ahmad_m_eid wrote:Still I have problems in force calculation.I often keep increasing the mesh density until the force results converge to whatever accuracy I want. One way to do this is to change the mesh size in all blocks by hitting F3, which halves the mesh size in all blocks, or F4, which doubles the mesh size in all blocks. 2- do you think that the block integral is better than the lineYes. 3- do you know any reference to calculate this force analytically?With most designs, you can get a pretty good notion of what the machine should be doing from a somewhat idealized but analytical approach. However, it's difficult to quote you an analytical _expression_ without knowing your machine's topology and geometry. A good reference for you might be Boldea and Nasar's "Linear Electric Actuators and Generators," Cambridge University Press, 1997. This book considers analytical design methods for many different sorts of linear machines. I notice something, when I change the area surrounding the problem,Check out appendix A3 -- "Open Boundary Problems" in the FEMM 3.3 manual. If you want to approximate an unbounded domain by just putting a lot of air around it, a "rule of thumb" from the literature (Q. Chen and A. Konrad, “A review of finite element open boundary techniques for static and quasistatic electromagnetic field problems,” IEEE Transactions on Magnetics, 33(1):663-676, January 1997) is that the distance from the center of the problem to the outer boundary should be at least five times the distance from the center to the outside of the objects of interest. Other techniques, which allow one to model an "open" boundary in more elegant ways, are also discussed in this appendix. Dave. |