[Top] [All Lists]

RE: [TowerTalk] Shields

To: 'Tom Rauch' <>,"Daron J. Wilson" <>
Subject: RE: [TowerTalk] Shields
From: Steve Katz <>
Date: Wed, 13 Oct 2004 07:47:48 -0700
List-post: <>
Actually, in my experience using hardline or 100% (or at least very good
double) shielded cables in VHF-UHF repeater systems and cavity interconnects
in duplexers does make a difference, not necessarily in notch depth but in
generated noise that can bother receivers.  I also see this "noise" issue
when using many antenna designs that are perfect for simplex operation and
stink in duplex operation with any power applied.  It seems the oxidation
between shield strands, which is always there even with brand new cable and
worsens with age, creates minute noise-generating interfaces (or something).
I've put up repeaters using brand new mil-spec RG213/U and running 100W
output power with a very good duplexer had desense problems resulting in
poor receiver performance.  Replacing absolutely nothing in the system
except the transmission line, going from 213 to 9913 or hardline or some
other 100% shielded cable, completely eliminates the noise -- it's gone.
This has happened to me so many times it cannot be coincidence.

I've seen the same thing with aluminum-to-aluminum clamped construction
antennas when used in repeater service, which is why you'll almost never see
this construction used in repeaters.  Soldered or welded construction
antennas work fine.  A brand-new Hustler G6 or G7, which are very good
"simplex" antennas and can last many years when installed with a bit of
NoAlox on the joints during assembly, are crappy repeater antennas.
Replacing one with a Stationmaster or RFS 4-pole made of heliarc welded
aluminum tubing and no clamps -- ahhh.  Perfect.  Zero noise, zero desense,
zero problems.  The only difference is the "repeater" antenna doesn't have
any clamped connections, other than the grounded ones that attach it to the


"Nobody needs more than 640k!" -Bill Gates, 1982

> -----Original Message-----
> From: Tom Rauch []
> Sent: Tuesday, October 12, 2004 6:55 PM
> To:   Daron J. Wilson
> Cc:
> Subject:      Re: [TowerTalk] Shields
> > Tom I'm not questioning your experience or testing, but
> what are the
> > manufacturers of duplexers missing?  They seem to think
> that double
> > shielded coax on their harnesses is of some value.  I
> would think it if
> > made no difference, they would use less expensive
> components.
> I don't know that they think anything. Many times I've
> decided to do something that isn't necessary simply because
> people (through popular opinion) think it is necessary. It
> isn't worth spending dollars being right to save a few
> pennies. I can prove the cable to cable isolation of typical
> RG8X exceeds the notch depth of the cavities and the loss in
> the duplexers is significantly higher than any loss in
> jumpers, but I'm sure customers would have a fit about
> seeing RG8X.
> The duplexers I have cost 1400 bucks, and use six feet or
> less of cable with UHF connectors at each end. I don't think
> they broke the bank in cable cost. I bet if they lost ten
> customers ($14K gross), they'd never come close to making
> the savings up.
> For an example of this thinking, look at trap dipoles.
> Almost no one builds them commercially, because we have it
> drummed into our heads that "traps are lossy". The fact is
> even mediocre traps have less loss than the alternatives
> some manufacturers use.
> Look at linear loading and low band yagi's. Same story.
> You just go with the flow, that's where the profit is.
> 73 Tom
> _______________________________________________
> See:  for "Self Supporting Towers", "Wireless
> Weather Stations", and lot's more.  Call Toll Free, 1-800-333-9041 with
> any questions and ask for Sherman, W2FLA.
> _______________________________________________
> TowerTalk mailing list

See:  for "Self Supporting Towers", "Wireless Weather 
Stations", and lot's more.  Call Toll Free, 1-800-333-9041 with any questions 
and ask for Sherman, W2FLA.

TowerTalk mailing list

<Prev in Thread] Current Thread [Next in Thread>