[1992] in linux-net channel archive

home help back first fref pref prev next nref lref last post

Re: Networking without a hub or a router/repeater

daemon@ATHENA.MIT.EDU (Dave Platt)
Thu Feb 29 12:58:33 1996

Date: 	Thu, 29 Feb 96 09:42:20 PST
From: dplatt@3do.com (Dave Platt)
To: linux-net@vger.rutgers.edu
In-Reply-To: <Pine.LNX.3.91.960228062906.4788A-100000@crimethink.bofh.1984.net>

> > I wonder why they don't like solid core coax?
> 
> I suspect that it has something to do with "skin effect".

I have my doubts about that explanation.  The skin effect applies to
stranded conductors, as well as to solid ones, because current can and
does flow between strands everywhere that they touch.  In order to get
each individual strand to behave independently (for the purposes of
skin effect calculations) you have to insulate the strands from one
another, and (I believe) keep them physically separate by an effective
distance at least that of the skin-effect depth.

I'd be more likely to believe that solid-core coax isn't specified for
thin Ethernet because it's usually more rigid than a stranded-core
coax of the same general type.  Solid-core coax would be less willing
to flex, and would probably put more stress on the transceivers or on
the damnclips that hold the transceivers to the AUI connector.

I haven't worked with the older "thick" Ethernet cable much...
thicknet and vampire taps and drop-cables had fallen out of favor by
the time I started geeking Ethernet - but my recollection is that the
thick Ethernet coax is solid-core.

Dave Platt    dplatt@3do.com
      USNAIL: The 3DO Company, Systems Software group
              600 Galveston Drive
              Redwood City, CA  94063


home help back first fref pref prev next nref lref last post