Towertalk
[Top] [All Lists]

[TowerTalk] Quad Users

To: <towertalk@contesting.com>
Subject: [TowerTalk] Quad Users
From: cebik@utkux.utcc.utk.edu (L. B. Cebik)
Date: Tue, 22 Dec 1998 06:28:51 -0500 (EST)
John,

I have not studied your last question, but the first two have
straightforward answers.

1.  5 bands is possible for a quad, but the individual quads for each band
need slight refiguring.  I have designed for study purposes a 5-band quad
on the hub-spider (also called the conical) principle where each
driver/reflector combination uses the same spacing in terms of
wavelengths, and the predicted performance on each band is about the same.
Over ground, the numbers will vary by band, since any particular height
above ground will be a different height in wl per band.

2.  My studies found two strong interactions with a common feed:  20-10
due to the harmonic relationship and 12-10 due to the close proximity of
frequencies that makes the 12 driver a reflector for 10 meters.  Properly
sized separate fed quads avoid these interactions if the unused driver
loops are effectively closed.  This can be accomplished by a central
switch to 1/4 wl sections to each driver which are open when not in use at
the switch end.

3.  The impedances for each driver will vary with the spacing you use
between elements in terms of fractions of a wl.  However, the following
pattern will emerge in the hub-spider type 5-band quad.  The middle bands
(17 through 12) will show a certain feedpoint Z.  The 20 driver--with no
lower frequency driver below it, will show a high impedance.  The 10
meter driver, with no higher frequecy driver beyond it, will show a lower
impedance.  Depending on spacing, you can design the middle group
impedance for 50, 60, 70 or so Ohms.  Essentially, only 10 meters will not
show full band coverage within a 2:1 SWR after matching.  Some designs
using a boom and fixed spacing appear to have overcome the impedance
difference on the band edges, although some by matching systems.

4.  GEM Quad, Lightning Bolt Quads, and Cubex contest the need for
separate feeds.  Antenna Mart (Bill Wall) has done empirical studies that
appear to support separate feed systems.  I have heard of some cases in
which common feed quads were redesigned by users for separate feeds.  Note
that the 20-10 interaction exists within the traditional 3-band quad.  If
one wants two quads, each with common feed, then perhaps a 20-17-12 and a
15-10 pair of combos would work.  A 17-12 combo with common feed should
show little interaction.

5.  A 5-band common feed system will work.  Whether it provides optimal
performance for the antenna type is exceptionally difficult to ascertain,
since having the requisite towers and other materials to make a valid
side-by-side test between separate and common feed antennas that are
otherwise the "same" is not normally feasible for most hams.  However, I
have received numerous reports of folks getting very good results.  I
cannot say that they would or would not get "even better" results with
separate feed due to the many variable involved.  The band that
potentially suffers most is 10 meters, at least in theoretic studies.
Hence, the question may come down to a decision between the convenience of
a single feed vs. the importance of 10 meters in the overall operating
schedule.

6.  Part of the intangible nature of the common vs. separate feed question
comes down to the gain one wishes to obtain from a 2-element quad.  Common
feed interactions appear to be highest when the antenna is designed for
maximum gain (about 7 to 7.2 dBi free space gain) and lessens for designs
satisfied with a gain averaging about 6.5 dBi free space--each with 18-22
dB F-B ratio.  I do not have a general demonstration to support this
point, but the principle seems to be that the higher the gain and more
peaked the performance, the more sensitive to interactions.

Even this long note is far from the full story, but hope the notes are
useful.

-73-

LB, W4RNL

L. B. Cebik, W4RNL         /\  /\     *   /  /    /    (Off)(423) 974-7215
1434 High Mesa Drive      /  \/  \/\     ----/\---     (Hm) (423) 938-6335
Knoxville, Tennessee     /\   \   \ \   /  / || /      (FAX)(423) 974-3509
37938-4443     USA      /  \   \   \ \       ||              cebik@utk.edu
                URL:  http://web.utk.edu/~cebik/radio.html




--
FAQ on WWW:               http://www.contesting.com/towertalkfaq.html
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm


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