[TowerTalk] Open wire

Martin Ewing martin.s.ewing at gmail.com
Tue Sep 29 06:56:45 PDT 2009

References to W9CF and W5DXP info would be helpful.  (Googling fails this

73 Martin AA6E

On Tue, Sep 29, 2009 at 9:02 AM, Paul Christensen <w9ac at arrl.net> wrote:

> > I'm not sure I would use the same language as Jim, but I *do* think
> > there is widespread misunderstanding amongst Hams about the performance
> > of "ladderline".
> Open wire transmission line can be the "near miracle" form of line if
> several electrical parameters are known.  All too often, open or ladder
> line
> is deployed without regard to line length from inefficient tuners.
> To optimize open wire lines, one should know: (1) the line to load
> mismatch;
> (2) length of transmission line and its velocity factor; and (3) the
> impedance at the shack end of the line for any given frequency in order to
> minimize coupling losses.  For example, it's entirely possible to connect a
> random length of open line to a loop antenna, and incur substantial
> coupling
> losses as the tuner tries to efficiently couple power from the transmitter
> to a line with a very low Z at the shack end of its terminals.
> Experimenting with the W9CF "T-Network Tuner Simulator" provides some
> valuable insight regarding tuner design.  While it's possible to achieve a
> 50-ohm VSWR with several iterations of the tuner, many combinations of
> adjustments -- and choice in component values result in very high coupling
> loss.  And, that's not taking into consideration an input or output balun.
> For a T network to couple efficiently into low impedance loads down to 160m
> without optimizing line length, one needs input and output capacitors *at
> least* 3,000 pF in value.  That's a factor of about 6X what most commercial
> tuners use in their designs.  Reasonably good Q of components is also
> important.  Don't believe it?  Try it yourself with the on-line W9CF
> simulator and pick a low Z value of say 10 ohms resistive at 1.8 MHz and
> run
> the calculation.  With 250pF In/Out caps, that's 2 dB of loss just in the
> tuner at the *best* possible matching combination of controls.  Now add in
> a
> balun, as well as the additional loss resulting from the antenna line to
> load mis-match and you've easily lost half your power in the system-- using
> open transmission line.  And yet, the rig's SWR meter shows 1:1 and all
> along you've been fooled into thinking that just because you're using open
> line, it somehow *must* be efficient.
> To make best use of open line: (1) the line-to-load mis-match should be
> reasonable; (2) The line length optimized for a reasonable Z seen by the
> tuner; and (3) good selection of tuner component values and Q.  These
> factors play an important role in tuner coupling losses, particularly a low
> Z seen by the tuner.  1 and 2 above are probably the easiest to manage by
> calculation.  Item 3 oftentimes requires extensive modification to an
> existing tuner, or use of a link-coupled tuner.  Link tuners without a
> capacitive voltage divider do best when the operator can select parallel or
> series feeds, the latter of which do best for low-Z loads (ref. Cebik).
> One of the most efficient systems I've seen to manage all of the above is
> from W5DXP.  He uses switched sections of open line, fed at current maxima
> point at the coaxial transition.  The variable length of line through the
> switch matrix performs the antenna tuning and matching function and results
> in a 50-Ohm VSWR of less than 1.5:1 even when the 450-ohm VSWR is 10:1.  If
> his matching system cannot attain reasonable results, it's an indication to
> the operator that something is unreasonable in the design, including
> antenna
> and/or transmission line length.  Unlike a T tuner, it does not give the
> operator a false sense of security when the SWR bridge reads 1:1.
> Paul, W9AC
> _______________________________________________
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk

More information about the TowerTalk mailing list