Glycol through Therminator/Plate Chiller: big batches

Attention all engineers (I should have paid more attention in skool):

I'm thinking of recirculating Glycol through the plate-chiller as the cooling solution to my large batches. On my 20 gallon system I run hose water through the chiller, while recirculating the wort back to the kettle to get the wort down to ~150, then use a recirculating ice bath on the plate chiller and pump from the kettle into the fermenter. No problem. BUT!

Now I'm moving to larger batches and don't want to waste as much and would like to speed this up... Wonder if I used glycol the whole time if I could go straight off the whirlpool kettle (still around 200ish) to the fermenter if it used Glycol as the chilling fluid?

I was thinking I could put a container (bucket like) in my chest freezer to hold a large glycol bath (20? 30?? gallon bucket in there, with a submersible pump) and two lines, one to and one from the glycol bucket through the chiller.

How would I calculate how much glycol I'd need? I really need to go single pass straight off the whirlpool kettle - and this for what will be my 51 gallon batches...

Thoughts? Seen it? Help!

Probably the simple answer would be to series connect 2 plate chiilers and pump from kettle to fermenter. First plate chiller would pull wort temperature down to 80 using water then second glycol cooled chiller would drop remainder of temperature. The method for the glycol system would be water filled tubes with glycol circulated through copper pipe hairpins, with water in frozen state you would have a larger energy sink than just liquid glycol solution. I will look into calculating what it will take to cool 50 gallons from 212 to 60 degrees as a starting point for a wort cooling system.

Kevin, using the method you mentioned above would a water bottle cooler be able to handle the second plates chilling needs for small batches?

If you used a storage container and circulated the coolant until it got down to temp before starting wort chilling it should work. The coolers thermostats bottom out at 40 F without some recalibration, so a bit of work would be needed to get temperatures down to 32 F. I have a cooling system built with 2 remote water fountain chillers and a 50L keg for coolant storage, running but not tested with a 10 gallon batch yet.

mikefromcu said:

Now I'm moving to larger batches and don't want to waste as much and would like to speed this up... Wonder if I used glycol the whole time if I could go straight off the whirlpool kettle (still around 200ish) to the fermenter if it used Glycol as the chilling fluid?

I was thinking I could put a container (bucket like) in my chest freezer to hold a large glycol bath (20? 30?? gallon bucket in there, with a submersible pump) and two lines, one to and one from the glycol bucket through the chiller.

Click to expand...

You could use brine solution instead of Glycol...less toxic if spilt and easily made at home.

I think you need about (depending on temperature of brine/glycol and efficiency of CF chiller) about two times the amount of chilling liquid than wort to be chilled. On a 51g batch you would need about 100g of brine at 0*F.


I'm only an electrical engineer so don't trust everything I say

mikefromcu,

It's difficult to say how much glycol you'll need, but I imagine quite a bit. You'd need to know how quickly your chest freezer can keep up with the heat you'll be dumping into it. (Don't forget that for every BTU of heat you dump into your freezer, that BTU has to be pumped back out).

You can calculate the minimum amount of glycol required using a heat balance. BUT! You want to speed this thing up, so you're going to need more than the minimum.

How quickly do you want to drain your wort? Do you know the UA on your heat exchanger from past runs or can you approximate it? (UA can be approximated if you have your old flow rates, ie how long does it take to drain, and temperatures going in/out). The manufacturer may have given the "design" UA or at least an area.

If you can answer those two questions then the needed amount of glycol can be calculated. (BTW, which glycol are you planning on using?)

Here is an ice bank example built by Yorg from HD parts http://picasaweb.google.com.au/YorgTheodore/ShareOnline#5356097670941569362, this lets the freezer freeze the water then the glycol coolant is circulated and transfers heat to the ice. This will increase the heatsink capacity of a freezer to a point where glycol cooling becomes practical on a small scale. The other advantage is the total quantity of the coolant is in the 5 gallon range not 50 gallons.

Here are some rough numbers for the glycol cooling systems for cooling 51 gallons of wort from 210 degrees to 65 degrees

1. All glycol at 40 degrees F starting temp, about 55 gallons at ideal heat exchange conditions.

2. Water pre cool to 100 degrees, Glycol to 65 degrees F., about 37 gallons at ideal exchange.

3 Water pre cool to 100 degrees,130 Lb Ice bank with glycol circulation (10-4"IDX30" cylinders in freezer), total melt of ice to bring wort down to 65 degrees F..

