JUST BRICKED LT1 PCM

[Terminal_Crazy]

Yay, another one.

This one has socketed chips so no problem, I've got a programmer what could go wrong.

Well my newer PC has no parallel printer port.

OK i'll fire the old PC up under the bench.

Pop bang went the PSU.
I've replaced 3 with the junk ones i've got & just about managed to get it to boot.
The BIOS screen is all messed up.

OK then i'll get a USB to parallel lead.
Checks usual suppliers, nope.
Checks electronics retailer, they have one, but not at the 3 stores half an hour away.
Ah it's not a DB25 parallel it's a Centronics plug.

Bloody marvelous.

Search online again for the usual electronic gadget suppliers.
Nope nothing.
Search for USB DB25 and they decide to list DAB radios.


I have found a local PC store now that lists them and is half an hour away, so we'll have to see in the morning.

That little lot has taken me 8 hours upto now.


Sometimes i wonder why i bother :-)


Mitch
Terminal Crazy

[steveo]

damn! that's rough!

[Terminal_Crazy]

damn! that's rough!

Hi Steve, I posted a bit of info to fbody tech.
I don't know and laptop has been turned off but are there any logs kept by EEHack. It flashed E side then couldn't complete. coms routine was in ram and wanted the PCM resetting.
EEHack mentions mode 13 security key not found then suggests it could be a glitch. Then stops and says everything is all right. It did communicate & log.

I had 1 chip marked up as E side so i swapped that, no difference but would still log.

I swapped another i had for the Tside but presume it's blank as All i get now is the usual fans on.

I have a pair of 64k E & T side rom images from Ion Soltan but they don't look anything like a .bin file.

Which half of the .bin file is E and which is T? I'm guessing from looking, the T is the 1st half 0-64K where it has the line Erasing the flash on the TIME side.
&
10000H starts: I G....... is the start of the E side?


Lets see what crap tomorrow brings.

Mitch
Terminal Crazy

[steveo]

making flash logs persistent kinda fell off my to do list

maybe i should get that working

reports of flash failures are few and far between but i'd sure like to analyze them when they happen

there certainly are parts about the programming process that i don't understand, too

[steveo]

I have a pair of 64k E & T side rom images from Ion Soltan but they don't look anything like a .bin file.

ok so it's a bit confusing but the rom is address scrambled, so when placed into a regular programmer for reading or programming, you don't get the normal bin back.

it's like you see ABCDEF when the bin is read from aldl based software like eehack, but when read with hardware the rom has FCDEBA or something since (im guessing) the address pins aren't traced normally to the processor. this might have been done on purpose.

that means you can't just get an LT1 bin file and split it, and use your programmer to flash them onto the chips. you'd need a clean read from the prom.

the rom images ion gave you would be a good starting point.

i got halfway to figuring out the scrambled addressing them gave up, but i'd love to write a conversion tool for it, that's for sure!

[Terminal_Crazy]

ok so it's a bit confusing but the rom is address scrambled, so when placed into a regular programmer for reading or programming, you don't get the normal bin back.

it's like you see ABCDEF when the bin is read from aldl based software like eehack, but when read with hardware the rom has FCDEBA or something since (im guessing) the address pins aren't traced normally to the processor. this might have been done on purpose.

that means you can't just get an LT1 bin file and split it, and use your programmer to flash them onto the chips. you'd need a clean read from the prom.

the rom images ion gave you would be a good starting point.

i got halfway to figuring out the scrambled addressing them gave up, but i'd love to write a conversion tool for it, that's for sure!

Update.
The one place that advertised the usb to parallel lead didn't have one.
Well what a suprise :-)
That the morning gone.

So, Today I've ordered
a USB to parallel lead (cheap, to see if it works) due next week.
an AGP graphics card to replace the one in my old PC with a parallel port. ( also cheap & also due next week).

The original ANf512 PLCC 32 chips do not appear to be available over here.
What other make/model chips will replace these or work ??

