Another unidirectional small loop: the Coplanar Twin Loop

[qrp-gaijin]

After seeing all the discussion and technical analyses of the very interesting SULA unidirectional small loop antenna, I thought I'd post a reference to another seemingly little-known, but rather mysterious and intriguing unidirectional small loop antenna: the Coplanar Twin Loop (CTL) antenna by the late Dr. Mike Villard, who was a noted antenna researcher and university professor.

The CTL is a very small antenna unidirectional that fits indoors on a tabletop, as seen on this image:


The CTL apparently achieves its cardioid pattern by combining an interior low-impedance loop antenna with an exterior and coplanar high-impedance loop, with the outer high-Z loop supposedly acting like a "bent whip" antenna that generates a so-called "magnetic shadow" region. It seems to be a rather unique antenna, and its detailed mechanism of operation (involving phase shifts, a variable damping resistor, a magnetic shadow region, and two concentric resonant loops) is intriguingly complex and subtle. A good introductory article can be seen here: https://web.archive.org/web/20201218104 ... planar.pdf .

Another detailed article can be seen at the links below (8 pages, as separate links).

http://web.archive.org/web/201912070924 ... ml/bg1.png
http://web.archive.org/web/201912070925 ... ml/bg2.png
http://web.archive.org/web/201912070926 ... ml/bg3.png
http://web.archive.org/web/201912071101 ... ml/bg4.png
http://web.archive.org/web/201912071149 ... ml/bg5.png
http://web.archive.org/web/201912071150 ... ml/bg6.png
http://web.archive.org/web/201912071150 ... ml/bg7.png
http://web.archive.org/web/201912071151 ... ml/bg8.png

The above article states that the magnetic shadow operation of the CTL was computationally and experimentally verified on 78-83 of a 162-page report by Dr. Villard, the bibliographic information of which is here: https://ntrl.ntis.gov/NTRL/dashboard/se ... 8180.xhtml

One experimenter who built this antenna reported on it here, but unfortunately without carefully testing its unidirectional properties: https://sv2yc.wordpress.com/2013/04/27/ ... -w6qyt-sk/ .

I was hoping that with all of the knowledgeable and interested people here on the forum, perhaps the above links about the CTL might encourage some discussion and analysis of this interesting and mostly-forgotten antenna. Some years ago, I did a preliminary 4nec2 analysis to confirm the unidirectional pattern. I'm not sure if my analysis is meaningful, but it may serve as a starting point for further analysis.

ctl-pattern-11MHz.png (165.09 KiB) Viewed 3356 times

The 4nec2 file for the antenna simulation is shown below.

Code: Select all

CM Model of Coplanar Twin Loop antenna.
CM Created by qrp-gaijin@yahoo.com (http://qrp-gaijin.blogspot.com).
CM 
CM Adjust model as follows:
CM 
CM 1. Adjust C2 (outer loop capacitance) to be very large (e.g. 1000 pF), so that the outer loop is far away from resonance.
CM 2. Use 4nec2 optimizer to adjust C1 (inner loop capacitance) to resonate the inner loop (X=0 is the optimizer criterion).
CM 3. Use 4nec2 optimizer to adjust C2 (inner loop capacitance) for maximum F/B ratio.
CM 4. Use 4nec2 optimizer to adjust RL (loading resistance of outer loop) for maximum F/B ratio.
CM 5. Repeat step 3 and 4 until no further improvement is possible.
CM 
CM When properly adjusted, the model shows a unidirectional null, both in free space, and above ground.
CM 
CM Note that the outer loop must be high impedance (achieved by a loading inductance), and the inner loop must be low impdeance.
CM 
CM The outer loop's high impedance allows it to respond to both E and H fields of the incoming wave, allowing for the unidirectional response.
CM 
CM The model has initially been adjusted for a unidirectional null at 11 MHz in free space.
CE
SY h=1.0
SY r=6.508e-4
SY c1=3.79e-11
SY c2=6.02e-12
SY rl=70.89846
GW	1	7	-0.5	0	h	0.5	0	h	r	'Wire 1, 9 segments, halve wavelength long.
GW	2	7	0.5	0	h	0.5	0	h+1	r
GW	3	7	0.5	0	h+1	-0.5	0	h+1	r
GW	4	7	-0.5	0	h+1	-0.5	0	h	r
GW	10	9	-0.642857	0	h-0.142857	0.642857	0	h-0.142857	r
GW	11	9	0.642857	0	h-0.142857	0.642857	0	h+1.142857	r
GW	12	9	0.642857	0	h+1.142857	-0.642857	0	h+1.142857	r
GW	13	9	-0.642857	0	h+1.142857	-0.642857	0	h-0.142857	r
GE	0
LD	0	2	4	4	0	0	c1
LD	0	11	5	5	0	0	c2
LD	0	13	5	5	rl	10e-6	0
LD	5	0	0	0	9.9E+99
GN	-1
EK
EX	6	4	4	0	1	0	0	'Voltage source (1+j0) at wire 1 segment 5.
FR	0	0	0	0	11	0
EN

