One of my current electronic projects is to build a distributed wireless temperature and humidity sensor network throughout my house and use that to eventually control my heater. One big restriction that I've forced myself into is to make it as inexpensive as possible. Well, after a few weekends of working an hour or so on it, I have something that works! Or sort of. But first for the summary of components (because that's important for what is going on):
The code is pretty simple: it uses a couple of libraries that already were provided out there (will provide the links soon, and post my modified code) and basically reads the two values and sends them to a receiver that is connected to my computer and logs the results.
Seems simple, right? Well, it is! So I put the sensor on the guest bedroom and I'm reading the information from my "lab" and it says: Temperature: 13ºC, Humidity: 27%. According to my measurements with other sensors, it should really say something like Temperature: 17ºC, Humidity: 52%! The interesting thing is that I have 2 of those sensors (they were $4 each) and they all say exactly the same thing, so if it's a calibration issue, it's consistent.
I'm not sure of what my theory should be. At first I thought that maybe humidity sensor was shifted by a bit and actually it's double that (which is consistent), but then the checksum should fail and the temperature should be off by a factor of 2 also, which it isn't.
What's next, then? Fortunately I have another humidity sensor: HH10D. A little more expensive and way more awkward to use (you have to read the calibration values from an EEPROM and then read the data and use those values to calculate the values), so I guess I'll leave this for some other time - probably next year (I don't know if I'll have time this weekend and then my parents will be in town for two weeks). Oh, well. We keep at it!