BME280 Humidity Reading Discrepancy

[o_o]

The soldering process completely dries out the humidity sensor due to the high temperature. Usually, the reconditioning process allows the sensor to re-hydrate and come back within spec.

So there are no way of soldering the device without reconditioning. As I said, since you did not manufacture the boards yourself, it is quite difficult to point out the issue. Have you tested with the official Bosch Shuttle board ?

[Alexei]

Unfortunately i have no Bosch Shuttle board. How can it help me?
But if i understand you correctly, exactly the negative offset of the RH may indicate that the sensor was contaminated during washing or any other violation of conditions during soldering.
What about exactly positive offset?
I simply hope that in my case the producer of the PCBs did not make reconditioning so if i will do it myself that would restore BME280.

[Alexei]

I have a good news!
I have performed reconditioning procedure with one of the BME280 which had positive +7% offset and now it reports almost the same humidity like the reference HIH4000 in the range between 10% and 50%RH.
I performed dry-baking at 120 centigrade for 2 hours. I could not control the conditions dyring rehydration. But i suppose that air temperature was close to 18 centigrade at ~70-90%RH for ~8 hours. For rehydration i used compact desktop air conditiner integrated with air humidifier.
Hope this information will be helpful for those who have faced with the same problem