[HCDX] K9AY terminated loop
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[HCDX] K9AY terminated loop

G'day all,

Ok I know that I am getting real old, grumpy and lazy, but it seems to me
that the K9AY "terminated loop" antenna is in theory a pretty miserable
receive antenna. Yet some very highly respected DXers are giving it rave
reviews from actual on-air listening. Guess my analysis of the theory must
be wrong but can't see quite see where. Can someone please enlighten me? I
can't understand Gary Breed's explanation at all, must be early onset
Alzheimers for me.

The production of a cardiod pattern by combining equal amplitude signals
from a loop antenna (figure eight response in the horizontal plane, but
opposite lobes 180 degrees out of phase) and a vertical antenna
(omnidirectional equiphase response in the horizontal plane) has been known
since time immemorial (well the 1920s anyway). The principle is and was
commonly applied for radio direction finding. Amplitude was equalized by
attenuating the vertical antenna signal with a series resistance (since the
vertical antenna is the more efficient antenna when physical sizes of loop
and vertical are comparable and there is no Q boost to the loop by
resonating) and the 180 degree phase relationship maintained by balancing
the loop with respect to ground (with a balun) and feeding the vertical
antenna (positioned at the centre of the loop) at the centre of the balun.

Because the usual implementation was for direction finding, loop antenna and
vertical antenna sizes tended to be small as sensitivity was not an issue
for nearby signals being df-ed, but the antennas can be much larger subject
to the less than quarter wave dimensioning limitation (actually at the top
end of the mw band a full wave loop and vertical could be designed but this
would lose the broadband feature of the antenna).

For DX purposes the small df loop/vertical (there is a designated
descriptive name for the antenna in the early literature but i can't recall
it at present) got a bad reputation because of its lack of sensitivity. The
resonated high Q multiturn box loop gave good sensitivity and directivity,
but a df loop/vertical of the scale of a K9AY loop or larger would be
undeserving of such a designation. I can see that a K9AY size loop has the
potential to be competitive with a Beverage, if narrow beamwidth is not a
major criterion and even that could be overcome by a broadside array of

But the K9AY antenna on first inspection seems to be an inferior
implementation of the df loop/vertical (apologies in advance to Gary if I
have this all wrong). The guys in the 1920s  had plenty of time on their
hands with a ceiling of 10 megacycles and discrete components and they did
not miss much in AM broadcast band antenna design.

It seems to me that the K9AY antenna largely uses the variable "terminating"
resistor to balance the loop antenna with respect to ground by matching its
value to the value of the reflected input impedance of the receiver on the
other side of the ground connection. Could this be the reason why some DXers
have not been successful with this antenna (unusual receiver input impedance
not matched by the "terminating resistor"). In any event, this is not the
best way to go because the output of a non-resonant loop antenna is already
quite low and the resistance will reduce the output signal due to damping.
The damping due to the receiver input impedance cannot be eliminated but can
be greatly reduced with an impedance step-up incorporated into the balun
(the three to one turns ratio ferrite cored balun giving nine to one
impedance step up would be typical for a 50 ohm input impedance receiver).

Again the vertical antenna component of the K9AY antenna seems to be largely
the connecting wire between the centre of the loop antenna and the ground
itself, but from the reports of useful sensitivity, I am forced to assume
that at least a portion (probably all) of the vertical component of the loop
antenna is acting as a vertical antenna also, with output affected by the
"terminating resistance". I just don't see any advantage in this type of
ground connection, while there are all kinds of potential problems with it
due to unbalancing, impedance transformations and phasing variations. Best
to get rid of it and use a low impedance earthing with multiple earth
straps, or no local earth at all and electrostatic shield in the balun.

Remember with a cardioid pattern the weakest antenna is the determinant of
overall sensitivity, since amplitudes must be equalised. Therefore greatest
sensitivity requires either a high-Q loop or a large loop. I know cardioids
have received a bad name in more recent years, because in our search for
greatest sensitivity we have begun phasing them against very high-Q (as much
as 1000 unloaded) resonant multi-turned box loops. The rate of phase change
for the output signal from such loops in the vicinity of resonance is
phenomenal and even with additional damping is quite high, such that nulls
are very difficult to adjust precisely and tend to drift. A broadband
cardioid such as the K9AY must be a real relief to weary knob twisters, and
the degree of rotatability with multi-loops (has anyone put one of these on
a rotator? - armstrong or otherwise) a great bonus.

I think what i am saying is that, if the K9AY loop works well, an extra 6dB
or so should be obtainable (while retaining the broadband cardioid pattern)
by going back to the 1920s version - just a single turn loop, centre mounted
vertical, balanced feed to loop and centre feed to vertical through variable
resistance (maybe some variable reactance to shift nulls).

If this proves to be the case then thought needs to be given to either (1) a
multiturn loop (say 3 turns, each separated by 5cm,  switchable to have
either one, two or three turns in circuit) for better sensitivity on the low
end of the band or (2) an even larger loop - all the designs on the net are
sacrificing size for good MW performance - a framework of two lightly guyed
15 metre telescoping aluminium masts, insulated from ground and separated by
15 metres with a centre hung vertical wire should provide signal levels
comparable to a Beverage antenna yet take up minimum real estate on a beach
and be quite portable or (3) both the foregoing. Again phased broadside
arrays could really tighten up the beamwidth but a linked rotator system
would be essential and it might be difficult to find a beach that was wide
enough at high tide.

73s gd dx de

sam dellit

Hard-Core-DX mailing list

Copyright (c) Hard-Core-DX.com

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.1 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Text. A copy of the license is available in http://www.gnu.org/copyleft/fdl.txt.

All the other copyright notices are VOID.