Hello Owen, thanks for the comprehensive explanation.

Which match type did you have problems with?

The 'problem' however was not directly related to lossy lines, but more
about why Nec2 would not correctly model the results as achieved by
Cebik's spreadsheet. This because both use lossless lines. I can
imagine the outcome has a spread around 50 ohms or so, I did not expect
results around 20 ohms or even worse.

At the moment I do not have the corresponding models available
overhere. I'll post them tomorrow, maybe someone else can figure out
what I am doing wrong...

Arie.