And, now that humidity thing in my chamber
I was worried about too low a humidity, but it turns out that I have the opposite problem. My little refrigerator runs close to 90-95% RH all the time. I think it's because when the compressor comes on, the moisture freezes out on the cold plate, but when it turns off, it just re-evaporates or drips off. The guys converting refrigerators are using frost free ones, which send the water out to an evaporation tray under the refrigerator.
You can see that the temperature is holding pretty well around 13C, with momentary dips
I was able to bring the humidity down by putting a tray of salt in (which is somewhat hygroscopic and gets damp), but that's not a good approach in the long run, because the salt needs replacing every few days. Since the run time on the compressor is very, very short, maybe just bringing in outside air would do. Some sort of tray to catch the melted ice off the "coils" would be needed.
Looking at the details of a couple cycles, it looks like the refrigerator cools down pretty quickly in about 20 minutes, but the humidity drops like a rock: the moisture probably condenses on the cold plate almost immediately (and then freezes). The warm up is much slower, more like 150 minutes. What's interesting is that the humidity starts coming back up before the temperature reaches the low point, I think that's because we're looking at relative humidity, which varies with air temperature. The dew point looks more consistent (since dew point is a proxy for "water content").
