Water cooling of steel 2

[Speedy]

I have an application requiring the cooling of a block of steel from 150 to 60degC, using room temp water through a central channel/bore.

I want to estimate the approximate cooling time?

I suspect I can simplify the problem by assuming turbulent flow and a very small rise in water temp from inlet to outlet. I also know the specific heat and conductivity of the steel. The shape of the steel is complex, but I could simplify it to a regular shape. I am not looking for really fast cooling time, 5 mins or so would do.

What value of heat transfer coefficient do I use?

Appreciate any help,

Speedy

Speedy

 

[rjw57]

Speedy -
Are you trying to remove heat from the block, as in to cool the entire part from 150 to 60 degC? If you are, this is a fairly complicated problem in that you must perform what is called a Heisler analysis (to keep it as simple a calculation as possible) or some form of finite difference method. Do you have access to a heat transfer text? You probably already realize that your coolant will initially boil as it passes thru the central channel/bore, but fairly soon after this, you will have cooled it locally in the bore so that water will flow without making steam (only to say, be careful). You will definitely have to simplify your geometry assumptions, i.e. block, cylinder or sphere. Depending on the size of this block, once you have finally cooled enough to flow liquid water (no steam), you will probably be less interested in the interior heat transfer coefficient than the conductivity of the steel since you will probably be conduction limited IF the block is big enough. That is to say you will not see much of a rise on the coolant temperature since you will be limited by the diffusion of energy in the block. All this babble assumes that you need the entire block below 60degC (at all locations) and that the block is big enough to be affected by internal conduction rather than the water side heat transfer coefficient. Good luck with this problem...

BTW, For the inside heat transfer coefficient (liquid ONLY), ~(0.023*Re^0.8*Pr^0.66)*(k/Dh) where Re = Reynolds number, Pr = Prandtl Number, k = Thermal Conductivity (of coolant), Dh = passage hydraulic diameter (passage diameter if passage geometry is tubular).

 

[witchdoc]

Try the "lumped heat capacity" equation found in most heat transfer texts. This will give you a quick and dirty estimate of the cooling time. Note that this method only gives accurate results when certain ratios of heat capacities, surface areas, etc. are applicable.

 

[poetix99]

The simplifications that are legitimate really depend upon (1) some of the details of your situation AND (2) just what it is that you want to know; what part of the "block" is supposed to reach 60degC?

If the "block" is in a 25degC room, the outer surface will (at least initially) be more influenced by heat transfer to the ambient air. Skin temperatures could reach 60C well before core temperatures. Then you stop cooling, and the surface temperature goes back up...

If the water flow rate is sufficiently high and the bore length to bore diameter ratio is not "too" large, constant temperature of the coolant core is not a big error.

 

[vtl]

Your first heat transfer coefficient is the mean temperature between the water temperature at the inlet(? C) and the tempurature of the block before cooling (which is 150 C). You can use the Log Mean Temperature method to approximate the heat transfer.