BMS280 reading far to low, PMS5003 reading too high

Support
1

Hi, as wasn’t successful @ Discord, I try it here as I am really stuck.

My enviroplus board is mounted piggy back style on Raspi Zero, th case ist like this one

My BMS280 shows much lower temp (°C) reading than references
BMS280 Reference1 Reference 2
1.75 6.1 6.1
-1.57 4.1 4.1
-2.12 3.9 4.0
-4.17 3.0 2.8

On the other hand the PM reading “seems” far to high. It looks like fog is terribly increasing PM values, which sound logical as fog can deflect/reflect laser. I read an article this is compensated in professional equipment !? Any chance to increase accuracy with the enviroplus?

Any help appreciated, as the current setup isn’t worth the electrical power spent to operate it.
Thanks

2

Although it seems I am the only one having this problem, I’d appreciate some help ;o)

3

I am not sure what you mean by “references”? I know that the temperature readings can be quite off if the sensor is too close to a source of heat, like the CPU of an RPi Zero or any RPi I imagine. In the example scripts they’ve included a ‘hack’ to calibrate the temperature based on the temperature of the CPU but personally I don’t find it very reliable.

The best thing to do is get the BMES280 away from the CPU which is of course very hard to do unless you make your own extension board that allows you to physically put some distance between the CPU and the BME280.

As far as the PM readings go, it sounds logical to me that if you create fog to have laser reflections, the fog particles are seen by the PMS5003.Not sure how the PMS can compensate for that… have you looked at the website of the manufacturer? I know they have other PM measurement devices like the 6003 and others, if I remember correctly.

What I’ve also done in the past is check out libraries from other parties like Adafruit for the same sensor but then it’s a library for C instead of Python and see how they read sensor measurements on a fairly low level. This does require some dedication because you have to be able to understand what they’re doing and then figure out how you can copy that into Python.

I hope this helps, it’s all I can give you right now.

4

Thanks, this helps. So I will take the piggy back off, relocate it and check out other PM sensors. As I am not the Linux expert I don’t go down the “C-route”.
The references I mentioned are simply two other digital thermometers.