As I mentioned earlier, I was suspect of the humidity values returned by the DHT22 sensors. So I used a hygrometer to get a separate reading – not that the hygrometer is perfect, but my experience over the past couple of weeks was that it agreed with another hygrometer, the temperature and humidity were consistent, and believable for cool winter conditions.
I placed the hygrometer close to the sensors and gave it a couple of hours to settle in. The temperature matched within a degree, so that’s great. Not so great on the humidity side, I applied a correction factor of -16 to get the reported values in the ballpark. Now the sensors are both reporting values that match the hygrometer within a percent. A significant improvement.
To move one of the sensors upstairs, I took some cat6 cable and soldered the twisted pairs together to make a 4 conductor cable. One conductor won’t be used with the DHT22 sensors, but I will need it for the BME280 sensors when I swap them. I ran the cable upstairs, which took a lot longer than it took to type this sentence. It’s a long story, don’t ask. Connected everything up and booted the pi.
Upstairs temperature was fine, but the humidity value (again) was off the charts. So I decided to try the +5v pin instead of the +3.3v pin on the pi for the upstairs sensor, as the connecting wire is now about 15 feet long. The humidity value came back to reality and has stayed that way for a few days now.
I’ll leave the DHT22 sensors in place for now and see how they behave. I am definitely going to swap the sensors, but there are higher priority tasks to do, especially since the current sensors seem to be behaving for now.