Immersion chiller - what's best - prechiller or recirculation ?

Recently there has been some discussion about using ice water with an immersion chiller to cool wort below tap water temperature. A couple of methods have been discussed: use of a prechiller submerged in a bucket of ice water to chill the tap water going to the IC; and recirculation of ice water through the IC and back to the cold water bucket or cooler by means of a pump. Either of these would presumably only be used after the wort has already been cooled to near tap water temperature by running tap water through the IC. Intuitively, I felt that the recirculation method would make more efficient use of the ice water, but I finally got interested enough to develop some equations in order to get a better handle on which method is best, and by how much. I wrote up my work in a pdf file which can be downloaded from here:

http://www.mediafire.com/?c9mkm19uxjiz9hj

I had to make some simplifying assumptions in order to be able to do this, so the equations developed won't be completely accurate, but should give a good idea of the relative benefit of the recirculation method over the prechiller method. I assumed complete heat transfer in both the prechiller and chiller; i.e., that the temperature of the liquid leaving the coil is the same as the temperature of the liquid in which it is immersed, and I avoided the complications of melting ice by assuming the use of cold water with no ice. I plotted the results of two "typical" situations, each with 6 gallons of wort at 90ºF, tap water at 80ºF, and starting with cold water at 35ºF with the only difference being the use of 4 gallons of cold water in one case and 8 gallons in the other. I hope some of you find this to be useful, or at least interesting. Sorry for the heavy math in the report, but it was unavoidable.

I, for one, do find this interesting. I'm considering both options at the moment. I had a question for you - taking into consideration one of my options. I have a keggle that I'm itching to use for the first time and my only means of chilling wort in the past has been through the use of a 25' x 1/4" copper IC.

To combat the expansion in size from my old kettle to the 15.5g keggle, I'm definitely going to buy/build a 50' IC (using either 1/2" or 3/8" tubing - I also don't know which is more efficient...I'm assuming the 1/2"). I'm considering using the 25' IC as a pre-feed into the 50' IC, where I'd submerge the 25' IC into a salt ice bath. Essentially it'd look like this: garden hose to 25' IC submerged in salt ice bath to 50' IC submerged in wort.

In the above scenario, is the 25' IC what you would call the pre-chiller? I'm just checking to see if my situation might be applicable to your equation.

Thanks!

Yes, that's exactly what I've plotted and labeled as "prechiller". Of course, my calculations didn't take into account such things as salt in the water, or the difference between tubing sizes - to make it possible to analyze mathematically I had to make a lot of simplifying assumptions, but it does make clear that for a given amount of cold water, you can get the wort to a few degrees lower temperature by recirculating that water from the cold water tank through the immersion chiller and back to the cold water tank than what you can get by running tap water through a prechiller then to the the chiller and then down the drain. Of course the few degree difference may not justify the cost of a pump needed for the recirculation method, but if you already have a pump, then maybe recirculation is the way to go.

DeafSmith said:

Yes, that's exactly what I've plotted and labeled as "prechiller". Of course, my calculations didn't take into account such things as salt in the water, or the difference between tubing sizes - to make it possible to analyze mathematically I had to make a lot of simplifying assumptions, but it does make clear that for a given amount of cold water, you can get the wort to a few degrees lower temperature by recirculating that water from the cold water tank through the immersion chiller and back to the cold water tank than what you can get by running tap water through a prechiller then to the the chiller and then down the drain. Of course the few degree difference may not justify the cost of a pump needed for the recirculation method, but if you already have a pump, then maybe recirculation is the way to go.

Click to expand...

The "few degree difference vs. cost" is the thing that I'm weighing...though technically the 25' IC is co-owned by myself and a friend - so it might make sense to get a submersible pump and recirculate. I'm currently investigating the cost efficiency of a 3/8" tube vs. a 1/2" tube.

I'm also assuming that you wouldn't start recirculating the ice water until the temperature in the wort (or at least the pumped water) drops to below 100, correct? I'm guessing that if you immediately started recirculating the water, you'd end up immediately melting your ice and turning your circulating water lukewarm.

