Keezer temperature controller considerations for longest life

[Earl_Grey]

So right now my keezer uses an inkbird single stage temp controller. The probe is in a glass of water on the hump portion of the inside of the keezer and I have the temp set to 39F with a 2deg differential so it kicks on at 41F and kicks off at 39F when it cools back down to this temp and the compressor delay is set to max time.

I've had it set up this way for years however current situations from working from home have made me ponder if this is the best solution for the refrigerator. I think we can all agree that fast cycling of the compressor will damage it quickly and I think I got that covered here but what I'm wondering now is if it better to run the compressor for longer times with a longer time in between cycles as the temp warms back up to 41 or is it better to shorten the times thereby decreasing the kick on differential to say 1degF?

I think in either case, fast cycling of the compressor would not occur however is there also a sweet spot for compressor on time as well?

 

[HarvesterOfBeer]

One question is "what is the normal compressor duty cycle like for a freezer"? Yes, cycling too often won't be good for it, but we don't really know what "too often" is unless we have a baseline to compare it with. Anyone have data?

 

[day_trippr]

It's so situational I'd be surprised if there was much convergence in anecdotal data. Along with unit differences in size and efficiency, there's the interior thermal mass varying - I have six 5g kegs nearly full right now, but when they get nearly empty the cycling goes up.

fwiw, I run my keezer using a BrewPi controller and typically see it cycle five to six times per day, with the duty cycle varying with ambient temperature. Right now with the ambient hovering around 80°F where my keezer is the duty cycle is up around 40% on. In the dead of winter when the same space sits in the low 60s it's closer to 20%...

Cheers!

 

[HarvesterOfBeer]

Sure. Several cycles a day (or maybe even a couple an hour) seem reasonable to me. On and off many times an hour is likely a problem...need to either adjust your limits, your PID loop inputs, thermal mass, or improve insulation.

However...maybe it's normal for a fridge to cycle that often? Dunno. If I had any spare time at all, I'd rig up a current clamp probe to a micro controller and log some data.

 

[day_trippr]

I can definitely say my three brewery fridges cycle way more frequently than my keezer. While my keezer worst case cycles as often as every 4 hours, the two 17 cf units cycle at least hourly, and the 22cf unit cycles every half hour. But, it also has the highest efficiency rating of the lot (all three are e-star rated) so I'm guessing smaller compressor managed tighter...

Cheers!

 

[mattdee1]

I've often wondered about this question, myself.

In lieu of hard facts and data, I've kind of settled into the assumption that the closer the operation of the fridge/freezer to its intended manner of operation, the longer it will last.

In practical terms, this means minimizing the frequency of the controller kicking on and off, because in normal use a fridge/freezer is just plugged in and left alone. For my fermenting fridge, I hold temperature at +/- 1 degree of the setpoint and set the compressor delay to the max (10 minutes). For the keg chiller - which is a converted chest freezer - I allow 1.5 or 2 degrees while also setting maximum compressor delay there as well.

I can't see any reason to do anything other than the max on compressor delay. The temperature of several gallons of fluid changes so slowly that it will barely move in 10 minutes; the compressor, on the other hand, will appreciate the downtime.

Seems to work well, and I've been able to run dirt-cheap 20+ year-old chambers reliably for a few years.

 

[day_trippr]

The one tip I'll add is strapping a controller's temperature sensor to a keg under an inch of closed cell foam (or, in a fermentation chamber application, to a carboy or even a bucket) will result in a dramatic reduction in frequency of compressor cycling. Thermowell'd sensors will do the same. Otoh, dangling sensors in the air is going to result in the opposite effect...

Cheers!

 

[mcblsb]

What setting do you put on the keezer itself (as opposed to the controller)?

 

[day_trippr]

fwiw, I have my 13cf Frigidaire chest freezer control set to its maximum cold setting, and control the unit (via the line cord) using a BrewPi instance.
But given most chest freezers have no temperature setting above "freezing" it probably doesn't matter what the oem thermostat is set to. I'm just aiming for 36°F

Cheers!

 

[Earl_Grey]

I like the thought of trying to mimic what a chest freezer would be doing in normal circumstances. Usually theyre full of food so it probably would not kick on very often but once it does its probably on for a while to get that mass of food down to freezing temps.