Water cooling something other than a PC

S

Soybomb

2[H]4U
Joined
Jun 12, 2000
Messages
2,171
Its been a while since I've done any extreme cooling on a PC, I moved into sysadmin and networking, but I figured the experience everyone here has should help me alot.

When I'm not doing computer stuff I like to brew my own beer. While the yeast go to work on it you need to be able to control the temperature of it pretty well. Lets say a summer day the room is 72, the yeast are adding another 4 degrees to the temperature from fermentation. I now need to cool that liquid from 76F to lets say 60.

Some people buy an old fridge and use a $60 temperature controller to regulate the temperature (and in all fairness colder temperatures since sometimes low 30s would be nice but I doubt I can do that without a fridge). I'm wanting a more compact assembly than a spare fridge. I'm working with 5 gallons of liquid at a time btw. Right now I have these old soda kegs (in the fridge on the left) that I use for beer that I think would be great http://johnmearns.com/gallery/kegerator/DSCN1444 They're made out of stainless steel so I have a conductive surface thats easy to move around.

I thought about wrapping the thing in a copper coil and running cool water from a resorvoir in the fridge trhough it, but that seems ineffecient even if I insulated it well. If I could do something with peltier that would be stand alone that would be awesome, but I think it'd take some whoppers for that much thermal mass. I'm starting to think of some kind of stainless tubing immersed through the top of the keg and the a resorvoir of cold water inside the fridge pumped contually through it on a temperature controller to turn off the pump when it hits the right temp. Would the standard pumps be enough flow for this you think?

Maybe no one cares but I thought this is the H if anyone has any good ideas or suggestions for a crazy project its here :D
 
:D not right off hand. I'd say on average it will hold 5 gallons of essentially water 4-5F above ambient. Most of the work would be cooling the liquid below ambient. Is there a handy calc to see what it would take? If I wanted to heat that volume 20 degrees hotter over 4 hours it would take a 64 watt element by the calculator I have. I would assume that'd be 64 watts of cooling power I'd need if I wanted the same drop over the same time. But my math could be off :D
 
Don't know of a calculator but I can give you the equations that you need to work it out.
The basic one is Q = m Cp dT
Where Q is the energy in Joules, m is the mass of the substance, Cp is the specific heat value and dT is the temperature change you re aiming for.
Water has a mass of 1kg/l so for 5 gallons is about 19 litres so it's 19 kg. Cp for water is around 4.2 kJ/kg/K and your dT is about 6 Kelvin.
Multiplying all out you get 19*4.2*6 = 478.8 kJ over 4 hours this gives you 33.25 watts
478800/(3600*4). This is done by dividing the amount in joules by the time in seconds as a watt is a joule per second.

Could be wrong with the unit conversions but I think it's roughly right. To answer the other question heating has the same energy difference as cooling just technically the dT is negative as you have to remove energy.

Bit of a long winded explanation really :eek:
 
You guys are calculating the heat it would take to bring that volume of water up to that temp, what you want is the power it takes to keep it at a certain temp. That's a complex problem, you have to account for air temp, airflow characteristics, and water (beer) flow characteristics. For practical purposes, it'd probably be easiest just to make an educated guess and experiment with it. For example - a water chiller (for those office-type cool water dispensers) chills a small amount of water to maybe around 40F. Something that size would cool a 5gal volume to maybe around 55F in a 75F ambient temp...? Sounds reasonabe to me. The exothermic yeast are going to require a little more power, you may need a small minifridge to accomplish this.
 
zer0signal667 said:
You guys are calculating the heat it would take to bring that volume of water up to that temp, what you want is the power it takes to keep it at a certain temp. That's a complex problem, you have to account for air temp, airflow characteristics, and water (beer) flow characteristics. For practical purposes, it'd probably be easiest just to make an educated guess and experiment with it. For example - a water chiller (for those office-type cool water dispensers) chills a small amount of water to maybe around 40F. Something that size would cool a 5gal volume to maybe around 55F in a 75F ambient temp...? Sounds reasonabe to me. The exothermic yeast are going to require a little more power, you may need a small minifridge to accomplish this.
Click to expand...
I've got this holding the chilled kegs http://johnmearns.com/albums/kegerator/DSCN1463.thumb.jpg and theres room in there for a large water reservoir. I'd like it to be a stand alone system but I think the cheapet and practical way would involve poking a couple holes in the back and running water in/out of there.
 
zer0signal667 said:
You guys are calculating the heat it would take to bring that volume of water up to that temp, what you want is the power it takes to keep it at a certain temp. That's a complex problem, you have to account for air temp, airflow characteristics, and water (beer) flow characteristics. For practical purposes, it'd probably be easiest just to make an educated guess and experiment with it. For example - a water chiller (for those office-type cool water dispensers) chills a small amount of water to maybe around 40F. Something that size would cool a 5gal volume to maybe around 55F in a 75F ambient temp...? Sounds reasonabe to me. The exothermic yeast are going to require a little more power, you may need a small minifridge to accomplish this.
Click to expand...

What you say is true but the same can be said for heating something aswell, no system can account for everything. The figure that I gave is what would be needed to cool it 20F below ambient temperature which is a much larger value than anything else in the process so therefore that is the what it should be based off. All the losses through convection conduction and radiation can easily be compensated using a temperature monitor which can switch the pump on or off.

As for the actually cooling problem soybomb running water in/out would be the easiest way because otherwise you might struggle to get the cooling coil in and out. If you can make the coil as big as possible because it's easier to correct the water flow than it is to correct the size of the coil or having to increase the size of the pump. The larger surface area of the coil and the colder the water is means less needs to be pumped.

I could throw together some more specific numbers if you know the temp of the water and what sort of tubing you are using but even in indsutry heat transfer problems tend to be solved by overcompensating and then slowing the water/steam down.
 
ChAotiCIllusion said:
What you say is true but the same can be said for heating something aswell, no system can account for everything. The figure that I gave is what would be needed to cool it 20F below ambient temperature which is a much larger value than anything else in the process so therefore that is the what it should be based off. All the losses through convection conduction and radiation can easily be compensated using a temperature monitor which can switch the pump on or off.

As for the actually cooling problem soybomb running water in/out would be the easiest way because otherwise you might struggle to get the cooling coil in and out. If you can make the coil as big as possible because it's easier to correct the water flow than it is to correct the size of the coil or having to increase the size of the pump. The larger surface area of the coil and the colder the water is means less needs to be pumped.

I could throw together some more specific numbers if you know the temp of the water and what sort of tubing you are using but even in indsutry heat transfer problems tend to be solved by overcompensating and then slowing the water/steam down.
Click to expand...


You're exactly right, the power figures stated above should be sufficient, just possibly overkill. It's not always a good way to base your judgment though, take for example a calculation done to find the power needed to cool the water 20F over 24 hours. This would most definitely not be enough to keep an exothermic fluid at that temp.
So again, use your brain to come up with something reasonable, and I'm sure other people can either back up your reasoning or tell you you're way off.
 
So lets be generous with losses and give some wiggle room and say I need 120 watts of cooling power. Does the submerged coil circulating fridge water sound like the only way to do it easily?
 
You must log in or register to reply here.
Back
Top