[Top] [All Lists]

[TowerTalk] Best vs. Maxwell -- Again??( LONG !!)

To: <>
Subject: [TowerTalk] Best vs. Maxwell -- Again??( LONG !!)
From: (Michael Tope)
Date: Thu, 18 Mar 1999 05:43:34 +0000

I haven't seen Dr. Best's article or even read Maxwell's book
for that matter, so I am coming somewhat from a point of ignorance.
My questions is, are the graphs that you see in the ARRL handbook
and ON4UN's Low Band DXing book, which relate mismatch loss to 
matched loss, correct generically or do they contain an implicit
assumption about the loss mechanism (dielectric loss due to 
displacement current in the dielectric versus I^2R heating losses in 
the conductors)? In other words, is the math kind to us in such a way 
that you can accurately predict the mismatch loss armed only with a "one 
size fits all" graph and a measurement of the matched cable loss, or 
do I need the additional knowledge of what is going on inside the cable
to make an accurate prediction of the mismatched loss?

If this is not the case that a "one size fits all graph" can be used,
then the handbooks should be revised to indicate this, and a family
of graphs should be included to illustrate the how the mismatch loss 
factors vary across a useful set of real world feedlines (open wire line, 
coax, ladder line, etc.). After all, the desired result is that Joe Ham
can go take a NEC prediction of VSWR for his new antenna and the matched line
loss data for the cable curled up in his garage, and figure out the 
penalty in decibels for having his tuner in the shack versus at the 
feedpoint. That is what I would call useful information.

73 de Mike, W4EF...........

From:   Tom Rauch[]
Reply To:
Sent:   Wednesday, March 17, 1999 3:36 PM
To:     Jim Reid; Jay, WX0B
Cc:     Walter Maxwell; Antennas; Tom Rauch; 'Towertalk';;
Subject:        Re: [TowerTalk] Best vs. Maxwell --  Again??( LONG !!)


What Best forgot, and what you also neglect to mention, is 
amateur antennas with MODEST SWR VALUES work just fine with 
some mismatch even using regular coaxial cables. 

Loss is nowhere near the "match efficiency" or "reflected power" 
would indicate, IF the transmitter or receiver end is matched with a 
matching network.

If the transmission line has low matched loss, it will also have less 
mismatched loss. That why open wire lines can be operated with 
extreme SWR and still have excellent efficiency.

Best, despite openly flaunting his "experience and credentials", 
initially had that concept totally wrong! I'd respect him today if he 
had been able to utter the words "I made a mistake".

With lines under 1/4 wl, minimum line loss can (and usually does) 
occur with an INTENTIONAL mismatch! That's because an 
overwhelming amount of line loss is due to conductor resistance 
and NOT dielectric losses (as many people seem to think).

Air insulation in feedlines offers less loss mainly because  
impedance is higher for a given conductor size. There is less 
current and lower I^2 R losses because the line has higher 
impedance for a given conductor size, or the same impedance line 
uses a LARGER conductor.  

The change by matching at the load in TYPICAL installations when 
using feeders with very low loss is negligible, just as is the change 
when the line is short.

There are some exception. Situations can add significant loss...

1.) Lines with significant initial matched loss

2.) Line where VAR power in the line becomes extremely high 
(current becomes extremely high in areas of the line).

Your use of the plasma coupling as an example is about like telling 
us about losses when feeding an unloaded six foot mobile whip on 
3.5 MHz with a tuner at the rig.

When a person spews several pages of complex math based on an 
extreme examples (as Best did), or gives an example based on a 
SWR that would make most calculators overflow (as you did), it 
probably does more harm than good to the people who are trying to 

The poor bloke trying to understand what it all means is left with a totally 
wrong impression.

Bottom line is.... if feedline loss is fairly low in a matched condition 
a system can tolerate a fairly high SWR with NO noticeable 
change in performance on receive or transmit if a matching network 
is placed somewhere in the feedline system.

Feeding a six meter dipole on 160 meters is more in line with your 
Plasma example. Feeding a dipole cut for 21MHz on 18 MHz is 
more in line with the real world.  

73, Tom W8JI

FAQ on WWW:     
Administrative requests:

FAQ on WWW:     
Administrative requests:

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