Ensuring Keezer Longevity

[jmhbutler]

I recently just finished a 5cu ft keezer build which I am extremely happy with. Now, after reading some stories, I'm worried of my compressor suddenly going and it dying. Right now the sensor is in 30 or so oz of water and set to average at about 9C/48F with a 5 degree swing. The freezer runs for about an hour and a half around 4 or 5 times a day.

Am I correct in my belief that the start-ups are whats most damaging to the keezer? How frequently and for how long can I allow the freezer to run before I start to impact its longevity?

I know theres a lot of posts similar to this but after hours of searching I've been unable to find the specific answer I'm looking for. Not sure if I should have started a new thread or dug one up from a few years ago.

 

[zachattack]

Those cycle times sound great to me, I wouldn't worry. You can also make sure the compressor delay is kept high on your temperature controller.

https://www.homebrewtalk.com/f35/keezer-cycle-time-292156/

 

[IslandLizard]

Is that a 5° swing either way or total?

If it's total, your keezer runs for 4-7 hours a day to maintain ~45-51°F (~48°F with a 5°F swing)? That sounds long to me. It means it takes an hour to bring the 30 oz of thermal liquid down 5°F, from 51°F to 45°F. Perhaps a smaller amount would be better?

Is the keezer in a tight enclosure? The hot side coils are in the outside walls. They need a way to chill themselves, preferably to the ambient air, not back to the inside.

The lids of some of the new freezers are a weak area, insulation wise, from what I've read.

 

[jmhbutler]

Island Lizard,

The keezer is not in a tight enclosure and I have re-insulated the lid. It kicks in at 13C/55F, kicks out at 8C/47F, and continues to drop to around 6.5C/44F.

My question is, am I better with fewer long periods when the freezer runs or more frequent, shorter periods?

 

[IslandLizard]

Longer, less frequent periods is generally better for the compressor. But I doubt it would be worse if it kicked on twice that amount for 1/2 the period, perhaps reducing the swing from 11°F to maybe 6°F. Given the time it takes yours to recover I have the feeling these modern day freezers have undersized compressors, and a minimum of insulation, and there's nothing you can do to change that, except for doubling up the lid insulation.

 

[jmhbutler]

Thanks Island Lizard,

I just did a second test since adding another keg and it seems as though the keezer is going around 8 hours off and 1.5 on - so its only running about 4h a day. I think my nervousness of wearing out the compressor has led me too far in the opposite extreme. Do you think 1.5h of consecutive running has potential to be damaging? If I was to reallocate the 4 or 5h of runtime throughout the day, what period length would you suggest aiming for?

 

[jmhbutler]

So we've determined that running a keezer too frequently can be damage the compressor. The analogy I was given is pedalling a bicycle - getting started is what is most exhausting and wearing.

However, can anybody comment on whether running the compressor for continuous, extended periods of time can also damage the unit?

 

[Zepth]

My understanding is that the short cycle time is most damaging to the compressor when it kicks on shortly after being shut off. The damage comes from trying to compress the refrigerant that is already compressed. If memory serves correctly "You're alright as long as it has at least half an hour off between cycles."

 

[IslandLizard]

What is the temp of the beer? How large is their swing? Your probe jar has a smaller heat capacity than kegs, and that is good. Your kegs likely don't swing more than 4-5°F, if that.

Make sure to cap and grommet that probe jar nice and tight so you don't get excessive moisture inside the keezer from evaporation.

Some people use a small computer fan to provide extra air circulation.

 

[jmhbutler]

The beer temperature is around 8-9C or 47F and surprisingly doesn't change more than a degree or two no matter where I am in the cycle. I do have a computer fan going in there and I've used some caulk to completely seal the jar so it doesn't ad moisture. Like I said before, my compressor runs about 4.5 or so hours a day and, given the amount of insulation I have, I don't think that can be improved upon. The only question I have left to be answered is if 1.5h of continuous running is okay for the 5cu compressor or am I better to split that (by dropping differential from 4).

 

[IslandLizard]

Yes, that makes sense since the thermal mass of the kegs prevents the larger swing the probe jar experiences. Although the whole system plays a role, the subsystems, i.e., kegs, probe jar, freezer chamber, have their own response and return that into the system mix. The fan is a good helper. I guess it's only on when the freezer is in cooling mode.

8 hours between cycles is not bad at all. I consider that pretty darn good actually. The 1.5 hours of run time each cycle still sounds a little long at that relatively high temp setting. But as I said, the compressor maybe a bit undersized, which helps in the overall efficiency rating. I've read people reporting shorter times, as low as half hour on-cycles, but their differential maybe set lower, so it runs more often.

Are 1.5 hour on-cycles going to kill the compressor pre-maturely? I seriously doubt it.

The chest is designed to deep freeze and the delta between ambient and -4°F (or lower) is way larger compared to your nominal 47°F setting. That means that at -4°F the freezer (compressor) has to work a lot longer and more often to maintain the lower temp within it's set range. Insulation is ultimately the limiting factor to how low the temp can get and how long the freezer can maintain it without cycling on.

