A/F gauge reading different than MS3.
 

Hey guys,

I just got my MS3 back from Greece today, and installed it in the car. I also have a new laptop to use for tuning with a larger screen. While taking a test run down the road I noticed that the A/F ratio on my laptop screen is somewhere between half a point and a full point leaner than the reading on the A/F gauge on my A pillar.

It may have been this way all along, and the A/F on the laptop was spot on for what I would expect out of my tune... but the tune was done mostly with the VEanylizer live so it would match what the MS3 thinks the A/F ratio is.

So, how other than going to a dyno with an A/F ratio gauge do I check to see which one is accurate? Keeping in mind that I don't trust my local dyno's A/F reading all that much. At this point I don't know if my A/F ratios are spot on, or pig rich.

The A/F ratio voltage input is done as described in the instructions (what wire goes where)... but I don't know if there are calibrations internal to the MS3 pertaining to the A/F ratio input that I should have adjusted when I first installed the MS3 and interfaced it with my Innovate MTX-L.

Thanks,

Keith

MS3 has ability for you to get it to read same as gauge. This has been addressed. You put in custom values instead of selecting a standard WB.

The big question is how to know which is right. I choose my gauge.

On the LC-2, one can also independently adjust the outputs, and make them match that way. I did it in the MS as it was already attached to the computer.

https://www.miataturbo.net/megasquir...-saving-78244/

.23v = 7.35
5v = 23.4

Ha Ha. Now, Twodoor, you have to decide if Brain is spoon feeding you good info, or BS. :giggle:

Also, if you don't trust your tuner, you need a new tuner.

Those are the correct values, or will get it very close. I have ran those same values on my car for years.

Got you. With mine from the link I posted, the equation is AFR = 3* volts + 7.2. Scott's is AFR = 3.36* volts +7.35. At 2 volts from the wide band, my MS would read 13.2, Scott's would read 13.3. Pretty close.

I didn't realize the corrections would be consistent between installations.

The best way to actually solve this issue is to get one of our CAN wideband modules: These connect to your wideband (any Innovate, AEM or Zeitronix) through the wideband's serial port, and then transmit the AFR data to the MS through the CAN bus, thereby eliminating any offsets or noise in the signal.

See a testimonial here: https://www.miataturbo.net/ms-labs-m...terface-78449/

because the voltage offset we all introduce is done in the same way.

Is there one of the data log inputs that shows the raw voltage from the wide band input? I just noticed that the max my A/F shows on my data logs is 19.5:1 when full lean decelerating... but the full lean on the MTX-L is 22.4:1

I am thinking of just doing a manual input of the calibration numbers into the MS3 and seeing if that corrects the problem if I can get the numbers to "stick" after seeing DNMakinsons problems.

In the mean time I have been putting weird offset values into my A/F target map in order to use VEanylizer live.

DNMakinson: did you ever figure out how to get a calibration for the wide band to "stick" in the MS?

Thanks,

Keith

I'm not the one who had issues with it staying. I was just giving input to the guy that was, on how he could verify. It should be straight-forward. Put in Brain's numbers and you should be good. Just make sure that you are on line and tell it to "write to the controller", the custom numbers.

I did some logging and then came up with my inputs based on how far my gauge was reading off from the MS3. The "official" numbers are 0V = 7.35:1 and 5V = 22.39:1 I am using 0V = 6.5:1 and 5V = 21.5:1 and as far as I can tell trying to look at my MS dashboard, and the MTX-L gauge at the same time they are virtually identical now. I'm still going to use the MS Labs canbus box though when it arrives :)

Keith

You went with more correction than I did. My MS still reads a bit lean above 15:1, so I may move in that same direction.

Will first see what it takes ($$, I'm cheap) to get Reverant Serial input.

Glad it's working for you.

Im using a Mslabs Ms3 just like you, using a Mtxl just like you. I have my mtxl wired to the db37 connector on my MS3. If I use the default 0=7.35 5=22.39 tunerstudio does not match the mtxl. i had to adjust in TS like you did to get them matched