Trunk line cooling

[Merkur]

I have two Komos kegerators which provide smack for eight kegs. I have six on a tower, one on hand-pump and the eighth smack is taken up with a keg of water with an submersible pump feeding 38F water up a home made trunk line to the tower. It works reasonably well but the water reservoir remperature increases over time to the high 40s and the kegerator that holds the water reservoir has to work harder than the other one.

I am looking to improve the design and have some ideas. A glycol line chiller is out of the question.

1) The water return line from the tower is not insulated and runs close to the back of the kegerators that get hot. I can easily insulate the return line which will keep the reservoir keg cooler.

2) I just ran a test with a similar submersible pump in a one gallon jug of water. The water started at room temp (72F) and over a couple of hours increased to 81F. I can replace the submersible pump with a regular pump exterior to the kegerator as the reservoir is obviously picking up heat from the pump.

3) I can add some 'water wetter' to the reservoir to increase the flow and more easily cool the product lines.

Any other ideas? The kegerators are under a 2" concrete bar top so cooling the tower using cold kegerator air is not an option.

Paul

 

[cfrazier77]

Is the pump size bigger than it needs to be? You could try a smaller one. And, insulation on your return line would help.

 

[IslandLizard]

! Definitely insulate the chiller return line, including the portion that gets heated by the backs of the kegerators.

Perhaps also have a small (computer type) fan (or 2 of them) blow the warm air away from the back area of the kegerators.
Insulate the (inside of) the tower too, if you can.
That all should help the pump work less hard, and so its kegerator too.

Does the pump run 24/7 or is it on a temp control?
Is the keg with chiller water mostly full or only partially filled?

BTW, your setup looks wonderful!

 

[Merkur]

! Definitely insulate the chiller return line, including the portion that gets heated by the backs of the kegerators.

Perhaps also have a small (computer type) fan (or 2 of them) blow the warm air away from the back area of the kegerators.
Insulate the (inside of) the tower too, if you can.
That all should help the pump work less hard, and so its kegerator too.

Does the pump run 24/7 or is it on a temp control?
Is the keg with chiller water mostly full or only partially filled?

BTW, your setup looks wonderful!

Thanks. Yes the tower is insulated pretty well and the cold water feed pipe is in contact with the beer lines up to the back of the taps. The water is in a corny keg in the fridge which holds 5 gallons and I don't believe the pump is oversized. It's one similar to this - sold to circulate glycol in chillers.

The computer fans are a good idea and I have thought about them - mainly to keep the sides of the Komos kegerators cool as that is where the condensers are located. I have been looking for a quiet tangential fan about 2' long for that.

 

[Merkur]

 

[Broken Crow]

That is a gorgeous area you've built there, kudos! Apologies if my wording is awkward, I'm not trying to be confrontational but:

I am looking to improve the design and have some ideas. A glycol line chiller is out of the question.

..is kinda like saying; "I have this problem, but the time-proven solution is out of the question". I can see that space is at a premium so my first thougt is simply you don't have the space for a glycol chiller,..if that's the case, what is under the floor? Do you have a basement or crawlspace you can use? Insulating your return line will likely help, as others have mentioned, but can you tell us specifically why a glycol-chiller is out of the question?

 

[Bobby_M]

It is true that glycol is the solution even if it's a huge pain in the rear to achieve. The compressor designs in kegerators and refrigerators are to run efficiently with almost no external heat load aside from maybe the introduction of one warm keg out of four and the fact that it can take hours to get that heat out of there. Running a cooling loop is just constant new heat coming in.

The other issue is that you want the tower, lines, faucets, etc to be at about beer serving temps; 38F. In order to get that, you can't cool it with the same temp fluid. Running 28F glycol in the loop allows for a temp delta that will actually move heat.

 

[cfrazier77]

It is surprising how much heat a pump makes. This little pump recirculates the glycol in my 3 gallon chiller tank. It will heat that 3 gallons to 85 degrees over time, with the chiller off and the fermentor pump off. It is in a 60 degree basement.

Insulating the return line will help and getting it away from the rear of your kegerator is possible. But, you may want to try a smaller pump than that to see if it makes a difference.

