FEMM - First improvement of the model

In the previous article we saw how to set up a magnetic problem with FEMM. The simulated core was very simple and we saw that the B field wasn't homogeneous across the section. To improve the field distribution we can design the core in a better way: when B distribution is uniform, we remove 'hot spots' in the core, allowing for a better control of its temperature. That is never done since core shapes with rounded corners cost a lot (extra tooling, higher production time)! The best shape for a core is a toroid, which guarantees better performance, lower power dissipation, lower flux leakage, and uniform field distribution.

Let's modify the corner shape, drawing quarters of the circumference at the corners, respectively of 2mm and 6mm of radius, then mesh and simulate: the results obtained are shown below B field has an intensity between 0.2 and 0.24T (0.22T from calculations).

Now let's change the core again with a different shape, findable on the market, with only the external corners rounded and simulate again with the same conditions. The results are shown below.

As you can see, the field distribution seems to be more uniform in the second example but it gives a maximum value along the magnetic path length which is lower than the calculated one. For this reason, in the next articles, I will use a core with both the internal and external corners rounded to have a condition more similar to ideality.