Magnet Myths and Misconceptions

[EMJunkie]

In a lossless transmission line the propagation velocity is:  c/(uR*eR)^0.5.  Most of the time in cables and etch inside printed wiring boards, uR is close to 1.0 and it is the eR that sets the velocity.  Typical printed circuit boards have eR values of 4-5 so the velocity is half or less of c inside the board.  The speed actually varies depending on how much of the fiberglass is glass and how much is resin.  At the ferocious data rates that we have today that can be a big problem because how traces line up with the glass bundles changes their timing.  In a coaxial cable that uses low density PTFE foam the average eR is only about  1.2 and the velocity is about 0.9 c.  Traces that run on the top side of the circuit board send some of the energy through air above the trace and other energy through the board material underneath the trace.  These components run at different speeds and that distorts the signals.  But what can really drive fast signals batty is the protective nickel plating that is often applied between outside traces and gold top finish.  Nickel has a high uR.

Ok, not bad MarkE. We can markE this as an answer because its close enough to what I was looking for.

See My point MileHigh!

[MileHigh]

In a lossless transmission line the propagation velocity is:  c/(uR*eR)^0.5.  Most of the time in cables and etch inside printed wiring boards, uR is close to 1.0 and it is the eR that sets the velocity.  Typical printed circuit boards have eR values of 4-5 so the velocity is half or less of c inside the board.  The speed actually varies depending on how much of the fiberglass is glass and how much is resin.  At the ferocious data rates that we have today that can be a big problem because how traces line up with the glass bundles changes their timing.  In a coaxial cable that uses very low density PTFE foam the average eR is only about  1.5 and the velocity is about 0.8 c.  Higher density of PTFE increases the eR and further reduces the propagation velocity.  Traces that run on the top side of the circuit board send some of the energy through air above the trace and other energy through the board material underneath the trace.  These components run at different speeds and that distorts the signals.  But what can really drive fast signals batty is the protective nickel plating that is often applied between outside traces and gold top finish.  Nickel has a high uR.

You are amazing Mark.  As you know, 25 years ago, clock frequencies were much lower and you could pretty much ignore this stuff.

[MileHigh]

Ok, not bad MarkE. We can markE this as an answer because its close enough to what I was looking for.

See My point MileHigh!

I gave you the real short answer - the impedance of the medium determines the velocity.

[MarkE]

You are amazing Mark.  As you know, 25 years ago, clock frequencies were much lower and you could pretty much ignore this stuff.

We are approaching 48Gbps over a single pair with goals to get to 100Gbps before the end of the decade.  We are talking very wicked fast.

[MileHigh]

We are approaching 48Gbps over a single pair with goals to get to 100Gbps before the end of the decade.  We are talking very wicked fast.

The entire Library of Congress in 17.9 seconds!  lol