WeatherDuino Forum

Maybe some will remember that some time ago we discussed here the observed relation between the increase of the CO2 readings with the increase of %RH on the air.

One reason that lead me to buy the Sensirion SCD30, was the fact that the sensor includes a Temperature / Humidity sensor onboard.
From that, I presumed that possibly Sensirion developed an algorithm to correct the behaviour we had observed with our MH-Z19, thus the use of the integrated T / H sensor.
However, seems I was wrong. I can observe exactly the same behaviour on the CO2 readings from the SCD30. They increase and decrease with a interesting correlation with %RH. Even small variations in %RH are reflected on CO2 readings.

Perhaps the images bellow will show it better than my words.

[attachment=2725]

[attachment=2726]


If this relation exists, and it seems to exist, then, even when used indoors, when we see an increase on the CO2 readings, what are we really seen? I real CO2 increase or a %RH increase?

Well, maybe the answer could be, that the main application field for this devices, as pointed for the manufacturers (Sensirion inclusive), is air conditioner (HVAC) equipment, specially large HVAC, where the air in the pipes have little %RH variations.

This is just thoughts... we learn experiencing and by thinking too.

Quote:The SCD30 from Sensirion is a high quality Nondispersive Infrared (NDIR) based CO₂ sensor capable of detecting 400 to 10000ppm with an accuracy of ±(30ppm+3%). In order to improve accuracy the SCD30 has temperature and humidity sensing built-in, as well as commands to set the current altitude. For additional accuracy the SCD30 also accepts ambient pressure readings!

It has been designed for the usage in various applications and devices such as:

Demand-controlled ventilation
HVAC equipment
Air conditioners
Air purifiers
Smart home
IoT devices

Hello,

I am not going to venture into an area I know very little about.
But one thing stands out to me: the operating temperature range: 0° - 50°.
The sensor is designed for indoor use, i.e. with lower humidity levels, especially here in Portugal.
Its design and operating range is similar to the MH-Z14.
One thing I don't understand: Why do you want to "correct" for humidity?
In the field we are concerned with, this seems to me to be counterproductive, as humidity is water vapour and therefore composed of oxygen.
Yes? No?

(26-04-2021, 14:13)tobyportugal Wrote: [ -> ]One thing I don't understand: Why do you want to "correct" for humidity?

Hi,

I don't properly want! The reason is the same why all dust measurements must be corrected in relation with humidity. That main reason why all dust sensors include a Temperature / Humidity sensor, is because we need the humidity reading to apply a correction on the raw readings.
All your dust sensors (no matter the brand), apply corrections to the raw readings from the dust sensor.

No going in to much details (also, I'm not the best one to do it), high grade CO2 measurement equipments, are very complex units. The fresh air before entering the measurement chamber is humidified or dehumidified until reach 50%RH, otherwise the readings may not reflect a real increase or decrease of CO2 in the air (as we humbly tried to demonstrate).

At fresh air, far from a source emitting CO2, the readings should always be always around 400ppm. Just the error margin of the sensors (+-30ppm Sensirion, +-50ppm MH-Z19) we are talking about, is enough to understand that they aren't suitable for serious measurement of CO2 in the atmosphere. However, the same way as dust sensors are playing an important role in bringing people attention to Air Quality, the CO2 sensors may also have a part on that role. Get people's attention to environmental questions is the main goal.

Happily, our small team can count with someone with a great knowledge on this matters, including weather, and we all had learned a lot from him.
Much before the appearing of the Purple Air and other popular dust sensors, we were already investigating and working on this field, using not well know dust sensors from Sharp (very hard to read and somewhat expensive). The same is happening now regarding outdoor CO2 measurements, we are trying to figure the maths that eventually may allow us to use inexpensive sensors, and get reliable measurements.
Connecting a sensor to a microcontroller and get a number from it, is a quite easy task, the hard part is the other things, like the maths involved when correction of the readings is needed.


The publication of this topic, regarding CO2 measurement, was more or less like launching a theme for reflection.
If at least one person is going to investigate about this matter, it has already been worth it. And I bet you will!

Rgs