Perhaps it wont come out of boot mode..... is soldering involved to get boot mode ? Bad connection or short ?

At the time of programming remove immo

in galletto 1 -bootmode does not work

MPPS - bootmode OK

Sorry
I did not notice, that the topic was already described.

About 29F400
maybe is defective.

Happened to me something like this
29F400 programmed sometimes OK, sometimes 70% or 55%
The programmer similar--- once ok, once bad

I change 29f400 on new and it was OK

Relic wrote:Perhaps it wont come out of boot mode..... is soldering involved to get boot mode ? Bad connection or short ?

It won't go into boot mode unless pin 24 is grounded, so it's not stuck in boot mode.

When out of boot mode the ECU just doesn't seem happy to be communicated with by any OBD protocol.


I have a feeling the seed key for OBD access has been screwed up (possibly by a slave KESS tool that tried to fix the original problem), and left locked or something?!
I have a sneaky suspicion that the ECU will work fine in the car now. But no OBD access will be allowed for things like ELM327/tuning tools etc.


Next job is to read the dil8 flash and see what the data looks like, and see if the seed key has indeed been changed from what it should be... (not sure how to check that but a friend does)

Maybe it's just a matter of setting seed key to allow OBD access again?



Yeah Glapsson, we figured it might just be dodgy flash data, but we wrong two ori files for this ECU on and off and checked them and they were exactly what was written on, then we wrote a file for a VAG EDC15, all checked fine, then we wrote random other files just to test data integrity.
We did all that via boot mode about 8-10 times and every file we read back off after a reset was in perfect condition.

We'll try the dil8 data check first, and if that appears ok then it might just need a new flash chip like you say!

Dave

This document help understand as boot_mode work. In short: CPU ROM have program named "bootstrap loader" (BSL). BSL activated by low level during reboot on CPU pin that connected to pin 24 flash chip. BSL allow load from serial interface short program (<=32byte) to CPU internal RAM and start it. Flashers (Galleto, MPPS etc.) this way load preloader that allow load to external CPU RAM program for work with flash memory and start it. This program communicate serial interface with flash chip.

So any reason it wouldn't work in boot mode?

Only thing I can try now is putting it into the car and seeing if it works, just won't comm over OBD any more.

Feeling is that the OBD interface seed key has been changed to not allow access...

Dave

Mr Whippy wrote:So any reason it wouldn't work in boot mode?

Something is wrong.

Mr Whippy wrote:Feeling is that the OBD interface seed key has been changed to not allow access...

Boot mode not use seed key and must work.

I have a sneaky suspicion the software is wrong - i know from your circumstances that it sounds implausible but as bootmode can succesfully write the ecu and the write can be verified it would suggest that software with valid integrity isn't able to boot on that ecu

Hi Dakta,

Thanks for helping me with these ECU's... your input and experience have been a great learning experience.


For those watching this problem, Dakta tried a Citroen Xsara software on the ECU, and that worked fine. The ECU communicated fine via boot, normal OBD tools, reading/writing the Xsara software on/off.

As soon as we went back to the C5 software it wasn't happy.

We then tried an 0281010808 ori files, the original (apparently) from the owner, a few from off the internet, another software version, and also a completely different C5 hw/sw type, and none of them worked except via boot mode.

Back to a Xsara file, and et voila, it worked fine.


Interestingly the person who bricked the ECU originally with a "Galletto" 1260 got a 44% write and it failed. BUT, he still managed to communicate with the ECU and get a 44% partial read off it with just a Galletto.

We wrote that to the C5 ECU and wow, it worked. It read fine, communicated fine over generic OBD. The file present (the 44% written) is identical to the ori file. The only things missing are chk and the end of the software section, and the data sections obviously.


Something to do with complete C5 software is not working basically. I can't fathom what is going on...

Part of me is wondering if the C5 software is very sensitive (ie, it needs the exact software it should have or checks in immo eprom to make sure it matches or something), and if the software isn't matching it won't work.
But that means either the owner didn't send me the right original file to recover the ECU, or the KESS tool changed something in the immo eprom... if that can even happen or exists.


Brain is fully fried. This is the danger of taking on an ECU to fix after someone else has had a 'go' at it... you just have no idea what is right or wrong, or if the owner is being fully open and honest about if it ever worked, or if the original file is really his original file etc etc...


Hmmm

Dave

The plot thickens.

ECU sent to owner, plugged in, works fine.

Now to see if his Galletto can even detect it!

Dave