CEP006 - Cellular Design Review with Roy Chen

:frowning: sorry for the late, late reply! Been _really busy the last few months - probably the most intense time in my career.

As for stubs - basically anything thatā€™s not a trace from point A to point B is going to (1) change the impedance at that point, (2) likely cause reflections and most importantly, (3) likely act as an antenna - with all the unintended transmission and reception that comes with that.

How large the discontinuity is, the frequency (actually dV/dt) of the signals transmitted and whether the lines are differential are just some of the factors that influence just how sensitive the circuit may be to the influence of any stubs.

My best guess would be that the ideal PTH/via would have the same total amount of copper as the trace. So, a 10 mil trace would have a PTH with 5 mil copper on each side of the hole. Depending on the trace width and manufacturing tolerances, that exact sizing may be impractical. But, it should be possible to get in the ballpark.

Also, it probably shouldnā€™t be made so small that itā€™s hard to probe it in the first place. :wink:

Does that help?