Thanks, very interesting results so far but a 20F difference?
3 things:
First, the frozen keg in the bottom is acting just like an ice pack inside a cooler, in other words, cooling the surroundings down and making a huge contribution for the stratification.
Second, do you actually have 2 identical thermometers calibrated for top and bottom? That is to make sure they agree with each other instead of over or undershooting the stratification.
Last but not least at all Are you taking the thermometer reads outside without opening the freezer lid while their temperature probes are inside? If not, when you open that door, you are rushing in 70F room temperature air (if I recall correctly your conditions) that is making a huge difference for the stratification.
First: The kegs are no longer frozen. I turned the freezer off and let it warm up for a day. It has been running for about three days now, so I figure everything is about as stabilized it will get for the first configuration.
Second: I'm using three digital thermometer and they all agree within one degree when tested, so my confidence level is high in that regard. One is the controller and the other two are digital indoor/outdoor thermometers. The controller probe and one thermometer probe are at the bottom of the freezer. The other thermometer probe is a about two inched below the collar.
Lastly: I have not opened the freezer at all for the past four days, so there is no issue with room air rushing in. The temp differential from top to bottom fluctuates from about 17 degrees to 20 degrees over one complete cycle (on period + off period = one cycle). I just checked it right now and there's a 17 degree delta. The compressor shut off a few minutes ago.
Room temp is a very stable 72F. Controller is set at 40F w/a 3 deg differential.
I've only had the patience to time the on portion of the cycle so far and it was right at a 30 minute run time. I got tired of waiting for the off portion to end as it is really long at more than an hour. I'm not sure how much longer, but it could be much more. One very startling discovery is how little power has been used so far. The meter indicates that in 66.5 hours the freezer has used only 0.47 kwh. I don't recall what I'm paying for electricity, but I think it's about 12 cents per kwh or so. What that translates to is a cost of less than two cents per day. I find that to be unbelievably low, but considering how infrequently the compressor is cycling, I guess it's correct. I plan to run the freezer in this configuration for at least a couple of more days to see how consistent the data is before I move on to the second configuration. I will get handle on the off portion of the cycle as well. At this rate, it may take awhile to complete the tests for all four configurations, but it will get done eventually.