[Andrew (grayhat)]

First of all, welcome to the forum !!

As for the CTL, I've played with it while designing the SULA and it's for sure an interesting design, I've just tried your NEC model and slightly changed it inserting the parameters for the ground

Code: Select all

CM Model of Coplanar Twin Loop antenna.
CM Created by qrp-gaijin@yahoo.com (http://qrp-gaijin.blogspot.com).
CM modifications by Andrew (GrayHat) :: inserted ground parameters
CM 
CM Adjust model as follows:
CM 
CM 1. Adjust C2 (outer loop capacitance) to be very large (e.g. 1000 pF), so that the outer loop is far away from resonance.
CM 2. Use 4nec2 optimizer to adjust C1 (inner loop capacitance) to resonate the inner loop (X=0 is the optimizer criterion).
CM 3. Use 4nec2 optimizer to adjust C2 (inner loop capacitance) for maximum F/B ratio.
CM 4. Use 4nec2 optimizer to adjust RL (loading resistance of outer loop) for maximum F/B ratio.
CM 5. Repeat step 3 and 4 until no further improvement is possible.
CM 
CM When properly adjusted, the model shows a unidirectional null, both in free space, and above ground.
CM 
CM Note that the outer loop must be high impedance (achieved by a loading inductance), and the inner loop must be low impdeance.
CM 
CM The outer loop's high impedance allows it to respond to both E and H fields of the incoming wave, allowing for the unidirectional response.
CM 
CM The model has initially been adjusted for a unidirectional null at 11 MHz in free space.
CE

SY h=1.0
SY r=6.508e-4
SY c1=3.79e-11
SY c2=6.02e-12
SY rl=70.89846

GW	1	7	-0.5	0	h	0.5	0	h	r	'Wire 1, 9 segments, halve wavelength long.
GW	2	7	0.5	0	h	0.5	0	h+1	r
GW	3	7	0.5	0	h+1	-0.5	0	h+1	r
GW	4	7	-0.5	0	h+1	-0.5	0	h	r
GW	10	9	-0.642857	0	h-0.142857	0.642857	0	h-0.142857	r
GW	11	9	0.642857	0	h-0.142857	0.642857	0	h+1.142857	r
GW	12	9	0.642857	0	h+1.142857	-0.642857	0	h+1.142857	r
GW	13	9	-0.642857	0	h+1.142857	-0.642857	0	h-0.142857	r

GE  1

LD	0	2	4	4	0	0	c1
LD	0	11	5	5	0	0	c2
LD	0	13	5	5	rl	10e-6	0
LD	5	0	0	0	9.9E+99

GN  2  0  0  0  13  0.005

EK

EX	6	4	4	0	1	0	0	'Voltage source (1+j0) at wire 1 segment 5.

FR	0	0	0	0	11	0

EN

as you see, I changed "GE" to 1 and changed the "GN" card for "average real ground", with such a config, which should better reflect a real world setup, running the model at the default 11MHz frequency we obtain this



that is, the expected cardioid pattern with pretty good characteristics and a decent gain figure, I'll need to play with the model a bit more, the only "problem" if we want to call it so which I envision is the fact that, while for indoor use the antenna is probably fine, for outdoor use (e.g. placed outside on a balcony or in a yard) it will need some way to remotely control the capacitors and the resistor values to mantain the cardioid pattern over the desired frequency range

Thanks for the idea, will play with the model a bit and check what comes out

[Andrew (grayhat)]

spent a (short) bit of time playing with C1, C2 and RL but probably I'm missing something, I mean, I followed the directions in the comments, that is

1. Adjust C2 (outer loop capacitance) to be very large (e.g. 1000 pF), so that the outer loop is far away from resonance.
2. Use 4nec2 optimizer to adjust C1 (inner loop capacitance) to resonate the inner loop (X=0 is the optimizer criterion).
3. Use 4nec2 optimizer to adjust C2 (inner loop capacitance) for maximum F/B ratio.
4. Use 4nec2 optimizer to adjust RL (loading resistance of outer loop) for maximum F/B ratio.
5. Repeat step 3 and 4 until no further improvement is possible.

but was unable to find a combo giving a cardioid pattern at 1.8MHz, any idea ?

