in voltage between the 2 for a certain amount of time (some milisec) during a crancking event with a set of battery not at 100% recharged on a large block or diesel could create this concern. Basically low battery and high starter motor current draw .

Modpcm

JoshH
May 30th, 2008, 11:31 AM
Hi , it is known by GM for a while , according to some GM engineers the IPC has a battery feed and a ignition feed , the difference in voltage between the 2 for a certain amount of time (some milisec) during a crancking event with a set of battery not at 100% recharged on a large block or diesel could create this concern. Basically low battery and high starter motor current draw .

Modpcm
That makes sense. I pulled my second batt around the same time it started. Thanks!