fuel trim variation between cylinder banks

[vtraudt]

I consistently see quite different readings for long term fuel trim (VCDS block 032) between the 2 cylinder banks on one of my 2.7T cars.
I have NOT seen anything like this on all the other cars (also tuned).
Initially thought coming from using one new one older Bosch O2 sensor. But replaced older with exact same one.

032/3: Adaption Partial Bank 1 Sensor 1 +7.8%
032/4: Adaption Partial Bank 2 Sensor 1 +0.8%

Any practical suggestions (least effort/cost) to determine if one the injectors might be causing the problem?

I am running Bosch EV14 750cc (and yes, I know now that those are not original Bosch, but rather modified; next time, I will get original Bosch (they make 627cc).
I have one extra injector (at some point earlier, i though one injector was bad, so bought an additional one from same source just in case).

All I can think of is:
a) take all injectors out and send in for flow testing (all 7).
b) buy new Bosch ($$$)
c) swap cyl 1-4, log, swap 2-5, log, swap 3-6, log. One of the logs should 'move' the imbalance; swap the spare injector into one of the respective positions. Log. If no change, swap in other position.

Other suggestions?

[phit]

Don't think your reading mvb 32 correctly. Display field 1 bank 1 fuel trim idle short term field 2 bank 1 fuel trim cruise long term. Field 3 bank 2 fuel trim idle short term bank 4 fuel trim cruise long term.
This should help.

[vtraudt]

Misprinted above. Field 2 and 4 are long term fuel trims for the 2 banks. 032/2=+7.8% 032/4=+0.8% ==> quite different between 2 banks.

[Robsa6]

I notice this too... I have a 5deg variation between the two banks with my camshaft phase and thought correcting that might help a little and from there I was thinking camshaft wear or valvesprings maybe. I'm interested to hear if it follows your injector swap

[mbagge01]

I also have about a 6 degree difference between banks. I am stage 3 on my 2.7t. Car runs fine and favors positive fuel trims which are good but hoping someone could explain the large variation

[mbagge01]

What I mean to saw above is it favors negative fuel trims (richer fuel mix)

[vtraudt]

I suspected off reading O2 sensor (upstream; downstream are removed and tuned out, no cat); when turbo died, exhaust was flooded with oil.
First cross switched the 2 sensors, and readings seemed to follow (until ECU gets confused trying to trim fuel).
Replaced one sensor randomly with a good used one on hand.
Now fuel trims are the same for both banks.