PCM Hammer P01 and P59 flash tool v015

[antus]

Yep. And you'll need to reverse engineer existing code to learn too. But first you need a platform that'll load the existing kernel then you'll need to confirm if it runs and if not see what addresses are different.

[darkhorizon]

Current plan to load the existing involves modifying the existing to some degree. I think that the mode 33 requires that I offset the memory addresses so that there is room for the mode 33 in the first block.

[antus]

yep, if ram is not at FF8000 you'll need to move the load address.

[JonnyB]

Hi everyone, and thank you to those developing this software! I've been following this thread since I discovered it about a month ago.

Since then, I made a nice little bench harness from parts Ive hoarded and info from Sloppy's site and have access to an mdi2 at work. Im completely new to aftermarket tuning whatsoever and my budget wont allow hptuners etc but I have an lm7 that I'd like to get running on more than starting fluid. I have a used 0411 pcm and just tried a read on it today with a fresh download of release 4 on my work computer. Upon clicking read full contents, it quickly makes its way to "...kernel uploaded to pcm sucessfully. Requesting data..." Thats when theres a seemingly long 35sec pause and it says "Something went wrong. Index was outside the bounds of the array. Clearing trouble codes. Elapsed time 00:00:35... Read failed. Error" The first error i see in the debug reads "ReadMsgs OBDError: ERR_BUFFER_EMPTY No payload following read request." Has anyone seen this? I hope its something Im just missing or doing wrong

Thanks in advance for any help!

[darkhorizon]

Having what feels like endless problems with this bit of code in the AVT receive. I get blockcopy errors all the time so I added a little try catch around it, but it still doesnt do much to help me debug why im having issues with the block. Most of these seem random, but I cant imagine its healthy. this is on low speed, with 2k blocks, and for some reason 4k blocks completely bomb in this (truncates the packet by a fair amount, upgraded the timer to 5 seconds to see if it would help too). Any suggestions on what could be happening here that I'm missing? Potential changes to this to make it a little more reliable at pulling the correct receive information off the serial bus cashe?

edit to add: This is not the current LS1 kernel though, i have changed from the random kernel to another random kernel tonight and it has largely solved my issues here.

Code: Select all

            // build a complete packet
           
            byte[] receive = new byte[length];
            byte[] packet = new byte[length];
            int bytes;
            DateTime start = DateTime.Now;
            DateTime stop = start + TimeSpan.FromSeconds(5);
            for (int i = 0; i < length; )
            {
                if (DateTime.Now > stop) return Response.Create(ResponseStatus.Timeout, (Message)null);
                bytes = await this.Port.Receive(receive, 0, length);
                this.Logger.AddDebugMessage("Total Length=" + length + " RX: " + receive.ToHex());
                try
                {
                    Buffer.BlockCopy(receive, 0, packet, i, bytes);
                }
                catch
                {
                    this.Logger.AddDebugMessage("failed receive: " + receive.ToHex());
                    this.Logger.AddDebugMessage("failed packet: " + packet.ToHex());
                    this.Logger.AddDebugMessage("failed bytes: " + bytes.ToString());
                }
                i += bytes;
            }
            

[DavidBraley]

Hi everyone, and thank you to those developing this software! I've been following this thread since I discovered it about a month ago.

Since then, I made a nice little bench harness from parts Ive hoarded and info from Sloppy's site and have access to an mdi2 at work. Im completely new to aftermarket tuning whatsoever and my budget wont allow hptuners etc but I have an lm7 that I'd like to get running on more than starting fluid. I have a used 0411 pcm and just tried a read on it today with a fresh download of release 4 on my work computer. Upon clicking read full contents, it quickly makes its way to "...kernel uploaded to pcm sucessfully. Requesting data..." Thats when theres a seemingly long 35sec pause and it says "Something went wrong. Index was outside the bounds of the array. Clearing trouble codes. Elapsed time 00:00:35... Read failed. Error" The first error i see in the debug reads "ReadMsgs OBDError: ERR_BUFFER_EMPTY No payload following read request." Has anyone seen this? I hope its something Im just missing or doing wrong

Thanks in advance for any help!

JonnyB,

I too am very new to this, and when I did my research for building my bench programming harness, I compared the schematics/pinouts of several bench harnesses and found that the bench harness that Sloppy uses didn't match the others. In the picture below is the bench harness I built. It's very close to the Sloppy harness so you will not need to change much. It works with both the Blue/Red P01's and the Blue/Green P59's. You might make yours like this one below and try to read again? Also, make sure your power supply is something reliable, above 12 volts, and able to put out at least 1.5 amps.:

offboard_harness1.png (283.62 KiB) Viewed 5142 times

If others have had good luck with Sloppy's bench harness, please speak up. I have no experience with it.

[darkhorizon]

Other news for the V6 stuff... The PCMHammer kernel loads and executes onto the PCM now, but does not run yet. This might be as simple as changing the load address, or just looking for different responses... Current response that the PCMHammer kernel gives is this:

[11:00:24:412] Trace: ReadAVTPacket
[11:00:27:530] Trace: 12
[11:00:27:535] RX: Header 0C
[11:00:27:538] Total Length=11 RX: 6C F0 10 7F 36 00 0F AC FF 91 12

Other news, the other kernel produced a good read on a 98 and a 03 v6 PCM, no errors, first try after I fixed some more stuff. So I think with this kernel i have it should work fine for reading, and it is something we could bundle with the PCMHammer legally one day soon after I do more testing.

[darkhorizon]

Oh, and this V6 read kernel i have built out, supports the old 97/98LS1 too, i was able to read one of those tonight as well.

[brandonwh64]

If others have had good luck with Sloppy's bench harness, please speak up. I have no experience with it.

I am currently using one I built from the sloppy wiki. I just followed the pictures

[DavidBraley]

I am currently using one I built from the sloppy wiki. I just followed the pictures

That's good to know!