NefMoto

I’m about to send my cluster for full rebuild and I was wondering if it’s possible to pull the 4 digit code that matches the ecu before I send it?  The last time I did something similar to this is when I got the code  from a beetle ecu, which I had to take apart and ground one of the pins on the eprom.  Is this similar idea on the cluster?  I’m hunting around, but maybe looking in the wrong place.

It is quite easy, something along the lines of VAG eeprom programmer (or similar software) can pull the EEPROM from the cluster through OBD using a blue kkl cable. Enjoy   :)

Please expand because I have not had any success reading mk1 TT clusters via OBD

It´s not VDO, it´s Magneti Marelli Jaeger M73. No external eeprom and not working with shit tools. Even original Vagtacho und Vagdashcan does not work correct, lot´s of tools read not complete range. I think cheapest solution for these is VWlog from codecard or just read processor with programmer. It´s some Motorola, nothing special, but you need to read the correct one because there are 2 inside.

I’m about to experiment.  I have a spare cluster I bought real cheap and going to wire up bench wise.  I’ll let you know if I get it to work. 

Well sorry - i was under the impression these were also VDO.

I have a friend with a TT, i'll see what my tools can do....

It always works on Seats, VW's, but tbh never tried on Audi stuff

So I found VAG K + commander 2.5 and I managed to read the cluster, but I'm not sure I got the whole file.  First off if you've ever used this program you'll see an open channel thing down at the bottom right.  You can select 17 for instrument all you want and open channel it won't work.  Do however select K-Line at the very bottom, this will allow you to press special function - instrument/immobilizer.  Then using the drop down there's a   Magneti Mareli shadow 3 (TT) option.  I tried the Magneti Marili Shadow -direct and it crashed my PC.  Anyway the shadow  3 TT option does something.   So now here's where I'm at.  I have a file read, but I'm not sure that's the full file.  I don't see anywhere to set the amount of hex lines to be read.   Or now that I have a file I'm not even sure what to do with it.  Anyone know if the 4 digit code is in there?  Or how I would translate it.  I'd also like to know how to switch from manual to automatic, but I think I can do that in vagcoms soft coding.  Maybe.....

read as direct

This program crashes in direct mode.  Looks like I may have to fish around for an updated version or another program. I did notice after there a login text box so I guess when it does work it provides the login code.  But yeah the problem is the program I think.

this option works fine i think you have a hardware, system, driver or laptop problem

Does your version of program have China at the top? And what version are you running?

I checked your bin file, its no good.  Should be 2K and within a hex editor you should see the VIN and AUZ... numbers in clear ASCII text.

I am currently working on the exact same problem and just posted a lengthy post on what I have tried.  Bottom line is I do not think I have cracked this nut.  See my post here on what I have tried. 
http://nefariousmotorsports.com/forum/index.php?topic=20166.0title= (http://nefariousmotorsports.com/forum/index.php?topic=20166.0title=)

Note all this is specific to clusters made by Magneti Marelli, ie "M73" clusters.

kw1281test_0.71-beta_Win10 - can read the cluster eeprom correctly
eepromtool_04.exe - decodes the eeprom.bin file nicely but in my case reports 00000 for the SKC so pretty sure it has it all decoding correct except for the SKC

numerous sources report that SKC is the 2 bytes preceding the "AUZ..." immo serial number.  You can read the .bin file in any good hex editor and get the 2 bytes easily enough.  Note they are NOT swapped ie they are big endian.  The VDO and RB manufactured clusters the SKC bytes are stored little endian and are swapped.  I have yet to prove this works however.  So far I have not been able to login to my cluster using what I believe to the the correct SKC and VCDS-Lite.  I have used similar tools to pull SKC from the ECU and login to that.

it doesn't matter which 1.4 or 2.5 both work

my friend, would you put your file here, then I will check if you have this pin correct

Thanks for the information.  And link to your post. Would you be able to send these programs to me? Or the links?  So I can try them? 
Also I could be wrong in this, but if you do the key programming routine your cluster might program at the same time.  You need both keys one in the door.  And I forget the rest.  I think it’s in the owners manual. Not saying it will work, but worth a try.
Did you also try to load the bin file to vag k commander?  After you uploaded from beta win10?  I have a funny feeling vag k might work if it looked at the correct hex addresses.

