Wort Chiller Just Ain't Chillin

[danestead]

Googling came across this.


http://homebrew.stackexchange.com/a/6613

Theres a post there that references a brewing magazine and some tests they did.

Whilst I can understand the laminar flow theory, Im not sure I believe the water can be so smooth flowing (not turbulent) that laminar flow issues with heat transfer could be a factor.

 

[lael]

Won't it also depend on the temp of the water in the coil? So as the wort cools down, I imagine longer exposure (slower flow) would be better.

 

[Parks]

I still can't see how it could possibly ever be quicker running it slower. You maintain a larger surface area of the coil at a lower temperature the faster it runs meaning a greater temp differential on average to the wort.

 

[pk.sax]

Let's just say that turbulence means energy spent.

Usually, any system is keen to lose energy to surroundings of lower energy and settle into a more entropic state.

However, for losing energy there has to differential. I.e., for flow to turn turbulent it will need a source of pressure to cause it to store energy that can be spent by turning turbulent. If the flow induced is smooth with little head pressure then there is little to lose and flow should remain laminar, besides, the regular shape of the pipe aids such flow.
If you were to introduce a kink in such a pipe, then sudden constriction causes reduction of area, I.e., rise of pressure that then suddenly drops past the constriction, in other words loss of energy can happen with a slow down of flow.

Narrow tubes are better for transporting highly pressurised fluid simply because the fluid acts more viscous under pressure and boundary layers shrink.
Wider tubes are better to be used at lower pressure with turbulent flow in them to keep breaking up the boundary layers sticking to the walls.

^ off the cuff reminiscence of high school science, might have errors in thinking.

 

[The Judge]

Copper beats stainless steel, and higher water rates cool more quickly than lower rates. Thermodynamics wins.

 

[Black n Tan]

Let's just say that turbulence means energy spent.

Usually, any system is keen to lose energy to surroundings of lower energy and settle into a more entropic state.

However, for losing energy there has to differential. I.e., for flow to turn turbulent it will need a source of pressure to cause it to store energy that can be spent by turning turbulent. If the flow induced is smooth with little head pressure then there is little to lose and flow should remain laminar, besides, the regular shape of the pipe aids such flow.
If you were to introduce a kink in such a pipe, then sudden constriction causes reduction of area, I.e., rise of pressure that then suddenly drops past the constriction, in other words loss of energy can happen with a slow down of flow.

Narrow tubes are better for transporting highly pressurised fluid simply because the fluid acts more viscous under pressure and boundary layers shrink.
Wider tubes are better to be used at lower pressure with turbulent flow in them to keep breaking up the boundary layers sticking to the walls.

^ off the cuff reminiscence of high school science, might have errors in thinking.

No idea what you just said, but surely the higher the flow rate the more turbulence, which means more efficient heat exchange. The link Danestead made, although not directly to the Zymurgy article, notes "His result was that the chiller worked best with the highest water pressure, and that swirling the chiller in the wort chilled the wort even faster." Surely that should be enough to put this to bed, although if someone has a copy of the Jan/Feb 2012 Zymurgy article that would be great.

 

[lael]

Any one know just how much difference stainless vs copper makes?

 

[Burt de Ernie]

No idea what you just said, but surely the higher the flow rate the more turbulence, which means more efficient heat exchange. The link Danestead made, although not directly to the Zymurgy article, notes "His result was that the chiller worked best with the highest water pressure, and that swirling the chiller in the wort chilled the wort even faster." Surely that should be enough to put this to bed, although if someone has a copy of the Jan/Feb 2012 Zymurgy article that would be great.

This is the summary from Zymurgy 2012 jan/feb




Nuff said!

 

[Burt de Ernie]

In the interest of all here I have uploaded the whole article for reference below.

 

[danestead]

Whats the summary of the article? Its too poor quality to read im afraid.

 

[Burt de Ernie]

Whats the summary of the article? Its too poor quality to read im afraid.

Laminar flows mean jack shit.

Run the water as cold and as fast as you can and increase surface area (AKA pipe length)

 

[Black n Tan]

Look at figure 1. In all cases increasing the flow rate improved the cooling performance.

 

[Ducatiboy stu]

Your analogy is a bit daft

Sorry for my simplistic view.

 

[Burt de Ernie]

Sorry for my simplistic view.

Don't worry mate....I loved it....

 

[Ducatiboy stu]

Copper beats stainless steel, and higher water rates cool more quickly than lower rates. Thermodynamics wins.

Yep. I did have a stainless coil.

Copper does win.

 

[Ducatiboy stu]

Don't worry mate....I loved it....

Please..dont...you will upset the uber brewers

 

[Ducatiboy stu]

Laminar flows mean jack shit.

Run the water as cold and as fast as you can and increase surface area (AKA pipe length)

Cold is good....fast wont give better results.

 

[pk.sax]

Ok, a few observations:

1. It does not address optimisation wrt total mass of water put through chiller - assumption is infinite source of cold water.

2. Look at the graphs, the effect of increasing flow rate yields diminishing reduction in chilling time.

3. Wider pipes chill better.

4. Flow rates used in the study aren't too extreme, all of them would lead to a low pressure situation. I.e., pressure effects aren't much in play. If you start turning the flow rate up higher you have to start taking into account flow layers and separation effects. Especially for narrower tubes.

Look up the term Reynolds coefficient. Every experiment needs to be standardised to mean something.

In practice, unless you got a dam full of cold water at your back or don't give a **** about how much you pour onto the lawn you are going to look at your comfort point on the curve where you accept the cooling rate vs flow rate.

Small tweaks that play with pressure can help, if you put some thought into it.

 

[Ducatiboy stu]

One thing I do want to try is the crimped/ribbed copper tube.

Its the ribbed tube which, after talking to an aircon mech gives a better surface area which should give a better heat transfer, both internall & external. But not cheap

 

[Burt de Ernie]

One thing I do want to try is the crimped/ribbed copper tube.

Its the ribbed tube which, after talking to an aircon mech gives a better surface area which should give a better heat transfer, both internall & external. But not cheap

This is what Rinnai use in their heat exchange systems for the same reason. Although they would probably be a lot harder to clean.