Solid Ice Recirculation Chiller

Hey guys, I thought I would post my own thread so I can start to get the skinny on recirculation chilling. I currently use a Therminator and regular tap water to get my kettle down to tap temperatures of anywhere from 60-75*F. This works perfectly for ales and I am happy, but I waste a lot of water. I have been researching threads and websites about recirculating ice water baths and they all seem great except the fact you have to use so much ice, and in most cases you have to use tap water to get under 100*F before you start using the ice bath. I am wondering about another approach.

What if you could freeze the entire cooler you were going to use for chilling and make it so that your pump had a place to fit down into the solid ice block. If you fed the return to an opposite side it would have to run over ice before reentering the pump and then the Therminator. I am more interested in getting below 140*F fast, but secondarily I would like longevity out of the chiller/cooler combo and wouldn't care if it took longer to get to lager temps as long as I could get there on one cooler. What do you guys think? I have been slammed with ideas from the a/c glycol chiller reservoirs to the constantly changed ice water baths. Do I have a feasible idea here?


Just for the record and sake of argument, I will recirculate back into the kettle until the final pass would get me to lager pitching temperatures.

I just tested a new setup with similar goals.

I had a 38 quart cooler filled with 50% ice and 50% and a 500 gph pump hooked to an immersion chiller. Then, I hooked up a pre-chiller, but used it for "post-chilling". I set this in a small bucket filled with nothing but ice. The idea being this would take the heat out before recirc-ing into the ice water. The goal here was to use only this water to cool my 3 gallons of wort to 75*.

It worked perfectly with no tap water to start. Using 3 bags of ice total, I dropped 3 gal. to 75* in roughly 17 minutes. The water going back into the cooler was definitely cool enough to not bring down that temperature at all during the cooling. Obviously a larger boil would have worse performance, but i was very happy with these results.

i have two things to add:


1) I think you are underestimating how hard it will be to remove the PVC mold for the shaft. It will be a real pain in the ass to get it out.

2) Performance wise, it will work much better if there were channels/grooves in the top where the water flowed. Ideally, it would be a long snake for the water to channel through. If the water just glides right over the top, I don't think it will cool anywhere near as well as it would if it had to travel down a pre-determined cooling path

I don't think the shaft will be that hard to get out, but I have thought about the channeling. I guess I could rig something up where it has a maze to get through on its way back to the pump.

What Berserker is saying in a roundabout way is more surface area/contact time the cooler the water gets. The best way to increase surface area is to use ice cubes. You could keep the maze idea by setting up some baffles in the cooler that would make the return water follow a circuitous route from it's inlet into the cooler to the pump.

Splitting the cooling between two locations ("post chiller" and main recirculation cooler) is not really buying you anything, your still transferring X BTUs of heat out of the recirc water whether it is in one pile/block of ice or two. You could just as easily have two coolers of ice and two pumps.

Obey Thermodynamics - it's the law!

OK, admittedly I was not an engineering major so I did not have to take thermodynamics - but here is my question. If you cool the output separately (post-chiller)from the source you pump from (ice water) overall wont that keep the source temp lower when the output is finally back in the source? Or at least stave off equilibrium long enough for effective cooling?

I'm no expert, but that was the most efficient cool I have had yet...

If your source container isn't big enough to hold all the ice then sure. I'm just saying that 50 lbs of ice has the ability to absorb just so much heat whether it is in one big (source)container or two containers (source + post chiller). You could even try this: instead of using a pre-chiller set the container you used for the pre chiller (I am assuming a cooler) higher than the source container with the drain open so the water runs out into the source container and just run the outflow from the wort chiller into the upper bucket. That way all of the melted ice water also goes into the recirc.

I agree that ice cubes are better. A higher temperature will form between what you are trying to cool and an ice block. The ice block will be further from the heat source so less speed in cooling. Ice cubes when agitated can continually be against the heat source for increase cooling. Again: surface contact.
Large blocks last longer but are not faster. Ice water(or very cold water) is how the fruit industry chills produce from the field.

First thing. I like the idea of the PVC sump. At the very least it's causing incoming water to spend more time touching the ice before it gets to the pump. I agree there's no reason to remove the pipe... let the water flow up and over the top. however, I'd also mold in a 1/2" tube that take your input water to the bottom left side of the vessel. You'd want to let the block melt for a while prior to use so that there's room for the water to flow around the block. It would also help to add a few cups of rock salt to the whole thing. I wonder how well the pump impeller would deal with 15F water.

However, to the point already made, smaller chunks of ice with so much more surface area would be much more efficient. you actually want the ice to melt as fast as possible. It would be really cool to find some really large cube trays so you can make 4" cubes or something like that. I'm thinking a bunch of 1qt chinese soup containers would be fine also.

After two summers of brewing, I'm at the point where I decided for next summer I'm going to make a similar system out of one of those 13 gallon screwtop barrels. The only different is that it's going to hold some pressure so that I can use hose water pressure to drive past the ice and into the chiller with no pump needed. The huge top opening allows for it to be filled with ice.

With this system I have proposed, I am not wanting quickness except for under 140*F. I want longevity, and do not want to replace/refill/do anything other than pump and wait till my kettle is close enough to get me to 45*F last pass through the chiller. I like the idea of leaving the PVC well allowing water to flow over the top. I know the hot inlet water will dig a lake and probably melt around the ice block allowing water to cool the whole way before going up and over the well. Like I said, I am not looking for fast I am looking for it to last the whole way through. I mean otherwise I wouldn't care to save water or trips to the store for ice (or make it). I think I will monkey with the well idea a little and run some tests on boiling water. As long as I can maintain a cooling water inlet temp below or at my tap temperature I am doing well in my mind. Do you guys agree that I should be at this temperature for all inlet water going into the Therminator??? The magic will happen as the kettle starts to come down in temp and the cooling water should do the same.