[qrp-gaijin]

Thanks for the warm welcome.

As for the CTL, I've played with it while designing the SULA

Excellent to hear! Did you ever actually get around to building one?

I have only simulated, but not yet actually built, the CTL.

I'm still at a bit of a loss to understand exactly what is happening in the CTL. As I recall from playing around with the simulations, the unidirectional aspect of the CTL is achieved solely by the outer loop; eliminating the inner loop entirely from the model still allows the same unidirectional pattern to be formed.

So I guess the signal could theoretically be extracted from the outer loop. But the CTL doesn't do that. I wonder why. Instead, the CTL uses the combination of two coplanar resonant loops and a "magnetic shadow" region caused by the outer loop to achieve the unidirectional response of the signal that is finally extracted from the inner loop.

I did an excitation test in 4nec2 where I had a small vertical dipole emitting an 11 MHz signal, and I observed the induced magnetic near field of the exterior loop (in this simulation, I omitted the interior loop). There was indeed a "shadow" effect -- when the transmitting dipole was located on the null side of the exterior loop, there was a very low magnetic field excited in the interior region of the exterior loop, but when the transmitting dipole was relocated to the non-null side of the exterior loop, there was a high magnetic field excited in the interior region of the exterior loop.

Then, I guess we could say that the interior loop is just a normal small resonant loop antenna that feeds the receiver. In its extreme near field zone, it responds to the magnetic field, not the electric field. But the outer loop blocks or shadows all incoming magnetic-field energy that comes in from one direction only. Conceptually, it makes sense, and somewhere I even have the 4nec2 files showing the shadow region, but it's still a bit of a mystery for me to really understand, from first principles, how the exterior loop is able to generate this shadow region.

So, to summarize my rambling: can we perhaps say that the SULA uses the same principles as the CTL -- resistive loading of a loop antenna -- to achieve a unidirectional response? The difference is that the SULA extracts the signal directly from the antenna, whereas Villard (creator of the CTL) recognized the interesting fact that the unidirectional loop antenna also creates a unidirectional magnetic shadow region, and thus the CTL extracts the signal by using another resonant loop from within the shadow region.

the only "problem" if we want to call it so which I envision is the fact that, while for indoor use the antenna is probably fine, for outdoor use (e.g. placed outside on a balcony or in a yard) it will need some way to remotely control the capacitors and the resistor values to mantain the cardioid pattern over the desired frequency range

Good point. The SULA is interesting in that it is a broad-banded solution. But the small size of the CTL makes it interesting for portable DXing or portable direction finding, as described in the articles previously linked.

One thing I've wondered about is... what happens if we apply regeneration to the interior loop in the CTL? Regenerative antennas are a very interesting topic because regeneration applied to the antenna actually increases the physical area (volume) of the EM field with which the antenna can interact, bending the EM field lines (somewhat like a lens, I believe). An old NASA paper describes this in detail and proves it with physical measurements: https://ntrs.nasa.gov/api/citations/199 ... 020710.pdf .

A simple line of reasoning could argue that regeneration applied to the inner loop of the CTL would reduce the effective resistance of the inner loop to almost zero, hence increasing the current flowing in the antenna -- current that results from incoming signals. This increased current also results in re-radiation and interaction with a larger portion of the EM field as mentioned above. But because the inner loop is within the magnetic shadow region, incoming signals from the null direction should still be nulled, even with regeneration applied to the inner loop. So, applying regeneration to the CTL might be one way to increase the performance while still maintaining the unidirectional property. Of course, regeneration adds yet another control that needs to be manipulated as the antenna is tuned.

