# HTC w/ adiabatic wall

 User Name Remember Me Password
 Register Blogs Members List Search Today's Posts Mark Forums Read

 July 26, 2007, 10:37 HTC w/ adiabatic wall #1 Kali Guest   Posts: n/a In my CFX case I have adiabatic walls with energy equation solved. When I ploted for wall heat transfer coefficients, it gave a heat transfer coeffient contour plot. I wonder what are those numbers..Does it calculate heat transfer coeffient using analogy from cf. Or its just a set of junk numbers. Please clarify. Thanks in advance. Kali

 July 26, 2007, 10:43 Re: HTC w/ adiabatic wall #2 Robin Guest   Posts: n/a It returns the heat transfer coefficient resulting from the turbulent wall function. This HTC is used by the solver to calculate the wall heat flux and/or temperature, depending on what boundary condition you have applied. Regards, Robin

 July 26, 2007, 11:00 Re: HTC w/ adiabatic wall #3 Kali Guest   Posts: n/a Robin, Is it independent of the type of BCs you give? Provided you near wall fluid temperature and fluid property remain the same. I saw in the manual that T+ is function of qw and Tw-Tf. With adiabatic wall how does it calculate heat transfer coefficient from turbulent wall function. Would you please elaboare more with reference to the manual. Thanks, Kali

 July 26, 2007, 11:18 Re: HTC w/ adiabatic wall #4 Robin Guest   Posts: n/a Hi Kali, T+ is defined as a function of qwall but it is modeled as eq. 183 in the v11 manual. The HTC is thus defined as: HTC = rho*Cp*u_star/T+ Note that this HTC is grid dependant, since the fluid temperature is the conservative value of temperature at the wall node (the wall node stores values for the near wall control volume). Keep this in mind if you intend to use this heat transfer coefficient for any other purposes, such as mapping to an FEA thermal analysis. If you are comparing to data, calculate an HTC the same way it was done for the experiment, don't use this one. Regards, Robin

 July 27, 2007, 11:47 Re: HTC w/ adiabatic wall #5 Kali Guest   Posts: n/a Robin, If you apply wall temperature, will the heat flux estimated will depend on the grid..paticularly the near wall Y+. please comment.

 July 27, 2007, 12:02 turbulence modeling for compressible flow #6 Kali Guest   Posts: n/a What are the extra variables/factors need to be taken care in CFX for turbulence modeling for compressible flow. Thanks, Kali

 July 27, 2007, 16:49 Re: HTC w/ adiabatic wall #7 Robin Guest   Posts: n/a Yes, in the sense that it is a function of the boundary layer and your grid spacing will determine how accurately you resolve the boundary layer. This is one reason why the SST model is preferred for heat transfer; it can give you a more accurate representation of the boundary layer (with sufficient mesh). Regards, Robin

 July 27, 2007, 16:52 Re: turbulence modeling for compressible flow #8 Robin Guest   Posts: n/a Use a compressible fluid material (Ideal Gas, Redlich-Kwong, etc.) and turn on the "Total Energy" equation. If your walls are moving relative to the frame (i.e. a spinning wall, counter-rotating wall, or wall with a specified velocity) turn on the viscous dissipation term. Everything else is handled automatically by the solver. -Robin

 July 29, 2007, 09:21 Re: turbulence modeling for compressible flow #9 Kali Guest   Posts: n/a Robin, I have taken care of those features. In particular to turbulenec meodeling, should I use some advanced option like production term modification etc.. Thanks, Kali

 July 30, 2007, 09:08 Re: turbulence modeling for compressible flow #10 Robin Guest   Posts: n/a No, the code takes care of the rest.

 August 7, 2007, 06:45 Re: HTC w/ adiabatic wall #11 kali Charan Nayak Guest   Posts: n/a Robin, When I ran the same model with KE turbulenec mode, my wall heat transfer coefficients quite low by 4-5 times? why? Thanks in advance. Quick reply will be helpful. Kali

 August 7, 2007, 10:20 Re: HTC w/ adiabatic wall #12 Robin Guest   Posts: n/a Hi Kali, The wall treatment of the k-e and SST models differ significantly, especially where the near wall resolution is fine. For low Y+ values, the SST model will use a linear wall function, resolving the viscous sub-layer directly, and blend to a logarithmic wall function as Y+ increases. The k-e model will always use a wall function and therefore overpredict turbulent mixing near the wall if the Y+ is small. This is probably the reason for the discrepancy but there are also the possibility that upstream b.l. and/or differences in the free-stream treatment have changed the local flowfield. I would suggest comparing the two results to see how and where they differ. Finally, the SST model is much better for heat transfer. In the absence of any experimental data to definitively determine which is correct, I would give greater credence to the SST results. Regards, Robin

 Thread Tools Display Modes Linear Mode

 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 OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Yr0gErG FLUENT 3 June 12, 2013 02:12 anita OpenFOAM Native Meshers: blockMesh 31 April 4, 2013 11:51 mehrdadeng CFX 10 February 25, 2011 06:25 MMACRB FLUENT 1 January 29, 2007 03:42 D.Tandra Main CFD Forum 2 March 16, 2004 05:29

All times are GMT -4. The time now is 19:56.

 Contact Us - CFD Online - Top