Pages: [1]
Author Topic: Cluster swap 2002 (8L0920932H) to 2000 (8L0920932N) oil level/temperature sensor  (Read 4612 times)
Awaken
Full Member
***

Karma: +3/-3
Offline Offline

Posts: 84



Hi guys,

I have a special case, all files are attached. I bought a new (actually older in age) cluster file for my 2002 BAM S3, because it has this colormfa display installed, which is the coolest thing (http://colormfa.ru/). I didn't want to modify my original one and this is why I bought the whole other cluster. The thing is that my cluster is 8L0920932H from a 2002 car with a oil level/temperature sensor and the new one is 8L0920932N from what I can guess is a 2000, which appears not to have a definition of this sensor. After I got the car starting and all, naturally I got the oil sensor error. On top of that I could not recode this cluster to tell it there's no oil sensor, since it is missing some channels in VCDS, but that's not really the point.

The first thing I tried is to write the whole 8L0920932H bin file onto the 8L0920932N cluster, but then I got the DEF error, I guess some checksums did not match or something and given I have no clue how to fix this I gave this up... 

Then after much trial and error I was able to guess the part of the code inside the 8L0920932H bin file which is responsible for the oil sensor (see oil_sensor_address_8L0920932H.jpg), noticed that this area of the code in the new one is all empty (FF) (see oil_sensor_address_8L0920932N.jpg) and to my surprise, by simply replacing that portion of the code everything worked (see oil_sensor_address_modified_8L0920932N.jpg).

Well it almost worked... I no longer get the error, the oil level measuring seems to work, but the oil temperature is really off, displaying around -50 when it is around 20 and + 120 Celsius, when it is actually 85-90. I tried different sensors to make sure mine was not faulty, same thing. It is not the end of the world because I have an external sensor and monitor my oil temperature from there, but it is really driving me nuts and I want to fix this.

Here is where I lack some knowledge and I hope someone can help:

1. Are there any available definitions to these cluster files where you can check what is what? (Obviously I was not able to find any on the internet, so the answer to this is obvious)
2. What would be a proper way to migrate the whole bin 8L0920932H file to the new cluster 8L0920932N?

Thanks! Hope someone finds this interesting enough to jump in.
Logged
macxxx
Sr. Member
****

Karma: +30/-1
Offline Offline

Posts: 498



Its strange every 920 cluster should work.with oil sensor - i can turn off oil sensor its just one byte to change
Logged
Awaken
Full Member
***

Karma: +3/-3
Offline Offline

Posts: 84



I am looking to keep the oil sensor though, I just need it to read correctly. Any idea how to find out what the highlighted area means byte by byte, a definition or something?
Logged
macxxx
Sr. Member
****

Karma: +30/-1
Offline Offline

Posts: 498



No defs , no ols etc
Logged
Black_A4
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 51



how do you read and write those clusters?

In my case i had to set the cluster to flexible intervall via VCDS. Otherwise it did not worked but this was an octavia cluster.
Logged
Awaken
Full Member
***

Karma: +3/-3
Offline Offline

Posts: 84



I used VAG Commander, but this is a bit crappy, since it was an old version. Now I am using EEPROM_Programmer (attached). Also there's a really great program called DashDump by some Russians, it is really good for setting your dash as new,and some other neat functionalities (also attached).

Anyone any idea why the DEF error on the tacho shows when I write the H file onto the N cluster and how this can be fixed?
Logged
macxxx
Sr. Member
****

Karma: +30/-1
Offline Offline

Posts: 498



Checksum is diffrent , you were lucky you didnt write the 2000 year dump to 2002 because the newer cluster is crypto and the def wouldnt be that easy to repair
Logged
Awaken
Full Member
***

Karma: +3/-3
Offline Offline

Posts: 84



I see. Well my original one was never meant to be in any way modified for any reason, this one including, even if  I didn't know that exactly.

What is weirdest to me is that the original 8L0920932N cluster bin file has values in the mentioned addresses and the one I got was all FF. So I am guessing someone was messing with it before that, maybe around the time they were installing that ColorMFA display. I was also getting some error (forgot exactly what,but I think DEF again) when I was trying to put on the original 8L0920932N file as well... anyway, in case there's a checksum fix somewhere available and I can try some oem file again, I will just leave it be, at least now I am not getting any errors...
Logged
macxxx
Sr. Member
****

Karma: +30/-1
Offline Offline

Posts: 498



Pre 2002 model year clusters have FF in the middle part od dump , thats normal
Logged
sonique
Sr. Member
****

Karma: +24/-12
Offline Offline

Posts: 283



DEFF error need directly write encrypted file

best read another same number cluster desoldered eeprom
Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.018 seconds with 16 queries. (Pretty URLs adds 0s, 0q)