Attached is the cluster .bin file.  I get 03997 as the SKC but its not working using unregistered VCDS-Lite.

elektronik13, I noticed you provided 2 SKCs here
http://nefariousmotorsports.com/forum/index.php?topic=19168.0 (http://nefariousmotorsports.com/forum/index.php?topic=19168.0)
for the same model cluster.  I checked the .bin files and the SKCs and in one case they matched but in the other the SKC was off on one digit.  Wondering if that was a typo or deliberate?

I am far from the point of teaching immo about the keys.  Putting new cluster in requires cluster adaptation which requires old and new SKC codes.  I have old ECU SKC but stuck getting new cluster SKC.

kw1281test_0.71-beta_Win10 - https://github.com/gmenounos/kw1281test/releases/tag/v0.71-beta (https://github.com/gmenounos/kw1281test/releases/tag/v0.71-beta)
eepromtool_04.exe -  http://nefariousmotorsports.com/forum/index.php?topic=6134.0 (http://nefariousmotorsports.com/forum/index.php?topic=6134.0)

Thanks!  Im doing something wrong hmmmm

Ignore this one the command I found they saying is close to beetle kw1281test.exe COM1 9600 17 DumpMem 14336 2048

When I try to dumpmarellimem it sends a 06x and times out waiting for acknowledgment.  Did you get that error?

Try 10400 baud rate and lmk how it goes.  My ECU is 10400 and while my M73 D14 cluster talks 9600 I believe older clusters may talk 10400.  Not that anyone cares but 10400 is a virtually unheard of baud rate and its incomprehensible to me that it was chosen for any reason other than to make the tools and access to these clusters as proprietary as possible.  "Right to repair" anyone?

I also noticed the "untested cluster version" warning which I have not seen before.  I ran kw1281test on a M73 D14 cluster without that warning and pretty sure the "D14" is a later version than yours.  My M73 D14 has a 2015 date of manufacture and my original with a 2002 date of manufacture is a M73 D01.

They should just have it wireless/bluetooth network.  All this OBD adapters is nothing but a waste of plastic.  And it seems like the only people making money from it is China lol. 
I have the cluster hooked directly on the bench using K-Line. it didn't like 10400 baud.  From the literature I've read in the past I think it's meant to be autobaud. Meaning what ever the program sends it with the 5 bit baud initialize, it should respond with the specified baud from the pc program itself.  I may need to read that again. But who does 5 bits per second also?  That's so dumb, but then again if you're going to fool emissions then you probably don't want anyone accessing your secret smoke you to death machine.   I tried two clusters, my own one 2003 and a 2002 cluster.  Maybe the ECU needs to be in circuit is the difference, I guess I could put an ECU on the K-Line.   The read eeprom thing looked like it worked, but it's all zeros.  This may take a while to figure out. I'm probably going to end up pulling the code from the ECU.  All I want is the code right now, but after thinking about how my cluster may just die at some point I may need to revisit this.   Are you able to read what numbers are on the chips in yours?  There is two motorollas in mine, but they are not real clear.

Some software will autobaud but kw1281test requires a baud rate on the command line and that's what it uses and I did verify that it matters.  10400 for ECU and 9600 for my 2015 cluster.  Do not know what would work for my 2001 cluster since its very dead.

pretty sure the ECU and the Cluster have the same SKC so does not matter which one you get it from.  The memory dump should not be all 0's.  Look at my other posts where I attached my dump and you'll see what a good dump looks like. 

Both my clusters have the Motorola chips.  The larger chip is laser marked with what looks like a firmware version and it is different between my 2001 and 2015 clusters.

I had no problem reading the eeprom on both the ECU and the cluster using the kw1281test tool and this cable https://www.amazon.com/dp/B017IG81SQ?psc=1&ref=ppx_yo2_dt_b_product_details (https://www.amazon.com/dp/B017IG81SQ?psc=1&ref=ppx_yo2_dt_b_product_details).  eepromtool_04.exe pulled the correct SKC from my ECU eeprom dump but not from my cluster eeprom dump.  I did it in the car, not on the bench. 

pin 03997 is ok 100%

DId you have to ground the Eeprom pin on your ECU to pull the SKC?  I had a program before that required to do that.  So that's why I don't want to pull it from the ECU.  It's a bit messy having to pull the ECU apart.