CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > FloEFD, FloWorks & FloTHERM

heat transfer coefficient value in floefd

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree4Likes
  • 2 Post By Chris_321
  • 1 Post By Boris_M
  • 1 Post By AlgoaBay

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   April 28, 2016, 21:24
Default heat transfer coefficient value in floefd
  #1
New Member
 
yongxin zhou
Join Date: Feb 2016
Posts: 5
Rep Power: 8
jason kid is on a distinguished road
I have finished a thermal simulation of the LED Fog Lamp with two conditions no wind and 0.5m/s wind face the lamp, in Floefd.
and I find that, from the simulations, the temperature results are almost the same with a bit different.

in the simulations, I use the out air, so after the calculation, I checked the HTC of the out walls of the lamp:

0m/s condition: HTC main in 5-6 W/m2K distributed
0.5m/s condition: HTC main in 11-16 W/m2K distributed

even the HTC changed a lot, but I think the HTC is also small to change the temperature obviously.

but the question is coming...
I consulted many papers, most of them tell me that, when the air speed is 0, the HTC experience value should be about 10-12, and double times in 0.5m/s.

they can not match completely,
in my opinion, the values from paper should be more exact, maybe some settings in the software are missed such as wall roughness? I am not sure.
jason kid is offline   Reply With Quote

Old   May 8, 2016, 16:47
Default
  #2
Member
 
zduno
Join Date: Dec 2013
Posts: 55
Rep Power: 11
zdunol is on a distinguished road
I would say that HTC depends on the turbulence (boundary layer thickness) so I think you can only say that HTC's value at this velocity is equal to something only when your geometry and its alignment to gravity is well defined (vertical/horizontal plane/cylinder). However if what you get from papers regards your geometry and hydrodynamics then I would resolve the mesh and see what happens :P
zdunol is offline   Reply With Quote

Old   June 16, 2016, 04:50
Default
  #3
Senior Member
 
Join Date: Jun 2011
Posts: 138
Rep Power: 13
CFDfan is on a distinguished road
Quote:
Originally Posted by jason kid View Post
I have finished a thermal simulation of the LED Fog Lamp with two conditions no wind and 0.5m/s wind face the lamp, in Floefd.
and I find that, from the simulations, the temperature results are almost the same with a bit different.

in the simulations, I use the out air, so after the calculation, I checked the HTC of the out walls of the lamp:

0m/s condition: HTC main in 5-6 W/m2K distributed
0.5m/s condition: HTC main in 11-16 W/m2K distributed

even the HTC changed a lot, but I think the HTC is also small to change the temperature obviously.

but the question is coming...
I consulted many papers, most of them tell me that, when the air speed is 0, the HTC experience value should be about 10-12, and double times in 0.5m/s.

they can not match completely,
in my opinion, the values from paper should be more exact, maybe some settings in the software are missed such as wall roughness? I am not sure.
I also noticed that floefd underestimates the heat transfer coefficient from the enclosure surfaces especially in free air models. And I am talking here compared with practice. The emisivity I use in the model is from the datasheets of the paint so I didn't make obvious mistakes. Also the heat transfer is laminar. Using finer mesh has never helped and I just accepted that the results got conservative and the temperature would be lower in reality. Would like also to know why Floworks fails to calculate correctly the temperature of free air models?

Last edited by CFDfan; July 4, 2016 at 11:43.
CFDfan is offline   Reply With Quote

Old   July 14, 2016, 11:19
Default
  #4
Member
 
Chris_321's Avatar
 
Christian
Join Date: Sep 2013
Location: Germany
Posts: 88
Rep Power: 11
Chris_321 is on a distinguished road
I get good results with FloEFD for temperature in LED free stream models. The difference in a validation model between measurement and simulation was round about 1K for temperature. ( High precision measurement with 0.1mm thermocouples)

The difference between the two results from your measurement and the simulation could be because of wrong parameters for the radiation.

One thing you should keep in mind. The definition of the HTC is in EFD a little bit different than in papers that use something like Nu correlations. In EFD you have to define a reference temperature, for calculating the HTC. Thats because (like every CFD software) EFD dont use the HTC for calculating the heat transfer. You should check the technical reference for more infos.
Boris_M and Sai Krishna like this.
Chris_321 is offline   Reply With Quote

Old   August 8, 2016, 05:17
Default
  #5
Senior Member
 
Join Date: Jun 2011
Posts: 138
Rep Power: 13
CFDfan is on a distinguished road
Quote:
Originally Posted by Chris_321 View Post
I get good results with FloEFD for temperature in LED free stream models. The difference in a validation model between measurement and simulation was round about 1K for temperature. ( High precision measurement with 0.1mm thermocouples)

The difference between the two results from your measurement and the simulation could be because of wrong parameters for the radiation.

