Reliability of temperature and humidity sensors.
I would like to get some feedback on reliability of temperature and humidity sensors in outdoors environment.
My experience is not very good. I live in quite humid and warm environment and time between failures can be as short as few month. So far I have lost 4 sensors. Two sensors were SHT15 and two were HTU21D.
One SHT15 reads 0 on both temperature and humidity.
One SHT15 reads temperature OK and humidity up to about 65%, then drops to 0%.
One HTU21D reads 999 on both temperature and humidity.
One HTU21D reads temperature but lower value by several degrees and full scale on RH (raw data 65532 which translates to 119% RH). This sensor has Teflon membrane to protect humidity sensor.
I have tried re-baking and re-soldering of sensors to breakout board with no success. I also noticed that breakout boards suffer from corrosion, quite a lot of green stuff on solder joins.
I am even wondering if these sensors are intended for outside use. Looking at the data sheets it is not really specified, if anything, it looks like they are not. Silicon Labs Si7020 specifies “Indoor weather stations”, HTU21D does not specify any outdoor use and could not find anything in Sensirion data sheet.
Maybe price of professional weather stations is good indicator of difficulty?
I’ve used a few SHT11 sensors for years and all of them still work perfect. I don’t recall the exact number but I think it’s about three sensors I used indoors and two outdoors. I’ve had measured values from about
25C° up to about 45°C and humidity all over the scale. Actually once the board was completely covered in snow and water dripped onto the sensor as the snow started to smelt. As soon as I cleaned it up it worked again. So absolutely no issues here. One of the sensors was outside for about three years and didn’t look too nice when I “recovered” it. Also some corrosion and a death spider inside but working.
Are you sure all other parameters, like supply voltage, are not an issue. Maybe there are some connectivity issues between sensor and uC? Or even some signal issues on one of the uC’s legs? I can’t belive SHT15 are so much worse than my SHT11.
Thank for comments. I believe if this was a common problem, we would have heard about it sooner. I agree with you that there should not be much difference between SHT11 and SHT15, they are most likely selected for performance at the end of manufacturing process.
In my case system is battery/solar powered, 4 NiMh batteries are usually charged to about 5.6 V. Output of voltage regulator is 3.29 V. I have tested sensors separately outside the system, on desktop set-up, using even different sketches, reading raw data as it comes from sensor and they are faulty. Enclosure I am using is like “Double enclosure”. It as a pagoda style, and inside is vertical tube opened at both ends housing T/RH sensor.