My objective however, was to couple two microstrip transmission lines of different lengths and widths. Is this possible?
I understand, but wanted to calculate the coupling for two parallel lines first.
The coupling between segments of different length seems undefined to me,
regardsless of the Sonnet modelling issue. This would means a short wire where
current starts "out of nothing" and then magically disappears at the end of the line segment.
Even if you could simulate that, it has no physical meaning.
From many years experience in RFIC inductors:
To get realistic results, I think that you need to include your closed current
loops,
not just a straight segment. Analyzing the magnetic coupling between straight segments of
different length seems strange to me.
The magnetic coupling depends on the area where the magnetic flux couples,
and the straight line segments with return path through the Sonnet box are probably
different from your real current path !? This will also change the coupling.