BME280 sensor variability

[Guillermo Schimmel]

Hi Guys

I just assembled two very simple nodes based on the BME280.

Same Standard sketch on both (we should add one in the main site, there is not an example), battery powered.

They both work fine, but look at this:

Pressure looks fine:

But temperature and humidity are out of declared Bosch specs:

Any recommendation? Could it be a case of fake sensors? They weren't exactly cheap.

Thanks

[boozz]

@guillermo-schimmel

Looks to me as if the sensors perform as specified (acc. to the datasheet etc.):

Relative Humidity: absolute accuracy tolerance of the RH: +/- 3

Temperature: in the range 0-65ºC: +/- 1.0 ºC

BTW: The datasheet says the following about the temperature:

BR,

Boozz

[Guillermo Schimmel]

I think you are right. I have misinterpreted +-3 as +-3% of the value. It seems that is +-3 absolute %RH.

Thanks a lot.

[JohnRob]

I'm not sure how your physical test was setup.

1. In the past I've found the best test was to put the sensors in a cardboard box, maybe 6x6x8 with the sensors on a plastic platform (an old food container), close it and let it sit for some hours in an area away from heat / cool vents etc.

2. Is the host board the same for both sensors?

3. For humidity you can put both in a plastic bag with a saturated solution of salt and water. Let sit for a few days. The humidity inside the bag will be almost exactly 75%. Becomes a cheap and dirty "reference"

4. There has been reports that sampling too frequently will cause internal heating to a small degree. At what rate are you sampling the device?

5. These suggestions are not to say what you see is not part to part variation.