 

[Merkur]

That is a gorgeous area you've built there, kudos! Apologies if my wording is awkward, I'm not trying to be confrontational but:

..is kinda like saying; "I have this problem, but the time-proven solution is out of the question". I can see that space is at a premium so my first thougt is simply you don't have the space for a glycol chiller,..if that's the case, what is under the floor? Do you have a basement or crawlspace you can use? Insulating your return line will likely help, as others have mentioned, but can you tell us specifically why a glycol-chiller is out of the question?

Understood. Good question. This is the basement so I cannot put anything below the bar. Furthermore it's a Townhouse with neighbors the other side of the wall so I don't have the option of putting a 'doghouse' housing a chiller the other side of the wall. I can make room for a small chiller but they are noisy and this is a bar so I want to keep it quiet. I have a Penguin chiller in the brewery next door and I have to close the door to the bar when that comes on. A silent Peltier cooler like the small wine fridges use may help.

Paul

 

[Bobby_M]

The one thing we didn't really bring up is that a fridge pulls heat from the air, then the air pulls heat from your water tank. That is a very inefficient process so that tank is always going to be quite a bit warmer than the fridge temp. The way that glycol units deal with that is they have the evap coil submerged in the glycol (cooling fluid) so that the heat gets pulled directly out of liquid into metal with no air gap transfer. I don't know what you'll do with that info, but it's just an explanation for why tank inside a fridge cooling systems are weak. Maybe an inline aquarium chiller would help. They don't have reservoir tanks, just chilling liquid inline.

 

[Merkur]

I had no idea that there was such a thing as aquarium chillers. That might not be a bad idea. I could replace the keg of water in the kegerator, with all the drawbacks of air to water temperature loss, with a small aquarium cooler. The amount of water in the lines is small - probably in the order of 1 pint so my gut feel says if the cooler has say 4 times that capacity and can keep it at 37F, it would work.

I will look into this.

Thx.

 

[day_trippr]

Look into the minimum temperature an aquarium chiller supports. I'm finding that value is typically 39°F, which would be borderline as the loop will gain temperature from that output level, plus the unit would be running at its limit...

Cheers!

 

[Merkur]

It is surprising how much heat a pump makes. This little pump recirculates the glycol in my 3 gallon chiller tank. It will heat that 3 gallons to 85 degrees over time, with the chiller off and the fermentor pump off. It is in a 60 degree basement.

Insulating the return line will help and getting it away from the rear of your kegerator is possible. But, you may want to try a smaller pump than that to see if it makes a difference.View attachment 871124

That's a very good point and kind of backs up what I have found - running the pump generated heat which will heat the water and make the Kegerator run more. A better solution would be a non-submersible pump that is located outside the kegerator.

In England, where I am from originally, pubs have 'shelf coolers' that chill the beer a few feet from the point of sale. These are glycol chillers with a loop of line immersed in the glycol. Similar in concept to a jockey box. These aren't as noisy as the homebrew glycol chillers made for chilling wort in a fermenter. I haven't found an equivalent in the US and kegs are typically kept cold in a cold room rather than a pub basement.

 

[cfrazier77]

That's a very good point and kind of backs up what I have found - running the pump generated heat which will heat the water and make the Kegerator run more. A better solution would be a non-submersible pump that is located outside the kegerator.

In England, where I am from originally, pubs have 'shelf coolers' that chill the beer a few feet from the point of sale. These are glycol chillers with a loop of line immersed in the glycol. Similar in concept to a jockey box. These aren't as noisy as the homebrew glycol chillers made for chilling wort in a fermenter. I haven't found an equivalent in the US and kegs are typically kept cold in a cold room rather than a pub basement.

Kegland makes one, it does not need glycol and chills two beer lines.
https://www.williamsbrewing.com/Hom...iZxtOV4jpwZlxkt9vzEAlRi497pfHMGWTBcR_CPhTR_6o

 

[Merkur]

Kegland makes one, it does not need glycol and chills two beer lines.
https://www.williamsbrewing.com/Hom...iZxtOV4jpwZlxkt9vzEAlRi497pfHMGWTBcR_CPhTR_6o

I have better things to spend my money on in the brewery! $900 to maybe gain another couple of degrees of cooling is a lot!

