[SI-LIST] Re: Gigabit fiber transceivers, EMI, trace lengths, and... antennas.

  • From: Chuck Hill <chuckh@xxxxxxxxxxx>
  • To: RMELLISON@xxxxxxx, si-list@xxxxxxxxxxxxx
  • Date: Sat, 01 Sep 2001 20:40:06 -0600

At 08:37 AM 9/1/01, RMELLISON@xxxxxxx wrote:
>Joel,
>you piqued my interest in talking about radiation of PCB traces.  I've been
>trying to get a better handle on this subject in order to give better
>guidance to my clients. I didn't see a reply to your questions relating to
>the half-wave dipole's radiation characteristics.  Did anyone reply offline?
>What's NEC?  Some type of antenna design software?

Numerical Electromagnetic Code (NEC).  Has been around for many 
years.  I've seen more recent versions available on the net but I don't 
remember where.

>I posed a question for the SI engineers that never got completly answered.
>The question was about the radiation characteristics of source-terminated
>lines, in comparison to end-terminated lines. It did get me to go back to the
>ARRL handbook and refreshed my memory of antennas.  I  finally decided that
>the radiation characteristics of a microstrip trace were governed by the same
>physics that make it susceptable to crosstalk, and had nothing to do with
>impedance (only indirectly).

Not exactly.  Radiation is due to the displacement current term in 
Maxwell's equations.  Inductance and capacitance are not the same term.

>Tighter coupling to the ground plane by way of
>thinner dielectrics reduces the radiation.  Increasing the width of the trace
>to lower the impedance of the line does not reduce the radiation.  So, as you
>said, removing the plane under the trace makes for a very good antenna.
>To get back to my original question--
> From my days as a ham radio operator, I can visualize that a
>source-terminated microstrip line would radiate the best at certain
>frequencies, probably where the line was equal to 1/4 wavelength, or odd
>multiples therof.  And, because of this, I would be better off using end
>termination on all of the microstrip lines.  But I cannot visualize the
>standing wave on the PCB trace when it is tightly coupled to ground.

The standing wave is not the same as coupling to another structure, even 
ground.  There are two components to the signal in the line: the standing 
wave due to mismatch with no power transferred to the load, and the 
travelling wave accounting for power transfer.

When there is an adjacent structure which is tightly coupled the current 
distributions in each nearly match.  The magnetic field depends on the 
current distributions (in the microstrip trace and the ground plane 
underneath it, or in the Yagi antenna elements if you like).  The net 
magnetic field is the sum of the fields due to both structures.  The more 
closely the magnetic fields from the current distributions match, the less 
the radiated field.  Thus the same currents and nearly the same physical 
location mean field cancellation.

The question to ask is: which configuration has higher currents?  Maxwell's 
equations dictate the current gives rise to a radiated field.  The series 
terminated case has some power transferred plus the currents due to the 
standing wave.  But the parallel terminated case has extra current that 
goes into the termination.  The load in the series terminated case cannot 
be purely reactive because no power would be transferred, thus no signal 
would be transferred.  What kind of impedance match is there at the 
source?  The answer lies in all these details and particulars to your 
situation.

Chuck Hill, consultant


>If it
>were a 1/4 wave vertical antenna where is could see the mirror image in the
>groundplane, it would be easy.  I would like to know your feelings on this,
>since you seem to have a good concept of antennas.
>Richard Ellison
>Ellison consulting
>214-544-1920  bus. ph.
>214-544-1924 fax
>
>
>------------------------------------------------------------------
>To unsubscribe from si-list:
>si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>For help:
>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>List archives are viewable at:
>                 //www.freelists.org/archives/si-list
>or at our remote archives:
>                 http://groups.yahoo.com/group/si-list/messages
>Old (prior to June 6, 2001) list archives are viewable at:
>                 http://www.qsl.net/wb6tpu
>


------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                //www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts:

  1. Home
  2. si-list
  3. Archive
  4. 09-2001