PCMHammer P04

[geep]

i have a few of the black box pcm's i can test if any one has the pinout's for them

[antus]

You can see pinouts on page 18 here: https://download.efilive.com/Tutorials/ ... erence.pdf

[geep]

Here is the log from the updated version

[geep]

log and bin from a black box 4 plug the pcm was from 4.3 v6 s10

[antus]

Great, thanks. Note your voltage is showing as barely 12v in the first log, and 11.8 in the second. You are likely to have problems at those voltages with erase and write, and should use a power supply or battery that can deliver higher voltage when writing. With 12.5v or so (and about 1 amp of current on tap) both PCMs should be writeable, but as always brick is still possible so dont risk it if you only have 1 pcm and need it for your car.

[antus]

We found a number of early PCMs that were P04 type, that had the early type hardware and 512KiB flash images. PCM Hammer development version was not able to deal with these as half the details matched the early, and half matched the late. We used to call them P04_256k and P04 (meaning late) types, but now we know this is not true I've renamed P04_256k to P04_Early, and documented the hardware types differently and its now working on all P04s that we have been able to test. I think this could mean, P04 is finally done! Here is hopeing nothing else comes up. So, A little bit more testing and maybe solve an edge case on the E54, then PCMHammer 2.0 should be ready.

[JohnCNY]

I have run into this issue - a P04 with a 512k bin:
Actually have 2 of these. 97 Lumina and 97 Century: Both are Serv No: 16217058, Hard No: 16218070

On the Lumina OS ID:16245158 and Cal ID:16245154 (there is an update for the OS to 09355440)

On the Century OS ID:16254347 and Cal ID:16254394 (there is an update for the OS to 09359553)

On the PCMHammer main screen, both come up as Hardware Type:P01_p59; BCC query fails and unlock fails on read. I suspect part of this failure is because the key algo is supposed to be 0x06 but since they are detected as hardware type P01_p59, I think it is probably using wrong algo.

I have a bench harness for these and would really love to be able to write to them. Looked at the pcmhammer on github and it doesn't even have os 16254347 in the lists.

I am willing to help mess with these. I have a spare pcm. I am not sure how many people are still interested in these. The 97 just got a 2005 3400 upgrade using that 16217058 PCM in stock form. The only change I made to the 3400 was that I built a different injector harness and swapped out and wired the dual output temp sensor from the 97 3100. Using the stock injectors from the 3400. The only issue I have with it at all is if it idles for 10-15 secs, I get a slight lag off the line. I think (I mean I hope) that if I change the injector flow rate to match the 3400 injectors it will be better.

[antus]

Try this test version, it has had limited testing, but 100% succesfull so far. If we dont have any issues reported it'll become pcmhammer 2.0. You are an early tester and risk is yours so dont try it on your daily but confidence is high. viewtopic.php?p=130513#p130513

[JohnCNY]

Spent a bit of time trying to get this to work. In the end, it appears that on a 97 P04_early 512k and 98 P04 both succeed to do a full read with a J2534 CarDaq plus3 and USB OBDLInk SX but FAILs with my ALLPro. I also tried with the AllPro with 4x disabled. It still failed. I have used that AllPro a lot with 411's with no issues. I attached the 4x disabled logs.

It seems like at the end, the read finishes, then begins to check CRC's from PCM and gets no response. File never gets saved but I assume that is normal behavior. I am pretty sure that I tried verify only and that failed as well.

Please let me know if there are any tests I can provide results from for you.

[antus]

OK Thanks. Allpro is pretty much abandoned at this stage as the hardware developer pulled the project, and my unit failed in hardware. I'll have a think about if I want to fix it or remove it. Good to hear it worked with the cardaq and obdlink. It sounds like the timeouts are occurring too quickly as set by the allpro code in the driver for the allpro to get the response from the PCM.