Now failure still occurs in freezers due to inferior manufacturing practices and lousy QC. So you're the ultimate field tester, and there's no saying where and on which side of the curve yours is. Only time will tell.

Ideally a superinsulated freezer would be less power hungry, and thus more efficient. I haven't seen one yet outside lab environments. Then realize the collar, rubber lid seal, and shanks sticking out are also not thermally the most efficient either.

For reference, I have a chest freezer here that is 25 years old (a Gibson) and still chugging away. At -4°F (it won't go lower) it runs at least 1-2 hours at a time and more than just a few times a day. And it's been doing that for 25 years. I really should put one of those Kill-a-Watt meters on it, or a graphic interface and get smacked in the face with reality.

Stop fretting and start enjoying your freshly tapped beer.

 

[Zepth]

So while watching some TV this morning got around to taking note of my freezer run schedule. Not a difficult task since the two are side by side. Bachelor living at its finest, the keg fridge is almost within arms reach of the couch.

Set close to default, 10C with 0.5 hysteresis, max delay (10 min)
Time-Temp (Celcius)-Event
0802-10.5-Turns on
0812-10.0-Turns off
0815-9.4-Residual cooling effect bottoms out
0843-10.0-Just checking for data point
0945-10.5-Turns on
0958-10.0-Turns off
1003-9.4-Residual cooling effect bottoms out
1015-9.7-Slowly warming, writing post
1023-9.8-Actually finished post

So I run for 10-12 minutes (possibly longer in the middle of summer), then have about 90 minutes off. Without doing any math that seems to be in the region of 12-16 cycles a day. Sounds good to me. Of course this will change with insulation properties as mentioned earlier by Islandlizard, ambient, what temperature you like your beer, and one other factor not yet mentioned in the thread. Thermal capacity. The more stuff if your freezer/fridge will take it longer to cool initially. However after it is cooled it will take longer to warm up towards ambient. Currently no collar or taps on the keezer just picnic taps inside. Content 2 kegs, 10# co2, handful of bottles. Space for 1 more keg. Sensor taped to co2 tank under a layer of insulating material.

Seems that to raise 0.6 degrees from 9.4 to 10 took a half hour, but the 0.5 degree rise from 10.0 to 10.5 took a full hour. I'll admit that one data point isn't exactly enough to make a solid conclusion on. Just an interesting observation.

I should also note that I do not have a heating circuit hooked up to the STC-1000. If I did it would have kicked on from the residual cooling effect. Set your hysteresis band accordingly! Since it is indoors the unit will always be below ambient, and the room will warm it up in due time. If in a garage that may go below freezing you'd probably want that feature.

Hope this has helped

 

[jmhbutler]

Thanks you two for the help. You've both provided me with valuable perspectives to look at this from. I went ahead and did the same analysis as Zepth with my keezer loaded with 2.5/3 cornies and my results are as follows:

10:00 11C
1:45 12C (Compressor On)
3:15 8C (Compressor Off)
6:30 9C
8:00 10C
11:15 11C

This means that at my current settings of 8C with a 4 degree differential, I'm running 1.5h on and 11.5h off. Interestingly, I'm running 11.53% of the day (or 2.75h) which is very similar to Zepth, who although at a lower temperature, is running 12.22% (or 2.9h).

Right now I think that my differential in combination with my probe insulation is a bit overkill and being afraid of killing my keezer has led me to the opposite extreme. I'm planning to my drop my differential to 2.5-3 This would give me approx 1h on and 6-8h off. The compressor would run an extra 0.25-0.5h a day but more periodically. Based on what you two have said, this certainly can't hurt and will make my beer a bit cooler (which I wouldn't mind).

I'm not sure if you two have any other suggestions based on the more detail info I've provided. If my goal is longevity, is 1h a decent runtime or should I push it lower? You've been extremely helpful and made me feel a lot more confident in my set up.

Final Question: Is it correct that the dial on the side of the freezer is for an internal thermostat which is overridden by the temp controller, making its setting have no impact?

 

[Zepth]

Final Question: Is it correct that the dial on the side of the freezer is for an internal thermostat which is overridden by the temp controller, making its setting have no impact?

Correct. My keezer never gets cold enough for the built in thermostat to trigger "off." The controller cuts off power to the circuit entirely. I've always wondered if digital thermostats in a fridge, air conditioner, heater, etc would fail miserably with an STC style controller. Long live analog!

 

[IslandLizard]

Correct. My keezer never gets cold enough for the built in thermostat to trigger "off." The controller cuts off power to the circuit entirely. I've always wondered if digital thermostats in a fridge, air conditioner, heater, etc would fail miserably with an STC style controller. Long live analog!

Exactly. With the mechanical thermostats the STC just takes over and governs the on/off cycle.