You definitely want to wait until your wort is below 100ºF before you start using your ice water - that's why my model starts with the wort at 90ºF. As to the difference between 3/8 and 1/2, you should be able to get the same amount of cooling from either, assuming that you run the water at a speed low enough for complete heat transfer to take place. Presumably the 1/2 inch tubing, or a longer length of 3/8 tubing would do the job faster, but not necessarily colder. I have a 3/8 inch, 50 foot chiller which I have been using with a 3/8 inch, 20 foot pre-chiller. Wort cooling is very fast from boiling down to a little over 100ºF, but in the summer it's very slow going after that due to the tap water temp. I don't think that going to 1/2 inch tubing would help that situation - I either need to get a longer pre-chiller and use more ice, or recirculate. I think I'm going to recirculate now that I have a pump and use my 10 gallon cooler MLT as the ice water tank instead of the 5 gallon bucket I've been using for the prechiller.

My biggest beef with a prechiller is that most people use such short coils that you'd have to run your tap water really slow to get a decent temp delta. If you run 80F water through a 25' coil in 35F water at 3GPM, it's not coming out at 35F or even 40F. No math, just a bold guess from experience. The other issue is that you have to keep that bath moving to keep the warm water away from the coils. Pumping icewater puts that cold icewater into the IC without any stirring and it's always that fixed temp no matter how fast you flow it.

You CAN use it when the wort is still hotter than 100F, but you should run it a little slower and discharge the IC output elsewhere (NOT back into the ice). If you need to makeup water volume, run tap water into the ice.

I agree about needing longer coils on the prechiller - 20 feet is not long enough, in my experience. And I can tell just by feeling the chiller input line that the water is warmer than 40 degrees with my prechiller. As for keeping the bath moving, I am constantly stirring the wort (slowly, but steadily) to keep wort flowing over the coils, and once I put the ice in the prechiller bucket, I use my other hand to agitate the prechiller by lifting it up a little and letting it fall back. Kind of like trying to pat your head and rub your stomach at the same time. So that's another way recirculating would make things easier.
Still seems a good idea to wait to use the ice until wort is down below 100 to get the most benefit from the ice.

I haven't made any fancy graphs but I have used both methods. In my experience it's no contest. Recirculating ice water with a pump works several times faster then using a pre-chiller. Quite honestly, a pre-chiller has a very limited effect for me. But my tap water is way colder then 80F. Even in the heat of summer my tap water never gets above 70F.

Bobby_M said:

You CAN use it when the wort is still hotter than 100F, but you should run it a little slower and discharge the IC output elsewhere (NOT back into the ice). If you need to makeup water volume, run tap water into the ice.

Click to expand...

You could but why would you want to? I think you'd be wasting your ice to use it before it goes below 100F. With either method.

Grasslands said:

The "few degree difference vs. cost" is the thing that I'm weighing...though technically the 25' IC is co-owned by myself and a friend - so it might make sense to get a submersible pump and recirculate. I'm currently investigating the cost efficiency of a 3/8" tube vs. a 1/2" tube.

Click to expand...

If your cooling 10+ gallons I'd recommend 50' of 1/2" refrigeration service copper tubing. The refrigeration tubing is cheaper but it's also thinner. It says 1/2" but it solders up to 3/8" fittings. I think it's really 1/2"OD. The more expensive type L & type M are fatter, thicker walls and much more expensive. For 10+ gallons I find that 50' of the 1/2" refrigeration service copper tubing does the trick. Expect to pay between 80.00 - 120.00 depending on if you roll your own or buy one pre made.

If using 50' of copper tubing, consider building a dual coil chiller running water through two 25' parallel coils, one coiled tighter than the other and sitting inside the other. No math, but those using a single 25' coils say they are getting hot water discharge at the end of the 25' coil. Merely increasing the length of a single 50' coil means you are only running hot water through the last 25' of coil which will have neglible effect to cool your wort. My chiller has 2ea 25' of 3/8" refrigerator coil. My water pumps into a 1/2" pipe which T's into the 2 coils and the discharge T's back into a second 1/2" pipe.

Also, if using a pump, don't bother pumping water into your coils directly from the tap or outdoor faucet and then, when the temp drops below 100, reconnect your plumbing to use your pump and ice water. Why waste the time rerouting your plumbing? Run your tapwater into a bucket with your pump and just adjust the tapwater flow to match the pump flow. When the temp drops below 100 (or whenever you choose), just start dumping ice into the bucket. When the chiller discharge temp gets close to tapwater temp, consider directing the discharge into the bucket and turning off the tapwater to conserve water.

rgray58 said:

If using 50' of copper tubing, consider building a dual coil chiller running water through two 25' parallel coils, one coiled tighter than the other and sitting inside the other. No math, but those using a single 25' coils say they are getting hot water discharge at the end of the 25' coil. Merely increasing the length of a single 50' coil means you are only running hot water through the last 25' of coil which will have neglible effect to cool your wort. My chiller has 2ea 25' of 3/8" refrigerator coil. My water pumps into a 1/2" pipe which T's into the 2 coils and the discharge T's back into a second 1/2" pipe.

