Distance to Link Coupling in a Loop Antenna
 

I have used Reg's fine loop design program rjeloop3 to help me with building
a couple of loops. It was a big help. Thank you Reg.

The shielded three-foot diameter loop I built is for receive only. The
primary loop is six turns paralleled by a variable capacitor, and the
secondary (link loop) is one turn (same diameter) fed to a preamp or
directly to the receiver. The band of operation of the loop is about 200kHz
to 500kHz (NDB chaser).

One question that is not answered in rjeloop3 is what should be the distance
from the output (link) loop to the main loop? The antenna's I built used
flat cable for the wiring and one of the turns was the link so the distance
was 0.050 inches. In other articles I have read that the Q, gain, and S/N
ratio can be varied by varying the distance between the two loops.
Approximately what is the distance? Are we talking a small fraction of an
inch or 2, 3, or perhaps six inches apart? I would just like a feeler
distance to work with. If the distance is to be over one inch, I'm
considering building the primary loop in one hoola-hoop form and the
secondary loop in a second hoola-hoop then varying the distance between the
hoops. Any thoughts on this or is this a waste of time and effort?

All input will be greatly appreciated. Thank you.

Al
KA5JGV

"Al" wrote
The shielded three-foot diameter loop I built is for receive only. The
primary loop is six turns paralleled by a variable capacitor, and the
secondary (link loop) is one turn (same diameter) fed to a preamp or
directly to the receiver. The band of operation of the loop is about
200kHz
to 500kHz (NDB chaser).

One question that is not answered in rjeloop3 is what should be the
distance
from the output (link) loop to the main loop? The antenna's I built used
flat cable for the wiring and one of the turns was the link so the
distance
was 0.050 inches. In other articles I have read that the Q, gain, and S/N
ratio can be varied by varying the distance between the two loops.
Approximately what is the distance? Are we talking a small fraction of an
inch or 2, 3, or perhaps six inches apart? I would just like a feeler
distance to work with. If the distance is to be over one inch, I'm
considering building the primary loop in one hoola-hoop form and the
secondary loop in a second hoola-hoop then varying the distance between
the
hoops. Any thoughts on this or is this a waste of time and effort?

=====================================
Hi Al,
To have a significant effect on operating Q it is necessary to shift the
coupling loop away from the main loop by more than about 1/10 of loop
diameter. This is mechanically inconvenient and anyway I've forgotten how to
calculate the coupling coefficient between the two coils. A more convenient
way of varying coupling between main loop and receiver is to place a smaller
loop inside the main loop in the same plane. Just like a Magloop
transmitting loop. Additional variation can be obtained by rotating the
coupling loop relative to the other. It is easier to calculate.

The two loops constitute a transformer. It makes little difference
wherabouts the small loop is located inside the large loop. So it can be
located near the bottom of the main loop near the tuning capacitor. It is
better to have no direct connection between the two loops. It may upset the
natural balance between the main loop and its surroundings and ground. One
side of the coupling loop is grounded at the receiver end of the line.

If the line is an appreciable fraction of a wavelength (very unlikely) then
use an impedance Zo in the same ballpark as the receiver input Z. It hardly
matters whether it's coax or very loosely twisted pair.

The transformer has an effective turns ratio of -

N = (N1*D1) / (N2*D2).

N1 = main loop turns.
D1 = main loop diameter.

N2 = coupling loop turns.
D2 = coupling loop diameter.

Which may be very interesting but will not be of the slightest use unless
the receiver input resistance is known.

RJELOOP3 calculates the parallel L and C resonant resistance of the main
loop. Call this R1. To match this to the receiver input resistance R2, make
N = SquareRoot(R1/R2).

When impedance-matched to the main loop the receiver input resistance damps
the intrinsic Q of the loop to exactly half of its unloaded value. The
received signal is then maximised.

Prefer to use a larger diameter for the coupling loop rather than a larger
number of turns.

To maximise operating Q and selectivity match the loop to 1/2 or 1/3 times
the receiver's actual input resistance. There will be a few dB loss in
signal strength.

None of the foregoing will have any effect on signal to noise ratio except
when the signal bandwidth is appreciably smaller than the bandwith accepted
by the loop. But it is best to manage channel bandwidth in the IF amplifier.

The natural Q of a good receiving loop at LF and MF is already high enough
to spoil the audio quality of broadcast stations.
---
Regards, Reg, G4FGQ

Reg, thank you for the detailed explanation. You have given me plenty to
work with and to contemplate on for my next loop project.

