A call for data...

My quest for lazy might be at an end (or as close to an end as I'm gonna ever get).

Today I stumbled across something that could prove to be rather time saving; I realized I could cannibalize RFinkle's polynomial MAFCAL smoother thing he made for me, and generate an entire (probably perfect) MAFCAL from just three (well, four) data points.

That's right kids, for the low, low price of a hot idle, steady state cruise, and WOT log, you to can generate a MAFCAL for your car by simply entering three separate voltages and three separate (corrected) massflow values.

You're probably wondering what this looks like.

Here's the input:

The low voltage cutoff is for big intakes that throw the ole crank angle code on cold starts. As pictured, any voltage below 0.9 will have a massflow value of, well, 0.

As for the four volts / massflow entries, the first one you probably won't ever need to edit as it's going to be known good 100% of the time. As for the others, well, the neat thing is that the voltages don't have to appear on your actual MAFCAL for them to be useful in generating the calibration. This alone is super, duper handy.

How does it compare? Have a look:


You'll likely have to zoom in to see it clearly, but Calibration 1 is what I currently have on the car, and Calibration 2 is the new calculated curve one.

Some notes on the comparison:
My logs show me running a bit rich in the 3.8v range (WOT), as well as fuckin hella rich in the 1.25v range (hot idle), and 2.15v ish range (light cruise) is only slightly less rich. The above graph seems to pan out on those corrections, and the numbers entered on the input side are exact in the curve (as they should be).

Unfortunately, since I'm not logging actual massflow, I've had to guess on correct values, but I'm pretty confident this will be closer than ever.

So, what can be done with this?
Well, for starters, you can plug in the voltages and massflow amounts into the inputs and generate a MAFCAL for any year car instantly, or even all years at once (more on that in a minute).


So, what do I need from yall?
Well, data for starters.

1. Year/model of car
2. Tuning device used
3. Voltage curve from the MAFCAL portion from your tuning software
4. Voltage and corrected massflow value for hot idle
5. Voltage and corrected massflow value for cruise
6. Voltage and corrected massflow value for WOT

NOTES:
It will be significantly easier for you if you have open loop trimming disabled (if supported) and you're also logging massflow.
VT People: Try to log as little as possible for the best resolution / response time for these logs.

Y'all provide me with this information, I'll poop out a MAFCAL that you can then compare to your existing one and if it looks good enough, test it out.


All goes well, I'll do some spit-shine on the sheet and release it for general consumption (ODF format so anyone can use it on any platform).

As always, questions, comments, concerns, and (constructive) criticisms are always welcome.

 

So, first try had it lean in the 3v range; looks like more data points are better (probably why this thing was being used for smoothing the entire curve vs generating one).

That said, I added another data point and populated it (two around the 3v range) and the curve is looking better:


I'm thinking 1 data point per voltage, closer to the even volt the better, so if you're going to provide data, try to match that instead.

 

I love this Enki science shit. Following. Will try and get some data to help out but I just started a new tune with Ryan@PD and I'm not sure we've fully ironed out mafcal yet, so idk if you'd want my current figures or not.

 

Doesn't really matter if the calibration is finished or not; it's basically just to see how close this method gets to whatever normal method the protuners use.

 

Hey so this needs to be done with an existing curve, cuz if ur just starting out with a new MAF housing WOT logs are a nono til u get it in the ballpark, right?

 

I suppose I could set up a calibration generator as well based off of stock intake calibration, size and new intake size.

If you could get me the values for 1v, 2v, 3v, and 4v from the stock calibration, the year of the car and the exact ID size of the new intake I can try to generate something that should be good enough for at least cruising.