CO2 Infrared Sensor

Aside from the MQ-series of sensors, there are a series of sensors with serial number starting MH that measure CO2 concentration using an infrared sensor very similar in operation to the to the GP2Y dust sensor.

The one that was tested for the sensor cocktails is an HM-Z14 infrared sensor that measures carbon dioxide concentration and offers various output interfacing options. The module can be operated via a serial connection, offers a pulse-width modulation output signal and is powered by $5V$. Note that our favorite IoT producer Gravity DFRobot has a module based on the MH-Z14 but that it is expensive relative to competitors. In principle, you cannot go wrong regardless the price given that the mode of operation of such sensors is the same regardless the brand and the sensitivity or precision is given by the sensor being used that is also common amongst all variations.

The datasheet offers a formula on how to calculate the actual value from the PWM signal, which is easy to accomplish in Arduino given that the computations involve just basic algebra. When such modules are mass-produced, sometimes the documentation does not exactly follow such that the calculations have to be amended. In the datasheet linked it can be seen that the formula is defined as:

$$ \begin{eqnarray*} C_{ppm} &=& 2000 * (T_{H} - 2ms) / (T_{H} + T_{L} - 4ms) \end{eqnarray*} $$

where 2000 is a factor that matches the upper bound that the sensor is said to be able to measure. In this case, the range is defined in the datasheet as $0ppm-2000ppm$, hence the 2000 factor used in the formula. However, depending on producer, the range might be different such that the $2000$ factor has to be replaced with other values, the most common value to be found on the Internet being 5000.

To spot whether an adjustment is necessary, simply implement the formula as it is and check the levels. Typically, the outside CO2 concentration should be around $600ppm$ and if the sensor shows too little then the 2000 factor should be scaled up to 5000 (in fact, these sensors, even though they advertise the zero point at 0, do not even go that far down and stop at about $400ppm$).

Otherwise, the sensor seems to be fairly accurate and reliable providing measurements without needing too long of a warmup time.

Index