pcmhacking.net

So I guess nobody was interested in taking a stab at converting my file...

Spring finally came here and many other projects presented themselves, unfortunately. I finally broke down and purchased a GQ-4X and adapter, I have uploaded the resulting file here.

I will post a link to my sourceforge project once I set something up there, for those who might be interested in a roll your own programmer.

Thanks again, and any help would be greatly appreciated.

Ivan

I got half way though it and had something sort of working but ran out of time and didnt get back to it. Thanks for posting the bin.

The read looks good. The reset interupt is the first 4 bytes of the bin, and every 4 bytes after that are the other vectors. The MFS blocks near the start look like the kind of place i'd expect the id stuff your looking for to be. I'd say try swapping 1 or all of the smaller mfs blocks near the start between your bins and you might find you can start the car on the other tune.

I'm not sure what you mean by the MFS blocks?

This is encouraging, I'm going to pull the eeprom off of the other PCM tomorrow and read it also.

Since my board is a flex circuit and not a PCB, it won't take too kindly to being put on and removed so many times. Hopefully with some help I can get this first time, or come up with a method where it can be read and written from the obd2 port, as other pcm's do?

Thanks for all the help so far.

Ivan

Yes, it might be best to fit a socket rather than reworking the pcb over and over. I expect you might need to do several flash cycles with trial and error to figure out what to swap. I attempted to install a TSOP 44 socket in an ls1 pcm once, but my soldering skills were not quite up to it. You sound like you might have the skills to do it though.

There are 4 blocks that start with MFS*. They are @ 4000-41ED, 4200-43ED, 6000-609F, 6200-62A0 as well as some other data between 4400-47F7. Due to the FFs the block ends ive named are somewhat arbitary. I think these 5 chunks are what you should attempt to swap.
.. or maybe ..
Keep in mind these are only my best guesses to point you in the right direction. My highlighting is not the only way to look at this, just for ideas.

The MFS blocks may also be an index in to other segments in the bin, in which case the data from 4400-47F7 is the most likely suspect.

Where did you manage to find a surface mount PSOP44 socket? Nearly all that I can find go to socket pins, and that doesn't really work for my application. I found one on a site from the UK, but I'm a little reluctant to order it.

Ivan

Unfortunately I dont remember where I got mine from, but the raw part is expensive. I would suggest you pull one off of or adapt something like this:

http://www.obd2tool.com/goods-3204-CHIP ... SOP44.html

I got one of those...but it's not smt I'm afraid..I could hack it in, but it would be a mess. I really like this:

http://www.inovauto.com/index.php/socke ... m_store=pt

but I haven't found any north american retailers of it yet..I could order from them I suppose, just not sure they ship here, and it would probably take ages.

I did order this...I don't know if it's SMT or not, it's really hard to see how it solders, but I think it's what I need:

http://www.ebay.ca/itm/201240149333?_tr ... EBIDX%3AIT

The price is reasonable, even just as an experiment.

Ivan

Ahh yes that second link is the one ive got. not much room to get it soldered down, but possible. My attempt failed because I lifted a couple of pads+tracks when lifting the second side of the original chip off the pcb. I tried to fix that with wire links but either there was damage i couldnt see/find or the signal integrity wasn't there due to changing wire lengths/capacitance. I dont know what the tolerances are for that kind of work.

the 2nd one is the one you have? cool..I had good luck taking the chip off of my pcb with a heatgun...that's really the only way to do it without lifting pads, the pads came out pristine, which I was surprised about considering it is flex..haha.

If the socket is solderable, I'm sure I can figure out a way to make it work.

It is surface mount, the rows of pins ive drawn the box around sit on the pcb pads so you can come in from the side with a fine tip iron (or probably other ways).