Discussion Forum

Hi everyone,

I am a new COMSOL user hoping to get some insight into why I am getting different results depending on my choice of Coil Geometry in my COMSOL applications.

I have a geometry representing a chamfered coil of copper wire that I built in SOLIDWORKS and imported into COMSOL via the CAD import module. I assigned the physics "Multi-Turn coil" to the geometry, chose a circular current geometry, and assigned the same current and number of turns each time I ran the calculation. When it came time to define a direction for the current, I tried a few different things to see if the result changed by more than a couple tenths of a Tesla... To my surprise, it did. This coil will be incorporated into an electromagnet design, so it's imperative that I understand how to define the coil before I add in the magnet core. It seems as though the biggest discrepancy is 0.5T, which seems significant.

I have attached several screenshots of the Coil Geometry directions (CoilGeometry_x.png) and the corresponding Magnetic Flux Density plots (Coil_x.png).

It seems like the results corresponding to "Coil_inner" are most consistent with the approximation of the coil as a solenoid, but that makes me worry that it is not a good approximation for the entire coil geometry, since the coil will have several layers of wire and the maximum width of the coil is comparable to the length of the coil. I have read that you should define the Coil Geometry around a loop that is close to the average radius of the coil, but I have no edge that is close to an "average" radius, and the thickness makes this tricky.

Does anyone have any experience modelling chamfered coils or electromagnets in COMSOL? Any advice on this matter is greatly appreciated!

Thank you!

Gulden