[Top] [All Lists]

Re: [TowerTalk] 43 ft vertical - feeding and balun

Subject: Re: [TowerTalk] 43 ft vertical - feeding and balun
From: jimlux <>
Date: Sun, 23 Nov 2008 16:59:50 -0800
List-post: <">>
Craig D. Smith wrote:
> I agree that a high quality general purpose remote auto-tuner suitable for
> KW+ levels over a "DC to Daylight" frequency range and undefined extreme
> impedance loads would turn out to be a quite pricy product!   But perhaps
> with the widespread acceptance and usage of the 43 ft vertical, a case could
> be made for a special purpose tuner product to service (and expand) that
> market. 

special purpose -> limited quantity -> expensive unit price (i.e. you 
can't spread the development cost over very many units)

  I'm thinking in terms of a set of L networks to get you in the
> right vicinity for each band.  Most people wouldn't use this vertical for 15
> and above due to the higher radiation angle.  So all you would need would be
> perhaps one network each for 80, 75, 40, 30 and 20 - a total of 5.  60
> should be OK w/o a network.  The philosophy would be to just use these
> networks to get you into a coax-friendly environment (SWR < 3 to 1 or so)
> and then use the tuner at the rig for the final tweaking if required. 

This is a good idea.  Perhaps it could be implemented similar to how 
folks do the band filters used in contesting situations.. that is, 
there's already some of the logistics figured out in terms of 
remotely/semi-automatically using relays to select filters.

Or, one could possibly devise a multiband matching network that would 
work without switches.  (QEX has an article on this, this month).  It's 
a non trivial exercise to design a multi section network and not have 
loss issues, etc.  However, since you can use a computer to do a lot of 
the tedious work of finding values and evaluating losses, etc., it's a 
lot more approachable than even 10 or 20 years ago...).

Then, you are basically building a box with some L and C inside and the 
appropriate connectors.  Totally passive, which means a lot of the 
complexity of an autotuner isn't needed.

And, making it handle the power is just a matter of making it bigger 
with lower loss componentry.

> networks could be directly switched via band decoder, etc. with no need for
> any feedback on the tuning process.  This should drastically reduce the
> complexity and cost of such a product when compared to a general purpose
> tuner.  I'd buy one!

Again, how much are you willing to pay for such a thing?  The remote 
controlled filters are in the >$500 range as I recall, and they're 
really receive only, I think.

High power is expensive, if you're buying the parts new.  And therein 
lies a big part of the problem... sticker shock.  A transmitting duty 
capacitor that can handle amps of RF current is a $50 part. Likewise, a 
high power air core inductor.  Say you needed 5 bands worth, and you 
needed 3 components for each band (for a pi or T network).  Even at $20 
each, that's 15*$20 = $300, just in parts, without the relays, 
fabrication, etc.

Note well that the relays might need some attention... you're talking 
either high currents or voltages, as opposed to the more mundane I and V 
in the usual antenna switch matrix working at 50 ohms with a 2:1 mismatch.

You can see why motor driven vacuum variables and such start to be 

The other challenge is that every installation is going to have a 
different Z vs frequency characteristic for the antenna (even if they're 
all 43 foot verticals, the ground systems vary), so a "cookbook" set of 
networks might not work, and then, you're back to the custom design.

I can see, though, that if someone wanted to set themselves up as a 
custom network builder you could go through this sort of process:

Have the customer measure the Z of their antenna over frequencies (if 
you're spending tens of kilobucks on the station, you can afford a 
decent antenna analyzer that will do this systematically)

You take their data and run it into a series of programs to synthesize 
the appropriate network. (the process needs to either by hand, or 
automatically, take into account standard values for things like 

You then price out the components selected, get a confirmation from the 
customer, and fabricate it to spec.

This is really the same model used in the commercial world.  I order 
filters and such like from a variety of vendors  (like K&L Microwave, or 
TTE, or RLC) for work. We send them the requirements, they run it 
through a program and come up with a quote, then build to order.

Maybe there's already companies catering to hams this way? (or hams that 
just use commercial companies to do this in the normal scheme of 
things..)  Think of Peter Dahl making transformers.

What you probably won't find is MFJ cranking these things out for $300.

Jim, W6RMK

TowerTalk mailing list

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