I have been trying to simulate reflections for impedance matching of TLINEs. The simulation works but somehow I can’t get the simulation to run with simulation steps smaller than 2ns.
If, for example, I lower the the timestep from 500ps to 50ps (“.trans 500p 100n 70n” to “.trans 50p 100n 70n”) the number of plotted data points increases as expected but the one can see that the points are just extrapolated between 2ns points.
To show this I exported the data and plotted the results together in Python, here it is clearly visible that the points are just extrapolations between 2ns steps.
I would expect a dampened oscillation. But, by my understanding, even if the resulting voltage had a different shape I should never get such ‘sharp’ edges in the function when going to such small time steps.
I am not a specialist in transmission line simulation, but I would assume the following:
When you are using a TLINE, that is an ideal, lossless tramnsmission line, there will be no damping.
You get a reflection at the line’s end. The input pulse is reflected as is, with a certain ratio, depending on the resistances. Sharp edges are reflected as sharp edges. And if you use smaller time steps, you will see the edges even sharper.
