[Beowulf] 1. Re: Jury rigged ethernet? (Robert G. Brown, et al)
Glen.Gardner at verizon.net
Sat Sep 3 15:07:27 PDT 2005
It is very easy for industry to make cabling which has consistent, and
reasonably precise impedance characteristics.
The big difference in cabling will be the type of dielectric between the
individual wires of the twisted pair, and the quality of the outer
With any "RF" cable the connectors and termination are important. For
very high performance applications at gigabit speed and beyond, it is
probably a good idea to test cables before installing them, as it is
very easy to end up with a termination which is unacceptable in terms of
reactive component and value of resistive impedance, resulting in
unacceptable performance characteristics.
You might be surprised at how bad the impedance matching in the feedline
itself has to be to end up with significant signal loss. Also, you might
be surprised to discover how easy it is to end up with a "bad"
terminating connector that causes very serious loss....
What I'm saying is that, barring a damaged cable, it is almost never the
cable itself and is almost always the termination at one end of the
Relying on mismatched cheap nic transceivers is spotty , at best and
probably not very productive because serious line losses will not result
from minor impedance mismatches. (A 2:1 VSWR or less is considered
acceptable in two-way radio work)
The thing you want to watch out for is the introduction of stray
reactances that throw things way off... like bad connectors or damaged
On Tue, 2005-06-28 at 12:50 -0700, Ed Karns wrote:
> Again, I would suggest simply trying the CAT3/5 wiring and get as many
> "clean" pairs as possible into use and/or as redundant circuits, and
> ignore what your impedance meters may tell you. Four (two twisted
> pairs) would do the trick and you get decent performance, maybe even
> 100baseT levels. The theoreticians and white paper writers usually
> don't take into account the irregularities of the real world. Truly it
> is almost impossible to accurately predict the real impedance of a
> cable set that is "dark". Modern impedance matching techniques built
> into modern ethernet circuits and chip sets can sometimes handle a very
> broad range of impedance mismatches once the wires are "heated up".
> Try it ... test the available wiring with a couple of laptops.
> Interestingly, the use of "mismatched" or "off brand" ethernet adapters
> (PCI or PCMCIA cards) will sometimes work better than the name brand
> originals, which are often manufactured with "tighter" specs. than the
> no name clones.
> " ... Honestly, I don't think that a 30 foot run (ten meters, barely
> longer than their five meter limit in toto) will be too much of a
> problem, although one might have to experiment a bit and get very clean
> connections to the terminators on both ends. ..."
> ... hear, hear ... (As mentioned off line, I have seen coat hangers
> work in a pinch.)
> Ed Karns
> On Tuesday, June 28, 2005, at 12:00 PM, beowulf-request at beowulf.org
> > Today's Topics:
> > 1. Re: Jury rigged ethernet? (Robert G. Brown)
> > ....
> > ----------------------------------------------------------------------
> > Message: 1
> > Date: Mon, 27 Jun 2005 16:52:19 -0400
> > From: "Robert G. Brown" <rgb at phy.duke.edu>
> > Subject: Re: [Beowulf] Jury rigged ethernet?
> > To: Ed Karns <edkarns at firewirestuff.com>
> > Cc: beowulf at beowulf.org
> > Message-ID: <cone.1119905539.89650.5767.500 at lilith>
> > Content-Type: text/plain; charset="us-ascii"
> > Ed Karns writes:
> >> ... below ...
> >> On Monday, June 27, 2005, at 10:22 AM, Reuti wrote:
> >>> Hi,
> >>> point to mention is the most likely unknown wave resistance of the
> >>> existing wires. You have
> >>> 150 Ohm for IBM Type 1 .... possible to use.
> >>> 100 (sometimes 105?) Ohm for CAT3/5 twisted pair ... The best
> >>> candidates ... if you have enough pairs between this cabling and
> >>> above
> >>> ...
> >> ..........
> >> These cables below are not recommended. Using coax will only make your
> >> connections unreliable unless you can make or use a breakout box with
> >> the proper connection converters (RG type coax conversion to RJ type
> >> connectors = bad news) ...
> > As I also pointed out to several people in offline remarks, impedance
> > matching is a known problem with known solutions for at least some of
> > the more common STP to UTP situations. The biggest problem occurs
> > where
> > one goes from STP (say, the lines into said "tight space") to UTP (say,
> > a UTP/Cat 5 patch cable). You tend to get reflections at these
> > junctures that can increase error rates if the UTP and STP runs are of
> > comparable length and the run lengths are "bad" lengths relative to
> > carrier wavelengths (where standing waves can be established via the
> > reflections).
> > One solution is to avoid such junctures and use STP end to end, primary
> > and patch cables alike. However according to this cisco white paper
> > (that examines the issue from the point of view of organizations
> > needing
> > to reuse old e.g. token ring or phone wiring):
> > http://www.cisco.com/en/US/netsol/ns340/ns394/ns74/ns149/
> > networking_solutions_white_paper09186a00800b4249.shtml
> > it isn't a big issue for short runs, either of the primary STP cable or
> > short patch cables on the ends. To quote:
> > ...short lengths such as 90 meters of Token Ring A-suffix STP cable
> > and 5 meters of UTP patch cords on each end should not cause
> > significant problems.
> > at least through 100BT.
> > They do discuss using impedance matching adapters for various cable
> > types, finding that in a lot of cases they don't really help much.
> > Honestly, I don't think that a 30 foot run (ten meters, barely longer
> > than their five meter limit in toto) will be too much of a problem,
> > although one might have to experiment a bit and get very clean
> > connections to the terminators on both ends. Termination per se is
> > likely to be the biggest issue, with or without adapters or baluns or
> > impedance matching as empirically determined to be necessary.
> > .......
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
More information about the Beowulf