MAF Calibration Query / Log Analysis

Background:

I'm starting fresh on my tuning after learning some more and have a query about my MAF calibration logging. This is 'on road' tuning and two pulls were done in 3rd gear. Both pulls exhibited the same shape / pattern with respect to the measured AFR's.

This is a Nissan VQ37 with the data coming from UpRev and the OEM wideband sensors. The fuel compensation tables were 'zeroed out' all to 100% (not relevant in open loop I'm pretty sure), the long term fuel trims reset and the target AFR tables set to 12.80 everywhere but idle.

I've created custom intake pipes but still use the OEM MAF sensors in their standard housings. I'm looking to validate the MAF tables are suitable to continue tuning on.

My Questions:

- Essentially I'm looking for guidance as to how much AFR variation against target is too much ? What is the threshold to tweaking the MAF tables and what an acceptable amount of variation ?

- How close to the start of the acceleration should be disregarded as 'unsettled' data ? The largest variation I have seems to be at the start of the pull.

You can never be too close, but an error of .1 - .2 AFR is probably OK. What is missing from your graph is the MAF frequency / voltage. That is what your MAF table probably uses an in input axis value, and it what you will want to know.

As for the settle time, I suggest 0.5 seconds after going full throttle should eliminate any effects from transient fueling.

Or use this technique: go to full throttle, but control the speed with the brakes (I do this at the start of a rollling-start race), when you release the brakes, the car will take off immediately with no transient fueling issues and you can use the data straight away. You can't do this for more than a few seconds, but it does work.

Apologies, the MAF tables are voltage vs % flow from memory ... the MAF voltages are plotted in the lower of the 4 graphs in my screenshot also.

Looks like I will want to tweak the lower and the top ends slightly to try to get things a little closer than they are currently.

Thanks for the guidance and tip re holding it back on the brake slightly, I shall try this at my next validation :)

I missed the units for the MAF A/F channels were in volts. That should tell you what table entries need to be modified. I bet you could get that just about perfect in 3 or 4 passes.

When you are finished, please post another graph showing your final results.