Board logo

Way off top: anyone know about thermodynamics?
BenB - 23/6/09 at 05:40 PM

If I'm using a copper coil full of cold water to cool a hot (but no longer heated) liquid is it best to have the coldest water entering the top of the coil or the bottom....



My thinking is that the water at the top of the bucket will be hottest due to convection so you actually want the cold water entering the bottom of the coil. It will gradually be heated as it goes up the coil but will ensure that there is a constant temperature differential between the stuff in the bucket (okay, it's homebrew ) and the water in the coil, ensuring maximum heat removal from the hot stuff.

Other people though say if you have the cold water going in the top it's best because then you've got the hottest water being cooled by the coolest water. But equally by that rationale by the time you get to the bottom of the bucket the water in the coil may be at the same temperature as the stuff in the bucket and so no cooling will take place.

Or do you work on the principle that there is X amount of heat energy within the bucket, that you are supplying Y amount of cold water at Z cc/second and that convection and thermodynamics will take care of the rest to the extent that it won't make any difference whether it goes in the top or the bottom..... In which case plan B is probably best (cold water in the top) as the convection will ensure good temperature mixing and no hot spots...

My brain hurts....

[disclaimer: that's not my photo- my copper coil will be much nicer ]


twybrow - 23/6/09 at 05:42 PM

Maximum temperature gradient, will give you the best heat exchange.... I would say go in cold at the top (but I hated thermo at uni!).


dinosaurjuice - 23/6/09 at 05:45 PM

its the exact opposite of a central heating cylinder. they have hottest water entering coil at top, so i think putting coolest in at bottom would work best. i doubt it makes a massive difference to be honest.


Jon Hazan - 23/6/09 at 05:46 PM

Im no expert but i would have thought cold water entering the bottom would be most efficient. Mainly due to natural convection within the copper pipe will want to move the heat upwards.

Also if you run from top to bottom if the heat difference between top and bottom of the pan is enough you could actually end up heating the liquid at the bottom of the pan!!


twybrow - 23/6/09 at 06:04 PM

But equally, if the brew does warm up as the bottom, it will rise to the top of the bucket and hit the cooler pipe.


JoelP - 23/6/09 at 06:15 PM

it makes no significant difference IMHO.

However, having a zip tie on the copper pipe means that some heat you've removed will go back into the bucket via the incoming cold! Insulate between them! Though again, it wont really make a difference.

I say no difference because there will only be a very small temperature difference in the beer between top and bottom. However, assuming the pipe was 100% efficient and the coolant exits at the same temp as what it went past last, you would techincally want it to start in the coolest part and finish in the hottest part.

However, a further complication is that you cant be sure its hottest at the top - afterall, most of the cooling will happen at the surface!


nib1980 - 23/6/09 at 06:17 PM

think of it as a kettle in reverse.

BUT it works by the same principle i.e element in the bottom of kettle, therefore coldwater in the bottom first.

enthalpy, entropy, SFEE, etc etc


matt_claydon - 23/6/09 at 06:18 PM

If you assume the fluid in the bucket in proximity to the coil is flowing upwards, you essentially have either a parallel-flow heat exchanger (cold in at bottom) or a counter-current heat exchanger (cold in at top).

Counter-current heat exchangers are well-known to be more effective. First para on Wikipedia article for heat exchangers has a diagram.

However, the convection current in the bucket will be complex and there will be significant cooling from the air/fluid interface at the surface to complicate matters.

[Edited on 23/6/09 by matt_claydon]


nstrug - 23/6/09 at 06:20 PM

Interesting question - try an experimental approach. Run both ways and measure the temperature difference between the input and the output of the cooling coil. Whichever direction gives you the largest temperature difference is the most efficient.

Nick


mookaloid - 23/6/09 at 06:43 PM

quote:
Originally posted by nstrug
Interesting question - try an experimental approach. Run both ways and measure the temperature difference between the input and the output of the cooling coil. Whichever direction gives you the largest temperature difference is the most efficient.

Nick


Ah yes empirical analysis


indykid - 23/6/09 at 06:45 PM

most importantly, what are you homebrewing?

looks like mead. mmmmm. mead. leave it slightly fizzy. it's delicious

my thermofluids lectures suggest cold should go in at the bottom for maximum efficiency
tom


b16mts - 23/6/09 at 07:35 PM

all this theory could be screwed up if we learn that the "brew" is first heated in the container it now resides, in which case there'll be plenty of residual heat in the tub.

either way, i shouldn't worry too much.


craig1410 - 23/6/09 at 08:48 PM

I'd say it will make no difference if...the rate of flow of water through the copper piping is the same in both directions (top to bottom or bottom to top).

However, unless your pump or source of pressure is much greater than you actually require then I suspect you will get slightly greater flow rate if you feed cold water in at the bottom as this will gain the benefit of thermo-syphon effect.

You wouldn't want to connect your engine's top radiator hose to the bottom of the radiator as you would then be pumping hot water downwards and then expecting it to rise through the radiator as it cools - not likely!

There may be other "secondary" effects involved in this but I'd say the primary effect is thermo-syphon and therefore cold feed to the bottom is the way to go.

Cheers,
Craig.


BenB - 23/6/09 at 09:47 PM

quote:
Originally posted by nstrug
Interesting question - try an experimental approach. Run both ways and measure the temperature difference between the input and the output of the cooling coil. Whichever direction gives you the largest temperature difference is the most efficient.

Nick


Oh you would not believe how tempted I am to do an experiment to find out. Strange how on the homebrew forums the answers to most questions appears to be "who f#cking cares, open another beer and choose an option randomly". Whilst I appreciate the thinking I prefer the slightly anally retentive "correct" answer to the usual reply I get of "you worry too much, just make the fricking beer"....


Nash - 24/6/09 at 07:07 AM

We need to establish more facts. Is the product succeptable to thermal shock? Does it react better to gradient cooling or quench cooling? Can you adjust the flowrate of one or both mediums? Is there an optimum time to cool eg 1 degree per 10 seconds?

I think the cooling rate of the product will have and effect on final taste and texture?!

Failing that drink it warm and give yourself a much more enjoyable headache

Regards............ Neil



BenB - 24/6/09 at 07:33 AM

I just remembered that viscous liquids convect heat differently to thin ones as well. I seem to remember from "earth story" that golden syrup convects in hexagonal cells....


dinosaurjuice - 24/6/09 at 07:42 AM

why not pass the beer through the copper pipe submerged in a saucepan of icey water. just to add ANOTHER option


BenB - 24/6/09 at 02:18 PM

That would be counter-flow-cooler design. Trouble there is trying sterilise the inside of 25m of copper tubing between brews..... but I like your thinking

Certainly in the CFCoolers they (as the name suggests) flow in the opposite directions to ensure a constant temperature transfer gradient...