FORD VE Calibration - alternate method?

In the Ford ECU "Worked Example", the VE calibration values are adjusted based on comparison of calculated vs actual MAP values at a fixed RPM. The demonstration is done using the actual MAP values from the dyno.

Let's say you are going from a naturally aspirated car and installing a power adder like a supercharger. The resulting VE calibration errors will be drastically different with increasing RPM and MAP. In a desire to get a better datalog of errors in the VE table would the following work better?

1) Tap the manifold and Install a 3 bar MAP. Wire correctly to the analog input of HPTuners MPVI,

2) Establish channels for "calculated MAP" and "MAP analog input" with correct units of measure (i.e. kPa),

3) Set up a custom Maths to calculate MAP error ( [MAPcalc - MAP actual]/MAPactual)

4) Set up a Histogram matching the VE table axis (i.e. RPM vs MAP vs MAP error Maths)

5) Datalog this histogram (dyno or street)

6) Correct as appropriate based on datalogged MAP error on all Map Point VE tables.

Generally, set it up so you read actual MAP and datalog it like you would on a GM vehicle.

Thanks for any insight.

Hi Victor, absolutely this technique will work. At the point we filmed this worked example it wasn't possible to add external sensors to the MPVI2 interface, hence we used the dyno. It's a lot messier than using the VCM scanner so if you want to add a sensor then I'd definitely say go for it.

That works well. I did it on my car and it took care of a lot of drivability issues. One key point though.

The MAF value is what starts the calculations and estimates / modeling of the cylair. you absolutely must turn off anticipation, and filtering first, and get really good MAF values to start with. Once your primary input is solid then you can dial in the other logic (after turning it back on). This of course assumes you have solid injector data and your fuel pressure is GTG.

Forget this and you will chase your tail. Ask me how I know...........

Are your cams locked or are you going to adjust each mapped point separately? I found that the approach of setting each mapped point with the same adjustments left a bit to be desired.

That worked example went a long way in making me realize what I had been doing wrong.