Water consumption for pre cool would be in excess of 65 gallons to remove most of the heat.

So...if you're making 50 gallon batches, a dedicated chest freezer filled with glycol might do the trick....

kladue said:

Here are some rough numbers for the glycol cooling systems for cooling 51 gallons of wort from 210 degrees to 65 degrees

1. All glycol at 40 degrees F starting temp, about 55 gallons at ideal heat exchange conditions.

2. Water pre cool to 100 degrees, Glycol to 65 degrees F., about 37 gallons at ideal exchange.

3 Water pre cool to 100 degrees,130 Lb Ice bank with glycol circulation (10-4"IDX30" cylinders in freezer), total melt of ice to bring wort down to 65 degrees F..

Water consumption for pre cool would be in excess of 65 gallons to remove most of the heat.

Click to expand...

This must be a rather large investment in glycol alone at this large of a amount. I see Texas so the pre cool ground water must be rather warm besides the amount available to let let flow and considered lost or yard water.
The 65 gallon amount isn't a worry with my brewing in the past as I have city water plus well water that I let run by the hour for the yard. I can go thru 1,400 gallons of well water every second day without any worries with the temp at 63*F summer, 59 to 61*F winter.
My friend has a 14 cu/ft freezer that had a 1/8" aluminum liner fitted inside which holds 135 gallons of well water starting at 34*F as a second chiller system for his 25 gallon batches. First run with well water then the keezer cooler water. He needs 4 days to chill 63*F water down first. This was way cheaper than the price of glycol, messing around with blocks of ice or any other chiller system with the larger brew batches.

Yea, so I'm thinking that a 50/50 mixture of glycol/water (gly-sol) in the freezer will be at about 0*.

My original idea was to run the gly-solution out from the bottom of a 50 gallon container, via submersible pump, through the Therminator, returning the heated gly-sol to the top of the bucket. Obviously the gly-sol will start at 0*, but will warm over the course of the cooling process. All I need is the NET of the wort in the fermenter to be at 65, so I would guess the first gallons would be damn cold and the latter gallons would be considerably warmer.

If I can get 51 gallons cooled via this method, that's what I'll go with, but otherwise, perhaps the two chillers in series, one on the hose, the second on glycol would be the best route.

My preference would be to do the all gly-sol method. What do you think?

Look at Yorgs example of an Ice bank method, it will absorb more heat than just the glycol alone and the amount of expensive glycol will be much smaller. A 5 gallon bucket with a pump as a resevoir would work once the tubing and HX were filled, you might be able to get by with 10 gallons or less of coolant instead of 50+ gallons. The ice can absorb 144 BTU's Lb in addition to same amount that the glycol will absorb as it warms. You could build one of the tube ice banks to prove the idea and then copy the number needed to match batch size.

There's no benefit to dumping the post-chiller glycol back into the reservoir because it immediately begins heating up the whole volume. You would be better off moving spent coolant to a holding tank so that the reservoir supply is always at zero. A chest freezer wouldn't provide any practical heat sinking during the time frame of a chilling operation.

OK, so I crunched the numbers from an energy balance standpoint. The answer is:

You need to use ICE (as other's have said).

Here's why.

First, let's calculate how much energy is required to cool the wort.

Mass Cp of wort @ SG=1.05 is about 0.882 BTU/(lb*F)

51 gallons * 1.05SG * 8.32 lb/gal (water density) = 445 lb of wort

Cp * m = BTU/F, so 0.882 * 445 = 393 BTU/F

If you start at 212F and cool to 80F, that's a change of 132F. Therefore:

132F * 393 BTU/F = 52000 BTUs to cool the wort.


If only glycol is used:

Cp 50/50 glycol = 0.8107 BTU/(lb*F), SG=1.0622
there's only sensible heat, so if we start at 0F and heat the glycol up to 70F (keeping a 10F approach) gives us

0.8107 BTU/(lb*F) * 70F = 57BTU/lb and

57 BTU/lb * 1.0622 * 8.32lb/gal= 504 BTU/gal

we need 52000BTU, so 52000/504=103 gal

You would need around 100 gallons of glycol!!