This group has some quality people in it and I, and I'm sure others, are
grateful that you are willing to share your knowledge with us. Thank you.
Al
KA5JGV


"Reg Edwards" wrote in message
...
=====================================
Hi Al,
...snip..
Regards, Reg, G4FGQ

"AL KA5JGV" wrote -
Reg, thank you for the detailed explanation. You have given me plenty to
work with and to contemplate on for my next loop project.

=========================

Al, I was able to respond in some detail to your query because you gave me a
clear idea of your objectives. It was also a topic likely to be of interest
to other readers. I do appreciate that many questioners are unable to
descibe a problem in terms which people like me (I'm no genius. Neither am I
a mind-reader) are unable to understand sufficiently clearly. Furthermore,
I have to be aware the enquirer is as likely to be just as unable to
understand my reply as I am in understanding him. I have to guess a person's
circumstances from an extremely limited amount of 'data'.

If we're not careful a thread continues with further questions, is extended
interminably by well-intentioned and other 'do-gooders' and eventually
fizzles out with everybody suffering from raised hackles. I mention these
aspects of newsgroup info-exchange, or often mis-information exchange, for
the benefit of other readers. Everybody has a different motive for joining
in.
.................................................. .................

Just to complete our technical discussion, if program RJELOOP3 does not make
itself clear, then if you have N turns on the main loop and one turn
closely-coupled turn of the same diameter connected to the receiver, then
the Q of the main L-&-C tuned loop is damped with a resistance across it
equal to N-squared times the receiver input resistance.

So, if there are 6 turns on the main loop and the receiver input resistance
is 1000 ohms (a bipolar transistor) then the equivalent resistance shunting
the main loop is 36,000 ohms. If the parallel resistance of the main loop is
100,000 ohms (a typical MF value) then the resulting operating Q of the
circuit is reduced to roughly 1/3rd of its intrinsic value. This is
over-coupling but may be acceptable.

But if the receiver input resistance is 50 ohms then you will feel obliged
to do something about it. As you cannot reduce the number of turns on the
coupling coil to less than 1 then, as described in my original reply, there
is no alternative but to reduce the diameter of the coupling coil, either a
square or circular, and locate it inside the main loop in the same plane.
Using 12 or 10 gauge enamelled wire (a better appearance) it will be
self-supporting.

Reply not needed. Just make an entry in your notebook. I have volumes A to
S. The bookcase shelf is sagging. You are sure to be successful with your
experiments. It is impossible for an experiment to fail. ;o)
---
Reg, G4FGQ

Al wrote:

I have used Reg's fine loop design program rjeloop3 to help me with building
a couple of loops. It was a big help. Thank you Reg.

The shielded three-foot diameter loop I built is for receive only. The
primary loop is six turns paralleled by a variable capacitor, and the
secondary (link loop) is one turn (same diameter) fed to a preamp or
directly to the receiver. The band of operation of the loop is about 200kHz
to 500kHz (NDB chaser).

One question that is not answered in rjeloop3 is what should be the distance
from the output (link) loop to the main loop? The antenna's I built used
flat cable for the wiring and one of the turns was the link so the distance
was 0.050 inches. In other articles I have read that the Q, gain, and S/N
ratio can be varied by varying the distance between the two loops.
Approximately what is the distance? Are we talking a small fraction of an
inch or 2, 3, or perhaps six inches apart? I would just like a feeler
distance to work with. If the distance is to be over one inch, I'm
considering building the primary loop in one hoola-hoop form and the
secondary loop in a second hoola-hoop then varying the distance between the
hoops. Any thoughts on this or is this a waste of time and effort?

All input will be greatly appreciated. Thank you.

Al
KA5JGV

I don't know if this will help, but perhaps you can adopt
the idea to your loop.

I built a receiving mag loop for 40 - 160 as follows:
Using hardline, I stripped the ends so the center conductor
stuck out about 1 inch on each end. I made the outer loop
out of it - it is ~4' in diameter. It is open at the top, and
a variable capacitor (900 pF) connects the two ends on the
shield of the hardline. (I switch in another 1650 pF for 160).
Inner loop is made from 52" of the inner conductor and
insulation of RG8 and is mounted at the bottom. The two ends
of the inner loop connect to the SO239. At the top of the
outer loop, the copper inner hardline conductor connects to
a 1N4148 with 1K in series and a 1 mA meter. When I excite
the inner loop with my MFJ 259B, I tune for minimum SWR with
the variable cap. The 1mA meter goes to maximum when the SWR
goes to minimum.

Ok, now to the coupling. The inner loop can be maneuvered
for maximum reading on the 1 mA meter. I found I get the
highest reading on the meter when I squash the inner loop
into an oval, decreasing its vertical diameter to about
1/2 what it was when it was a circle. I taped the inner
loop to the outer loop to hold the shape with the maximum
coupling. Perhaps some variation of this technique can
be used on your antenna.