 

[cfrazier77]

I have better things to spend my money on in the brewery! $900 to maybe gain another couple of degrees of cooling is a lot!

Agreed, that is for if people are not using a fridge.

 

[IslandLizard]

Agreed, that is for if people are not using a fridge.

I had never heard of that before, so needed to see what that is.

Well, reading through the ad hubbub, it turns out... that unit works like a refrigerator!
Isobutane (R600a) is its coolant. It contains a compressor, and an evaporator as in any regular fridge.

The difference with a fridge is, it chills the beer as it's being dispensed, going through the machine. Basically being an electrically powered jockey box.

 

[cfrazier77]

I had never heard of that before, so needed to see what that is.

Well, reading through the ad hubbub, it turns out... that unit works like a refrigerator!
Isobutane (R600a) is its coolant. It contains a compressor, and an evaporator as in any regular fridge.

The difference with a fridge is, it chills the beer as it's being dispensed, going through the machine. Basically being an electrically powered jockey box.

Bingo, these are not a new thing. When soda came premixed in 5 gallon kegs, we are very familiar with those, there were dispenser machines with taps and cooling coils that went through a water bath. Here are two examples from different decades.
https://www.worthpoint.com/worthopedia/cornelius-premix-750-soda-fountain-458344089
https://www.fiftiesstore.com/coca-cola-premix-dispenser-machine-very-rare-original.html

There are many ways to get to a goal, cold fizzy drinks dispensed. I don't recommend this way, the beer will age faster and getting the co2 levels right would be a headache. But, I enjoy learning about stuff like this.

 

[Broken Crow]

Dunno if this might fit well as a solution, but it seems like it might be a useful option:
https://www.rigidhvac.com/mini-water-chiller

 

[sibelman]

Dunno if this might fit well as a solution, but it seems like it might be a useful option:
https://www.rigidhvac.com/mini-water-chiller

That's a cool product (oops, pun now intended!). And tiny! It requires a fairly hefty DC power supply, likely pushing the total cost over $600. I wonder how much sound this little guy makes. At 1880 BTU, it could be a fine alternative to a purpose-built glycol chiller, no?

 

[Broken Crow]

That's a cool product (oops, pun now intended!). And tiny! It requires a fairly hefty DC power supply, likely pushing the total cost over $600. I wonder how much sound this little guy makes. At 1880 BTU, it could be a fine alternative to a purpose-built glycol chiller, no?

Yeah, I was wondering about the sound too but what most grabbed my attention is that this unit could fit nicely in the cavity we see in the above pictures behind the kegerators....If it's noisey, it's not really that hard to build a vented acoustic housing. Just a thought and yeah; $, but it might be worth it in the long run....just needs some more investigation.

 

[shoengine]

I'm not sure how the math works out, but my fridge cools drinking water with nothing more than a coil of hose in the body of the fridge on a wall. You could put an inline pump outside of the fridge to keep the heat from affecting your chilling capacity.

It is a lot less thermal mass than a tank reservoir, but as long as the air temp doesn't warm the lines more than the loop can cool it, it may work.

 

[Broken Crow]

I'm not sure how the math works out, but my fridge cools drinking water with nothing more than a coil of hose in the body of the fridge on a wall. You could put an inline pump outside of the fridge to keep the heat from affecting your chilling capacity.

It is a lot less thermal mass than a tank reservoir, but as long as the air temp doesn't warm the lines more than the loop can cool it, it may work.

@Bobby_M points out the issue in post #7. That's what put me on the search for an alternative. Owing to brain-injury I can only deal in simple data-sets, but all the math required to work the problem was recently provided by @doug293cz in this other thread: https://www.homebrewtalk.com/threads/sizing-rims-heating-element-for-mashing-scenarios.736014/
The compact chillers I linked to claim to chill down to -4°C and given that it's only chilling the line with losses on the way and within the tower, it seems like it might be an ideal solution..with a mix of water and glycol anyway as per:

Running 28F glycol in the loop allows for a temp delta that will actually move heat.

