Calculating heat transfer Coefficients

Shafmk2 · June 2, 2016, 03:28

Hello Guys

This will be a really crude and novice question, I'am really new to CFD modelling. I use Flow3D for my doctoral project.

How do you accurately calculate the heat transfer coefficient's(forced convective and/or natural convective) for complex geometries? I need this as an input for my model. I read you have to do them empirically by experimentation, but i was never taught how to do that.

I'd greatly appreciate if someone can guide me on this.

mprinkey · June 2, 2016, 09:37

Heat transfer coefficients are useful zero-dimensional approximation, but outside of a handful of classic flow conditions (flat plate, cylinders), they are more fudge-factors than approximations. For a complicated geometry in a complicated flow (especially if free convection is important), you are going to have to run experiments...either physical or numerical to cover your range of operating conditions and then apply dimensional analysis to try to get those experimental results to collapse into a number or a curve of some sort in your parameter space.

To do that, you need to define a pair of reference temperatures...one in the free stream and on at a point in your solid. If you are lucky and your solid is highly conductive, you may be able to assume that the entire solid is at the single reference temperature. If so, then you don't need to do the conjugate heat transfer analysis in the solid.

For free convection, you will need to examine flow under several different temperature differences by changing one of the other. If you have force convection, you will need to define a free stream flow velocity as a boundary conditions and run simulations for several different free stream velocities. If you have both free and force effects, you will have to run simulations varying both temperature difference and flow velocity.

When you are done, you are going to look at the integral of heat flux from the object and plot that versus the temperature difference and/or free stream velocity. If you are lucky, those data will collapse to a line (or plane) and you can fit the curve and come up with your heat transfer coefficient. If you are unlucky, you may need more than a single heat transfer coefficient value, but an entire curve fit.

If you are doing free convection and/or low-speed force convection, do be careful that look at the temporal behavior. Both situations are well known to produce unstable phenomena (vortex shedding, Rayleigh-Taylor instabilities) that are not steady. So, if you are in that flow domain, you will need to run transient simulations and then time-average the heat flux.

Good luck.

June 2, 2016, 03:28	Calculating heat transfer Coefficients	#1
Shafmk2 New Member Shafayat Rashid Join Date: Sep 2015 Location: United Kingdom Posts: 2 Rep Power: 0	Hello Guys This will be a really crude and novice question, I'am really new to CFD modelling. I use Flow3D for my doctoral project. How do you accurately calculate the heat transfer coefficient's(forced convective and/or natural convective) for complex geometries? I need this as an input for my model. I read you have to do them empirically by experimentation, but i was never taught how to do that. I'd greatly appreciate if someone can guide me on this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Error finding variable "THERMX"	sunilpatil	CFX	8	April 26, 2013 07:00
Water subcooled boiling	Attesz	CFX	7	January 5, 2013 03:32
Heat Flux at wall in a conjugate heat transfer problem	Chander	CFX	2	July 9, 2011 22:22
Calculating overall heat transfer coefficient (U-value) of fenestration products	Alaska	Main CFD Forum	0	February 25, 2011 10:09
Heat Transfer Coefficients	Ashish Kumar Gupta	Siemens	6	November 22, 2001 02:50

June 2, 2016, 09:37		#2
mprinkey Senior Member Michael Prinkey Join Date: Mar 2009 Location: Pittsburgh PA Posts: 363 Rep Power: 25	Heat transfer coefficients are useful zero-dimensional approximation, but outside of a handful of classic flow conditions (flat plate, cylinders), they are more fudge-factors than approximations. For a complicated geometry in a complicated flow (especially if free convection is important), you are going to have to run experiments...either physical or numerical to cover your range of operating conditions and then apply dimensional analysis to try to get those experimental results to collapse into a number or a curve of some sort in your parameter space. To do that, you need to define a pair of reference temperatures...one in the free stream and on at a point in your solid. If you are lucky and your solid is highly conductive, you may be able to assume that the entire solid is at the single reference temperature. If so, then you don't need to do the conjugate heat transfer analysis in the solid. For free convection, you will need to examine flow under several different temperature differences by changing one of the other. If you have force convection, you will need to define a free stream flow velocity as a boundary conditions and run simulations for several different free stream velocities. If you have both free and force effects, you will have to run simulations varying both temperature difference and flow velocity. When you are done, you are going to look at the integral of heat flux from the object and plot that versus the temperature difference and/or free stream velocity. If you are lucky, those data will collapse to a line (or plane) and you can fit the curve and come up with your heat transfer coefficient. If you are unlucky, you may need more than a single heat transfer coefficient value, but an entire curve fit. If you are doing free convection and/or low-speed force convection, do be careful that look at the temporal behavior. Both situations are well known to produce unstable phenomena (vortex shedding, Rayleigh-Taylor instabilities) that are not steady. So, if you are in that flow domain, you will need to run transient simulations and then time-average the heat flux. Good luck.