J
John Larkin
Guest
I have a customer who wanted to review the PCB layout that we are
doing for them. Unfortunately, one of their engineers has been reading
HoJo's silly Black Magic book, and was harassing me about "return
currents" and plane discontinuities and such.
I just got the solder sample of the PC board. It has some TDR/TDT
impedance test traces. I checked it on my Tek 11802 20 GHz TDR scope.
It looks pretty good... see pics. Trace impedances are close to 50
ohms.
https://www.dropbox.com/sh/xnbjw0y2stycpn4/AACm580uYyexbmtAhjY3Jf0ea?dl=0
There are a few deliberate insults, namely a layer 6 trace crossing
over different L5 power pours, and a layer 1 trace crossing a slit in
the L2 ground plane. Neither seems to be visible on TDR. The only
anomalies are some extra capacitance at the launch connector, and some
small bumps at the four vias. The hard right angle on L6 doesn't do
much.
Through-signal risetime (upper scope trace) after 8.3 inches of trace
is about 300 ps. Cheap standard FR4, no impedance control called out.
The 3.3 volt power pour looks almost like a pure capacitor. It's about
2.9 nF with no bypass caps installed.
doing for them. Unfortunately, one of their engineers has been reading
HoJo's silly Black Magic book, and was harassing me about "return
currents" and plane discontinuities and such.
I just got the solder sample of the PC board. It has some TDR/TDT
impedance test traces. I checked it on my Tek 11802 20 GHz TDR scope.
It looks pretty good... see pics. Trace impedances are close to 50
ohms.
https://www.dropbox.com/sh/xnbjw0y2stycpn4/AACm580uYyexbmtAhjY3Jf0ea?dl=0
There are a few deliberate insults, namely a layer 6 trace crossing
over different L5 power pours, and a layer 1 trace crossing a slit in
the L2 ground plane. Neither seems to be visible on TDR. The only
anomalies are some extra capacitance at the launch connector, and some
small bumps at the four vias. The hard right angle on L6 doesn't do
much.
Through-signal risetime (upper scope trace) after 8.3 inches of trace
is about 300 ps. Cheap standard FR4, no impedance control called out.
The 3.3 volt power pour looks almost like a pure capacitor. It's about
2.9 nF with no bypass caps installed.