VAG-COM/VCDS EVERYTHING

[hid3]

The problems are two:
1) hell knows if only the pinout is different between 0x44 and 0x46. May be different logic on some pins too
2) hell knows if the cables which come as "0x46 hardware" do actually have corresponding 0x46 dump inside the GAL. It might be possible that they use widely available 0x44 instead...

PS: Elektrik1, greetings from Vilnius

[Elektrik1]

I do have 6 working cables, 4 of them is found as 0x46 and 2 as 0x44 by kolimer tool. Will dig a bit later and will summarize what are the differences.
From GAL datasheet it is clearly seen which pins are inputs only, so I can generate thruth-tables for corresponding outputs for them, then from schematics it can be seen which I/O pins are used for inputs and outputs. This way logic can be compared for 0x46 and 0x44 GAL`s.
A007739.jpg

hid3, hehe, nice

[Roma77]

Well, then we can arrange brute-force of 0x46 GAL chip - generate truth-table for each input-output combination (using arduino or any other fast solution), and maybe then someone can write a code for GAL.

Or maybe C3H8 decompiled 0x44 logic can be rearranged for different 0x46 pins (if only pin-out is different).

I do not know whether you know about this article or not, it will be useful to read ch. http://www.bhabbott.net.nz/atfblast.html

0х44 logica
JEDEC file for GAL16V8 created on Tue Jan 04 01:01:11 2011
*QP20*QF2194*QV0*F0*G0*X0*
L0256 11111111111110111111111111111011*
L0288 11111111111111111111111110111110*
L0768 11111111101111111111111111111011*
L0800 11111111111111111111101111111110*
L1024 11111111111111111111111111111011*
L1536 11111111111111111110111111111111*
L1568 10111111111111101011111111111111*
L1600 11111010111111111011111111111111*
L1792 10111111111111101111111111111111*
L1824 11111010111111111111111111111111*
L2048 11111111*
N UES 0C 0C 04 00 00 00 00 00*
L2056 00110000001100000010000000000000000000000000000000 00000000000000*
L2120 11111111000001010111111111111111111111111111111111 111111111111101001000011*
N PES 00 00 00 00 00 00 00 00*
C2C23
*8BF4

-------------------------------------------------------------------------------------

CUPL(WM) 5.0a
Device g16v8as Library DLIB-h-40-2
Created Fri Mar 17 10:55:11 2006
Name atf16v8 from AGK
Partno 00
Revision 01
Date 10.03.2006
Designer AGK
Company
Assembly None
Location
*QP20
*QF2194
*G0
*F0
*L00256 11111111111110111111111111111011
*L00288 11111111111111111111111110111110
*L00768 11111111101111111111111111111011
*L00800 11111111111111111111101111111110
*L01024 11111111111111111111111111111011
*L01536 11111111111111111110111111111111
*L01568 10111111111111101011111111111111
*L01600 11111010111111111011111111111111
*L01792 10111111111111101111111111111111
*L01824 11111010111111111111111111111111
*L02048 00000000001100000011000000100000
*L02112 00000000101001001111111111111111
*L02144 11111111111111111111111111111111
*L02176 111111111111111110
*C2C1E
*B8F2

[Elektrik1]

Roma77, now those dumps needs to be checked somehow if compiled C3H8's decompilation produces same result

Ok, so let's work on this together! What I can offer is the logic for 0x44
chip vcom GAL16V8

nc1=1 i2=2 i3=3 i4=4 i5=5 i6=6 i7=7 i8=8 i9=9 GND=10 !i11=11
o12=12 o13=13 f14=14 o15=15 o16=16 f17=17 o18=18 f19=19 VCC=20

equations

!o18 = !i5 & !i9
# !i8 & i11
!o16 = !i4 & !i9
# !i7 & i11
!o15 = !i9
!o13 = !f14
# !i2 & !f17 & !i6
# !i3 & !f19 & !i6
!o12 = !i2 & !f17
# !i3 & !f19

If it does, I will try to generate such logic for 0x46 GAL, and then if compilation of JED produces same result, we could compile it for 0x46 as well.

[Roma77]

Yes, I am waiting for my programmer. XGecu PlusTL866II

[hid3]

Elektrik1, before jumping to straight conclusion that some of your cables are really 0x46, better check them manually.

0x44: ATmega 34 and 35 pins are connected to GAL pins 5 and 4 accordingly and GAL pin 9 is connected to some pin on the FTDI (use multimeter in diode testing mode to confirm).
0x46: no connection from ATmega 34 and 35 pins exist to any pins of GAL and GAL pin 9 is NOT connected to FTDI and GAL pin 5 is connected to FTDI.

If you have anything between those two, then you have a cable-mutant

[Elektrik1]

hid3, they are, you can see it in screenshots and my reverse engineered schematics, I have 3 of such cables
20181229_010018.jpg

[hid3]

Yep, this one seems to be correct 0x46 one as per schematic. Thumbs up for the protective diode on the OBD2 #16 pin

[eliteland786]

Interesting read...The cable I have, pictures few pages back reports 0x46 but it's pins 34 and 35 are connected to the chip on other side...so basically software shows 0x46 but in reality it's 0x44...It auto detected 46 and flashed with no problem....Thanks to Kalis.

[eliteland786]

Need to check the last bit ftdi not connected to gal 9...and gal 5 connected to ftdi....good post.

[hid3]

eliteland, if your interface is the one as shown in Kolimer's Info.pdf example, then it's definitely 0x44 just mistakenly flashed with 0x46 dump from the factory. To be absolutely correct, you'd need to reflash it to 0x44

[eliteland786]

eliteland, if your interface is the one as shown in Kolimer's Info.pdf example, then it's definitely 0x44 just mistakenly flashed with 0x46 dump from the factory. To be absolutely correct, you'd need to reflash it to 0x44

Sorry but a dumb question is atf16v8b a gal chip cause can't find gal chip..cause it bells out.

[hid3]

Yes, it's the same thing. ATF/GAL/Lattice/Palce are all the same logic chips

[eliteland786]

0x44: ATmega 34 and 35 pins are connected to GAL pins 5 and 4 accordingly and GAL pin 9 is connected to some pin on the FTDI (use multimeter in diode testing mode to confirm).
0x46: no connection from ATmega 34 and 35 pins exist to any pins of GAL and GAL pin 9 is NOT connected to FTDI and GAL pin 5 is connected to FTDI.

If you have anything between those two, then you have a cable-mutant[/QUOTE]

Just to add mine is 0x44....also to add to 0x44 is connected from gal 9 to pin 1 on FTDI...just checked....also when I opened the reader it didn't detect if bootloader is enabled or not....but flashed the cable automatically didnt see option to select anything but pressed write and it worked might have got lucky....don't know if it selected 0x44 or 0x46 automatically.

[hid3]

Not pin 1 of FTDI but rather pin 25, I believe. Check the chip orientation!

The EEWrite tool reflashes it to the same HW as it was flashed before. It's impossible to flash it to different one because of different bootloader size/different fuse bits.
In order to reflash from 0x46 to 0x44 you will need to do that using AVR programmer (e.g. USBASP)