Ivar KJELBERG
                                                                                                                                                    COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
                                                         
                            
                         
                                                
    
        Please login with a confirmed email address before reporting spam
     
    
 
                                                Posted:
                            
                                1 decade ago                            
                            
                                2010年12月16日 GMT+8 18:06                            
                        
                        
                                                    Hi
if you have a heat source of 25[W] for a given volume, the average heat density you should apply is for a 3D domain
Q[W/m^3] = 25[W]/(your_ domain_volume[m^3])
the your_domain_volume is obtained by a integration variable  (or volume operator in V4) of "1" as integrand.
What you must catch with COMSOL is that the values you enter in the GUI fields are defined "per FEM element" of size dV=dx*dy*dz (in Cartesian) so you enter body forces in [N/m^3] and heat in [W/m^3] etc with the assumption that you integrate over the applicable domains.
for 2D nothing changes, but COMSOL assumes its working in 3D with a depth of ".d" or ".thickness" (by default = 1[m] for most physics) so then you must multiply you integrand (on a surface by 1*solid.d for a ST case.
In 2D-axi it is the "loop length" that matters which is "2*pi*r"
--
Good luck
Ivar                                                
                                                
                            Hi
if you have a heat source of 25[W] for a given volume, the average heat density you should apply is for a 3D domain
Q[W/m^3] = 25[W]/(your_ domain_volume[m^3])
the your_domain_volume is obtained by a integration variable  (or volume operator in V4) of "1" as integrand.
What you must catch with COMSOL is that the values you enter in the GUI fields are defined "per FEM element" of size dV=dx*dy*dz (in Cartesian) so you enter body forces in [N/m^3] and heat in [W/m^3] etc with the assumption that you integrate over the applicable domains.
for 2D nothing changes, but COMSOL assumes its working in 3D with a depth of ".d" or ".thickness" (by default = 1[m] for most physics) so then you must multiply you integrand (on a surface by 1*solid.d for a ST case.
In 2D-axi it is the "loop length" that matters which is "2*pi*r"
--
Good luck
Ivar                        
                                                
                                                                                                            
                                             
                                            
                                                
    
        Please login with a confirmed email address before reporting spam
     
    
 
                                                Posted:
                            
                                1 decade ago                            
                            
                                2010年12月17日 GMT+8 20:42                            
                        
                        
                                                    hey ivar,
thanks for you quick reply. theoretically i understand what you told me, but i have a problem in v4 to find the volume operator you mentioned, because i'm new at v4
thanks for your help and have a nice weekend!
jaques                                                
                                                
                            hey ivar,
thanks for you quick reply. theoretically i understand what you told me, but i have a problem in v4 to find the volume operator you mentioned, because i'm new at v4
thanks for your help and have a nice weekend!
jaques                        
                                                
                                                                                                            
                                             
                        
                        
                            
                                                                                        
                                Ivar KJELBERG
                                                                                                                                                    COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
                                                         
                            
                         
                                                
    
        Please login with a confirmed email address before reporting spam
     
    
 
                                                Posted:
                            
                                1 decade ago                            
                            
                                2010年12月17日 GMT+8 23:45                            
                        
                        
                                                    Hi
the coupling variables are in v4 defined as "Definitions Model Couplings Iintegration" on a domain 
--
Good luck
Ivar                                                
                                                
                            Hi
the coupling variables are in v4 defined as "Definitions Model Couplings Iintegration" on a domain 
--
Good luck
Ivar