AVCS, End of Injector Timing, VE, and Fueling Targets - Subaru STI

Hey Guys,

So I finally got around to messing around with AVCS/VVT on my Subaru STI and based on what Im seeing in the logs Its left me with a few different questions.

1.) Determining the ideal AVCS and overlap targets when road tuning.

So far, my approach was to:

- Pull timing everywhere for safety when finding best AVCS settings

- Set to forced open loop fueling for safety as well

- Try out different settings and log the results. Mainly used calculated VE (Note: Im still MAF based though), and how quickly the turbo spooled up.

With this process I dont feel overly certain that my current settings are ideal but its sort of all I have to go off of without a dyno.

My current AVCS Intake/Exhaust table is attached. Just wondering if this seems reasonable based on anyone else's experience? From what I understand it would be considered aggressive on stock cams (~ peak 65 degrees overlap) for an EJ257, but from my logs... these settings are what yielded the highest calculated VE.

2.) Overlap and EOI

Given I was a bit concerned that the overlap might be "too much" for stock cams, I was looking at what "too much" means and it seemed like it mainly centered around creating a scenario where you have a lot of overlap and EMAP > IMAP so you are just getting exhaust reversion... OR you have so much overlap that you are blowing air and fuel straight out into the exhaust before its combusted. To (partially?) mitigate this I was looking at the settings for injector timing with respect to crank angle which I've adjusted somewhat to inject earlier when overlap is low, and inject later when overlap is high.

The idea being that if the injector fires later into the cycle... even if overlap is high(er)... You wouldn't have as much of an issue with the fuel just getting blown out the exhaust port? Can anyone confirm this approach is valid (see attached EOI table).

3.) AVCS Impact on Fueling

Since I am still MAF based, I didn't really expect to have to make any fueling changes to accommodate increased VE from the AVCS changes I made. This has seemed to be the case for most of the time I've been working on it, however most of that time I was also hard forcing open loop fueling. Once I re-enabled closed loop fueling, I've noticed some odd behavior with the fueling. From one of my most recent logs, it looks like target fueling and fueling as measured by my wideband are tracking as you would expect... But then right where overlap and VE peak, and then begin to taper back down... the target fueling starts tracking leaner back towards stoich for a few hundred RPM even though I'm in open loop fueling and those fuel targets would not match up with my open loop fueling table. Its almost like the increased VE or overlap has triggered some weird logic in the final fueling calculation that made it apply some significantly lean fueling targets under load. Has anyone ever encountered this before or know how to resolve it? (log and screen cap attached)

Edit: This fueling issue just ended up being some CL/OL delay behaviour.

Any input appreciated!

Note: Turbo is not an oem VF48, full turbo back exhaust, headers, free flowing intake, etc. So EMAP probably doesn't spike as fast as it would on a stock setup.