Heat transfer coefficient for a rectangular duct

[elpepe2]

Hi,
I haven't got luck with my previous post (thread391-197659). One thing I need to know is the heat transfer coefficient for a rectangular duct fluid (air) inside, three of the walls are isolated from the enviroment and the remaining is an isothermal wall. The Reynold number corresponds to a turbulent flow. I begin my calculations considering a flow parallel to a flat plate and using the Nusselt number as:Nu = h*L/k = 0.037*Re^(4/5)*Pr^(1/3) but I know the actual Nu should be different. Can anybody help me?

 

[daubery]

elpepe2 wrote:

...heat transfer coefficient for a rectangular duct fluid (air) inside, three of the walls are isolated from the enviroment and the remaining is an isothermal wall. The Reynold number corresponds to a turbulent flow. I begin my calculations considering a flow parallel to a flat plate and using the Nusselt number as:Nu = h*L/k = 0.037*Re^(4/5)*Pr^(1/3) but I know the actual Nu should be different. Can anybody help me?

desA replies:

What do you think the Nu -> h value should be?

For a first pass, I'd use:

Nu = h*L/k = 0.027*Re^(4/5)*Pr^(1/3)*(v/vs)^0.14

where:
v = mean flow viscosity
vs = viscosity at wall

Works for gases at Re>10,000, for round ducts. For a rectangular duct, you should use hydraulic diameter in your computations.

With this h value, you can then compute the heat extracted from the single isothermal side:

q' = h * A,s * (Tw - Tsur)

where:
A,s = isothermal surface
Tw = wall temperature
T,sur = core (mean) duct temperature.

You may need to take a guess at the Tsur (haven't checked) & iterate until your solution settles (if outlet temp limited).

A good book is "Fundamentals of heat and mass transfer", Incropera F.P. & DeWitt D.P.


Des Aubery...
(adTherm Technology - adthermtech dot com - des@adthermtech dot com)

 

[elpepe2]

Hi Daubery, thanks for your answer. I already have used hydraulic diameter because I needed to make the coupling of a circular 6" diameter duct from which the air come, to the rectangular duct where I need to work!. The idea was to have the same hydraulic diameters for both ducts avoiding to the minimum any pressure drop. But, I don't know where to use this diameter in the heat transfer. I mean,Prandtl's number is 0.71 for air at my conditions, and viscosities (are you meaning dynamic viscosities?) are known. So, the only parameter with geometric influence is the Reynold's number. I use:

Re = p*v*L/u

Where p is air density, v is velocity, u is dynamic viscosity and L is the length (not a perimeter or width) of the plate. Well, this apply for a flat plate with a flow parallel to it. How would you calculate the Reynold's number for the actual problem?

 

[daubery]

Use the hydraulic diameter (D,h) in the equation, as follows:

L -> D,h

Re = dens * Vel * D,h / visc

This should get you to h.

From h, compute q.



Des Aubery...
(adTherm Technology - adthermtech dot com - des@adthermtech dot com)

 

[elpepe2]

Ok, I will try it. Thanks again.

 

[prex]

As you have turbulent flow, I would consider the duct as a pipe exchanging on the whole periphery to determine h, then using it for the uninsulated wall only to calculate the heat flow.

prex

http://www.xcalcs.com

: Online tools for structural design

http://www.megamag.it

: Magnetic brakes for fun rides

http://www.levitans.com

: Air bearing pads