[Antennas] Mult-band Quads- Matching
Joe Giacobello
k2xx at swva.net
Sat Nov 28 11:55:41 EST 2009
I, like Norm, prefer separate feed lines for my multiband quads. Any of
the commonly available or easily constructed remote antenna switches
makes this easy to accomplish.
I'd also like to add that when one goes beyond two elements, the feed
impedances "naturally" drop to the vicinity of 50 ohms. By using
modeling software, one can adjust the element lengths to get around 50
ohms while sacrificing little, if any, forward gain. Using that
approach, I have successfully modeled and built a six band quad (four
elements on 10-20M and seven on 6M using equidistant spacing on a 24'
boom) without using any kind of matching system. (I should point out
that I was only interested in the lower ends of each band - CW and lower
end of SSB - as opposed to having an acceptable match across an entire
band.)
OTOH, I have found that it is important to electrically close the driven
element loops on the inactive bands in order to get acceptable, if not
optimum, performance on any given band. I was alerted to this
requirement by W4RNL after I had built my first multiband four element
quad and it turned out to be a real dud. He recommended using quarter
wavelength coaxial links between the remote switch's terminals and each
feedpoint for each band. The switch end of each link is left open when
it is inactive so that the feedpoint end was effectively (electrically)
closed. When I rebuilt the same quad at my current QTH, I followed that
recommendation and the results were quite gratifying. Before putting
that quad in place on the tower, I checked the input impedances and SWRs
of each band with and without shorting the inactive bands' driven
elements. There were major differences on most, but not all, bands. I
have meant to model this effect but I never got around to it. Some have
reported that when the driven elements are left open, the inactive
elements show higher currents than when they're closed. (As K4DLI
pointed out to me, one can use half wavelength links and keep the switch
end grounded and accomplish the same electrical effect at the feedpoint
end.)
73, Joe
K2XX
NPAlex at aol.com wrote:
>
> I would like to add some comment regarding mult-band Cubical Quad matching
> based on some of my experience. Historically there have been basically two
> approaches, one was parallel all driven elements and feed them with coax
> or balanced line and use an antenna tuner. Or individual feeds for each
> band switched with a remote antenna switch. The problem to be resolved is the
> variation in source impedance from band to band with a constant element
> spacing. The impedance ranges from near 50 ohms (typically 20m) to over 100
> ohm for 10 meters.
>
> The "Spider" designed quads purports to solve this problem by distributed
> spacing (optimum for each band) and therefore each now is 50 ohms. Not
> achieved in my analysis of this form, and I believe there is more interaction
> when the elements are distributed versus co-located ( I think LB Cebik also
> concluded this).
>
> So what is the solution to single feed of a multi-band quad? Lightning
> Bolt (SK), Gem (SK?) and Cubex (alive an well) offered matching
> transformers. The Lightning Bolt model was a two level transformer, with Low and Hi Z
> terminals. Cubex (still offers) a family of transformers depending on the
> number of elements (2, 3 or 4) which offers three levels of termination Lo,
> Mid and Hi. One limitation is power, 1500w is the maximum. The matches
> are not perfect, but in general the antenna will exhibit less than 2:1
> across each band.
>
> However, my preference is individual matching cables for each band.
> Typically 10, 12 and 15m respond well with odd quarter wave 75 ohm calbe (I
> found 3/4 wave the best choice as it allowed the remote switch to mount on the
> mast) and 1/2 w 50 ohm cables for 17m and 20m. There is always the
> question as to how should the unused matching cables be terminated? Open or
> shorted? I have done both and still have not found a consistent position - cut
> and try.
>
> Norm W4QN
>
>
> Message: 1
> Date: Tue, 24 Nov 2009 18:52:06 -0600
> From: Wendell Wyly - W5FL <w5fl at sbcglobal.net>
> Subject: Re: [Antennas] Quad matching
> To: Tom Horton <k5iid at sbcglobal.net>
> Cc: antennas at mailman.qth.net
> Message-ID: <77FE249A-B56E-419F-961E-5E9C5B616198 at sbcglobal.net>
> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
>
> Hi Tom. I have the same experience you have with the matching
> sections. I simply put a home-made relay box on the bottom spreader
> to switch all 5 quad loops on the driven element and feed it with a
> single coax up the tower.
>
> I did get a transformer/balun from lightning bolt quads several years
> ago, but never tried to use it and he got out of the business! Hope
> to get my quad back up in the air after lying on the ground for the
> last 5 years, but not willing to try anything different than described
> above.
>
>
> On Nov 4, 2009, at 10:34 PM, Tom Horton wrote:
>
>
> I have been using quads for the better part of 50 years and I have
> always used 1/4 wl transformers of RG-11 to RG8.
> Has anyone ever built a single point transformer for one? I know
> that Cubex among others have them for sale, but was interested in
> trying to build something to use with a single feedline for a change.
> Any ideas or thoughts or refs?
> Thanks,
> Tom K5IID
>
>
> "If It Weren't For The United States Military There Would Be NO United
> States of America ."
>
> September 11, 2001
> We can never, never forget !
> If we aren't free and alive,
> absolutely nothing else matters !
>
>
> Tom Horton K5IID
> Hillsboro, TX
>
>
>
>
> ______________________________________________________________
> Antennas mailing list
> Home: http://mailman.qth.net/mailman/listinfo/antennas
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Antennas at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
>
>
More information about the Antennas
mailing list