[qrp-gaijin]

spent a (short) bit of time playing with C1, C2 and RL but probably I'm missing something, I mean, I followed the directions in the comments, that is

1. Adjust C2 (outer loop capacitance) to be very large (e.g. 1000 pF), so that the outer loop is far away from resonance.
2. Use 4nec2 optimizer to adjust C1 (inner loop capacitance) to resonate the inner loop (X=0 is the optimizer criterion).
3. Use 4nec2 optimizer to adjust C2 (inner loop capacitance) for maximum F/B ratio.
4. Use 4nec2 optimizer to adjust RL (loading resistance of outer loop) for maximum F/B ratio.
5. Repeat step 3 and 4 until no further improvement is possible.

but was unable to find a combo giving a cardioid pattern at 1.8MHz, any idea ?

First of all, I made a typo -- step 3 is correct in referencing C2, but this is the outer loop capacitance, not the inner loop capacitance as mentioned.

Secondly, at 1.8 MHz, maybe you need to significantly increase the both (1) the load inductance of the outer loop and (2) the intrinsic inductance of the inner loop. The CTL article says that for (2) you should use additional wire turns on the inner loop to increase the inductance, but that is a very bad idea in 4nec2 (very small multi-turn loops are not simulated well). So you might need to add some loading inductance onto the inner loop as well. But still, the outer loop should be higher impedance (with higher inductance) than the inner loop. I haven't tried this myself, so these are just guesses.

[Andrew (grayhat)]

Thanks for the warm welcome.

Well, I've been "following" you for some time and it's a pleasure to see you here !

Excellent to hear! Did you ever actually get around to building one?
I have only simulated, but not yet actually built, the CTL.

No, I just played with some NEC models but then I gave up, since I was trying to find a simple design which could be easy for anyone to build and, at the same time, offer good performances, so after some fiddling I skipped it and went on with other ideas

I'm still at a bit of a loss to understand exactly what is happening in the CTL. As I recall from playing around with the simulations, the unidirectional aspect of the CTL is achieved solely by the outer loop; eliminating the inner loop entirely from the model still allows the same unidirectional pattern to be formed.

Didn't put too much thought on it, so bear with me, what I'm writing are just some "coarse ideas" or, if you prefer, I'm just "thinking aloud"

My take is that we may see the CTL as an evolution of the classic tuned loop using an internal pickup loop for feeding, the difference in this case is that the pickup has a size comparable to the one of the main loop, and this is probably causing the effect we're observing on the antenna lobe, I haven't much time at the moment, but I want to try experimenting a bit with the CTL model varying the size and orientation of the inner loop to check the effect; that said, I think that the principle on which the SULA and the CTL work are similar, check this design, now the idea is almost the same of the "loop and whip" antenna used for direction finding, in that case the loop has a bidirectional (classic "8" shaped) pattern so, once we find the loop orientation we'll still need to determine "at which side" the transmitter is, to do so, we add the whip to the circuit; the linked antenna (and the SULA) work with the same principle, think at the SULA as a pair of verticals (the two halves splitted by the resistor) and you'll see what I'm saying, now, I think that the CTL is working in a similar way, although it uses inductive coupling, then again maybe I'm wrong (as I wrote I'm just "thinking aloud", so please bear with me)

So, to summarize my rambling: can we perhaps say that the SULA uses the same principles as the CTL -- resistive loading of a loop antenna -- to achieve a unidirectional response? The difference is that the SULA extracts the signal directly from the antenna, whereas Villard (creator of the CTL) recognized the interesting fact that the unidirectional loop antenna also creates a unidirectional magnetic shadow region, and thus the CTL extracts the signal by using another resonant loop from within the shadow region.

exactly so, at least if my idea is correct

One thing I've wondered about is... what happens if we apply regeneration to the interior loop in the CTL? Regenerative antennas are a very interesting topic because regeneration applied to the antenna actually increases the physical area (volume) of the EM field with which the antenna can interact, bending the EM field lines (somewhat like a lens, I believe). An old NASA paper describes this in detail and proves it with physical measurements: https://ntrs.nasa.gov/api/citations/199 ... 020710.pdf .

