Tuning the freezer loop, or "I was told there would be no math"

So I got home yesterday and went to look at how well the temperature is regulating in the freezer with the new Johnson Controls A419 temperature controller attached. The setpoint is 64 F, the temperature reading at 6:00 PM yesterday was 58 F. Hmm.

The controller's default configuration is "cut in," meaning that when the temperature exceeds the setpoint the controller cuts in the power and the freezer starts to cool. The controller has a "differential" setting (by default, 5 F) which determines the point below the setpoint at which the power is cut out. It's a coarse method for adjusting temperature.

Here's what I see happening:

• the freezer warms up to 65 F and the power cuts in, turning the compressor on and starting the chill process.
• the freezer stays on until the controller reads 59 F (64 - 5) and cuts off the power
  
• the temperature reading continues to go down briefly as the coils absorb a little more heat, but then it starts to gradually rise as the inefficiency of the insulation allows convective heat exchange with the outside air and the heat of the compressor

I didn't measure how long this process takes. (The cooling is pretty quick. I don't know how long it is between cut out and cut in.)

The questions I have are - a) is a regular 7 degree swing in temperature OK during fermentation, as long as the range is within the appropriate band for the yeast? b) is this constant cycling bad for the compressor? For (b) the answer is likely dependent on the cycle time. For (a) I need to consult some of the experts at HBT.

Edit 4/30/2009 21:00 CDT: Check out this thread at HBT for some detailed commentary on this issue.