What could have happened to my refrigeration?

Had a bit of a weird anomaly yesterday, where my system’s refrigeration ability seems to have abruptly disappeared for no apparent reason. My setup is a slightly more sophisticated version of the classic “Son of a Fermentation Chamber” system, but instead of loading a cold chamber with ice, I use a mini-fridge for the cold chamber (which is not managed by the Spark 3). The system has always been a bit underpowered, but it has generally worked fine for maintaining lager temperatures as long as I use another fridge to cold crash wort down to the desired temperature.

Last night before going to bed I noticed that my beer temperature was 0.4 degrees C higher than the setpoint, which was unusual. I then noticed that the chamber temperature and fridge temperature where strangely high as well. After reopening my computer, I discovered that my cold chamber had pretty much completely lost refrigeration capabilities for the better part of the day. In the graph below, the red line shows the temperature in the cold chamber. Up until midnight on July 17th it shows a fairly typical duty cycle for an active fermentation. Then, suddenly, just before midnight, and in the middle of a compressor cycle it appears to have suddenly turned off, there are a few strange fluctuations (which could result from a makeshift glycol system in that chamber transferring heat around) but the temperature rapidly rises until around 9:30 am on the 17th when the compressor appears to kick in again for at least part of a cycle. At this point the beer temperature is steadily rising and the cooling fan PWM is at its max of 60%.

I finally notice the problem around 2am on the 18th but couldn’t figure out what had caused it. The compressor appeared to the running and transferring heat (hot evaporator coils on back). There were no doors open or leaks. The “Freezer” corner needs deicing, but that doesn’t seem like it would create such a sudden problem. The glycol loop wasn’t leaked and seemed to be functioning normally.

Ultimately, because I needed to get to sleep I just threw a bunch of ice in the cold chamber and set the system to fridge constant to avoid having to worry about integrator problems from a day of unresponsiveness. When I woke up this morning, other than very long duty cycles trying to get the fridge back down to the right temperature things seem much more normal.

Any ideas what might have caused this? In the long run, I know I need to upgrade my system to something with more cooling power; however, when I overstress this system I’ll usually notice the fridge gradually lose control of the cold chamber, never anything as sudden as this.

For context, here some (old) pictures of the system in question. The second one is more representative of the current state of the system (just imagine more ice buildup on the freezer part).


First of all, cool setup!

We can’t really see whether the compressor was running or not.
Is it on the same power circuit as the spark?

Fridges have a bimetal strip as thermostat usually. If it makes contact, the compressor gets power.
It is possible that there is some kind of internal protection circuit, possibly thermal protection.

A compressor can get too hot if it is cycled too quickly, which doesn’t seem to be the case.
Is it the original fridge circuit or some other temp controller that’s driving the compressor?

Check how hot the compressor is and make sure you have enough ventilation. Perhaps aim a fan at it.

They (and the raspberry pi) all get power from the same wall socket, but the spark and fans run off of the 12V DC rails (left side in picture) while the fridge (and power adapter for the pi) run off of the 120V AC rails (right side in picture, the fridge power is the black cable on the floor towards the back).

If the compressor did lose power, it seems that it did so without anything else losing power, and then the power came back on by itself.

The original fridge circuit controls the compressor. The graph suggests that the off time of the duty cycle was about 80 minutes. The on time is pretty long, and rose to ~2.5 hrs in the cycle before the anomaly. So the fridge is probably underpowered, but I definitely am not short cycling the compressor.

The long on cycle tripping a safety of some sort (either temperature or maximum on time) is an interesting theory. The evaporator coils are about 26-27cm away from the wall–so while a fan would probably help, I don’t think I’m abusing it any more than the average fridge that is shoved against a wall.

I’m mainly just trying to get a sense for the extent to which this was a fluke or shows that my fridge is dying and I should start thinking about how to replace it (ideally with a more appropriately sized solution).

I switched to water/glycol cooling myself and use a lindr AS-80, which also serves as my tap line chiller.
Works very well. I have a SS brewtech conical with coil, but I think you could DIY something too.

I’d like to also cool my kegs a bit with the recirculating water, so I’d like to DIY something for that.