If we use 30 gallons of ice:

solid ice needs 16 BTUs to heat from 0F to 32F
the heat requred is 144 BTU/lb to melt the ice
heat capacity is 1 BTU/(lb*F) after that

30 gallons of water (before it is frozen) yields

30 gal * 8.32lb/gal = 250 lbs of ice

0-32F
16 BTU/lb * 250 lb = 4000 BTUs

32F-32F (melting)
144 BTU/lb * 250 lb = 36000 BTUs

32F-70F
1 BTU/(lb*F) * (70F-32F) * 250lb = 9500 BTUs

Total= 4000+36000+9500=49500 BTUs

This will provide the majority of the cooling. You now need to know how much glycol is required to provide the additional 52000-49500= 2500 BTUs

So from the above, the Cp for 50/50 glycol is 504 BTUs/gallon and

2500 BTU / 504 BTU/gal = 5 gallons of glycol



So your choice is to either use 100 gallons of glycol OR 5 gallons of glycol in a 30 gallon chunk of ice. The latter will probably be better since it will give you a better driving force in the exchanger (allows the wort to flow more quickly).

Hope this helps!!

And BTW, the above doesn't have much of a "margin" in it. You're going to gain some heat in the lines as the glycol goes over to the exchanger, but you'll lose some heat as the wort gets to the exchanger. I don't know that these will really offset, so if you can go with more than 30 gallons of ice, that's what I would do!

Moral of the story, use tap water to reach 100F, then just pump some icewater for the final run to the fermenter.

Wow, thanks to all you guys who paid more attention in Themodynamics than I did. I went into this thinking glycol was the magic bullet I needed, but it appears it helps but not much, and probably not as much as the hazard/headache.

I think my best bet may be to go back to the recirculating ice batch. Perhaps if I had ice water going through two chillers at the same time, so that wort flowing through would hit ice water as the coolant in both. First one knocks it down, 2nd one finishes it up. I would need a little more than 30 gallons of ice, but goatchze included his formulas so I should be able to work them.

John Blichmann looked at this too, suggested that it would take a lot of glycol too and suggested hose water would probably still be a better bet. One thing he did suggest was working up a test batch of just water though, to see if it worked before getting into a problem with 51 gallons of wort.

Thanks again for the help!

If you pump ice water into the plate chiller, you should be able to pump wort at full bore while reducing the coolant flow to a mere trickle. The wort will come out cool enough in a single pass and you'll run through the batch quick enough that DMS production will be negated. You should get at least 3 gpm out of the March pump which would drain 50 gallons in 17 minutes. You'll reduce the poundage of ice required if you collect the coolant output into a separate container for cleanup use. Make up the icewater level with water from the tap.

Bobby_M said:

If you pump ice water into the plate chiller, you should be able to pump wort at full bore while reducing the coolant flow to a mere trickle.

Click to expand...

I discussed this with the engineer i work with at our vendor for plate chillers. we plunked in different numbers for wort temp and water temp.

Per engineering, your best efficiency (Thermal and water usage) will be by reversing this scenario. monitor the out temp of your wort and control its flow till you get an outflow of the proper temp.

Recirculate the chill fluid as fast as possible for best thermal transfer.... the reason is that you want the highest temp differential inside the cooler as possible.

Not sure if this was mentioned before. Could you use your cold glycol to chill tap water that is already cool to make things go quicker and just get a large vessel of COLD water to do the work for you?

oldschool said:

Not sure if this was mentioned before. Could you use your cold glycol to chill tap water that is already cool to make things go quicker and just get a large vessel of COLD water to do the work for you?

Click to expand...

Water isn't going to be much of an improvement. The reason that ice works so well is that you are using the heat of fusion of the ice to pick up most of the heat (much like water evaporiting cools better than just water...or cools with less mass of water).

The fundamental problem is that you need so many BTUs removed from the wort in total. You'll still need a lot of water. Whether you remove some heat from the water ahead of time or not, the net effect is the same. This wouldn't be an improvement of prechilling the wort with tap water and then using the cold glycol or cold water to finish it off.

goatchze said:

Water isn't going to be much of an improvement. The reason that ice works so well is that you are using the heat of fusion of the ice to pick up most of the heat (much like water evaporiting cools better than just water...or cools with less mass of water).

The fundamental problem is that you need so many BTUs removed from the wort in total. You'll still need a lot of water. Whether you remove some heat from the water ahead of time or not, the net effect is the same. This wouldn't be an improvement of prechilling the wort with tap water and then using the cold glycol or cold water to finish it off.

Click to expand...

