[NLRS] Favorite horizontal loop design?

[Dr. Gerald N. Johnson]

On Sat, 2007-07-07 at 21:47 -0500, John P. Toscano wrote:
> 
> Paul Beckmann (WAØRSE) wrote:
> 
> > I'd like to make up a stack of 3 horizontal loops for 6, 2, and .7
> > meters and am wondering what tried-and-true designs the NLRS crew
> > might recommend. I'm especially interested in matching section
> > designed used. I'd also like any advice on spacing that will keep me
> > out of trouble. I'll likely build all of them out of copper
> > pipe/tubing for fixed outdoor use. Not anything for great performance,
> > just to have weak signal capability from the home QTH.
> 
> This has started an interesting thread.  At home, I have an old M 
> Squared SQLOOP on 6M, and a stacked pair of M Squared SQLOOPs on 2M.

>From the EZNEC computations I've done this evening, you could do a lot
worse than feeding your 2m with some 432 or 1296 RF. The antenna I have
modeled is a lot like the M2 and the R of the feed is close to 50 ohms
but its a bit reactive. I think minor tweaks of shorts or open gap
spacing might improve the 3rd and 9th harmonic impedances without
changing the fundamental too much. The 1/4 wave power divider will work
fine in 3/4 wave mode with the same results.

A gamma match loop won't go to the third harmonic, but the stub fed one
will.

> For 
> the 2M stack, I went the "Cadillac" route with a real 2-port power 
> divider and all 50 ohm coax (one half-wave section and one 3/2 wave 
> section to the two antennas from the 2-port PD).  Not sure how much 
> penalty I would have paid by using two odd multiples of 1/4 wavelength 
> 75 ohm phasing lines to a simple T, which is the other option that M 
> Squared was selling at the time (the Yugo solution instead of the 
> Cadillac solution).

There is little difference in these solutions. One uses only 50 ohm coax
OUTSIDE the splitter (which uses a quarter wavelength of 35 ohm coax),
the other uses those odd multiple quarter wave lines of 75 ohm coax
which isn't quite as common though it has a bit lower loss than 50 ohm
and for a prefect match ought to be 70.7 ohms. The bandwidth of the 75
ohm lines when 3 or 5 or more quarter wave lines is less which means the
length has to be more precise and that precision can be messed up by the
coax itself, particularly the velocity of propagation which is messed up
by variations on the dielectric constant of the inner insulation.

Both schemes work and work well. The 50 ohm power divider and the all 50
ohm coaxes can be any length so long as they are equal. The 75 ohm lines
to the T need to be equal AND close to that odd multiple of 3/4 wave.
The bandwidth isn't horrible at 5 or 7 quarters I don't think just
requires one to be more precise than for a single quarter wave where the
bandwidth my be +/- 20% for a reasonable match.

> Anyway, I realize that when you talk about stacking 
> them, you mean three different single-band antennas stacked on the same 
> support mast, not 2 or more antennas per band stacked for extra gain.
> 
> But, sort of on topic, I *DO* have these three antennas stacked on the 
> same support mast.  The 6M loop is at the top, and the pair of 2M loops 
> are below it, spaced 1/2 wavelength (about 40 inches) apart from one 
> another.  I guess the optimal spacing of a pair of 6M loops would be 12 
> feet apart (per M^2 web page on the 6M HOLOOP), so the capture area of 
> the 6M loop at the top of the mast would extend down 6 feet, and the 
> capture area of the 2M loop below it would extend up about 2 feet, so 
> the interaction between them would be negligible if they were 8 feet 
> apart (their capture areas just touch), and excellent if they were 6 
> feet apart (the capture area of the 6M loop just reaches the actual 2M 
> hardware.  Anything closer is a compromise, and the closer they get the 
> worse the compromise gets.  As Jon has pointed out, a test would be in 
> order to see how close they can get without an obvious performance 
> penalty, but as Gerald points out, your performance testing is probably 
> going to be awfully crude.

The 6m loop is probably not going to be affected by the 2m loops because
the little loops are no where near resonant at 6m. But the third
harmonic resonance of the 6m loop may absorb energy from the 2m loops if
they get too close.

> 
> I did want to mention that the horizontal loops have been "lifesavers" 
> for me in a couple of circumstances.

I'm noticing that so I'm working on a triband version, at least on the
computer for now.
> 
> One, when my rotator was out of service for half a year being repaired, 
> I still had SSB/CW capability on those 2 bands for which I had the 
> loops. I had a lot of fun with Sporadic E and tropo openings on those 
> two bands when my rotatable antennas weren't rotatable.
> 
> Another, if I want to quickly check for 6M sporadic E in *ANY* 
> direction, I can just flip the coax relay to the 6M loop and listen in 
> every direction at once.  In a really powerful opening, many stations 
> can be worked on the loop with no pointing hassle at all.
> 
> As was also pointed out, working close-in stations is much easier with 
> the loops, since they have "enough" gain with zero pointing effort, 
> whereas a close-in station to your NorthWest may be hard to work with 
> your high-gain beam pointed NorthEast or SouthWest. How fast does your 
> rotator rotate?  :)  And, in a multi-station ragchew where the other 
> folks are all in different directions from you, the omnidirectionality 
> is a blessing.

All three good points we miss from concentrating on the stacked 5wl
yagis that have to be within 10 degrees of perfect to be heard.
> 
> Don't get me wrong, I like antennas with a lot of gain!  But there are 
> times when the horizontal loop comes in handy. I never got one for 222 
> or 432, and after looking at this thread, I may get motivated to build a 
> couple instead of sending more money to a "real antenna" builder...
> 
> 73 de W0JT

I think my 2m to 1296 square loop is going to be built from a length of
1/4" OD copper tubing with a shorting strap and a coax connection on the
stub. Probably need about 50 inches of tubing. It might work better not
being square, but none of the NECs do well with angles between wires
other than square corners or parallel wires. And then they need to be
spaced some distance. I was looking at so called 1/4" rigid which is
3/8" OD, but EZNEC didn't like the segment lengths and changing
segmentation changed the characteristics, especially at the higher
frequencies. That would have been stiffer, but at about 10 inches square
the soft copper tubing should stand up pretty good. it can always be
braced with some pieces of fiberglass electric fence posts (about 5/16"
or 3/8" diameter).
-- 
73, Jerry, K0CQ,
All content copyright Dr. Gerald N. Johnson, electrical engineer