Topband
[Top] [All Lists]

Re: Topband: K3NA Loop Array

To: Luis Mansutti IV3PRK <luisprk@tin.it>
Subject: Re: Topband: K3NA Loop Array
From: Eric Scace K3NA <eric@k3na.org>
Reply-to: eric@k3na.org
Date: Sat, 19 Jun 2010 19:16:14 -0400
List-post: <topband@contesting.com">mailto:topband@contesting.com>
    I'll be glad to answer question, Luis.

    In general the design goal for the matching network was to achieve 
an extremely low SWR (for 75 ohm cable).  An excellent match means the 
feedline can be used as a precise delay line for phasing the elements 
together.

    Because 80m is such a wide band, I decided to use a matching network 
that has two points of near-perfect match: one in the CW section and one 
in the SSB section.  The multi-stage matching network is what does this 
trick.

    With a little bit of tweaking, almost the entire band 3.500 to above 
3.800 could be kept within a very small SWR circle on a Smith chart.  
This means the delay lines work almost perfectly across the entire band 
-- which is very helpful in keeping the pattern clean of side lobes 
across the band.

    [This is usually the challenge: designing a system that works 
consistently across the band, not just a one design frequency.]

    The monte carlo exercise has a different purpose.  It's one thing to 
design a network on paper, and to build a prototype.  It's another thing 
to be able to build many units consistently.  Maybe the matching network 
only works with near-perfect, high-tolerance (expensive) components?

    The monte carlo analysis varies each component (mathematically) of 
the circuit a little bit in random directions from the ideal... and then 
calculates the resulting match behavior across the band.  It then 
selects another set of random variations for the components in the 
circuit, and does another calculation of match behavior.  After many 
such calculations, one can see how badly a match is affected by 
unit-to-unit component value variations within a specific tolerance.  
This particular analysis let me to choose fairly high tolerance parts.

    I have some board layouts used for the first several systems.  But I 
would modify the next version so as to put out two signals 
simultaneously (forward and reverse directions -- like a 2-wire beverage 
system delivers) so that an operator can listen in multiple directions 
on his K3 radio.

73,
    -- Eric K3NA

on 2010 Jun 19 12:55 Luis Mansutti IV3PRK said the following:
> Hi Topbanders,
>
> I read the very fine presentation by Eric Scace, K3NA, at 2009 Dayton
> Antenna Forum on a new Rx Loop Array.
> http://www.kkn.net/dayton2009/dayton-2009-antenna-forum.html
>
> I like to test every kind of Rx antenna, but the matching of this array
> seems (to me) not as simple as with verticals or flags.
> I don't understand the 4-stage network and the related K2TJ monte carlo
> analysis.
>
> Did anybody get inside this project, or tried to build it ?
>
> Thanks in advance for any comment.
>
> 73
> Luis IV3PRK
>
>
> _______________________________________________
> UR RST IS ... ... ..9 QSB QSB - hw? BK
>
>    
_______________________________________________
UR RST IS ... ... ..9 QSB QSB - hw? BK

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