I am attempting to setup a Lamba sensor (O2 sensor) on a biomass boiler flue to continually monitor fuel:air mix ratio (via residual O2 left in the flue products).
I am trying to work out the best cost effective route to achieve this using an Arduino.
I was thinking of using a Zirconia based sensor such as the one below. They are relatively cheap and readily available however I need to work out how best to interface this.
In an ideal world I would be able to get a 0-5V linear signal from the sensor that can feed directly into the Analogue inputs of the Arduino but from my research so far it does not seem to be that simple.
I have found the following interface module but it is prohibitively expensive for the home project I plan to use it for so am looking for more cost effective solutions.
The transfer characteristics of lambda probes is highly non-linear. Therefore you you won't get a linear signal 0V-5V but a non-linear one in between 0V and 0.8V. You can use a OpAmp to amplify this to 0V to 5V and after measuring the voltage you need to calculate the oxygen concentration from the measured voltage. You either need a very accurate description of the test gas or you need to measure the characteristics using test gases of diferent oxygen concentrations. I am not sure if there is a standard for the response curve of lambda probes. The board you linked does this linearization for you. Most probably there is a standard for this characteristics. But you need to find it. If you find it it would be nice to share your information.
Thanks rw950431, unfortunately this sensor does not have the high temperature characteristics required to be used in a flue / exhaust.
Dande80 I suspected this to be the case. There is also an embedded heater element within standard lambda probes that needs activating as well. In some cases there is also an on board pressure sensor to do barometric pressure correction too. Over lay this with the calibration side of things at it starts to get quite complicated hence I would prefer to use an interface board that looks after that headache for me and provides a nice sanitised linear signal! But I don't want to have to pay hundreds of pounds for it....the potential benefits just would not pay back!
If I wasn't bothered about the actual value and just needed to use it as a reference signal this ie. control air flow via a PID loop to hit a signal voltage of XXmV which I know from using a calibrated reference then I may be able to get around the elaborate interface circuitry....just a thought.