|February 28, 2014, 08:53||
Re: Magnetic field around a current carrying square bar
Join Date: Mar 2013
Posts: 67Rep Power: 4
I am working on the aforementioned problem in ANSYS FLUENT. The aim is to validate this case by implementing UDS - 3 equations for magnetic vector potential and 1 for electric scalar potential.
I have modelled the conductor as a box and made an enclosure around it.
For the V(electric scalar pot) equation I need to define 2 separate electrical conductivity - one for air and for copper. I have defined both of these as fluids and unchecked the flow equation option.
1. I have tried to do it using Define_Diffusivity UDF and Thread_ID, but the problem is that the conducitivity for copper is not being taken into account. It should be taken as 2700 but I am not able to comprehend why it is being taken as 85.
2. While giving the boundary conditions, FLUENT is slitting the conductor up and down walls into coupled walls and I am getting confused on how to give the boundary condition - I need to give 670 V on conductor_up and 0 V on conductor_down. How should I apply the boundary conditions
3. When I model the copper bar as solid, and solve the UDS over all zones, while initializing I am getting Unknown UDS property method error. What does this mean? Also could someone kindly tell what does "coupled wall" mean?
4. Lastly, I would like to ask does zero normal gradient of field imply that the flux is zero?
I am attaching the UDF I have written, boundary conditions to be applied and a the plot of diffusion coeff of V to highlight my problem.
Any help would be really appreciated.
|March 13, 2014, 08:51||
Join Date: Dec 2013
Posts: 41Rep Power: 3
Dear friend yashganatra,
i haven't read over the thesis you list to me yet, but here i have some opinion and advice about your problem,you can think about it.
1.In the UDF, you used "C_UDSI_G" macros in the source term,but you have to know, when fluent compute for the first step, there isn't any gradients.So if C_UDSI_G was called in the first step of iteration, an error may occur.
in my UDF, i usually put C_UDMI and C_UDSI_G in ADJUST macro, and call them when one or two steps of iteration were done.
2.You said you have defined both of air and copper as fluids, i doubt that, shouldnot copper be taken as solids?
3.If i were you, i would choose another software(like ansys or ansoft) to do electromagnetic field problem like this,and if you need to couple electromagnetic field with thermal or flow field ,you can import the result to fluent.
In my case of plasma arc, i define V and A in fluent,and just set V in the UDS terms in boundary condition panel.And i haven't met a case(couple air and copper) like this.May my advice will help you.And I will be glad to discuss with you.
|boundary condition, fluent - udf, magnetic field, uds diffusive|
|Thread||Thread Starter||Forum||Replies||Last Post|
|particles in a localized magnetic field||Arturo Ortiz||FLUENT||1||March 10, 2010 11:26|
|magnetic field, workbench||alireza ramezani||Main CFD Forum||0||March 13, 2008 15:42|
|ACCESS VIOLATION||MHDWill||FLUENT||1||September 23, 2007 02:51|
|where to find velocity field for square pipes???||enrico||Main CFD Forum||2||February 4, 2005 12:32|
|Divergence Free on the Magnetic Field||Valdemir||Main CFD Forum||1||March 19, 2003 19:33|