Click to expand...

Sounds like a very efficient design

I use a Recirculating system with a pump and no ice. I go from boil to 70f in 15-17 mins using this method. I can hit 60f in about 30 mins times if I let it go that long.

rgray58 said:

If using 50' of copper tubing, consider building a dual coil chiller running water through two 25' parallel coils, one coiled tighter than the other and sitting inside the other. No math, but those using a single 25' coils say they are getting hot water discharge at the end of the 25' coil. Merely increasing the length of a single 50' coil means you are only running hot water through the last 25' of coil which will have neglible effect to cool your wort. My chiller has 2ea 25' of 3/8" refrigerator coil. My water pumps into a 1/2" pipe which T's into the 2 coils and the discharge T's back into a second 1/2" pipe.

Also, if using a pump, don't bother pumping water into your coils directly from the tap or outdoor faucet and then, when the temp drops below 100, reconnect your plumbing to use your pump and ice water. Why waste the time rerouting your plumbing? Run your tapwater into a bucket with your pump and just adjust the tapwater flow to match the pump flow. When the temp drops below 100 (or whenever you choose), just start dumping ice into the bucket. When the chiller discharge temp gets close to tapwater temp, consider directing the discharge into the bucket and turning off the tapwater to conserve water.

Click to expand...

The speed you can cool the wort is dependent on the volume of water flow per minute through the chiller. Splitting the chiller into two shorter coils would reduce the pressure needed to force a given volume through the chiller, but if enough pressure is available to get the same flow rate through the single 50' coil it should do the same job. If you're only running hot water through the last 25' then you either need to increase the pressure or split the chiller in two as you have done.
Running the tap water into the bucket is worth some consideration. For the initial cooling from boiling down to about 100 degrees, I want the greatest flow possible and I suspect I can get more from my tap than from my pump.
If so, the faster cooling may more than make up for the time spent reconnecting. I may need to do some experimentation here.

Brew-boy said:

I use a Recirculating system with a pump and no ice. I go from boil to 70f in 15-17 mins using this method. I can hit 60f in about 30 mins times if I let it go that long.

Click to expand...

I really hate you people in the frozen North with cold groundwater.

The speed you can cool the wort is dependent on the volume of water flow per minute through the chiller.

Click to expand...

I would think it is possible for the water to run too fast to efficiently pick up heat. Too slow and it heats up too soon in the coil, thus reducing the temp differential that is the biggest factor in the heat transfer rate.

Running the tap water into the bucket is worth some consideration. For the initial cooling from boiling down to about 100 degrees, I want the greatest flow possible and I suspect I can get more from my tap than from my pump.

Click to expand...

Unless you have a hellacious pump, you are probably right. I think my pump is rated at about 285gph at zero lift, which I almost achieve by putting the bucket with pump on a table next to the keggle. The discharge is slow enough for me to alternately fill and dump a bucket into my rain barrels while another bucket is filling.

rgray58 said:

I would think it is possible for the water to run too fast to efficiently pick up heat. Too slow and it heats up too soon in the coil, thus reducing the temp differential that is the biggest factor in the heat transfer rate.

Click to expand...

You will get more heat transfer per unit time by running fast because more of the chiller coil will have a greater temperature difference between the chiller water and the wort. Same for the prechiller. But you will have less heat transfer per unit volume of chiller water. If you are recirculating, open the valve all the way and "let 'er rip". If you are using a prechiller, then you have to make a choice between chilling with a minimum amount of water vs. chilling in a minimum amount of time. If you want to minimize the water usage, run slow enough to get almost complete heat transfer - if you want to minimize time, run fast, but not so fast that you run out of ice. The chiller with prechiller case is complicated if you account for different flow rates and heat transfer efficiencies - could be interesting to model, but I'm not sure I want to try!

