[cre]

I haven't the patience for everything in this build that I disagree with or that could be handled much more practically and provide greater reliability and response... Perhaps someone else will step in.

Vf is indicative of many things and there's actually a LOT more in that signal than just fuel trim feedback. Hook it up to an oscilloscope if you have one and check it out. It's also not a reliable indication of AFR as a narrowband isn't really capable of providing reliable information about your AFR. The narrowband is essentially disregarded in open loop operation and Vf represents changes made for other reasons... The only thing you know is that it's making changes that it 'thinks' need to be made; This could be because of detected knock, an OX signal that isn't within spec (for a narrowband signal that means it has to be at the correct frequency with relation to injector and spark timing and within 400mV and 500mV), or due to learned information from driving conditions four weeks ago or any other of three dozen reasons. Vf is unreliable except for situations where you're trying to get one of the more advanced piggyback EMS dialed in as close as possible so they'll live together in piece and it's still not what I'd call trustworthy. (I just re-read your response concerning tuning with Vf) I'd really not rely on it while the TCCS is in it's initial learning mode... it doesn't even know what it's going to do next yet in this state.

In the stock location the narrowband should not cool at idle to such a degree as to skew readings notably. If it is cooling to that degree you may want to invest in insulating your header as that heat's probably ending up in the engine bay and the gasses in the exhaust piping are ending up cooler, denser and thus slower moving than is preferable.

Adjusting the tension spring in the AFM causes a non-linear shift in the measurement of airflow. The vane bypass creates a linear shift, IIRC. In both cases you adjust the entire scale and can not just tweak idle or whatever. Closed loop operation helps as the ECU corrects the shift you've caused, but it's limited in response time, how much it can correct and so on. But again I recommend you do it right and get a fuel controller. You may also screw with your tip-in/out response as the plate is prone to bouncing now due to the unanticipated loss of tension (only really an issue if you make a significant adjustment). I'll grant that one or two teeth either way isn't going to create a significant shift but it isn't without possible side-effects.

Opening the crank case to atmosphere allows more pressure to build within, water vapor and corrosive hydrocarbons to accumulate in your oil (reduced oil and bearing life), reduces the efficiency of your oil seals and piston rings, etc... Fair trade off just to HIDE another problem? FIX the real problem.

Anyway, good luck with the project and have fun with it!