My Keezer Alternative: The Electric Jockey Box

[ryan_george]

Backstory:

We have started brewing bigger batches and as such have moved to using larger (50 L) kegs for ease of storage and reduction of cleaning. The issue with this is that my keezer cannot fit very many of these. I’m used to being able to hold 6-7 20 L kegs in the keezer, but now I can only get 2 big kegs in there… This seriously limits its flexibility and has hindered my beer selection.

Current keezer:



I didn’t want to buy bigger/more chest freezers because they are big, ugly, expensive to run, and require a lot of heavy lifting to get full kegs into it.

Several months ago I asked the forum about the feasibility of a system which uses a window AC unit to chill the water bath in a Jockey Box (https://www.homebrewtalk.com/f51/possible-kegerator-alternative-glycol-cold-plate-428673/). I didn’t get a lot of feedback from that which meant either: it was a dumb idea, or, nobody had really done something like this before. I was hoping it was the latter.

After a lot of thinking and planning, I decided I would give it a shot. There were some unknowns that could only be answered by testing it in real life.


The Build:

The idea for the project is this: I want to build a jockey box with a cold plate, and use the evaporator coils from an air conditioner to chill the water (or glycol mix) bath. I would control the temperature of the water bath with an Arduino to give the most in terms of flexibility. I will store the kegs in my basement crawl space at the ambient ~10-15°C (50-60°F). I’ll need to increase my CO2 pressure to adequately carbonate at this temperature, but you generally need to run higher pressures anyway when pushing through a cold plate.

I bought a used AC unit for $25:




Started taking it apart and bending the evaporator coil out:




Fitting it into the jockey box cooler:




Making sure it still worked properly after my modifications:



I have placed 2 small aquarium pumps in the bath to ensure the water keeps moving.


Here’s the final, assembled cooling system controlled by an Arduino. I made a manifold on the cooler lid with flared bulkhead fittings so I can easily attach my in/out lines to the cold plate. I have 6 circuits total in this system. I only have one circuit hooked up so far as my proof of concept.




Here’s the back side of the wall where my taps will come through. I have not found a need to chill the lines between the cooler and the taps, as you would with a typical keezer trunk line. This is because the beer is carbonated at the ambient temperature so there is no real tendency for the CO2 to come out of solution in the lines. I no longer have to worry about being able to fit all my kegs into the keezer, because I have tons of room in the crawl space. If I get desperate, I could even build shelves to stack the kegs. This gives me the ultimate flexibility that I wanted.




I decided to spend the money and get the Perlick 525ss faucets. I then made a backsplash from aluminum, brushed with a scotch-brite pad, and made a drip tray for everything. The inaugural first pour:






I have the system set to 5°C (40°F). I have not actually measured the temperature of a pint that comes out, but it feels/tastes just about perfect! With the Arduino I’m calculating the duty cycle that the AC unit is running at, and it is only on about 2% of the time. This is because (in addition to ambient losses) it only has to cool the beer that has passed through it.

As you can see there are still some empty taps, but I am working to get them filled up The end goal is to have 5 regular taps and one nitro.

Cheers

 

[iijakii]

Fantastic!

 

[Jmarsh544]

This is pretty cool.... Very interesting implications....

 

[ryan_george]

I've now got the remaining taps installed and the rest of the plumbing done. I've moved all of the kegs from the keezer over to the new system, so I now have 5 of the 6 taps pouring.








I've put an English Mild on the "nitro" tap for now. I use "nitro" in quotes because it's still just hooked up to CO2, but I'm not even sure if I'll bother using nitro or beer gas... It pours a wonderfully creamy head as it is (although it must sit for a few seconds to cascade down).

Now I just need to make some more tap handles, and it's pretty much done! Further down the road I'll be building this into a simple dry-bar with some cabinets and shelves, but for now this will do the trick

 

[njs170]

Looking good! Do you notice any differences between the standard setup and using a cold plate?

 

[ryan_george]

There are no obvious negative differences between this and my keezer. One issue I do foresee is that when summer comes, the basement will likely be a bit warmer. Which will screw up the carbonating pressure a bit. But we'll see how that goes....

