Both of our BME820s have saturated over the last 9 months we have tried to reset them by heating to 120 degs C in our home oven then leaving at room temperature but this hasn’t improved the output as they still read 100% humidity. It isn’t possible in the home environment to use the exact conditions suggested by Bosch to reset the BME280 but I don’t think that would make too much difference. I note a large number of the BME280s are reading 100% and this is in line with the observation given in this report (PDF) Comparison of a Computational Method for Correcting the Humidity Influence with the Use of a Low-Cost Aerosol Dryer on a SDS011 Low-Cost PM-Sensor
Would it be better to ignore the met sensors and use data from other sources?
1 Like
Hi @sensorsalnorth12,
are you sure that you are using BME280s? This phenomenon was seen on the DHT22, but not on the BME280. There are many sensors in Bulgaria with the BME280 (they were the first using it in a larger amount of devices) and there we don’t see this effect.
Hi ricki-z
Thank you very much for the reply.
Yes we are certain it is a BME280, I have also attached an image of one of the chips. (We took the image when we were setting the systems up). We bought the devices from Nettigo as a complete package so the BME280 came pre-soldered.
With respect to your other comment about the BME280 not having the same problem I draw you attention to page 4 of the document I linked in my original post (I supply a screen shot for convenience). Perhaps Bulgaria has a lower humidity than some other parts of Europe so the (possible) issue hasn’t been as obvious there?
I am not sure if the linked document has been peer reviewed so I would not necessarily accept all of the comments the author has made but the high humidity readings are undeniable.
The author comments that the BME280 is designed for indoor use (not sure how he has come to this conclusion). This is a surprise to us as it is used in other outdoor systems. It is possible they protect the sensor in some way?
The author also uses an outdoor met sensor HY221 which might be a good replacement but in the U.K. it costs £50.
I think this needs some more consideration and perhaps a phasing out of the BME280s especially in areas of high humidity.
Can I see the reverse side and the solderings on the BME280 ?
Maybe the oxidation disturbs something.
Sorry about the image definition of the first photo of the back of one of our BME280 sensors when it was new. To my unskilled eye the soldering looks perfect. The second photo I have just taken
there is a difference in lighting conditions but there seems to be a difference in the appearance of solder on the pins which could be oxidation. The sensor as a whole has a weathered look to it.
Some are using some kind of ‘muffler’ in order to avoid condensation but still be able to get accurate readings. See here.
That said, it would probably be cheaper to get a sensor better suited for outdoor use like the Sensirion one with the coating. I wonder, since the humidity factor reliability is more important than barometric readings wouldn’t it better to also make it the reccommended sensor?
Thank you for the link. We are considering building a complete new sensor system and definitely agree with your last comment on the more robust sensor.
Please keep us posted. 
Old thread but I try my luck 
You mentioned the reset by heating to 120°C sadly I did not find the specifications or details for this.
Can you tell me how long you have to heat it up or what the suggested procedure is? My sensor shows almost exclusivly 100% humidity now for some days and i’d like to try that without breaking anything more 
Hi
We consulted page 8 of this document
I hope this helps. Sadly, we couldn’t control the humidity in our oven so it didn’t work for us. I hope you have better luck.
Hey,
thanks a lot for that, the oven-approach at 120°C did not work though sadly… I’ll try on
We ended up buying new ones.