I'm still experimenting with my loop. The tuning on 40
is difficult - I need a smaller variable cap. I plan to
try switching out 450 pF of the variable (it is currently
2 450's in parallel), and switching in some series fixed C
to see if I can cover 40 with a 180 degree rotation of the
variable - right now a few degrees of rotation of the
variable makes a very large change in the frequency.

Sounds like you have made your own variometer
In the case of a loop antenna it is a coupling of two circuits each
having resistance, capacitive and inductance .Each of these values
have two forms,
reactive whuch means that it is a combination of two reactive
components and in the case of resistance there is real resistance and
radiative resistance.
You can insert distributed loads for each of these components or vice
versa
so that these two circuits can be made into one circuit by capacitive
or inductive coupling means.( read up on solving complex circuits)
Obviously the inductive coupling values will change along with the
other loads so the use of a variometer helps to not tie one in
absolute terms of inductance coupling to the thickness and length of
material used which is representitive of distributed loads as well as
losses.
If one chooses compatable material thicknesses for a 50 ohm impedance
one can physical connect both circuits which is a inductive coupling (
you can also do capacitive coupling) and thus use only the variable
capacitance for what will be a limited total band spread.
Remember that when designing any antenna it is to get maximum current
flow
So the introduction of TWO radiating circuits gives you extra leverage
to obtain this. If you are sufficiently skilled then you can add a
series of circuits and thus obtain efficient radiators with a
controlled input impedance while choosing what circuit has the high
current flow as well as selection of any individual bandwith.This
should explain why physically coupling two loop circuits can restrict
choice to determine what is required as opposed to your choice to be
able to change the coupling mode which gives you more choice for
better options.Increased variations and design changes for ultimate
rotational design of antennas for any frequency without being bound to
wavelength which will soon be available via the PTO.
Good luck and please keep us informed of your progress
Art Unwin KB9MZ...XG



wrote in message ...
Al wrote:

I have used Reg's fine loop design program rjeloop3 to help me with building
a couple of loops. It was a big help. Thank you Reg.

The shielded three-foot diameter loop I built is for receive only. The
primary loop is six turns paralleled by a variable capacitor, and the
secondary (link loop) is one turn (same diameter) fed to a preamp or
directly to the receiver. The band of operation of the loop is about 200kHz
to 500kHz (NDB chaser).

One question that is not answered in rjeloop3 is what should be the distance
from the output (link) loop to the main loop? The antenna's I built used
flat cable for the wiring and one of the turns was the link so the distance
was 0.050 inches. In other articles I have read that the Q, gain, and S/N
ratio can be varied by varying the distance between the two loops.
Approximately what is the distance? Are we talking a small fraction of an
inch or 2, 3, or perhaps six inches apart? I would just like a feeler
distance to work with. If the distance is to be over one inch, I'm
considering building the primary loop in one hoola-hoop form and the
secondary loop in a second hoola-hoop then varying the distance between the
hoops. Any thoughts on this or is this a waste of time and effort?

All input will be greatly appreciated. Thank you.

Al
KA5JGV

I don't know if this will help, but perhaps you can adopt
the idea to your loop.

I built a receiving mag loop for 40 - 160 as follows:
Using hardline, I stripped the ends so the center conductor
stuck out about 1 inch on each end. I made the outer loop
out of it - it is ~4' in diameter. It is open at the top, and
a variable capacitor (900 pF) connects the two ends on the
shield of the hardline. (I switch in another 1650 pF for 160).
Inner loop is made from 52" of the inner conductor and
insulation of RG8 and is mounted at the bottom. The two ends
of the inner loop connect to the SO239. At the top of the
outer loop, the copper inner hardline conductor connects to
a 1N4148 with 1K in series and a 1 mA meter. When I excite
the inner loop with my MFJ 259B, I tune for minimum SWR with
the variable cap. The 1mA meter goes to maximum when the SWR
goes to minimum.

Ok, now to the coupling. The inner loop can be maneuvered
for maximum reading on the 1 mA meter. I found I get the
highest reading on the meter when I squash the inner loop
into an oval, decreasing its vertical diameter to about
1/2 what it was when it was a circle. I taped the inner
loop to the outer loop to hold the shape with the maximum
coupling. Perhaps some variation of this technique can
be used on your antenna.

I'm still experimenting with my loop. The tuning on 40
is difficult - I need a smaller variable cap. I plan to
try switching out 450 pF of the variable (it is currently
2 450's in parallel), and switching in some series fixed C
to see if I can cover 40 with a 180 degree rotation of the
variable - right now a few degrees of rotation of the
variable makes a very large change in the frequency.