 

[Cromwell]

This is fascinating. Great job, first of all - looks awesome.
But it raises a question for me. I'd never considered taking the cooling coils out of an A/C and putting them in water. I built a big lager chamber with an A/C in the side to cool it, and it works, but the A/C cycles pretty fast. It cools the chamber down quickly - like in 3-4 minutes since it's small, and turns off. But then the chamber heats up again in about 7-8 minutes. So the thing is constantly cycling. Yes, I should look for leaks.
But I'm wondering if I put the AC coils in a water bath, if it would run longer to cool the water, and then stay off longer because the heat mass of the water would be much larger?

Anyone think this is a good or bad idea?

 

[ryan_george]

This is fascinating. Great job, first of all - looks awesome.
But it raises a question for me. I'd never considered taking the cooling coils out of an A/C and putting them in water. I built a big lager chamber with an A/C in the side to cool it, and it works, but the A/C cycles pretty fast. It cools the chamber down quickly - like in 3-4 minutes since it's small, and turns off. But then the chamber heats up again in about 7-8 minutes. So the thing is constantly cycling. Yes, I should look for leaks.
But I'm wondering if I put the AC coils in a water bath, if it would run longer to cool the water, and then stay off longer because the heat mass of the water would be much larger?

Anyone think this is a good or bad idea?

I think you could solve your problem by simply placing the temperature probe into a volume of water. The chamber would heat up the water slower than the air, and similarly the air would warm the water slower as well.

This would achieve your goal of adding some thermal damping to the system, without the complexity of tearing apart an AC.

 

[jrm21086]

This is really a great idea and I'd love to do something similar, but in a portable version. Do you have any pictures of the inside of the cooler once you are done? I'd imagine you would have lengths of tubing inside the water so that by the time it reaches the tap it is cold.

EDIT: I see the cold plate now. I understand better how it all works but still would love to see more pics!

 

[BobzillaBrew]

Does the beer serve cold? It seems it might be better to have the jockey box closer to the taps otherwise the first pour will be at the temp of your crawl space.

Any idea on the efficiency or electirc cost of using an ac unit like this?

I have a small keezer now that will fit 3 corny's. Do you think it would work to fill the bottom of the keezer with the water glycol and put the cold plate in there? That way the plate gets cold and the airspace above keeps the beer lines cold. also leaves room for cold storage. As small as the cold plates are you could even stack them and have 12 taps in your living room if you wanted to. thoughts?

 

[ryan_george]

Yes, the first amount drawn from the taps is at crawl-space temperature, but this amount is very negligible... there's about 5' of 3/16" line between the cold plate and the tap, which calculates to 27 mL. Not enough to worry about A glass (or even partial glass) is at a nice drinking temperature.

As far as efficiency, the Arduino is monitoring the duty cycle of the A/C unit. Since it's been hooked up (3-4 weeks now) it is reporting an overall duty cycle of 1.9%. You can do the math on this based on your electricity cost and what your keezer is using.

Your idea of a cold-plate in a water bath inside of a keezer will likely work for most scenarios. The only downside is the poor heat transfer rate from the freezer walls, to the air, to the water bath. I cut out the middle-man (the air) and cool the water directly. But this would only be a problem if you were drawing off a ton of beer through the cold plate without giving enough time for the water bath to drop in temperature.

 

[ryan_george]

Here's a current shot of the inside of the cooler. You can see all the in/out lines running to the cold plate, with the AC coils resting on top of the cold plate. The blue lines are my temperature probes (one mounted to the coil to detect if ice has built up, and one measuring the water bath temperature. There are two small aquarium pumps loosely placed in there which provide some circulation of the water.

And all that crud in the top right corner seems to be dust/dirt/etc that has made its way out of the used AC coil. Thankfully that stuff never touches the beer

 

[jrm21086]

Thanks for the reply. Looks great. It looks like you have the AC coil partially out of the water. Is that just to take the picture or does it sit like that all the time? I have a few old AC units that are itching to be converted!


Sent from my iPad using Home Brew

 

[ryan_george]

Thanks for the reply. Looks great. It looks like you have the AC coil partially out of the water. Is that just to take the picture or does it sit like that all the time? I have a few old AC units that are itching to be converted!


Sent from my iPad using Home Brew

Yes, the evaporator coil is sitting slightly our of the water. I forget the reason why I didn't fill it up more I'll put that on the to-do list!

 

[psubrewer]

Great build!

How many feet of beverage line need to sit in the cooler to achieve the correct temperature at the tap?

Edit: nevermind. I see the cold plate in the bottom. Makes sense now!

Sent from my SPH-D710 using Home Brew mobile app

 

[chad041]

I assume there is a reason for building it this way, but couldn't your just put a cold plate in a small temp controlled freezer and achieve the same results?

