The inductance computation included in the tool is a guideline at best, and not at all dependable or stable. It would make sense to either change the algorithm (internally use a magnetostatic solver on the line segments or something along those lines) or investigate the inaccuracies of the current algorithm using a commercial solver and some test cases to improve this feature.
The inductance computation included in the tool is a guideline at best, and not at all dependable or stable. It would make sense to either change the algorithm (internally use a magnetostatic solver on the line segments or something along those lines) or investigate the inaccuracies of the current algorithm using a commercial solver and some test cases to improve this feature.