I bought a 25' x 3/8" Immersion chiller when I started all grain this May, I do 5 gallon batches in a 10 gallon BC. Being so close to hell, the ground water here is about 90 degrees this time of year. I struggled through batches, bringing down the wort to 90-ish degrees after quite a long time and then racking and putting the bucket in the fermentation chamber till cool enough to pitch yeast some hours later. I finally broke down about 10 days ago and planned my recirc system. I bought a pond pump that had a 1/2" female inlet. I bought a $0.75 1/2"x3/4" male x male adapter, to fit my 5' hoses I use pre=IC so I can test for leaks. I got my temp down to 100 or so, unhooked the hose from the pre-IC hose and hooked straight to the pump that I submerged in a tub with 20 pound of ice and water. Placed the outlet back in my ice water to recirculate and I went from 100 or so down to 65 in about 10 minutes. I was shocked as I have never seen the true benefits of my IC. The recirc system kept the IC ice cold and also another important thing for me, didn't waste a lot of precious water (people here paint rocks green so they don't have to grow grass). It is a perfect system and I am only pissed that I didn't do it 6 batches ago knowing the ground water temp was an issue.

I did some more math modeling - this time I wanted to see the difference between running tap water through a prechiller then through a chiller - both slow and fast flow. I assume 80 degree F. tap water, 6 gallons of wort starting at 90 deg. F, and 4 gallons (not enough really) of prechiller water (no ice) starting at 35 degrees F. The chiller is assumed to be longer or more efficient than the prechiller (probably representative of most people's systems). In the slow case, the water exiting the prechiller has gotten 80% of the way down to the temp. of the water in the prechiller bucket and the water exiting the wort chiller has risen by 90 % of the difference between the wort temp. and the chiller input temp. In the fast flow case, the prechiller % is 60 % and the chiller is 73% (not simple to figure this one). Note that the x axis on this is time, not gallons as my other charts.



It seems clear from this that while you are better off to run fast while cooling without the prechiller, once you hook up the prechiller, or put the ice in the bucket, you are better to run slow. If you run fast, too much of your "cold" water goes down the drain. I am convinced that the best option, if you have a pump is to recirculate the ice water, and here you can pump as fast as your equipment will allow. This study has convinced me to abandon my prechiller and recirculate.

Revisiting this thread after a poor purchase of a submersible pump. Anyone have a suggestion as to a good recirculation pump that has a 3/8" or 1/2" attachment? I can find some good ones (1/6 HP) but the only exit valve attachments they have are 1" and above.

I suppose I could simply use my racking cane too, but that seems like it'd be a bunch of work...

I've got a different take. Here in Phoenix, the tap water in the summer is 100-105* (Not joking on that, i've measured), which as you can imagine doesn't do the best job of bringing your wort down to temp. So every brew day, I buy three 10# bags of ice (99c/bag at the local gas station).

I fill up the sink with tap water, and then dump in a bag of ice in, and let the water cool down. This gives me a sink full of 40* water, I then start a siphon to pull the 40* water through the IC to cool the wort. The water coming out of the IC is in the 160* range when i first start chilling it, which is a pretty effective heat transfer for my 25' coil. Using the 3 bags of ice, I can bring 5.5 gal of wort to 75* in around 25 minutes, which for $3 works for me.

Brewing in Arizona in the summer is a pain in the behind.

scottland said:

I've got a different take. Here in Phoenix, the tap water in the summer is 100-105* (Not joking on that, i've measured), which as you can imagine doesn't do the best job of bringing your wort down to temp. So every brew day, I buy three 10# bags of ice (99c/bag at the local gas station).

I fill up the sink with tap water, and then dump in a bag of ice in, and let the water cool down. This gives me a sink full of 40* water, I then start a siphon to pull the 40* water through the IC to cool the wort. The water coming out of the IC is in the 160* range when i first start chilling it, which is a pretty effective heat transfer for my 25' coil. Using the 3 bags of ice, I can bring 5.5 gal of wort to 75* in around 25 minutes, which for $3 works for me.

Brewing in Arizona in the summer is a pain in the behind.

Click to expand...

I'm probably going to do something similar with my auto siphon. While the tap water isn't quite hwat it is in AZ, FL aint all that fun either. I might dump some rock salt into my icewater vessel to bring down the temp even further. According to DeafSmith, slower circulation may actually work better. I might have to pump the siphon a few times, but I'm pretty certain this'll work.

Mine flows about a gallon per minute, maybe a touch less with just gravity. I actually just suck on the end of the outlet tube to start the siphon, it takes less than a second before gravity takes over, and there's no need to worry about sanitation there.

A pre-chiller would possibly work for me, but what I currently do is nice and simple, and only costs $3 per batch, and I'd probably need to buy at least a bag or two for a pre-chiller anyway.

Being that this is a old thread. I was wondering what you are all doing now for wort chill'n. Thanks for all of the posts and ideas on chilling & prechilling.

