Hey guys, new member here but not new to armature ecu mcfkery.

I have a 2.7t I'm working on, ko4s, 630 injectors @ 4 bar, 83mm hitachi maf, and catless exhaust - completely stock otherwise. I've been having trouble getting his idle fuel trims to not swing completely negative with little success, however part throttle cruising and WOT is almost perfect (fuel trims will vary +/- 5% short term while long term will head towards and stay at 0%).

I'm using injector data from another forum for the particular injectors on this car (0 280 158 117) as a baseline and have MLHFM scaled up 35.6% to account for the change in diameter for the larger housing, again, fueling seems to be almost there except at idle. The car ran very good with this data unmodified, much better than how it was when I started working on it.

If I reduce TVUB, fuel trims at idle will go a little more towards zero, but the car then gets a horrible hiccup/stumble/miss at idle which is especially pronounced at 1500-2500, the car will also buck at low speed off throttle. Curiously, I've had this exact problem before on a completely different motor - a 3.0 ford v6 which was completely stock running a standalone, which would be cured by running a higher offset at the expense of burning my eyes out, I wound up giving up on that and tuning a ford ecu instead...This particular car had small ~350cc injectors which should have had in theory no problems with idling.

So, what could cause this? My guesses are:
-There is some sort of noise coming somewhere for fueling, resulting in injector load varying around, which upon swinging lean causes a miss that can be fixed by increasing offset
-The injectors are simply so large that on gasoline they can't meter fuel in a small enough quantity, and the low injection time isn't enough to open the pintle
-The O2 sensors are lying/defective - kinda wonder this since the car despite being catless really does not stink or smoke at idle whatsoever, though could be from the -25% trims doing their best.
-Similar to point #1, there is a major problem in my tune which is causing the ECU to use one load value which causes the car to run good rev up lets say, which then causes the ECU to read a table such as MLHFM and read a value that then causes it to go lean, and so on

Unfortunately I do not have the car in front of me and am limited in what I can do right now, and the owner currently does not have a wideband or any other better means of confirming fueling 100%.

As another side note, I am working the tune that was on the car from a previous tuner (horrible practice I know), but the car behaves exactly the same when a stock M-box tune has only MLHFM, KRKTE, and TVUB modified for his hardware, so I doubt anything else weird is going on in the file.

id also verify that the injectors are genuine. many many fakes now adays. sets for 40 bucks on ebay make me lol. but on a serious note your dead times are majorly wrong most likely. look at the ev14 thread it has dead times for those injectors

I don't know if it will help anything but when I had my DEKAS installed .
I changed my FKVVS axis to help with misfires at idle.
This is how I had mine set  0.85 0.95 0.98 1.00 1.50 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00.
You probably just have oriental injector issues.

if they say 550 on them they are 100% fake