I had my Bank 1 EGT sensor die on me the other day, so I decided to take a look into a way to retain ATR and just work off the readings from my one good sensor.  Unfortunately, I didn't find an easy way to do this, so I had to get dirty.

The best way that I found to do this was to just go into GGATS and change all of the references to the input variables from the bad bank (bank 1 in my case) with the input variables from the good bank.

It ended up being 11 references total, I did all of them just to be safe. Five for tcchats_w, five for tccpats_w, and one for taats_w.

Nothing should be changed for ATR, DATS, DATR, or anywhere else. Diagnosis and everything else will just reading the one bank for both banks.

I have attached an XDF with everything defined that is required to switch either bank. All of the ones labeled EGT_B1_* would need to be changed to run off of bank 2 only. The ones labeled EGT_B2_* would need to be changed to run off of bank 1 only. The descriptions explain the 16 bit value that should be entered.

I have also attached a stock M box bin set to run ATR off of bank 2 only as an example and a ME7L ecu file containing GGATS variables.

This modification is being applied to code, so proceed with caution if you choose to attempt this. I have not tested this yet.

Is there anything that I am overlooking here?

Tested this tonight and it worked perfectly.

I attached a short log. Bank one is bad and tats_w was reading a constant 1035* before. Note in the log that tcchats_w and tccpats_w are different between banks, but everything after is exactly the same. This is because bank 1 is reading tcchats2_w and tccpats2_w instead.

Nice, this is exactly what I was doing before I got stymied by my lacking ability to disassemble and find the locations of those variables:
http://nefariousmotorsports.com/forum/index.php?topic=1592.msg17746#msg17746

I just got your PM, and now that you've found the locations of these variables, I'll be doing this myself shortly as well.