One thing you should keep in mind. The definition of the HTC is in EFD a little bit different than in papers that use something like Nu correlations. In EFD you have to define a reference temperature, for calculating the HTC. Thats because (like every CFD software) EFD dont use the HTC for calculating the heat transfer. You should check the technical reference for more infos.
Which parameters (for radiation) do you mean - the reference temperature is not a parameter asked for when specifying radiation in SWFW. It is not available either in the results section when specifying surface parameters -> heat transfer coefficient.
CFDfan is offline   Reply With Quote

Old   August 16, 2016, 06:47
Default
  #6
Senior Member
 
Join Date: Jul 2009
Posts: 616
Rep Power: 22
Boris_M will become famous soon enough
Chris is correct.

It is very important to provide correct radiation properties to solids and surfaces. The absorption in transparent plastic for example has to be accurately defined if the band definition is used and a deviation in the band limits can result in a reduced average absorption value for the band. Same goes for the reflection and emission coefficients.

As for the reference temperature, it is not shown by default on where to set it. Just like some boundary conditions are not shown by default in the analysis tree, you have to right click on the project name in the analysis tree, the very first name directly above the "input data" and "results" part of the tree. Here you can select "customize tree" and in the bottom section which is for the results there should be the point "Reference parameters". Simply activate that and then it shows in the results part of the tree. Here you can right click on it and set the default reference parameters such as fluid temperature and others. The pressure for example is important for correct force calculation as it will be used if a body is selected with some of the surfaces that are not exposed to the fluid and therefore the pressure on those surfaces is not known and need a reference value.

I hope this helps,
Boris
Boris_M is offline   Reply With Quote

Old   August 17, 2016, 16:16
Default
  #7
Senior Member
 
Join Date: Jun 2011
Posts: 138
Rep Power: 13
CFDfan is on a distinguished road
Quote:
Originally Posted by Boris_M View Post
Chris is correct.

It is very important to provide correct radiation properties to solids and surfaces. The absorption in transparent plastic for example has to be accurately defined if the band definition is used and a deviation in the band limits can result in a reduced average absorption value for the band. Same goes for the reflection and emission coefficients.

As for the reference temperature, it is not shown by default on where to set it. Just like some boundary conditions are not shown by default in the analysis tree, you have to right click on the project name in the analysis tree, the very first name directly above the "input data" and "results" part of the tree. Here you can select "customize tree" and in the bottom section which is for the results there should be the point "Reference parameters". Simply activate that and then it shows in the results part of the tree. Here you can right click on it and set the default reference parameters such as fluid temperature and others. The pressure for example is important for correct force calculation as it will be used if a body is selected with some of the surfaces that are not exposed to the fluid and therefore the pressure on those surfaces is not known and need a reference value.

I hope this helps,
Boris

Boris this is good rhetoric, but where in reality to get this "correct" radiation property data from. In enclosure cases one can only get the paint emissivity data. If one is persistent he could also get manufacturing data about the texture (roughness) of the paint, but how to make SWFS accounting for the roughness of the model surface. I know the roughness of a boundary condition "wall" can be specified in SWFS but don't know how to specify it for a model surface.

Thank you for your explanations about the reference temperature but this is the (ambient) temperature specified in the initial wizard. Is that temperature always used by SWFS to calculate HTC?
My understanding is that HTC value should be calculated referencing to the fluid temperature adjacent to the radiating surface in question. Such fluid temperature can be much higher that the ambient temperature for radiation surfaces, say, inside the enclosure. What I am saying is that the reference fluid temperature is different for the different surfaces of the model. Does the software account for this variation when calculating the HTC of model surfaces
CFDfan is offline   Reply With Quote

Old   August 18, 2016, 04:14
Default
  #8
Senior Member
 
Join Date: Jul 2009
Posts: 616
Rep Power: 22
Boris_M will become famous soon enough
Hi CFDfan,

Yes, you are right, it is not always that easy to get material properties but you shouldn't expect CFD tools to deliver them all. For that there are just to many variations and mixtures possible. Usually you can get some material properties from the vendor, in particular for optical elements such as filters and glasses for lenses etc. They are provided by BASF for example or other manufacturers such as Schott.
As for paint properties it is probably a little harder but I usually never had such a big problem defining a lighting application project with boundary conditions. In many cases I use black body as a wall condition in default and some reflective condition with around 90% reflection and 10% absorption for reflectors. Only the transparent material should be considered more closely as here the absorption coefficient can heavily depend if you use PMMA or Quartz Glass or other plastic.

Also in most cases I don't even define roughness for the walls. In most cases the surface is either too short to have a big influence or the channels are not that narrow and long as well. I usually consider it more in some longer channels or pipes where the roughness starts to play a larger role.

