P05 PCMs

[antus]

Is there anyone on the forums who can help me understand the P05 PCM Hardware?

I currently think there are 3 variants. VPW, VPW+CAN, and CAN. Is this correct?
Is it fair to refer to these as P05 a b and c?
Is the hardware different for protocol support, or is it in the OS?
Universal patcher XMLs define P05c and v6-1mb. Are these really the two types? If so, do a and b exist at all?

Working with another forum member we are close to the point where we'll start to cross flash OSs and see if they brick or the protocol follows the hardware. But it's not worth taking the risk if that information is already available.

In particular we have access to two PCMs with service number 12600930. One appears to be only talking CAN, one only VPW. We are not sure if one is miss-labled, they have different operating system protocol support, or a hardware fault on one or both. Cross flashing may answer the question, but also has a brick risk.

[MPC001]

Is there anyone on the forums who can help me understand the P05 PCM Hardware?

Please forgive me being blurry on this, but is P05 the family the E40 is part of? At least appearance wise.

If so great question!

[antus]

Yes, its the crossover model. The platform is nearly identical to the platform in the former shape, but it has a newer generation of flash chip, and duel OBD protocol with VPW, CAN or Both depending on the version. Exactly how is the detail I am trying to figure out so that I can describe them properly in PCM Hammer (though only with VPW support at this stage).

[antus]

I think I have the P05 file formats figured out. Up until now the P05 files I've looked at use the same config as P04. I just needed to modify p04 a fraction to handle the 1Mb file.
Looking at universal patcher, it has different ID logic for v6-1Mb and P05c. It seems to use v6-1Mb for the files that PCM Hammer could handle, and P05c for the one that I have so far that it can not.
PCMHammer itself was used to read the file that fails the test, so it looks like security algos and hardware compatability is there between the v6 1Mb and the P05c, and I think the p05a and p05b must be variants that together are handled by v6 1Mb.
So I have decided to the internal type of P05 for all of them, as that will handle the flash side of things, and add some clauses and some new logic to handle the P05c variant when reading OSID and calculating checksums on the P05c format. The P05c has operating system and calibration checksums, whereas the others do not.
This seems to be working nicely now and has been pushed up the github.

I still need answers about VPW, CAN, and what service numbers can be interchanged with each other. I am hoping protocol is part of the operating system and all the hardware will do VPW with a VPW operating system. If this is the case I'll need to make pcmhammer capable of detecting protocol support from the bin, and blocking a flash of a CAN only bin because even though that may not brick, another CAN based tool would be needed to restore the VPW functionality. Any help with regards to this topic appreciated.

[darkman5001]

I posted this answer in the P05 support forum a little bit ago.

viewtopic.php?p=135889#p135889

"I think all of the P05s speak VPW, however on certain cars that uses them the car speaks in CAN so I think the VPW pins are not used in some cases. So flashing a CAN wired car with PCM Hammer wouldn't work because there is no VPW wire run from PCM to OBD2 connector. Now does the OS dictate whether VPW pins are enabled, or are they always enabled? I don't know the answer to that. I just know that they all have the VPW pins on the PCM connector. Would be nice to bench test with a P05 out of a CAN only car and see if it will talk on those VPW pins as is."

[darkman5001]

https://youtu.be/WPzT0fdESjw?si=CH6VNULdMMvNRxwx

[antus]

Thanks. I have two 12600930 and they only talk can. I thought the first one was faulty and so I bought another one to find the same result. At that time I thought they all had vpw and so it was only then I hooked up the can pins and found they were alive. Theyre not that useful to me and I have 2 so if I cant find the answers here i'll try and flash a vpw bin in with a can capable tool and see if the vpw pins come alive or it soft bricks needing bdm (which we dont have figured out) to restore can comms. If it dies it'll probably be a bin job or flagged as dead and live on a shelf. At least it'll be answered.

If I can find a bin with both can and vpw i'll test with that as hopefully it would keep talking can even if vpw doesnt start up after swapping the OS in.

I need to figure this out because if there will be can only bins that ID as good P05 files and people flash them in with vpw but dont have a way to flash with can they'll soft brick for their application. So i'll need to find a way to identify bin capabilities and block or strongly warn before writing it in.

[antus]

I hooked up my can only service number 12600930 and I cant find any tool that will read them, can just retrieve vin and basic info. So maybe getting a can bin is a dead end in the first place, as is flashing a vpw bin in to see what happens.

Can anyone confirm if service number 12600930 is a P05? That could explain things. Commercial software IDs it as P05c over can by vin, but can't do anything with it. There might be an issue there, same as how I picked this one by eye from ebay listing based on car type. Maybe that car had a platform cross over at the end of its run that nobody has picked up on yet and its something else, newer, that is can only. The vin on my can only 12600930 is 1G4HD57297U116713.

[darkman5001]

I hooked up my can only service number 12600930 and I cant find any tool that will read them, can just retrieve vin and basic info. So maybe getting a can bin is a dead end in the first place, as is flashing a vpw bin in to see what happens.

Can anyone confirm if service number 12600930 is a P05? That could explain things. Commercial software IDs it as P05c over can by vin, but can't do anything with it. There might be an issue there, same as how I picked this one by eye from ebay listing based on car type. Maybe that car had a platform cross over at the end of its run that nobody has picked up on yet and its something else, newer, that is can only. The vin on my can only 12600930 is 1G4HD57297U116713.

Yes Antus, that is a P05 and that vin belongs to a 2007 Buick Lucerne 3.8L. It does look like it communicates using CAN. I do not see it using VPW according to its connector view for this particular vehicle. What pins are you using to communicate VPW?

I also have two P05s here in my collection of PCMs. So, I can do some testing if need be.

[antus]

Using the pinouts here: https://www.customecm.com/tune-file-rep ... Up3QhysIUi
I am getting VPW on service number 12581501 and can on service number 12600930 with these connections. Neither of those service numbers are communicating on both. We (rdrew is helping) have found 2 different file formats on the 12581501 and pcmhammer can handle both. So far there is no way to read the can. I'd like to find an OS that is talking can and VPW that we could use to cross flash in to both of these to enable both protocols. But we also dont have a way to write the can version at this stage. That may be solvable but it'd be good to read it first. At this point, I am considering calling P05 done and just ignoring the can PCMs. If I can work out what service numbers support vpw I can document that, and that is what pcmhammer will do. There are no current plans to add can support, I just want to make sure nobody locks themselves out by flashing in a can osid to a vpw pcm.

If you want to test you can get the latest 'green tick' version from https://github.com/PcmHammer/PcmHammer/actions and try that. I believe its fully working or can vpw.
I've also just added an attempt at p11 support but it'll take a few weeks for the hardware to get here, and is has no P11 OSIDs listed. That is still the latest and greatest to test for p05 as well.

If you have any P04/P05/P08/P10/P11 bins laying around a couple of random file verifications would be a good test too. There are a lot of similarities internally, but I had trouble with pcm hammer identifying p11 as P10, and universal patcher identifies P10 as P08. I think its sorted now, and from my random tests against bins I have on hand its working ok, but it'd be good to have some more spot checks on those formats on other files. P05 has 1 checksum for the whole file, P05c has a checksum for operating system and one for calibration. It should be able to tell the difference and show the appropriate sums for both.