With digital thermostats the best thing is to splice the STC into the circuitry, replacing the built-in one. So the fridge/freezer and it's fancy display stay on, but the STC controls the compressor, not the internal thermostat. A bit more involved. "Analog" is the easiest.

 

[Zepth]

Then again the default on such a digital thermostat is likely cold enough that as soon as the "standard" method of controller wiring gives it power it would think "I've just been plugged in!" and start doing its thing. Or it would beep at you like an infant just begging for settings. That would be a PITA.

 

[jmhbutler]

Thanks guys, thats what I was under the impression for the dial setting. After doing a bit of appliance reading, it seems IslandLizard is bang on when he says the newer, smaller, compressors are designed to run longer but more efficiently.

To follow up, I've finally got my keezer to a serving temp of around 8C where it runs for roughly an hour and then spends 6 hours off. (I'm sure this will vary slightly when the keg volumes change.)

Focusing on preserving the unit, would you two say this balance is good or should I continue adjusting my settings to either reduce the length of the run period or increase the length of the recovery period?

 

[IslandLizard]

Sounds like you got it tweaked nicely. The more full kegs in there the longer the lag times will be. Iow, it takes longer to warm up above the threshold, and then longer to chill it down, because of the larger mass inside. As I said before the jar with the probe plays a role too, as that has taken over the function of the thermostat.

How long will your freezer last? Good question! Time can only answer that. Say average lifetime is designed to be 20 years under "normal" use, then 67%, with a standard deviation of say 5 years, will last between 15 and 25 years. 95% between 10 and 30. This is assuming a standard distribution, and where yours is positioned on that curve is anyone's guess. By not having the unit cycle on every hour you likely increase its potential lifespan, pushing it toward the right side of that curve. That's about all you can do. Unless you believe talking nicely to it each time you fill a glass increases the odds.

 

[jmhbutler]

Thanks Island Lizard. Although you've soothed the concern of my freezer dying, your statistical analysis fails to include the fact that my poor luck with electronics severely skews our bell curve towards failure. Jokes aside, what your saying makes sense. The only other concern I have is, with such a big jar insulating my probe and with such infrequent runtimes, the ambient temperature in my keezer must be getting pretty warm, and in turn so are my lines. This doesn't bother my pour (I can toss the first ounce if necessary) but is it a concern for infection or can lines also be considered sanitary?

 

[Zepth]

I would say the lines are "clean enough" that you don't need to worry about it. The only opening into the system is at the faucet, and even then it's a system under pressure. They're also being constantly flushed every time you pour.

As far as cleaning and sanitizing goes between kegs, that's a no brainer. As long as it's connected, I have no worries.

 

[IslandLizard]

Thanks Island Lizard. Although you've soothed the concern of my freezer dying, your statistical analysis fails to include the fact that my poor luck with electronics severely skews our bell curve towards failure. Jokes aside, what your saying makes sense. The only other concern I have is, with such a big jar insulating my probe and with such infrequent runtimes, the ambient temperature in my keezer must be getting pretty warm, and in turn so are my lines. This doesn't bother my pour (I can toss the first ounce if necessary) but is it a concern for infection or can lines also be considered sanitary?

+1^ on lines being clean.

I understand your concern about luck and electronics. My 4-year young Samsung kitchen fridge (side by side) falls in that category and is under constant supervision now, after the fridge side froze up once. It's one of the few brands that uses double evaporators. I found and replaced one of the thermal sensors that had failed prematurely. Yet I have the feeling something else isn't quite working right either, it runs too much IMO.

For peace of mind, if you have a second thermometer/sensor or a Raspberry Pi or so, you could monitor or even plot the temp of the "ambient air" inside the keezer, at different locations. But I doubt the local temps go up that high. There's a lot of cold mass in there (filled kegs). If you happen to see lots of foam from the taps, that could indicate warm lines, and/or too much serving pressure. Generally the top part of the keezer will get warmer first, but the fan circulates cold air from the bottom to compensate. And the beer that is being poured comes from the bottom of the kegs, most likely the coldest area in the system.

One thing about fans, they do generate some heat, sometimes enough to function as the sole "heat source" in fermentation chambers, aside from imperfect insulation. You can calculate the wattage of your fan. P=V*A.

How's the beer?

 

[jmhbutler]

This all makes sense, pour and temperature is perfect. The only concern is if I don't pour a beer for a while the first couple ounces I get has an off, oxidized (wet cardboard) taste then everything else runs smoothly. I'm running about 7ft of Bevlex200 lines which have been recently cleaned and have a small computer fan going as well. Any thoughts on an explanation? I thought perhaps poor practice has lead to minor oxidation which is only detectable in beer that has warmed up in the line...

 

[number40fan]

Kenmore 7.7. Temp kicks on at 39 and off at 38 with a cool down to 36.2. Off for 2 hours and 50 minutes and on for 10 minutes. 2x10 collar with three taps out of the side. STC sensor in a pint jar of water.