Yes, the reference temperature is used by default from the ambient temperature but if you consider an enclosure where the temperature in the interior is higher and you want to know the HTC of that side of the model then the ambient temperature on the outside is the wrong temperature to use. However this only is of interest if you want to evaluate the HTC as a value from your results, the solver doesn't work with the HTC in this way, this is usually a parameter that the user is interested as it is a value he can understand and compare. Similar to an aerodynamic calculation where the solver calculates pressures and shear stress on the surface. The solver doesn't care about forces on the body or the drag coefficient. Those are parameters the user understands and can use to compare results.
So for radiation cases the ambient temperature also only plays a role for the radiation to the outside of the computational domain as here this value is assumed to radiate back and lets the solver define in which direction the radiation energy goes and to what amount. As soon as there is another surface opposite to the radiating surface, the fluid temperature in between basically becomes irrelevant as it will be a surface-surface radiation. The fluid itself practically doesn't participate in the radiation, or at least to a negligible amount in most cases.

Yes, of course the surfaces can be many and the fluid temperature varies in the model. The question is which fluid temperature would you want to use in what distance of the model. In all CFD tools this is the same question.
If you refer to any literature then that literature should also reference their reference temperature used to estimate the HTC for that velocity. The HTC by definition is the ratio between the heat flux and the temperature difference between solid wall temperature and the surrounding fluid. But as you can tell that temperature is not constant and it also is not defined in which distance. Ideally the far field is used, where the temperature is undisturbed. But in reality in an enclosure, that temperature is hard to define.

My advise is to use not all faces of the model if the location varies too much and therefore also the surrounding fluid temperature, but only some selected faces and select the reference temperature for those surfaces according to the approximate air temperature a little away from the direct surface.
You can also create reference temperatures for specific surfaces if you right click the reference parameter point in the result tree and select "insert..." and then you can apply it to selected surfaces individually.
A possible way of defining the fluid temperature would be using the surface parameter and check for the average fluid temperature of the desired surfaces. It will be the fluid temperature directly next to the surface but it will be an easier selection than in a complex fluid temperature distribution around the surfaces. Or you can build a volume around the body or for the interior of the enclosure and use the volume parameter and check for the average surface temperature this way. Of course the hotter surfaces near the body will influence the average of the usually desired far field but what do you want to do else?
You can also always assume a fixed value if you want to compare design changes.

As you can see, this is almost like a chicken and egg problem.

Hope this helps,
Boris
CFDfan likes this.
Boris_M is offline   Reply With Quote

Old   August 18, 2016, 20:20
Default
  #9
Senior Member
 
Join Date: Jun 2011
Posts: 138
Rep Power: 13
CFDfan is on a distinguished road
Thank you Boris. Your explanations answered a couple of questions I always struggled with. This will also help jason kid who started the topic

1. No, I am not expecting SWFS to deliver all the answers of course. I just expressed my observation of temperatures from SWFS being higher than reality in cases of enclosures in free air cooling conditions. At the same time SWFS was pretty accurate in forced air conditions where the radiation plays very little role (and was not enabled). So my conclusion was that in free air something was not quite right with the radiation portion of heat transfer. And since the variables there were temperature difference and emissivity I was struggling to figure out which of them was the troublemaker.

1. As a designer I need to built a "database" of the HTC value in various conditions to evaluate roughly (in front) if a certain design is thermally feasible. After your explanations the HTC values in my "database" for internal surfaces and heatsinks have to be corrected
CFDfan is offline   Reply With Quote

Old   October 12, 2016, 09:20
Default
  #10
New Member
 
G.
Join Date: Mar 2016
Posts: 4
Rep Power: 8
AlgoaBay is on a distinguished road
Hi there,

Air velocity =0 m/s means natural convection plus radiation will determine the HTC value.

I did studies on natural convection and the HTC matches with literature/experience.

Run a simple simulation with a cube at T=const, with and without radiation. Results should be self explaining.
Boris_M likes this.
AlgoaBay is offline   Reply With Quote

Old   October 16, 2016, 17:55
Default
  #11
Senior Member
 
Join Date: Jun 2011
Posts: 138
Rep Power: 13
CFDfan is on a distinguished road
Quote:
Originally Posted by AlgoaBay View Post
Hi there,

Air velocity =0 m/s means natural convection plus radiation will determine the HTC value.

I did studies on natural convection and the HTC matches with literature/experience.

Run a simple simulation with a cube at T=const, with and without radiation. Results should be self explaining.
Dear AlgoaBay

You are just posting this for the sake of posting.

The only useful thing from you post would have been experimental results matching simulated.
CFDfan is offline   Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Heat Transfer Coefficient Alberto Schroth FLUENT 17 February 10, 2020 09:40
Surface Heat Transfer Coefficient CFD Post MJ1105 Visualization & Post-Processing 0 February 19, 2016 05:57
Ansys CFX problem: unexpected very high temperatures in premix laminar combustion faizan_habib7 CFX 4 February 1, 2016 18:00
Heat Transfer Coefficient in Compressible Flow 3D turbine cascade Karkoura CFX 0 March 10, 2011 16:35
Constant velocity of the material Sas CFX 15 July 13, 2010 09:56


All times are GMT -4. The time now is 14:48.