AFR Air Or Oxygen

The calculations of the pulse width that is calulated in this module is relative to the air mass, but If the oxidizer component of the air is the oxygen (fuel burns when mix with oxygen), the AFR wanted, is relative to the air introduced into the cylinder,or to the oxygen introduced?

It is based on air mass assuming that oxygen content inthe air is about 21 percent at normal conditions. That is why we are using WBO2 sensor to adjust for any changes. For instance I was raicing once in a few hours after heavy rain and that morning it was very hot . My actual AFR had chaged as much as 1 point(!) rwacher because of very high humidity which ECU had no means of compensating for so I had to do it manually... Steam displaced some air in combustion chamber and MAF could not compensate for that.

On the other hand when i changed my stock ECU to aftermarket standalone one it had an option of closed loop operation even under the boost so air/fuel mixture could be adjusted automatically and brought back to the target AFR after a few runs.

As Georg* said, it's the relative mass of fuel to air, but some atmosperic conditions can affect the density of the air and so a correction mechanism is required.

*BTW, do you prefer Georg or Shota - assuming the latter is also your name and not the dubious description?

My mame is Shota) georg is short for georgian, my granddad was from small country called Georgia)

OK if I use Shota then?

Absolutely)

I understand, AFR theoretical calculations are based on air, but for better precision, the ideal would be to be able to measure the oxygen that the engine consumes, but it is a very complicated thing, since air is not a perfect gas, i'm right?

Your wideband O2 sensor is measuring residual oxygen content in exhaust gases and translates it backwards to air/fuel ratio assuming it is 21 percent of oxygen in the air.

Yes, i mean, install a oximeter sensor at the intake to measure the oxygen concentration of the introduced air.

From OEM point of view it is not worth it as they would still have to have at least one 02 sensor in the exhaust for emission control purposes. If you have one sensor that already measures 02 why would you want to introduce the second one? I think they figured it out that the accuracy of converting O2 sensor output backwards is good enough. Moreover O2 sensors have to be quite hot in order to operate properly so not only you introduce a point of restriction in the intake but also a source of significant heat which is not what you normally do ( unless you are Smokey Yunick)...

As Shota said, there isn't any point in doing so as the engine is already self calibrating by analysing the exhaust gas.

However, it does reflect on something that I don't ever recall being mentioned. While 'free air' oxygen percentage is ~21%, in heavy race traffic some of the air ingested by the following cars will already have been through some of the vehicles ahead, both heating it and depleting the oxygen - it will be by a small amount as it blends with the air passing around the vehicles, but is it actually of any significance?

I think it would be making sense for them to be operating in closed loop mode to compensate for any afr changes but i wouldn't think the difference is that significant...

There are devices for measuring the percentage of oxygen in gases, in medicine or in diving, to check the oxygen concentration that this gas contains, if you install a sensor of these in the intake, before performing the gasoline injection, you can know how much oxygen concentration your air has in the cylinders to make a more accurate calculation of the injection time before combustion occurs.

But they already measure oxygen content and they cannot change the situation because of emission control requirements. If there was no such requirement then they could be thinking about other options. There's no sense to have two devices serving the same purpose from cost point of view...

Yes but you speak in terms of original engines from the manofacture, and i speak about high performance engines, where the emission control don't exit.

I see what you're saying but again you need to measure oxygen in the exhaust anyway. And the reason why is that although with your device you will know how much oxygen is getting past it in the intake you will have no data how much of it actually gets inside the combustion chamber. What if there is a leak somewhere, how do you compensate for it? For instance i have completed leak test on Chevy Camaro recently and i was very surprised to find the air leak from the engine front cover right side! I would never think it would be leaking there. And that air was already counted for by MAF and ECU in terms of fuel delivery. Having O2 sensor placed past combustion chamber allows ECU to see the difference between air coming to the engine and oxygen remaining in exhaust gases and compensate the difference by changing the fuel trims. That would not be possible is there was only one sensor measuring oxygen in the intake thus actual AFR could differ from comanded AFR without any means of correcting it.

At the intake you have usually, air temp sensor, map sensor, or maf sensor to evaluate the density and quantity of air that you introduce into the combustion chamber, If you add an o2 sensor, you can know your o2 concentration in the combustion chamber due to humidity variations, which cannot be seen by a temperature sensor, or map and keep the o2 sensor in the exhaust to finish adjusting or see other faults. It is one more sensor that would help to improve the calculation of the injection pulse.

))) but there is no sense of doing that since you are measuring oxygen in exhaust anyway)) you can always adjust injection pulse based on that actual data and you do not need anything else to duplicate))

But i do understand your logic - you want to make sure that ECU knows exactly how much oxygen present at the intake to calculate precised amout of fuel for injection in first place rather than find out how much of oxygen actually got in combustion chamber and deal with errors afterwards. Well, apparently having O2 sensor in exhaust provides quite accurate resolution and feedback so it outweighs the need of additional sensor in the intake.

Fundamentally if you added an industrial oxygen sensor it would have to be in addition to MAF and MAP and would be probably be used as a single line calibrateable as a fuel multiplier base on O2 %. i.e. if O2 was 25% then the fuel multiplier would be 1.19 or if O2 was 18% then the multiplier would be 0.85. If this was the case however, we would have other rather more pressing matters to deal with than if our fuelling was correct.

As mentioned above any minor changes in atmospheric conditions are compensated by the lambda sensor and since our atmosphere is somewhat stable monitoring the oxygen content pre-combustion is not required; they are used in lab test engines to validate combustion modelling and study thermodynamics etc.

From experience it would be more beneficial to compensate for relative humidity as this have a larger impact on combustion and can lower the effective knock limit etc.