Dumb question perhaps, but here goes. If you arrange 4 biquads in a rectangular formation and used a power combiner, would you get a decent omnidirectional antenna? Looking at the biquad pattern
On Sat, 2 Feb 2008 22:26:54 -0800 (PST), snipped-for-privacy@sushi.com wrote in :
What would be the point?
Like I already stated, high gain without a squished donut, i.e. compared to a colinear.
On Sat, 2 Feb 2008 23:57:24 -0800 (PST), snipped-for-privacy@sushi.com wrote in :
What makes you think it won't be a squished donut, albeit a crude one (due to uneven horizontal gain pattern)? The coverage of a biquad is more like 45° than 90°.
The other problem I see is if you use some sort of combiner or phasing harness the loss associated with such a device would be more or equal to the gain of any one antenna.
Adair
huh - just a random google search yielded this typical construction ?
No, the loss is very close to 6dB. Gain should be around 14db prior to loss. What might make a difference is the aperture.
The vertical response should be that of the biquad itself.
On Sun, 3 Feb 2008 11:50:53 -0800 (PST), snipped-for-privacy@sushi.com wrote in :
Which will be pretty much a squished donut, albeit a crude one.
On Sun, 3 Feb 2008 11:49:59 -0800 (PST), snipped-for-privacy@sushi.com wrote in :
Gain will probably only be 10-12 dB at best, with holes of much worse gain. A standard hi-gain omni is a much better way to go.
ok... just so we all are reading the same page -
Based on what technical analysis?
-
On Sun, 3 Feb 2008 15:08:33 -0800 (PST), snipped-for-privacy@sushi.com wrote in :
Published data on biquad antennas, both gain and antenna pattern; e.g.,
The BiQuad also makes a useful stand-alone antenna, with wide bandwidth and about 11dBi gain ...
Gain pattern:
On Sun, 03 Feb 2008 23:39:26 GMT, John Navas wrote in :
See also
I'm familiar with both of those websites. At the 45 degree point, the antennas would be in phase, boosting the signal 3dB. It would dip, but nothing is perfect. For instance, the omni is only at full gain over a narrow portion of the donut.
As you know, some sector antenna omni schemes use a similar design, though they use individual transceivers per sector.
snipped-for-privacy@sushi.com hath wroth:
Well, yes... it's kinda dumb the way you presented it, unless it's a theoretical question. What's missing are:
Actually, yes. That's roughly what would happen. The roughly 60 degree horizontal beamwidth would not yield a continuous omnidirectional pattern. There would be some holes every 90 degrees, where the pattern of two of the antennas meet. See patterns at:
However, there's a rather nasty power divider/combiner effect you need to consider. In order to get gain, you have to feed it with a 180 degree power splitter (Wilkinson Combiner). A 90 degree splitter will result in a rather ugly pattern. Let's assume for a moment that there are no combiner losses, and that the power splits equally (-3dB) per 2 ports.
In transmit, 1/4 th of the power goes to each antenna, resulting in a
-6dB loss in gain. In receive, all the signal received by any antenna goes to the receiver, so there's no receive loss. Therefore, if a single biquad has a gain of about +10dBi, the array will have a gain of +4dBi in transmit, and +10dBi in receive.
The -3dB bandwidth of a biquad is about 60 degress, not 90 degrees. At
+/-45 degress, that pattern appears to be -8dB down from the peak. That means you'll have a -5dB hole in the pattern every 90 degrees if you use a power splitter/combiner.Yep. -6dB per port for a 4 port splitter plus whatever internal loss the combiner adds. Typically about 0.5dB per port. See specs on a typical combiner at:
Did you ever even look at the Amos/Franklin antennas I've often suggested?
Note that there are several photos showing back to back antennas to form an omni pattern.
Incidentally, I blundered across this "curtain quad" antenna, which looks interesting:
It was mostly a thinking outside the box exercise. Also, I have need for a high gain vertically polarized L-band antenna, and have seen enough poorly made coax colinears that I figured something else should be examined.
I will look into that "curtain" for wifi, but I need an omni. Nothing COTS exists for my other project. L band is an odd place to be. Too low in frequency for some designs and to high in frequency for others (J-poles for instance).
Have something to add? Share your thoughts — no account required.
Ask the community — no account required