[Top] [All Lists]

[TowerTalk] XQ quad a question

To: <>
Subject: [TowerTalk] XQ quad a question
From: (L. B. Cebik)
Date: Mon, 9 Jun 1997 07:18:23 -0400 (EDT)
On Tue, 27 May 1997, Johan Van de Velde,ON4ANT wrote:
> Hi there
> I received some dimensions for an XQ quad.
> 12el on 28ft boom for 50 MHz.
> Claimed gain over 18dBd wow

I ran the 12 element XQuad on NEC-4, using 11 segments per side, in free
space, as a quick check on the calculated/claimed gain.  I think you will
be disappointed.

The feed Z was about 133 + j21 ohms, which indicates the design is close
to correct in terms of resonance.

However, the XQuad consists of 2 wl long loops or 1/2 wl per side.  This
yields a split pattern.  At short lengths (2 or 3 directors) the lobes are
40 degrees each side of center with no gain at the forward center line.
With all 12 elements, the lobes are 35 degrees each side of center with no
centerline gain.

The F-B ratio is between 15 and 16 dB, with a ragged rear pattern, due in
part to the split forward pattern.

The gain of each of the split lobes was only about 7.6 dBi.  This amounts
to a 15.2 dBi gain, but split into 2 directions, each with half the gain.
In terms of dBd, this is only about 5.4 dBd per lobe.

The trouble with the XQuad 2 wl loop concept is that the 1/2 wl sections
are not correctly space to establish phase relationships for a single
pattern with maximum gain along the centerline.  Changing the spacing and
number of directors will not change this situation.

The split pattern is also often encountered at HF when a 20 meter and 10
meter quad are fed together.  Since the 20 meter quad also presents a low
impedance at 10 meters, it hogs a good part of the power and the 10-meter
pattern is then split, with much reduced gain along the forward

Although the XQuad may have a niche in radio work--especially if one needs
to maintain contact with two stations about 70 degrees apart--for general
radio work where we expect the signal to go where we point the antenna,
the models suggest that the XQuad 2 wl loop concept might be replaced with
a quad using 1 wl loops and optimized director placement.

I hope this modeling data is useful to you.



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      /  \   \   \ \       ||    

FAQ on WWW:     
Administrative requests:

<Prev in Thread] Current Thread [Next in Thread>
  • [TowerTalk] XQ quad a question, L. B. Cebik <=