Amps
[Top] [All Lists]

Re: [Amps] *** SPAM *** Re: Alpha 87a

To: amps@contesting.com
Subject: Re: [Amps] *** SPAM *** Re: Alpha 87a
From: Ian White GM3SEK <gm3sek@ifwtech.co.uk>
Reply-to: Ian White GM3SEK <gm3sek@ifwtech.co.uk>
Date: Tue, 2 Jan 2007 21:10:07 +0000
List-post: <mailto:amps@contesting.com>
Peter Chadwick wrote:
>Tom said:
>>Currents on the
>shield can't change SWR inside the cable <
>
>This is very true: however, what they can do is to change the indicated 
>SWR, depending on the SWR meter. So you can have a cable that in 
>reality has a 1:1 SWR, but has an indicated 1.5:1. This fact is not 
>always appreciated..

That can only happen if either the cable, the SWR meter, the 
transmitter, the ATU or a dummy load has a substantial break in its 
shielding. Without any of those, the indicated SWR is not affected by 
common-mode currents on the outside surfaces.

The one case where it definitely *does* happen is when the "load" is an 
antenna. Common-mode currents on the feedline will change the feedpoint 
impedance from what it should be, and that will also change the SWR; but 
the actual change is taking place at the feedpoint, and not local to the 
SWR meter.

The tool to detect such effects is not an SWR meter, but a clamp-on RF 
current meter.

Altering the length of cable between the driver and the PA may have some 
effect if the PA input does not present a good 50-ohm load. Varying 
lengths will categorically not affect the SWR (unless there is a 
shielding fault as indicated above) but they will transform the load 
impedance that the driver sees, rotating it around a constant-SWR circle 
on a Smith chart. When that impedance is not exactly 50 ohms, a 
solid-state driver with no tuning or loading controls may perform better 
with some lengths of cable than with others. However, a "special" length 
of cable will only improve the situation on one band (unless you're very 
lucky) and even then, it's only a patch. The real problem is with the PA 
input matching.

The Collins book gives some justification for making the total phase 
shift between the driver anode and the PA cathode into an exact multiple 
of 90deg, and for avoiding odd multiples of 45deg. Part of this phase 
shift occurs in the driver output tank circuit, part in the PA input 
tank, and the rest is made up by the cable.

Both tank circuits are band-switched, so their phase shifts won't vary 
much from band to band; but the phase shift in the cable will vary from 
band to band. This means that the special length of cable can only make 
up the correct phase shift for one band, and it will be valid for only 
one combination of driver and PA. Switch bands, and the whole 
phase-shift story falls apart.

I'm not disputing that a 20.5ft length of cable between a KWM-2 and the 
input of a 30S-1 does *something*; but I don't believe it works the way 
Collins claim.


-- 
73 from Ian GM3SEK

_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

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