Re: [SI-LIST] : Skin Effect

About this list Date view Thread view Subject view Author view

From: Doug McKean ([email protected])
Date: Wed Jan 24 2001 - 12:48:15 PST


Two definitions are necessary.

Skin depth is the depth current travels in the material related to
frequency.
Skin effect is the resistance of the material related to frequency.

Differential lines can display a proximity effect whereby the skin
effect
is different in one part of the trace than expected by the proximity
of the
other trace. A non-uniform bunching of the current in the part of the
traces which are closest to one another. Instead of a uniform current
flowing around the conductor according to it's skin depth, more
current
gets bunched to one side of the conductor than the other.

You're loss is actually what Larry has pointed out, an (R*I^2) thing.

- Doug McKean

----- Original Message -----
From: Farrokh Mottahedin
To: [email protected]
Sent: Wednesday, January 24, 2001 11:44 AM
Subject: [SI-LIST] : Skin Effect

There seems to be a phenomenon that on a differential transmission
line, an increase in the characteristic impedance (Zo) will help to
reduce IR losses due to skin effect.

Now, we know that Zo = sqrt(L/C). Likewise the IR losses due to skin
effect can be summarized generally as 4.34 (R/Zo+GZo) in dB/meter. R
and G are the load resistance and admittance.

Conceptually, it also makes sense that if a transmitter sees a larger
Zo (the transmitter does not see the load directly, but only sees the
line ahead), less current will flow, and since the load doesn't
change, there will be less power loss. But if the drivers are current
sources, then the current should be constant, and a larger Zo serves
only to cause more IR
loss. Here I am looking for some math to clear all this up rather
than to rely on intuition.

==============================================
Farrokh Mottahedin
Cisco Systems, Inc.
Gigabit Systems Business Unit
170 W Tasman Drive
San Jose, Ca 95134-1706
Phone: (408) 525-9185
Fax: (408) 527-8254
email: [email protected]

**** To unsubscribe from si-list or si-list-digest: send e-mail to
[email protected]. In the BODY of message put: UNSUBSCRIBE
si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Tue May 08 2001 - 14:30:43 PDT