[Top] [All Lists]

[TowerTalk] Dielectric losses

To: <>
Subject: [TowerTalk] Dielectric losses
From: (Tom Rauch)
Date: Thu, 31 Jan 2002 06:27:31 -0500
> The ARRL Antenna Book 17th edition, page 24-16 says:
> "The power handling capability and loss characteristics of coaxial
> cable depend largely on the dielectric material between the
> conductors...."

Of course it does. No one said it doesn't.

The reason the power rating changes is some dielectrics have 
lower voltage ratings, and some allow the use of a larger center 
conductor for a given power.

HF and VHF loss in good quality cable is mostly caused by 
conductor resistance, not dielectrics. At HF and lower VHF virtually 
all loss is due to conductor resistance. 
> Fig 22 on the same page shows attenuation vs. frequency. If one looks
> at coax like RG8 "regular" dielectric and RG8 foam dielectric curves,
> one can see that slope of the curves for both coaxes is different,
> indicating that while they have similar or close physical dimensions,
> at higher frequencies "regular" dielectric coax has steeper slope,
> therefore loses increase with frequency. Curves (lines) for most foam
> coaxes closely follow slopes of air dielectric coaxes or lines.
> Polyethylene "regular" dielectric has a steeper slope indicating
> increasing loses with frequency.

RG-8 style cable with foam dielectric uses a #13 conductor.

RG-8 style cable with a foam dielectric uses a #10 center 

Both have the same outer diameter and impedance.

> Most of us know that when coax gets contaminated, insulation between
> the conductors deteriorates and this shows as sometimes significantly
> increased loses. This is clearly caused by change in the quality of
> dielectric and not by thinning of the wires or increased resistance in

Most cables are closed cell, and water does not ingress into the 
foam any more than it leaks through a foam coffee cup.

While water inside the cable will certainly increase loss, even if you 
dry the cable out loss remains high at higher frequencies. The 
reason for the loss is the oxidized shield becomes a series of high 
resistance connections every place it weaves in and out. 
Remember skin effect causes the current to flow on the inner 
surface of the shield. 

The resistance-per-foot of oxidized braid is many times that of a 
similar sized solid smooth conductor, even though the dc 
resistance remains virtually unchanged.

> them. Velocity factor in the cable (due to dielectric properties) has
> effect also, more "slowing" of the wave in the cable, more wavelengths
> needed to pass through the lengths, more loss. Goes higher with
> frequency, more "waves" present.

I'm not aware of any physical or electrical property where the 
number of "waves" in a cable has something to do with loss. The 
velocity factor does not relate directly to cable loss.

It is true that a higher velocity factor cable, for a given outside 
diameter, can use a larger center conductor and have less loss. 
But that effect has nothing to do with the "number of electrical 
waves", whatever that might mean.

> So W8JI "gospel" that there is "virtually no difference" is dumbest
> statement of the year #2. ("Explanation" by W8JI to follow, dancing
> around the what "is" is. :-)

I'm not so sure I hold the record for dumb statements.
> I was glad to see Tom W6XR "correcting" W8JI statement about
> "inferiority" of Force 12. Contesters and DXers I know, have nothing

There was nothing to "correct". I never said Force-12 antennas 
were "inferior". I simply stated one should NEVER depend on 
anything written that trys to sell or promote a product for technical 

That is true, and "the article" published in QST was full of technical 
inaccuracies. Most of the inaccurate statements centered around 
hyperbole that was designed to sell antennas.

It is the quality of the technical information I question.

To set the record straight, I never said anything about the quality of 
the product. That was someone else, not me.

> about things I am not sure about. Some might remember Tom's attacks on
> some of my postings about my findings about 160m propagation, etc.

But Yuri, you didn't "find" anything. You claimed a low dipole (1/8th 
wl high) beats a four-square because of ducting, and I said it was 
bad science to form a theory based on the results you saw 
VE1ZZ's QTH for one weekend, which:

1.) Is about half the distance to Europe from people in my area

2.) Was only one weekend 

3.) Never included blind A-B tests

4.) Is on a rock ledge a few hundred feet above the ocean

I pointed out I have dipoles at low and high heights, and have 
NEVER in hundreds and hundreds of blind tests seen a low dipole 
"win" for DX. As a matter of fact, the low dipole only wins within a 
few hundred miles.

Virtually all of the time the verticals win over any horizontal antenna 
at any distance beyond a few hundred miles on 160 meters.

Most other people in the world using good verticals with real ground 
systems share my experience, although people with small ground 
systems sometimes find dipoles to be more competitive than their 
poorly-grounded verticals.

That's why the vast majority of successful DX'ers on 160 use 

I'm sorry you take any critical comment personally, but there is no 
reason to carry a grudge over a simple questioning of the science 
used in forming your theory to the point that you feel compelled to 
insult me on a personal level at every chance.
> If you care to read comments on the coax subject by one of my mentors
> who was removed from the reflector, here it is:
>  He claimed "the lower loss of polyfoam is more due to the larger
>  center 
> conductor than the polyfoam dielectric without any qualifications such
> as the frequency--the major factor."  A blanket statement without the
> necessary qualifiers is "technically irresponsible."  

I'm glad your friend agrees with me.
73, Tom W8JI 

Where do you get ICE bandpass filters & beverage matching boxes?  The
same place that pays for the hosting of this list:  The eHam Store.
Order online at

FAQ on WWW:     
Administrative requests:

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