You will quickly melt a hole right through the end you pump back too. Just a data point. Even room temp water will melt a block of ice quickly.

MixnMatchBrew said:

You will quickly melt a hole right through the end you pump back too. Just a data point. Even room temp water will melt a block of ice quickly.

Click to expand...

could be a good thing. It would create a nice "cold water well"

That is what I was hoping for, that with the well not being removed around the pump might work great.

Also, the reason I am not going with individual cubes of ice (which I know would get me faster lower temperatures) is because of the fact of surface area. A block of ice will melt considerably slower than cubes. Take a bag of ice and dump it on a table, then take the same weight in a block. Which one will disappear first? I want longevity, not flash cooling.

Give it a shot, but doesn't everyone want flash cooling? The colder that water gets, the faster you chill overall. I wouldn't even put this thing in the cooling loop until your wort is under 140F. Even with 80F tap water, it shouldn't take more than 5 minutes to get there, then switch over to the ice.

Bobby_M I am hoping that any and all water going through the chiller would be below tap temperature so as to have one system for everything. This would get me the same time for under 140*F as now, but then I would have the extra ice there to slowly but surely get me to 45*F.

I'm going to make a similar system out of one of those 13 gallon screwtop barrels. The only different is that it's going to hold some pressure so that I can use hose water pressure to drive past the ice and into the chiller with no pump needed. The huge top opening allows for it to be filled with ice.

Click to expand...

I've been thinking about doing the same thing let me know how your design works (new location the pipes are too close to the surface winter my CFC takes the wort down to 60, summer it has to sit and cool for a couple hours)

OK, so now I am entertaining the idea of a radiator and fan post terminator and pre-reentry for the chiller water. I wonder though how much a rad would take off of 200*F water before it went back on top of the ice?

This is a good thread so I will put my 2 cents in here. Most commercial brewers use glycol in their cooling systems for good reason. The water does not freeze and stays a liquid. You can use it to cool anything including an insulated fermenter or wort using a plate chiller. A home brewer can do this by using a chemical grade plastic container in their freezer with a stainless coil immersed in the glycol and use glycol in the tubing with a pump. You could also use a pump directly in the glycol in the container (imersion pump). The glycol must be food grade glycol! Any other type is poisonous.

I am going to do this in the near future and will not have to buy ice. If you did not want the freezer running all the time you could turn it on the night before brewing to cool the glycol tank and back off when finished.

I live in California and the ground water gets too hot.

Ok, knowing there are other types of chilling liquids out there has been covered in the threads I have been researching. Specifically talking about freezing ice and using it for chilling and then cleaning (since that has to be done anyway) is what I am talking about. I worked in a brewery and know glycol is the best. It doesn't freeze at the temperatures you need it at and it is reusable. For a homebrewer... not so much. I have read the threads on homemade glycol chillers and the reservoirs are too large to accommodate what I am wanting from this project. I even saw the one that had a a/c unit to chill the glycol (which was very very cool, but...).

I am liking the idea of a little rock salt in my solution to melt the solid block and hopefully maintain ice in the cooler throughout the entire cooling process. I need information such as how long a certain size of block would last (with the obvious situation), and if it would cool to below 45*F before the cooler equalized to whatever wort temperature. I would be willing to bet my solid block would last the whole way through compared to a equally sized reservoir of glycol chilled to 14*F. Now I know the glycol would chill faster, but with no cooling power when I am wanting below 45*F due to the whole solution heating up above this temperature (14*F glycol returning would be much hotter just like the water I am talking about and would have no way of cooling back off). Now show me the math and set volume needed for a single pass for boiling to 45*F and I will listen right up. I also know that using water first and then the glycol or ice would work...but that is not the "wanted outcome" of this thread. I am looking for hard facts about what I am proposing. If this works then I have just used the right amount of water (not caring about electricity for making ice or anything else other than that). Like I said before, I will be happy if I can get below 140*F fast and then don't care how long it takes to get to 45*F-ish. I only care that I can get there. If it isn't possible it isn't possible, but I think one could remove the heat from the returning chilling water to where the ice lasts longer and you can use the solid block (without the surface area of cubes) to get down to temp.

Now, I'm thinking that since you get better efficiency out of your cooling water if the returning water is colder. Since I don't want to add any extra water other than what is needed to start pump flow at the beginning, I would like to see what would get water temperature down as best as is possible. I am thinking a radiator type of device that is fan powered. This way no more water and only more power is used. I do not know, but suspect it would get the water temperature down quick when I needed it at the first. Then, it would not be so efficient once the temperatures got closer to air temperature in which case it could be tossed into the cooler or unhooked and not used. Either way, the chiller would rely on the solid ice block and recirculation of wort back to the kettle until it could be at 45*F for the last pass though the chiller.

Ideally what you want to figure out is how much ice you really need in order to reach the wort temp you want without any leftover. One reason why glycol wouldn't be overly advantageous in this situation is the rather inefficient heat transfer of a resevoir being inside a freezer. First, you couldn't possibly hold enough to get your wort down to 45F. Barring doing the math, you'd probably need 20 or so gallons of 14F glycol to chill a 10 gallon batch and you couldn't count on the freezer keeping up to take any appreciable heat away. It's chill evernight, then use it up in one shot.

My gut feeling on the fan/radiator to get the exhaust water temp down is that it's a diminishing return. That's a lot of BS to mess around with in addition to the cooler, ice, pump and chiller. I know it's not exactly eco-friendly but water is really cheap compared to the cost of making ice.