I use my immersion chiller with tap water until the wort is down to about 85-100 ºF, depending on the season, then switch to a 1/6 HP submersible pump (from Harbor Freight) in a 20 gallon tub with about 10-12 gallons of ice water and recirculate. In summer, when I only cool the wort to about 100 before recirculation, I need more ice than in winter when I can get it down to 80-85 pretty quickly. Also, of course, I try to get the temp. as low as possible with tap water when doing a lager, and then use a bunch of ice. I also gently stir the wort constantly.

Greetings!

Luckily up in WA the groundwater temps aren't a problem. For 5 gallon batches I've just used a 25' 3/8" immersion chiller. My 60L pot I picked up is much shorter and wider than a keggle, so I made up a 50' 3/8 immersion chiller and then made a simple jumper to add my 25' in the middle making for 75' total. This worked fine but was awkward to handle.

Recently I got a brewstand with an elevated burner for the HLT. Problem is that it was built for a coleman cooler MLT and the HLT isn't high enough to gravity sparge into a keggle. SO (rather than adding a 2nd pump) I had to use my short/wide boil pot for MLT and keggles for HLT and boil. The brewstand also came with a counterflow chiller and I've used it twice now but I plan on doing it a little differently.

An accomplished brewer friend of mine does it like this: hose water into the counterflow, then a jumper hose carries that water into an immersion chiller in the boil kettle. His theory is that 160 in the immersion is a lot colder than 212 in the kettle. The thing you want to do is cool down to under 190ish as quickly as possible to stop breaking down the hops. My counterflow experience had me worried that the beer waiting in the kettle to be drained would just sit there over 190. Now what my buddy will do is take the beer exiting the counterflow and recirculate back into the boil pot for a while. I guess if done properly you could call this a whirlpool? Anyway, once the boil kettle has cooled below 180ish he'll just proceed to fill his carboys using a ball valve to adjust the wort flow through the counterflow as usual.

I haven't re-bent my 50' yet but I plan on bending it to fit in a keggle. Until then I'm going to do the re-circulate/whirlpool thing to get below 190 as quick as possible.

Does anyone have any experience with the recirculation method and putting salt in the water? In the beginning this was mentioned and I was wondering if there was any concern about corrosion.

Thanks

DeafSmith said:

I use my immersion chiller with tap water until the wort is down to about 85-100 ºF, depending on the season, then switch to a 1/6 HP submersible pump (from Harbor Freight) in a 20 gallon tub with about 10-12 gallons of ice water and recirculate. In summer, when I only cool the wort to about 100 before recirculation, I need more ice than in winter when I can get it down to 80-85 pretty quickly. Also, of course, I try to get the temp. as low as possible with tap water when doing a lager, and then use a bunch of ice. I also gently stir the wort constantly.

Click to expand...

Thank you for the quick answer.
What are you using in the tub of water for the heat exchange? Thinking about a pre chiller (2nd Immersion chiller) in the ice water tub. The flip side, are you just pumping the water from the tub to the IC in the wort? And salt in ice bath, yes or no? If yes, have you experienced any corrosion?

JerseyJoe said:

Thank you for the quick answer.
What are you using in the tub of water for the heat exchange? Thinking about a pre chiller (2nd Immersion chiller) in the ice water tub. The flip side, are you just pumping the water from the tub to the IC in the wort? And salt in ice bath, yes or no? If yes, have you experienced any corrosion?

Click to expand...

I'm just pumping the ice water from the tub - the pump has a screen at the bottom so it doesn't suck up any pieces of ice large enough to hurt anything - actually, the ice floats to the top and the pump sucks from the bottom, so no problem anyway. I don't use salt because I don't want to corrode the pump.

This is the pump I use:

http://www.harborfreight.com/16-horsepower-submersible-utility-pump-68422.html

I got that one because it has a fairly high lift capacity (23 ft) so I figured it would give me enough pressure to force water through the narrow tubing of my chiller. Even though the pump is rated at 1350 gallons/hour (with no lift or restrictions on the output), I actually get about 60 -70 gallons/hour through my 50 feet of 3/8 inch OD tubing - just a bit over 1 gallon per minute, but that works fine.

JerseyJoe said:

And salt in ice bath, yes or no?

Click to expand...

Salt in the ice bath will just lower the temperature by melting the ice quicker, perhaps shaving several minutes off the process...not really worth it IMO...you will still need the same amount of ice so not worth it for me.