[Top] [All Lists]

[TowerTalk] Petrezewski (sp?) array ..

To: <>
Subject: [TowerTalk] Petrezewski (sp?) array ..
From: (Tom Rauch)
Date: Thu, 5 Apr 2001 14:21:56 -0400
I'm not trying to re-invent the wheel here, but to make a point we 
should always keep  in mind when dealing with coaxial lines.

>         If I recall correctly, the feedlines must be open-circuit (OC)
> because ~3/8 wavelength OC adds inductance to make the undriven
> elements look like parasitic reflectors in the system.  I guess they

I think what we often miss is the inside of the shield and outside 
are always two different conductors when the shield is several skin
depths thick, which would cover any shield on any cable we have 
at HF.

The can be no current inside the shield unless we allow a 
differential current to be applied to the center conductor and shield. 
To do anything with that current, the system has to do something 
BETWEEN the shield and the center....not between the shield and 
the outside world. All of our systems (hopefully) operate that way. 
We strive to make them that way, since that is only real advantage 
to using coaxial lines! That is the single effect that allows us to 
bury the cable, put the cable in with a group of cables, and do what 
we like with the cable shield grounding without affecting our 

If we open the center conductor, the impedance across the 
feedpoint of the dipole is the same no matter what we do with the 
shield as a single long as the feedline has no 
common-mode current. That is the job of a balun, and feedline 

A properly located balun can do that job more effectively than 
simply opening the shield at the tower. If we dress the feedline so it 
picks up minimal radiation, and if we used a good choke balun (and 
correct length of the cable from the antenna to the point where it is 
grounded), there is no need to switch the shield. 

If we do have to switch the shield, it clearly indicates the shield is 
part of the radiating system. Now we might get away with common 
mode current by band-aiding around it, but I always like to learn 
why it is a problem and fix the problem. Then shields can remain 
grounded, cables dressed together and taped to tower legs....or 
whatever I like without worry that the whole system will quit working.

That's actually the only reason I use coax.

> P.S.  Personally, I would go with a driven system (Comtek) if I were
> using full-size only need 4 instead of 5 and the results
> are more predictable using a forced system IMHO.

So would I, if I was serious about the band. With a four-square of 
dipoles you get, in effect, a three element end-fire array. With a 
single dipole fed you really have a two element array with what we 
might consider a three or four wire reflector. I think the results 
would be noticeably better, and the hardware addition is slight. 

Like many things, it is a trade of performance and cost. Now I'm 
thinking of installing one on 80 meters, maybe at first as a 
Petrezewsjki(sp?) array but with the common mode problems 
removed. Then I can just use an RCS-8, because I rarely work 80 

I'll use N connectors, put rope in the elements, point it at true 
north, and use an owl to house each balun. With enough luck, it 
will heat the ionosphere and make the path open and be published 
in Radiosporting Magazine.

73, Tom W8JI 

List Sponsor: - Trylon self-supporting towers, safety
equipment, rigging gear, LOOS tension guages & more!

FAQ on WWW:     
Administrative requests:

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