Could anyone tell me a way of computing the heat transfer coefficient near a solid wall by using a numerical code?. Many thanks

Peter,

Determine what T_reference you're going to use (near wall cell, T_adiabtic, T_??). Get a Q, Know your Area then:

HTC=Q/A(Twall-T_whatever)

Good luck, you're now half way there to deriving a proper parameter such as Nusselt Number

Fred.

Thanks Fred, my main obstacle is how to apply that formula to complex geometries in order to achieve a HTC and therefore a Nusselt number distribution for the geometrical domain. Some people compute with adiabatic condition on the wall, and then take this temperature as a reference value, but I heard this is not the only technique. Do you have any other clues.

(1). For the wall heat transfer to occur, there must be a temperature gradient at the wall. So, the wall temperature is all that is needed to calculate the heat flux at the wall. (2). If the temperature profile is continuous and monotonic, then you can quantify the temperature gradient by the wall temperature and the temperature at a neighboring point. It is also convenient to move it further outside the thermal boundary, where the temperature is T,free stream. (3). Now the heat transfer can be related to the temperature difference between the free stream and the wall, and you can define the so-called heat transfer coefficient using these two temperatures. (4). This approach is in general acceptable for low speed flow, where the temperature profile is monotonic. For high speed flows (supersonic), the temperature profile will take a different form or profile because of the viscous heating effect. In this case, the adiabatic wall temperature is used. (This is normally higher than the free stream temperature). (5). For other flow problems where it is difficult to define the free stream condition, you are free to define a temperature which is easy to identify, so that the subsequently defined heat transfer coefficient can be applied in the similar flow problems later. At least the heat transfer direction has to be preserved when the wall temperature gradient is replaced by the difference between the wall temperature and another properly selected temperature. It is important to know that the heat transfer coefficient is defined and used for the similar flow problems.

So heat transfer coefficient choice depends actually on the reference temperature that one has to specify, and then use this heat transfer coefficient for similar kind of problems... What if considering the adiabatic wall tmeperature for reference. I have thought of computing first a case with adiabatic walls obtaining the adiabatic wall temperature. Then computing two more cases with specified temperature on the wall (say 100K and -100K the value of the adabatic temperature) obtaining then two heat fluxes q1 and q2 for the two temperatures T1 and T2. The HTC can be obtained doing HTC=(q1-q2)/(T1-T2). Are you aware of this technique? have you done it before? Many thanks

(1). The formula you mentioned is not quite right. (2). I think, there are other factors missing. So, why no derive the formula directly from the definition of the heat transfer coefficient, and express the Q1 and Q2 in terms of the heat transfer coefficient. From there you can easily find the difference and then divide it by (T,wall1-T,wall2). Then Check the result.

Peter, feed your head with the following:

http://www.electronics-cooling.com/h...00_may_a2.html

http://w3.arizona.edu/~thermlab/publica.html

http://www.flotherm.com/technical_papers/t261.pdf

Happy reading! Fred.

(1). Take a look at the book "boundary layer theory", chapter thermal boundary layer in laminar flow, by Schlichting. (2). For incompressible flow, (T,wall-T,free stream) is used. For compressible flow, (T,wall-T,adiabatic wall) is used instead. (3). For other applications, you have to be very careful and follow the author's specific definition. Otherwise, follow the textbook definition.