Getting CFD Nusselt No to correlate with gnielinski theory

jonheb · January 28, 2018, 12:14

Hi there

I am working on a project which is modelling smooth flow through a rectangular micro-channel. (The project does go further but for now this where I need help).

I am trying to obtain a Nusselt number that correlates with the use of Gnielinski theory but thus far I have had little success. I am trying to find htc by doing q/(Twall-Tbulk) but I don’t understand at what point I should take the temps and how I should. I have entered a heat flux BC.

I am using k omega SST model. I am using ANSYS fluent and their post.

If anymore info is needed just ask

raghulvr · February 16, 2018, 21:33

For example, if you want to calculate Nusselt number at a surface, you need to find Twall- max temp at that point in a surface and Tmean-Avg temp at the surface

You can use your temperature contour to estimate at which point max temp occurs.Later, in CFD post you can create a point by specifying the co-ordinates of that point.
Then create a expression
maxVal(Temperature)@Location(in our case specify the point)

For mean temperature,
areaAve(Temperature)@Location(in our case specify the surface)

Cheers!

jonheb · March 2, 2018, 12:23

Hey there

I'm trying to get Nusselt for in the centre of the channel flow.

In Fluent I have been drawing a line across the wall and using surface integrals, area weighted average, wall temp for Tw

and then for the Tb I go into post and draw a plane at the same location of the line from above and then areaave temp plane.

I have a constant heat flux of 50000

The problem is these results aren't even close to what the Gnielinski correlation says I should be getting.

January 28, 2018, 12:14	IMPORTANT PLEASE HELP!! Getting CFD Nusselt No to correlate with gnielinski theory	#1
jonheb New Member Jonathan Join Date: Jan 2018 Location: Belfast Posts: 6 Rep Power: 8	Hi there I am working on a project which is modelling smooth flow through a rectangular micro-channel. (The project does go further but for now this where I need help). I am trying to obtain a Nusselt number that correlates with the use of Gnielinski theory but thus far I have had little success. I am trying to find htc by doing q/(Twall-Tbulk) but I don’t understand at what point I should take the temps and how I should. I have entered a heat flux BC. I am using k omega SST model. I am using ANSYS fluent and their post. If anymore info is needed just ask Last edited by jonheb; January 28, 2018 at 16:02. Reason: recieved no answers

February 16, 2018, 21:33		#2
raghulvr Member Rahul Join Date: Sep 2015 Posts: 39 Rep Power: 11	For example, if you want to calculate Nusselt number at a surface, you need to find Twall- max temp at that point in a surface and Tmean-Avg temp at the surface You can use your temperature contour to estimate at which point max temp occurs.Later, in CFD post you can create a point by specifying the co-ordinates of that point. Then create a expression maxVal(Temperature)@Location(in our case specify the point) For mean temperature, areaAve(Temperature)@Location(in our case specify the surface) Cheers! jonheb likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFD Meshing Theory. Where can I read something more???	Stabum	Main CFD Forum	4	October 26, 2012 07:48
is there any money in CFD?	T	Main CFD Forum	35	May 9, 2001 20:35
ASME CFD Symposium - Call for Papers	Chris R. Kleijn	Main CFD Forum	0	September 8, 1998 09:19
ASME CFD Symposium - Call for Papers	Chris R. Kleijn	Main CFD Forum	0	September 3, 1998 09:45
salary range	Frank Muldoon	Main CFD Forum	7	August 3, 1998 20:04

March 2, 2018, 12:23		#3
jonheb New Member Jonathan Join Date: Jan 2018 Location: Belfast Posts: 6 Rep Power: 8	Hey there I'm trying to get Nusselt for in the centre of the channel flow. In Fluent I have been drawing a line across the wall and using surface integrals, area weighted average, wall temp for Tw and then for the Tb I go into post and draw a plane at the same location of the line from above and then areaave temp plane. I have a constant heat flux of 50000 The problem is these results aren't even close to what the Gnielinski correlation says I should be getting.