Yup, total BTU's removed is what you are after. A large Delta T will do it faster. Only two reasons to go after this, Local water supply is too warm to give efficient or cool enough end temp, or to save billed water or water mess.

Recirculating cooling water will give more opportunity to transfer heat, so contrary to earlier posts dumping coolant back to the reservoir will be productive until the output side of the coolant line is within 20ish degrees of the wort temp.

In a properly set up loop, the coolant side will be circulating quickly. Coming in at say 4o-50 degrees and going out at 60-80ish possibly less. With a wort temp between 160-190 degrees, the temp differential will still be significant enough to give good transfer until the coolant reservoir temp is raised to a point you must slow your wort flow excessively to maintain a desired output temp.

Hopefully by this point the remaining wort dumping in will be small enough that its 80-100 degree mass will not significantly impact the overall temp. If it does, the remaining BTU's will have to be finished off with tap water coolant or a reserve coolant of some sort, and you will know you need either a colder starting coolant or more total volume in your coolant reservoir.

The reason that i said what I said is because i know a brewer that does this. He says that their chiller isn't large enough...well it is but the hot wort would deplete all of the cold glycol reserve needed for fermentation. The have one small heat exchanger and one large. When it comes time to cool the wort, glycol is used in the small exchanger to chill the already cool city water. Cold city water then travels to the large heat exchanger; once it's warmed by the wort it then goes to the HLT and reused, saving energy.

oldschool said:

The reason that i said what I said is because i know a brewer that does this. He says that their chiller isn't large enough...well it is but the hot wort would deplete all of the cold glycol reserve needed for fermentation. The have one small heat exchanger and one large. When it comes time to cool the wort, glycol is used in the small exchanger to chill the already cool city water. Cold city water then travels to the large heat exchanger; once it's warmed by the wort it then goes to the HLT and reused, saving energy.

Click to expand...

That's a bit of a different issue. When someone says an exchanger is "too small" what they mean is that it is "too small for the flowrates I would like to have". You can cool your wort with the smallest of heat exchangers, but you would have to slow down the flowrate of your wort and coolant. Likewise a larger exchanger will allow you to have higher flowrates.

Precooling the water changes the driving force (temperature differential). This allows a smaller heat exchanger to transfer heat at a higher rate or, in other words, would allow you to have higher wort flowrates.

I think this is the problem they are/were trying to fix, which is different from the OPs.

BTW, the equation for heat transfer in an exchanger is

Q=UA(dTlm)

where
Q=rate of heat exchange (BTU/h)
UA=property of the heat exchanger, A being "AREA"
dTlm=the log mean temperature difference, is a function of the hot/cold stream temperatures on each side of the exchanger

Q will determine how fast you can push wort through the exchanger and get the end point temperature you want. Prechilling the coolant allows the dTlm to be greater, which gives a larger Q for a smaller UA.

oldschool said:

The reason that i said what I said is because i know a brewer that does this. He says that their chiller isn't large enough...well it is but the hot wort would deplete all of the cold glycol reserve needed for fermentation. The have one small heat exchanger and one large. When it comes time to cool the wort, glycol is used in the small exchanger to chill the already cool city water. Cold city water then travels to the large heat exchanger; once it's warmed by the wort it then goes to the HLT and reused, saving energy.

Click to expand...

He would probably have better success knocking off the majority of the heat with city water and finishing with the coolant down to the desired temp... sounds like he is adding complexity by cooling the cooling water with other coolant... kind of like using a gas engine to drive a generator that charges the batteries of your electric car... it will work, but overall efficiency is lost at every step.

What's easier? Doing it that way or, cooling all you can with city water, recirculating BACK in to the kettle and switching the line over to glycol? Sounds like a better one shot deal and preheating water for the HLT

mikefromcu said:

Attention all engineers (I should have paid more attention in skool):

I'm thinking of recirculating Glycol through the plate-chiller as the cooling solution to my large batches. On my 20 gallon system I run hose water through the chiller, while recirculating the wort back to the kettle to get the wort down to ~150,

Click to expand...

What's wrong with your plate chiller? If you run water in the correct way, it flows in the opposite direction of the wort. The typical hose water runs through the ground, which is typically around 50°. If I run my wort through the chiller at a normal pace, it cool downs to about 65-75°. Do you need it cooler than that? Try checking the water temp direct into the plate chiller, slow down the wort flow and see what you get.

Also, be careful, the plate chiller is a source of contamination if not cleaned properly.


Cheers