PCM Hammer Release 021

[kur4o]

I used pcmhammer to flash some 99 vette in car, and got some weird conditions. After the flash completed 2 of the car modules refused to wake up[and needed hard reset to revive-battery disconnect]. It happened twice so it is not a coincidence. I went further to investigate and did some script dump of pcm using 3f and modes 28 and 20 to exit, and the modules didn`t bogged down. I don`t think reading and writing differs in any way in terms of bus control.

NOt sure what commands pcmhammer use on exits, but it may definitely improve a bit. On test script for reading exit sequence was.
mode 20 to exit read routine and crash the pcm to resume x1 mode and modules to get active.

[Cincinnatus]

I have a 97 vette with a 2002 0411 swapped in and have experienced the same problems with other modules after just reading pcm in car. I also had issues with 2004 vette which is a stock p59 after a read. I then started only flashing on bench to avoid problems.

[antus]

Yeah car reading and writing should work, but some cars/modules do strange things and have symptoms like that. Usually power cycle fixes them. If you have a car like that and want to flash in car, then that is what you need to do. Bench reading avoids that but means you need to build a bench setup so at the end its up to the person doing the flashing.

[darkman5001]

I have a thought on this that might help. Alot of times when I was using a Tech2 to program GM vehicles, just before programming the PCM, it would tell me to remove a couple of fuses. This information should be easy to get per vehicle. You just need to know what the reprogramming procedure and if it says to remove certain fuses depending on the vehicle then remove them. When programming is complete put the fuses back in.

[kur4o]

I will try to debug the issue and see why modules fails to return to normal communication. So far it seems that it is exit to normal communication issue, since with other programs reads, modules works fine without hard reboot.

[Bingus]

I've got a P12 that is soft bricked (?). The PCM is from an 06 trailblazer OS number 12606400

Using PCM hammer 021 I grabbed a bin successfully from my computer. I'm not using a bench harness but the computer is the only module on the bus. I made a calibration change and attempted to write calibration and got a "Unable to erase flash memory: Error". Unfortunately I didn't save the log from that run. After that my PCM is unhappy - not responding to any messages, fuel pump relay doesn't kick on, etc.

Successive calibration writes are able to identify the pcm is in recovery mode, and can upload the kernel. The kernel is able to retrieve the chip number and compare checksums but it still fails at the erase. Makes me think the erase was successful but pcm hammer doesn't believe so?

This is my first post but I have been lurking for a while and am a programmer by trade. I have built the master branch locally and have poked around with the VPW messages before. Hopefully I can get y'all more debug info than the average person

Anyways - here's the logs and bin - any help is greatly appreciated.

[antus]

I cant see why that would not be working from the log. I agree it looks like it has erased but the car appears to have stopped talking at that point. What is the power and voltage like? The intel chips are quite sensitive to this and usually erase problems are due to low voltage or electrical noise. This is the AMD chip though. I'd consider power first, the P12 should be working fine in PCMHammer 0.21.

Once you have considered power, we do have an update for pcmhammer to use new assembly based kernels. I was testing this with success yesterday on my P12 and my P12b. I can send you a development version if you want to try that and understand and accept the risk. Failed test will be inconclusive though as you dont have a working "before" state. But the updated XPro driver in the development version will show us the data coming back from the PCM which the cal_fail.txt log does not show.

[Bingus]

Voltage is 12 or so volts when on battery, 13 when on the battery charger. I'm going to the junkyard to get connectors to build a bench harness to remove any unknown variables tomorrow. I have an ac adapter that I know works for P01's and has done P01 flashes multiple times on multiple computers. Feel free to send me the branch and I'll gladly give it a test and send you the logs.

[antus]

12 volts is borderline for intel, not sure if AMD is as strict. Chargers are electrically noisy and should be avoided. The one that works for P01 sounds like a good choice.

[Bingus]

No luck with the bench harness. I did pick up another computer while I was at the junkyard - I'll try to be more careful with this one! It's an 07 which I believe are 2MB?

Anyways - while I have this bricked one I'm more than happy to use it as a test bed for any PCM hammer features. Can't break it anymore than it already is!