Thanks
Mitch

[steveo]

usually some guy in china on ebay has a collection of new old stock ones. if you're careful de-soldering the old ones, you can just re-use them

[Xnke]

Not as easy as it sounds to re-use them. You need to remove the conformal coating, and that shit don't come off easy.

Unless you get ALL the coating off it makes for no end of programming errors. Pins don't make good solid contact in the programmer OR the new socketed PCM.

[Terminal_Crazy]

Not as easy as it sounds to re-use them. You need to remove the conformal coating, and that shit don't come off easy.

Unless you get ALL the coating off it makes for no end of programming errors. Pins don't make good solid contact in the programmer OR the new socketed PCM.

I found some on ebay, so they are coming in a few weeks.

After replacing the PSU the PC booted, ok, the screen was still mangled
The AGP card arrived Saturday.

So today i've spent the morning getting the Willem programmer to do anything.
Hardware error blah blah blah.
After several hours it dawned on me to check the system devices, for some reason the COM & LPT port don't show up on this machine. Great
Another morning wasted.

Let's see if the USB->parallel converter works ok.

Mitch
Terminal_Crazy

[steveo]

After several hours it dawned on me to check the system devices, for some reason the COM & LPT port don't show up on this machine.

check in the bios, they're probably disabled.

[Terminal_Crazy]

check in the bios, they're probably disabled.

No they are enabled :-(.

The floppy/floppy controller has also died, not sure if it is related to the same issue. (PSU let the smoke out)

Everything else seems to work OK.

This old PC is going into the garage to run my milling machine & software rather than keep using the cars laptop.

Either the USB-> Parallel converter works or I get a Parallel port card or a new programmer.

Thanks
Mitch
Terminal Crazy

[Terminal_Crazy]

Well the PromProgrammer turned up today & I reflashed the two spare roms, Broke the leg off one.

I've just emailed the EEHack comm log to Steveo to see if he can say if i have any comms.......

Both chips reflashed. & compared to Ion Soltan's 64k bins.

I've removed & replaced them several times, polished the legs & sprayed cleaner on them.

I'm fairly positive that they aren't sat correctly as they were shimmed slightly with cardboard when i took them out.
Even the flash programmer needed them shimming a little.

One socket has a baseplate, the other hasn't.

I'm just going to rethink where i am now.


Which side do i concentrate on getting working first, the T side ?

I presume with these flash images, the fans shouldn't come on with ignition ?

EEHack originally reflashed the E side and the car communicated. I tried several other roms in both side but fans ran every time with ign& no comms.
Will comms work with just one rom connected so i can narrow down the error.
Any way i can check the rom leg connection ?

Any thoughts please people.

Thanks
Mitch
Terminal_Crazy

[steveo]

both sides need to boot up for anything to really happen, but if the t-side alone is working, you'll get some rudimentary comms

seating those things in aftermarket sockets is a real pain in the ass, i made my own plastic shims. you want the chip surface to be almost 1mm above the top of the socket.

[Terminal_Crazy]

both sides need to boot up for anything to really happen, but if the t-side alone is working, you'll get some rudimentary comms

seating those things in aftermarket sockets is a real pain in the ass, i made my own plastic shims. you want the chip surface to be almost 1mm above the top of the socket.

Cool, I'll try that over the weekend.

Thankyou
Mitch
Terminal_Crazy

[Terminal_Crazy]

OK tried again.
Reflashed the roms & verified them against Ion Soltans images.

Fans run when ignition switched on.

EEHack connected and read vin no etc. Had a malf 43 (knock sensor show up, cleared and kept coming back, not an issue i don't think).

Tried testing comms. EEHack connected and read packets but fails at the stability test.

Steveo: What does
Executing programming Stability Test
Disabling programming lockouts(mode 13)
Can't get the security key seed (mode 13) giving up.
This might just be a communications glitch. try key on/off
go again.
------------------------
FINISH
actually do ?

This is as far as i've got. :-(


Thanks
Mitch
Terminal Crazy