Even though I have a wideband in my car, I've been trying to find a reliable formula for calculating the actual afr from just the 02 volts. I googled the topic and found a bunch of formulas out there that do not seem to work at all. So, I started messing around with some numbers in excel. Here is the formula I came up with:


AFR = 17-(02Volts*7)


This formula is very simple, yet seems to be extremely accurate. I checked the numbers from this formula against the readings from my wideband, and they seem to be very close.


Sorry if this formula is old news or if there are better ones out there. I couldn't find anything better so please let me know if there is a more accurate formula to accomplish this.

 

If you know how a narrowband O2 sensor works, then you'd know it's impossible to calculate true A/F from narrowband O2 voltage. It does not have a linear output like a wideband does. O2 voltage will vary from vehicle to vehicle, even at the same A/F. If it was that easy, widebands would be obsolete.

 

Yep, narrowband O2 sensors are only accurate very close to stoich. Anything richer than about 14:1, and the output becomes very temperature sensitive and has low resolution.

 

Yea, no way that will work at all.

I've seen AFR from 10.5 to 11.5 and the narrowband stays at .91 the whole time

 

let me guess, u came up with this formula while driving around part throttle? lol!

 

Might help diagnose bad O2 sensor in closed loop

 

Of course I do not rely on this method...I just noticed that it seemmed extremely close to my wideband readings. That would be pretty nice if there were a way to calculate AFR from 02 volts.

 

Say you have a Target A/F of 12 to 1. Your A/F is 11.3. Simply take 11.3 divided by 12 which is .941. Multiply your current fuel table by .94%. This should get you close.

 

too much voodoo for me. I only trust a wide band for this. its just too dangerous to try to "calculate" a reading from a sensor operating way out its intended range.

 

Unfortunately Narrowband readings are only good for "Ballpark" values.. Nowhere near accurate enough to tune 100% safely.. I have to admit I have done it in the past.. Unfortunately EGT temps affect the accuracy of the sensor which is why they are strongly discouraged from being used.

 

Even though its not 100%, I'll include it in the new XML defs for my logger tool..

 

Someone who doesn't know the difference between a narrowband and a wideband is going to think its the real deal.

 

Yep thats why its under an advanced calculation.. and disabled by default..

 

Jack, maybe you could take a look at the Dsmlink source code and see what they used for a formula for the A/F estimate. It seems to be very accurate.

 

If its a narrowband, its not really possible to be accurate.. the fact that a narrowband's output can vary wildly in voltage with exhaust temp renders it useless for anything other than rough tuning..