Heat Transfer Coefficient

Dilraj · November 18, 2011, 11:26

Hi all,

I am running a simple analysis of turbulent forced convection (cooling flow over a plate at high temperature). Right now I am struggling to get the HTC to match with the relevant published correlations when using the mean film temperature for the properties. When I use the properties at inlet temperature the match is a lot better.

The method I am using to obtain the HTC is to run two analyses and use the difference of the wall heat flux from both divided by the wall temperature difference.

I expect a small error but I am about 50% out at the moment. I am using the low-Re SST turbulence model and have a y+ of 1. I am also reasonably confident that the solution has converged.

Any help on the matter would be useful. Thanks!

Dilraj

xrs333 · November 20, 2011, 06:14

Hi, Dilraj,
The ref. and qualitative temperature should be same as those used in published correlations, usually the bulk temperature, which is often close to inlet temperature, instead of film temperature should be used, some correlations use average temperature, arithmetical or logarithmic.

Dilraj · November 21, 2011, 03:54

Thanks for the reply xrs333,

I am using correlations that suggest the use of the mean film temp for the properties. However when I compare the numerical results to the published correlations there is a much better agreement with the correlation using properties at the bulk temperature.

Is there an error with the numerical simulation or is this inaccuracy to be expected?

xrs333 · November 21, 2011, 04:36

Sorry, I misunderstood the words 'mean film temperature', my English is so poor...
In heat transfer, the mean film temperature is defined as algebraic average of bulk temp and the wall temp.

sandeep_tu · November 21, 2011, 05:17

Hi,

I didn´t exactly understood your problem other than you want to calculate the HTC. Normally in Fluent for HTC calculation it takes the reference temperature the value which you give in reference conditions panel.

Dilraj · November 21, 2011, 06:17

Thanks for the reply guys.

What I've done is I have extracted the total surface heat flux from the cfd analysis and used the Nusselt Number correlations available on this page http://courses.cit.cornell.edu/fluent/plate/step6.htm.

When I plot the curve for the correlation against the curve for the cfd analysis there is a significant difference.

I have evaluated properties at the inlet temperature.

I have tried changing the analysis type to compressible but it doesnt seem to make a huge amount of difference. Any other ideas?

Cheers.

November 18, 2011, 11:26	Heat Transfer Coefficient	#1
Dilraj New Member Dilraj Sehmi Join Date: Nov 2011 Posts: 4 Rep Power: 13	Hi all, I am running a simple analysis of turbulent forced convection (cooling flow over a plate at high temperature). Right now I am struggling to get the HTC to match with the relevant published correlations when using the mean film temperature for the properties. When I use the properties at inlet temperature the match is a lot better. The method I am using to obtain the HTC is to run two analyses and use the difference of the wall heat flux from both divided by the wall temperature difference. I expect a small error but I am about 50% out at the moment. I am using the low-Re SST turbulence model and have a y+ of 1. I am also reasonably confident that the solution has converged. Any help on the matter would be useful. Thanks! Dilraj

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Heat transfer coefficient - what is waht	Stan	FLUENT	28	December 29, 2021 16:29
Water subcooled boiling	Attesz	CFX	7	January 5, 2013 03:32
Constant velocity of the material	Sas	CFX	15	July 13, 2010 08:56
Convective / Conductive Heat Transfer in Hypersonic flows	enigma	Main CFD Forum	2	November 1, 2009 22:53
Convective Heat Transfer - Heat Exchanger	Mark	CFX	6	November 15, 2004 15:55

November 20, 2011, 06:14		#2
xrs333 Senior Member xrs333 Join Date: Aug 2010 Posts: 125 Rep Power: 15	Hi, Dilraj, The ref. and qualitative temperature should be same as those used in published correlations, usually the bulk temperature, which is often close to inlet temperature, instead of film temperature should be used, some correlations use average temperature, arithmetical or logarithmic.

November 21, 2011, 03:54		#3
Dilraj New Member Dilraj Sehmi Join Date: Nov 2011 Posts: 4 Rep Power: 13	Thanks for the reply xrs333, I am using correlations that suggest the use of the mean film temp for the properties. However when I compare the numerical results to the published correlations there is a much better agreement with the correlation using properties at the bulk temperature. Is there an error with the numerical simulation or is this inaccuracy to be expected?

November 21, 2011, 04:36		#4
xrs333 Senior Member xrs333 Join Date: Aug 2010 Posts: 125 Rep Power: 15	Sorry, I misunderstood the words 'mean film temperature', my English is so poor... In heat transfer, the mean film temperature is defined as algebraic average of bulk temp and the wall temp.

November 21, 2011, 05:17		#5
sandeep_tu Member Sandeep Join Date: Apr 2009 Location: Munich, Germany Posts: 30 Rep Power: 15	Hi, I didn´t exactly understood your problem other than you want to calculate the HTC. Normally in Fluent for HTC calculation it takes the reference temperature the value which you give in reference conditions panel.

November 21, 2011, 06:17		#6
Dilraj New Member Dilraj Sehmi Join Date: Nov 2011 Posts: 4 Rep Power: 13	Thanks for the reply guys. What I've done is I have extracted the total surface heat flux from the cfd analysis and used the Nusselt Number correlations available on this page http://courses.cit.cornell.edu/fluent/plate/step6.htm. When I plot the curve for the correlation against the curve for the cfd analysis there is a significant difference. I have evaluated properties at the inlet temperature. I have tried changing the analysis type to compressible but it doesnt seem to make a huge amount of difference. Any other ideas? Cheers.