Do widebands measure AFR or Lambda?

[dan7055]

I know most widebands display an AFR reading. But are they actually measuring the air to fuel ratio or is it measuring a lambda value then calculating the AFR based on the stoichiometric point of gasoline?

[cliffyk]

Somewhat overly simplified, they measure Lambda and the display mechanisms scale the AFR displayed value about the user specified stoichiometric ratio air/fuel--to make the humans viewing the display happy.

An practical example of how this impacts displayed values is that the stoichiometric AFR for gasoline is approximately 14.7:1 (actually closer to 14.64:1), however for E10 (90% gasoline/10% ethanol ) it's 14.0 to 14.1:1.

If a WB system like the LC-1 is programmed to believe that Lambda 1.0 = 14.7:1 AFR, and the engine were operating on E10 fuel, then it will display Lambda 1.0 as 14.7:1.

In fact you could program the LC-1 to believe that Lambda 1.0 = 13.0:1 AFR, then run it on straight gasoline, and at Lambda 1.0 it would display 13.0:1 AFR.

[dan7055]

Originally Posted by cliffyk

Somewhat overly simplified, they measure Lambda and the display mechanisms scale the AFR displayed value about the user specified stoichiometric ratio air/fuel--to make the humans viewing the display happy.

An practical example of how this impacts displayed values is that the stoichiometric AFR for gasoline is approximately 14.7:1 (actually closer to 14.64:1), however for E10 (90% gasoline/10% ethanol ) it's 14.0 to 14.1:1.

If a WB system like the LC-1 is programmed to believe that Lambda 1.0 = 14.7:1 AFR, and the engine were operating on E10 fuel, then it will display Lambda 1.0 as 14.7:1.

In fact you could program the LC-1 to believe that Lambda 1.0 = 13.0:1 AFR, then run it on straight gasoline, and at Lambda 1.0 it would display 13.0:1 AFR.

Thanks

[Dr.Mike]

What they are actually measuring is the partial pressure of free Oxygen in the exhaust stream. That's why the sensor is called an Oxygen sensor.

When there is 0psi of free Oxygen in the stream, then the mixture is assumed to be sociometric.

In "free air" there us about 3psi of Oxygen ( 14.7psi x 20.9% ) .

The Oxygen sensor (narrowband) puts out about 450mV when it sees a 3psi difference in the partial pressure of Oxygen between the exhaust and the outside air. The voltage is created by the Oxygen ions moving from the high concentration side to the low concentration side on the sensor.

And. Yes. You can effectively have a negative partial pressure of oxygen in the exhaust, if there is unburnt fuel in the stream, consuming whatever Oxygen moves across the sensor as it burns.

Soo...

The sensor measures the difference in the partial pressure of Oxygen between the exhaust and the outside air. From that, it knows where the lambda=1.000 point is ( exactly 0psi free Oxygen ).

The sensor has no idea what type of fuel is being burnt. So, a multiplier is applied to the output so that the display will be consistent. 14.7 for gasoline, etc.

[graham]

Thanks for the informative information on how the O2 sensors determine lambda and how it is converted to AFR.

On our early Mercedes, the setup is based on %CO in the exhaust. My question relates to use of current fuels that contain ethanol - usually E5 but can be up to E10. In winter fuels, there could also other oxygenates.

Original exhaust spec for car (at idle) was 0.5% - 2.0% CO. (In 1972 presumably for regular gasoline with stoic. of 14.7). They did not give numbers for cruise or under load.

I would like to know how to translate the original Mercedes %CO spec to the equivalent for E5/10 fuels. And then what AFR as displayed on wideband meter (based on 14.7 stoic.) would be equivalent to that %CO.

One of the charts I have found shows 0.5% CO for lambda=1 and the other shows an AFR of 14.28 for 0.5% CO, presumably both based on straight gasoline. How SHOULD I do the conversion from lambda (or AFR)? (As a Chem Eng, I should know these things, but memory is rusty!)

[cliffyk]

Originally Posted by graham

Thanks for the informative information on how the O2 sensors determine lambda and how it is converted to AFR.

On our early Mercedes, the setup is based on %CO in the exhaust. My question relates to use of current fuels that contain ethanol - usually E5 but can be up to E10. In winter fuels, there could also other oxygenates.

Original exhaust spec for car (at idle) was 0.5% - 2.0% CO. (In 1972 presumably for regular gasoline with stoic. of 14.7). They did not give numbers for cruise or under load.

I would like to know how to translate the original Mercedes %CO spec to the equivalent for E5/10 fuels. And then what AFR as displayed on wideband meter (based on 14.7 stoic.) would be equivalent to that %CO.

One of the charts I have found shows 0.5% CO for lambda=1 and the other shows an AFR of 14.28 for 0.5% CO, presumably both based on straight gasoline. How SHOULD I do the conversion from lambda (or AFR)? (As a Chem Eng, I should know these things, but memory is rusty!)

I am an ME, so my chemistry is far beyond rusty. However my initial reaction (no pun intended) is that while I am certain there exists some relationship between Lambda and %CO, I suspect that trying to quantify the latter using a Lambda measurment system and application of a scaling factor, or other conversion, is akin to measuring apples with an "orangeometer".

An exhaust gas analyzer would get you right to this number directly...

[graham]

Originally Posted by cliffyk

I am an ME, so my chemistry is far beyond rusty. However my initial reaction (no pun intended) is that while I am certain there exists some relationship between Lambda and %CO, I suspect that trying to quantify the latter using a Lambda measurment system and application of a scaling factor, or other conversion, is akin to measuring apples with an "orangeometer".

An exhaust gas analyzer would get you right to this number directly...

From what I have read, it is possible to calculate %CO from Lambda. The best info I found was on this website:

http://www.megamanual.com/PWC/index.htm

Part way down the page there is even a computer program that will do the calculations.

I now just need to figure out how to properly enter a typical E5 or E10 gasoline. When I have tried, it shows very little effect as compared with regular gas. But elsewhere I have found charts that indicate that E10 would reduce the CO in exhaust by 50%. So something is wrong.