Honestly I'm just spitballing and hoping someone with a working brain can do the math... I just think that @Merkur has a sweet looking setup and it'd be worth the time and dollars to have it fully optimized.

 

[Bobby_M]

I'm not sure how the math works out, but my fridge cools drinking water with nothing more than a coil of hose in the body of the fridge on a wall. You could put an inline pump outside of the fridge to keep the heat from affecting your chilling capacity.

It is a lot less thermal mass than a tank reservoir, but as long as the air temp doesn't warm the lines more than the loop can cool it, it may work.

That only works because a glass's worth of water sits in there for a while/hours and then you put it in your glass. If you fill 3-4 glasses rapid fire, it comes out warmer and warmer. When you're always pumping it out of the fridge and adding heat to it, it can't keep up.

The one design that would have some promise is if you could get a tall thin rectangular tank that you could slide behind the back evap coil "plate" on the back wall of the kegerator so that plate was partially or even fully submerged. The glycol would get VERY cold if it were only passively cooling the air in the box. When you think about it, the current iteration has the fridge air cooling the tank of water and then the evap coil is taking the heat out of the air but only down to 36F. The tank can't possibly ever be colder than 36.

Doing it with coil submerged in the tank, the coil takes heat directly out of the water/glycol very efficiently and then that is cooling the trunk line AND taking heat out of the air of the box until the thermostat senses 36F. By rule, the glycol will always be colder than 36F.

 

[Merkur]

That only works because a glass's worth of water sits in there for a while/hours and then you put it in your glass. If you fill 3-4 glasses rapid fire, it comes out warmer and warmer. When you're always pumping it out of the fridge and adding heat to it, it can't keep up.

The one design that would have some promise is if you could get a tall thin rectangular tank that you could slide behind the back evap coil "plate" on the back wall of the kegerator so that plate was partially or even fully submerged. The glycol would get VERY cold if it were only passively cooling the air in the box. When you think about it, the current iteration has the fridge air cooling the tank of water and then the evap coil is taking the heat out of the air but only down to 36F. The tank can't possibly ever be colder than 36.

Doing it with coil submerged in the tank, the coil takes heat directly out of the water/glycol very efficiently and then that is cooling the trunk line AND taking heat out of the air of the box until the thermostat senses 36F. By rule, the glycol will always be colder than 36F.

This has got me thinking now. I agree that the only way the colling water in the keg is going to get cold, is by the evaporator plate cooling the air and then that air cooling the keg and eventually the cooling water. Very inefficient. I don't think there is any room behind the Evaporation plate on the back wall of the Komos, but that plate gets mighty cold. Enough for a frost to form so <32F. There are also six screws holding it in place. I could make up a coil of copper pipe from and using a couple of those mounting screws holding it in contact with the Evap plate. Then the cooling water (which would have to have glycol added at that point) could circulate through that cooling loop. The keg of cold water could be eliminated potentially.

 

[IslandLizard]

I don't think there is any room behind the Evaporation plate on the back wall of the Komos, but that plate gets mighty cold. Enough for a frost to form so <32F. There are also six screws holding it in place. I could make up a coil of copper pipe from and using a couple of those mounting screws holding it in contact with the Evap plate.

I'm not familiar with Komos kegerators, and how they're constructed. But they are basically refrigerators, with modifications.

Just be careful when working with the evaporator system. You've got to make sure there's enough slack in the connecting (copper) tubing before (gently) pulling the evaporator plate forward and tinkering with it.

Before doing this, any ideas on how you could fasten the copper coil to the plate, to ensure there's close contact, without compromising the cooling system?

 

[mac_1103]

I don't really have anything to contribute, except to say that this seems like an awful lot of trouble to go to just to have a chilled water tap. Obviously I must be missing something. Maybe if I keep following the thread I'll figure it out eventually.

 

[IslandLizard]

this seems like an awful lot of trouble to go to just to have a chilled water tap. Obviously I must be missing something.

Sorry, yeah, you missed that the OP is operating a trunk chilling line here. Not a chilled water tap, although that could be added.

It's also in the thread title.

 

[shoengine]

I'm not familiar with Komos kegerators, and how they're constructed. But they are basically refrigerators, with modifications.

