Towertalk
[Top] [All Lists]

[TowerTalk] 1/4 wave coax "Balun"?

To: towertalk@contesting.com, jim.thom@telus.net
Subject: [TowerTalk] 1/4 wave coax "Balun"?
From: Norman Alexander <npalex@bellsouth.net>
Date: Tue, 28 Dec 2010 20:31:50 -0800 (PST)
List-post: <towertalk@contesting.com">mailto:towertalk@contesting.com>
Jim,

The use of an ODD electrical quarter-wave of coaxial cable as a BAL-UN is not 
correct.  The correct description would be "series coaxial transformer".  And 
there are certain other conditions to be satisfied when constructing this 
impedance matching network.  An odd quarter-wave of 75ohm coax is a special 
case 
of the Series Transformer (which uses two different pieces of coax) and is 
nominally a 2:1 impedance transformer.  Its application for Cubical Quad 
antennas was popular as often the loop impedance of a quad was around 100 ohms, 
and required a 1:2 impedance change to match a 50 ohm source.  It does not 
perform a BALUN function (Balance to Unbalanced).  Coaxial baluns can be 
constructed using electrical 1/2w piece of coax (Same Z)and connected so that 
the shields are together, the feed coax center to terminal 1 of the antenna and 
the 1/2w piece center from T1 to the other terminal of the antenna - this also 
is a 4:1 impedance transformer as well.  


Your description of connecting the "braid" to the tower or mast top doesn't  
apply in the applications I am familiar with.  One of the most common 
arrangements is in a multi-band quad. where several of the quad driven element 
loops would be connected to a remote antenna switch via 1/4 or 3/4 wave 
sections 
of 75ohm cable for a 2:1 impedance match.  


It is not clear what the actual arrangement of cables, tower and antennas are 
in 
your comment regarding W8JI.  Without a more complete description of the 
antenna 
arrangement and feed points it is not possible to determine what the function 
of 
the 1/4w cable sections are.   You speak of Hi Z, again not clear what that 
value is.  I recommend looking up "Series Coaxial Transformer" (ARRL Antenna 
Book)as this will discuss a more general solution for impedance matching using 
a 
length of coax equal to the source Z, and a larger value of Z for the other 
section.  The lengths and value of cable Z is determined by the frequency and 
impedance differential being matched.  These are single band solutions.

Summarizing - 1/4w transformers do not provide a balanced feed.  The 
characteristic impedance of a series transformer is different then the source 
impedance.  


A 1/4w of 75 ohm coax is a special case series transformer and provides a 2:1 
impedance transformation.  


An ODD quarter wave of 75ohm cable is also used in coupling two identical 50 
ohm 
antennas into a phase array.

A half-wave of same characteristic impedance coax is required to create a 
Balance feed and results in a 4:1 impedance transformation.  This arrangement 
is 
frequently use in large mono-band yagi's where the driving Z may be 10 or 12 
ohms and the 4:1 provides a balanced match to 50 ohms.

Most amateur applications of the odd quarter-wave transformer use 75 ohm cable 
which results in a 2:1 transformation.  

    

Norm  W4QN

==============================================================
Date: Tue, 28 Dec 2010 07:04:13 -0800
From: "Jim Thomson"    <jim.thom@telus.net>
Subject: [TowerTalk] Baluns...again.
To:    <towertalk@contesting.com>
Message-ID:    <52EE1FDD307B4F319F0EB430BFCF95D2@JimboPC>
Content-Type:    text/plain;   charset="iso-8859-1"

Looking through some old    notes  plus the info on Jim Brown's  site.... there 
was  a    question about using
an electrical   1/4 wave or electrical 3/4    wave  of coax.   The  1/4 [3/4]  
wave of coax     was run between the
DE  of the yagi  and the tower... with the    braid of the coax  bonded to the 
top of the tower.  


The    theory is... if a  1/4  [3/4]  wave of coax has it's braid    bonded to 
the top of the tower... the Z  at the  DE
should be sky    high..and there should be no requirement for a balun.    
Apparently    there  was an article out
about doing just that. The ant was a quad...    and individual  1/4 wave [ and 
3/4 wave]  pieces  of 50 ohm    coax
were  brought to a remote switch box at the top of the    tower..where of 
course 
all the braids are bonded to the tower top.     The author claimed that no 
baluns were needed on the various loops  used    on each band.   I have not 
seen 
the article, but it was     brought to my attention by a local ham. 


W8JI  also mentioned the    use of  1/4  and 3/4 wave lines  from DE  to tower  
  
top...and bonding the braids to tower top.  Seems to me that    W8JI  mentioned 
that if a balun was used in conjunction with the     1/4 [3/4]  wave coax, that 
the balun would makes things    worse.   There was some discussion on one of 
the 
other    refelctor's  years ago  about the  1/4 and 3/4 wave    concept.  It 
had 
something do do  with the height of the    tower.   If the tower was not the 
'correct height'.... the concept    would not work.    Does anybody remember 
this ?  


In one    of Jim browns write ups.... there is a depiction  showing     
feedlines  of various  lengths... from  0  up to... but    not  quite  1/4 
wave.  Then the chart starts up again...    just a bit longer than a 1/4 
wave..... and up to... but not  quite     a  3/4 wave.    Then it starts up 
again.. just past a    3/4 wave.  OK, what gives?   Is an  electrical    1/4  
wave  and odd  quarter waves  the real answer to a    high Z  on  one band ?    
In the case of something    like  monoband yagis.... this may prove to be a 
huge 
benefit... if the    theory  actually works.    If I remember correctly, the    
 
main coax braid was bonded to the base of the tower as well.   IF    the theory 
does work... and a high Z  is presented to the DE... how high    is this Z?    
Is is mainly reactive... mainly resistive..or a bit    of both ?      Why would 
a bead balun or torroid balun, when    used in conjuntion with a odd qtr wave
of coax makes things worse ? 

later........ Jim   VE7RF    
_______________________________________________



_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk

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