PowerPCM_Flasher: my tool for E38 and E67

[beeradoc]

e38 needs VATS delete patch[available in universal patcher]. And write entire for sure. There is some gm programs that might give you a headstart on write entire with mdi, but it is not preferred to be publicly discussed here.

Thanka kur40, i will keep investigating

[antus]

FWIW I looked at this powerpc flasher that this thread is about, and note that 'calibration' flash includes the OS, but not the boot sector. So code patches in the OS but not the boot sector for vats would be written with it. I think it would work.

[beeradoc]

FWIW I looked at this powerpc flasher that this thread is about, and note that 'calibration' flash includes the OS, but not the boot sector. So code patches in the OS but not the boot sector for vats would be written with it. I think it would work.

Thanks Antus,

I have a vehicle that I did with Hpt so I will flash that back with Vats enabled via Hpt and then give it a go I gi guess

[antus]

Re-reading my own message, and just to be clear, because the boot sector is not written you cannot use it to change the operating system to a different one. But you can use it to write the same OS, with patches in the OS segment (eg VATS). I seem to recall that for E38 they are code patches but I havn't looked at an E38 bin a long time.

[ironduke]

I'm not saying anyone is wrong, but E38 and E67 OS starts at 0x10000 and goes to 0x1BFFFF. PowerPCM has been unable to flash that area. I can try again but I had even been communicating with Daniel on this because I was using his kernel to try and be able to write with an elm327 and he verified that this program wrote the calibration files only from 0x1C0000 to 0x200000. I remember it being a little weird because when it's pushing the calibration it's showing how much data is being pushed, not the addresses.. 0x8000, then 0x12000,etc all the way up to 0x40,000. That's only the calibration section being done, OS is not touched when writing.

[V8fan]

Tactrix openport 2.0 clone J2534
Connected successfully to E38(2007) but reads it with "no reply following frame. Read.FF" Between uploading process percentage time to time
At the end -Validating upload -NOT valid
Win 10 x64
DotnetFX enabled into win10 features
Any advice how to fix it ?

[clyde]

I'm new to this and I'm trying to connect to an e38 (actually two) with no success.
OBDXPro VX (updated to latest firmware and J2534 driver installed)
Bench harness with 12.6v PS
I can connect and read the VIN with OBDXplorer but I'm getting same types of errors with both e38s and two different Win10 computers.
--
initialising Tool J2534 PassThru
OBDX Pro - VX...
ok.
connecting...
start diagnostic session (0x10 0x03)...ok.
disable normal communication (0x28)...ok.
security access (0x27)...
request seed...NOT ok, no valid seed received.
Wait 30s after power on or failed attempt.
connecting...
start diagnostic session (0x10 0x03)...ok.
disable normal communication (0x28)...ok.
security access (0x27)...
request seed...ok.
seed: 0x00000D30
key: 0x00009BD2
sending key...not ok, key wrong?

Any thoughts?

[ironduke]

That's the key I come up with as well.. Do you know the year? Are you USA or AUS? I think I remember Tazzi stating some of the earlier E38's on his side of the pond using a different algo?? Other possibilities are tuner locked or eeprom corrupted?

[clyde]

I am in the USA, this e38 is from a Canadian 2008 Chevrolet Silverado 1500, the other e38 is from a USA 2008 Chevrolet Express 2500. Both have the same (similar) error and both I got off eBay and they should be junkyard pulls. I tried to use the ECU brute force unlocker, but I get the mscorlib error on both computers too.

ECU brute force unlocker-tested with E38,E78, and E92 ecu and gm MDI
Should work with any GM ecu with a 2 byte seed/key and a 10 second timeout between keys
This could take up to 7 days so make sure you have a stable power supply.
using dll from https://github.com/BrianHumlicek/J2534-Sharp
1 - Unlock Ecu
2 - Change starting key from decimal:0 / Hex:0000
3 - Connect to MDI and check connection by reading Vin
4 - Change Send timout from 600 milliseconds
5 - Change Read timout from 600 milliseconds
6 - Enable debug- more feedback
7 - Close Program
8 - Read vin - check connection and ecu
8
1 - OBDX Pro VX

Input Device Number:1
SendCommand Error source: mscorlib
SendCommand Error source: mscorlib
SendCommand Error source: mscorlib
SendCommand Error source: mscorlib
SendCommand Error source: mscorlib
SendCommand Error source: mscorlib
Press any key to exit.

This is the only J2534 device I have (OBDX Pro VX). Since the driver just got released, I'm not sure where the issue might be.

[ironduke]

Honestly not sure why your getting the error, usually I think that's because the .net library isn't installed. Did you build and compile it yourself off github? or did you get the .exe somewhere else? Did you build it on one computer and run it on another? If so build using the release version and use that exe.
Past that I can only guess it's an incompatibility with the OBDX Pro VX and that would be on my end, I'm 100% sure that Tazzi knows what he is doing 100% compared to that brute force code I wrote a few years ago.