Just be careful when working with the evaporator system. You've got to make sure there's enough slack in the connecting (copper) tubing before (gently) pulling the evaporator plate forward and tinkering with it.

Before doing this, any ideas on how you could fasten the copper coil to the plate, to ensure there's close contact, without compromising the cooling system?

A copper spiral could lay flat against the plate and adhered with a thermally conductive adhesive.

 

[Bobby_M]

I have a Komos at home and keep failing to take a look at it. Maybe if the screws are all loosened, you could very carefully let it hinge away enough to get a coil of copper behind there. I don't have any idea how cold the glycol solution would actually get, but I predict it will be colder than any water leaving a tank in the airspace.

 

[Merkur]

Before doing this, any ideas on how you could fasten the copper coil to the plate, to ensure there's close contact, without compromising the cooling system?

I am thinking that the input & output of the copper tube will be held by P-clips under the head of two of the screws that hold the Evap plate. If I straighten the malleable copper tube, it should be in contact with the plate or within a couple of mm. I could get some of that heat conducting foam tape that is used for 3D printer beds or under-floor heating. The Evap plate itself is not flat as it has passageways molded in it for the refrigerant. I could put a slight set on the tube so that when the screws are cinched down the tube is in contact with the plate.

I could drill through the Evap plate in between the pipes and hold the copper tube in place with zip-ties but that seems risky. I could also JB Weld the tube in place but that could get messy.

 

[Bobby_M]

I'm assuming you would just use an inline pump rather than a submersible and just run it as a continuous loop. You'd need to bleed the air out of the system but that would give you space back for the extra keg.

 

[Merkur]

This is what I am thinking to start with. An externally mounted inline pump will circulate the glycol.

It's a difficult balance of cooling loads. This kegerator already runs harder (more often and hotter) than the other Komos I have alongside it. I could add a much longer copper cooling loop and undoubtably that would result in lower tower temperatures, at the expense of putting more load on the kegerator's refrigeration system - which wasn't designed to handle it. The good news is that the kegerator is rated to work at ambients up to 80F and this is in a basement at around 68F.

 

[Broken Crow]

I could add a much longer copper cooling loop and undoubtably that would result in lower tower temperatures,

still just spitballin but I gotta say; that coil on coil idea sounds brilliant, in terms of efficiency though, I'd go with more mass, ie; @shoengine s coil and thermal adhesive suggestion and top it of with a temp controller, maybe an STC1000 coupled with a diaphram pump: https://www.brewhardware.com/product_p/supersucker.htm connected via a tee https://www.brewhardware.com/product_p/tee12.htm and a couple compression fittings https://www.brewhardware.com/product_p/comp38tx12mnpt.htm with a temp sensor https://www.auberins.com/index.php?main_page=product_info&cPath=20_131_15&products_id=249 so that it doesn't run 24/7. ..if noise is an issue, take a look at what @stamandster did: https://www.homebrewtalk.com/threads/transfer-pump-housing-diy.734201/
Whatever you finally do...please keep this thread updated... you got me invested

 

[Merkur]

I have the existing system plugged into a WiFi outlet that turns off every day at midnight. I turn it on manually when I need the cooling.

Once again, this is with the existing system that uses 5 gallons of water in a corny keg and a submersible pump providing the trunk line cooling. When I turn on the pump, the corny keg has 5 gallons of 39F water in it and it will chill the inside of the tower down from 72F to 51F in 10 minutes. The problem is that after an hour the reservoir temperature has increased to 50F and the tower temp starts climbing.

 

[cfrazier77]

Here is an idea. Keep the keg, put the pump outside the kegerator.
https://www.amazon.com/bayite-BYT-7..._csd=d2lkZ2V0TmFtZT1zcF9kZXRhaWxfdGhlbWF0aWM=

Then add a small radiator with computer fans to the output and input of the keg, or at least one of them.
https://www.amazon.com/DIYhz-Comput...29&sprefix=small+radiator,aps,174&sr=8-4&th=1


Moving the pump out will give you a couple of degrees and the radiators may also as well, getting you to 40 without modifying the kegorator.