 

[Stealthcruiser]

I'm thinkin', because of temp transfer efficiency issues............Water conducts heat about 25 times better, ( or faster), than air, best I remember from my Scuba divin' days.

 

[njs170]

How is the 40F set point working out for you? Have you really pushed the system yet with a party or something? Pint after pint?

 

[ryan_george]

Sorry for the slow reply. We haven't been able to really push the system yet with a lot of throughput. Next weekend there will be some heavy usage, though. I imagine it will just cause the AC to run a bit more.

I've made some more tap handles to fill up the rest of the taps. Also added some more kegs to the inventory. Still waiting to figure out what to do for a more aesthetically pleasing labelling system!

 

[ryan_george]

Thought I'd post an update about this. The system has been working quite well, although some of the lines tend to develop gas pockets if left sitting for a long time. This is likely due to the difference in temperature between the cold floor and the ceiling of the crawl space (where the lines are coiled) causing some CO2 to come out of solution. I'm going to add some small PC fans in there to get some air movement.

It has seen a few parties with a good through-put, and there have been no complaints about warm beer. The last stress test I gave it (and probably the most harsh), was filling 12 growlers at once. The tap was essentially open continuously for 6 gallons (~23 liters) of beer. The beer started flowing at 5C (the water bath temperature) and after the last growler it was coming out at 7C. Not too bad, in my opinion...

 

[philipCT]

Okay, so the system is very impressive, and I'm also impressed with the tap handles. Can you tell me how you make those and what you're making them from (looks like old CO2 carts). It's a nice, simple look.

 

[ryan_george]

Thanks for the kind words. The tap handles are just 1" aluminum rod, worked on a lathe to add the round end and chamfer at the tap. Then drilled and tapped to screw on to the tap.

 

[philipCT]

^^skills^^

 

[Indygunworks]

Back to the top... with the same question other has asked.... Could this be done with a freezer? I understand the water to air exchange is not efficient but I was considering a super small freezer and filling it up with treated glycol/antifreeze.... that would have full contact with the sides and should allow for proper heat transfer.... It would also be a significantly larger thermal mass which would take longer to heat up. I would want to wire my pumps in with the compressor, or attached to the thermal adapter so they only run while the compressor is running...... Or possibly hook one up on the exchanger/s to a timer of some sort.... that would save on it running 24/7

 

[ryan_george]

I don't see why that wouldn't work for a "casual" drinker at home. The more thermal mass the better....

 

[mcgeebc]

I apologize for bumping a really old thread. I've been scouring the internet for a setup like this. Wondering from the OP if this is still working? Has anyone else tried this or similar?

I can fit 2 kegs in my current kegerator. I'd like to expand to 4-6 and run taps to the wall on the other side of my brew room.

I also have a new Spike CF5 and looking for temp control options.

I could:
1) Build a keezer and run air cooled lines to the wall for taps AND a buy a standing freezer for my CF5
2) Build a DIY glycol chiller. Create a glycol chilled jockey box and use the chiller for my CF5.

The glycol chiller and jockey box would take up less space, allow for expansion of other conicals in the future and would likely be cheaper. I already have a few cold plates and could source a used window A/C on the cheap.

My concern is that this seems approach rather unproven. I couldn't find many references to people doing this. Just this post and reference to setups in Japan that are similar in approach. Again, looking for other people that have attempted this or some insight into downfalls I'm not thinking of.

 

[BeerBoss]

WoW!!!...
That is alot of work!!!.. how much has this build cost you?

 

[ryan_george]

I apologize for bumping a really old thread. I've been scouring the internet for a setup like this. Wondering from the OP if this is still working? Has anyone else tried this or similar?

Sorry, missed this post from a few years ago

Yes, as of Christmas 2022 the system is still working as originally built. No modifications or replacements.

WoW!!!...
That is alot of work!!!.. how much has this build cost you?

I can't remember the budget for this (from 2014), but it wasn't a lot. The most expensive item was likely the cold plate.

Okay, so the system is very impressive, and I'm also impressed with the tap handles. Can you tell me how you make those and what you're making them from (looks like old CO2 carts). It's a nice, simple look.

For a shameless plug.... I make these now and have them advertised on Etsy (Beer Tap Handle Machined Aluminum Made in Canada - Etsy). I can also make other variants or custom parts if there is interest.