Industrial Reliability - DHT11/22 vs. AM2305/6 vs. ???

My Humidity/Temp logging project is virtually complete using Mega, Ethernet/SD shield, and DHT22. As part of some final reliability testing, I allowed the DHT22 sensors to get wet (rain) to see how they'd react. Not good. After drying out and running for weeks, the temperature values are fine but the Humidity is wildly off on both sensors.

So, my question is: Can anyone provide some feedback on an "industrial" capable HT sensor that can withstand potential wet environments?

Of course there is the obligatory:

  • not too expensive
  • digital output
  • multiple sensors on same MEGA (which I understand disqualifies some I2C bus sensors without unique addresses)
  • 5V
  • (nice to have) uses same library as DHT22

Thanks,
Ron

In the absence of any recommendations, I just decided to try an AM2305. I'm not yet sure about its environmental resiliency, but here are a couple notes on my initial experience:

  • mine came with no documentation at all - not even labels on the connectors (red, white, yellow)
  • guessed (apparently correctly): red=V+, yellow=data, white=grnd
  • without a pullup resistor (on a 20' wire), the sensor would only randomly respond with accurate values
  • with a 2.5k pullup resistor it seems to respond pretty consistently
    (I should try a higher ohm resistor. 2.5k might be a little low)
  • I'll have to wait and see how it does in a more harsh environment

Chinese product at low cost: there is nothing to add.

If you want reliability you must go to the catalog of REAL founders as Texas, Vishay etc.
Obviously you will pay the product 5 or 10 times more but reliability has a price.

Aosong site (founder of AM2305) http://aosong.com/en/home/index.asp

To have an "idea" of fabrication have a look here : https://sigrok.org/wiki/Protocol_decoder:Am230x