Compare different i2c temperature and humidity sensors (SHT2x, SHT3x, SHT85, ..)

I have a couple of AHT10s and they seem fine for household use. I'm actually looking at upgrading to the newer AHT21s.

I am not a fan of AHT. But I believe AHT will replace DHT in cheap sensor segment and will be most common sensors for basic DIY projects. And it does not really matter if me or you like it or not. DHT was never the most accurate sensors, but anyway it is still in use. Most Arduino All-in-one training sets got DHT in the box. That is the main reason.

There are some other reasons:

  • Previous board is just go no more space and multiplexers lines to add sensors
  • Previous board was not very stable
  • AHT was unstable when more than one device is on i2c line (see previous board version). So It actually consume multiplexer line and makes it hard to place other sensors.
  • large numbers of sensors and multiplexer is hard to make stable
  • two voltages on the same board is also add some instability.

So that is why only AHT. Only 1 sensor per line. Only 5v. And the board survive all test including freezer and condensation. two "old sensors" was added on root i2c line just to have some baseline to compare.

Also I was thinking of SHT only board (incl. SHT4x). Or kind of "good sensors board". But not sure I will do. There is actually enough data from v1-v9 boards to see "the picture". And good sensors board is much more expensive actually than "cheap sensors board".

The code is actually open so any person may put couple of multiplexors and sensors together and continue this data collection with any new or old sensors available

You know, It might sound strange, but I not feel like it is time to replace AHT10 and AHT20 to newer sensors. Reviewing the test data (trends) I feel like old AHT sensor is got some stability in last 1-2 years I would say, and new is not yet. And there is no "more accuracy" found in new models. So I would wait couple of years. or at least use AHT20 not AHT21.

There are some other test to review like freezer for example

https://wiki.liutyi.info/display/ARDUINO/Test+4+v10b+Freezer+1+draft
https://wiki.liutyi.info/display/ARDUINO/Test+6+v10b+Freezer+2
https://wiki.liutyi.info/display/ARDUINO/Test+7+v10b+After+freezer+high+humidity
https://wiki.liutyi.info/display/ARDUINO/Test+8+v10b+High+to+Low+Humidity
https://wiki.liutyi.info/display/ARDUINO/Test+9+v10b+Low+to+High+humidity

Those are no changing the "whole picture". But might be also interesting to look into.

Good to know. I was getting ready to buy a dozen AHT21 from AliExpress.

I know this topic may have stalled, but did you come to a certain definite conclusion that certain sensors are better (best) in terms of long term performance and accuracy? I've tried many BME280 sensors and can tell for sure that their humidity measurement lifespan is about 6 to 9 month outside in Florida humid environment. After that they just become useless for RH reading although temp readings are pretty solid for over a year or two. DHT22 is a good "toy" for very first sensor measurement exposure in Arduino world and then can be thrown away.

Hello...
I'm right now planning to build a weather station for a nearby farm field. after going through the thread I think most people have a better experience with these sensors. so could someone specify which sensors would be better?
Like for :
Temperature
Humidity
UV index
Light intensity
Pressure

a comparison with the price range too would be good in terms of cheap or expensive. having better accurate reading is important for this project
Thank you in advance!

I reckon that's completely expected as this type of polymer humidity sensors are very rough and ready in their operation (especially without individual calibration). They are also history-sensitive especially if subjected to 100% RH.

For instance the SHT20 datasheet gives accuracy as +/-3% at a fixed temperature ignoring hysteresis and long-term drift
In other words you don't actually know the repeatability and accuracy as you cannot ignore hysteresis or drift in an actual application - or put simply the technology is pretty flakey if you want actual accuracy rather then a rough guide and trend-spotting.

You need a wet/dry bulb system to measure relative humidity with reasonable accuracy, or something really high tech using IR absorption spectroscopy.

1 Like