Regenerating the loop may be interesting but I think isn't an easy thing, see, if we're using the loop as the "L" part of the regen circuit, we'll need to tune both the antenna and the regen and that will be quite "touchy" and probably the frequency range will be limited (assuming the loop size remains the same), so I'm not sure it may be practical

[Andrew (grayhat)]

Secondly, at 1.8 MHz, maybe you need to significantly increase the both (1) the load inductance of the outer loop and (2) the intrinsic inductance of the inner loop. The CTL article says that for (2) you should use additional wire turns on the inner loop to increase the inductance, but that is a very bad idea in 4nec2 (very small multi-turn loops are not simulated well). So you might need to add some loading inductance onto the inner loop as well. But still, the outer loop should be higher impedance (with higher inductance) than the inner loop. I haven't tried this myself, so these are just guesses.

Hmmm... so to get down (or up) in frequency we'd have to change the loop size... not exactly easy/practical imHo

[Andrew (grayhat)]

as for the directional lobe, check out the currents and phase of the CTL, that should tell us something, I believe

ctl_curr.png (118.11 KiB) Viewed 3303 times

[qrp-gaijin]

Hmmm... so to get down (or up) in frequency we'd have to change the loop size... not exactly easy/practical imHo

It's not necessary to change the loop size. You simply need to increase the loading of the outer loop by increasing the lumped inductors, and increase the loading of the inner loop by adding additional turns in series with the inner loop. Because it is difficult to simulate additional turns in 4nec2 (very small parallel loops cause numerical inaccuracies), you can also use (in 4nec2) lumped inductive loads to increase the inductance of the inner loop.

The outer loop should be resonant at the desired frequency with a low capacitance of maybe 10 pF (meaning the loop is high-impedance); the inner loop should be resonant at the desired frequency with a high capacitance of maybe 300 pF (meaning the loop is low-impedance).

Have a look at the updated 4nec2 file. The outer loop is the same size as before, but now has an additional 400 uH of lumped inductance in series with the inductor. In the simulation the inductor is lossless. In reality you would probably need to use a ferrite core to achieve the required inductance with low loss. The inner loop is also the same size as before, but has an additional 25 uH of lumped inductance. Again, in a real build it is recommended to increase the inductance of the inner loop not by using a lumped inductance, but instead by using additional turns.

Also note that in the simulation, the wires are lossless conductors, so the gain figures are unrealistically high. Add some wire resistance to get more realistic gain figures.

Finally, due to the lower frequency, the segment size is too small, which causes a 4nec2 warning, but the results still seem plausible.

ctl-topband.jpg (59.03 KiB) Viewed 3295 times

Code: Select all

CM Model of Coplanar Twin Loop antenna at 1.8 MHz.
CM Created by qrp-gaijin@yahoo.com (http://qrp-gaijin.blogspot.com).
CM 
CM Adjust model as follows:
CM 
CM 1. Adjust C2 (outer loop capacitance) to be very large (e.g. 1000 pF), so that the outer loop is far away from resonance.
CM 2. Use 4nec2 optimizer to adjust C1 (inner loop capacitance) to resonate the inner loop (X=0 is the optimizer criterion).
CM 3. Use 4nec2 optimizer to adjust C2 (inner loop capacitance) for maximum F/B ratio.
CM 4. Use 4nec2 optimizer to adjust RL (loading resistance of outer loop) for maximum F/B ratio.
CM 5. Repeat step 3 and 4 until no further improvement is possible.
CM 
CM For use at a different frequency f:
CM 
CM 1. Estimate the loading inductance required to resonate the outer loop at frequency f with C2 at 10 pF. This ensures the outer loop is high-impedance. Add this load to the outer loop, opposite of the capacitor.
CM 
CM 2. Estimate the loading inductance required to resonate the inner loop at frequency f with C1 at 300 pF. This ensures the inner loop is low-impedance. Add this load to the inner loop, opposite of the capacitor.
CM 
CM 3. Detune C1. Place the feedpoint on the outer loop, opposite of C2. Optimize C2 for resonance such that X=0. If this is not possible, then the loading inductance for the outer loop is wrong and needs to be adjusted.
CM 
CM 4. Detune C2. Place the feedpoint on the inner loop, opposite of C1. Optimize C1 for resonance such that X=0. If this is not possible, then the loading inductance for the inner loop is wrong and needs to be adjusted.
CM 
CM 5. Set C2 back to its resonant value. Optimize C2 for best F/B ratio.
CM 
CM 6. Optimize loading resistance RL for best F/B ratio.
CM 
CM When properly adjusted, the model shows a unidirectional null, both in free space, and above ground.
CM 
CM Note that the outer loop must be high impedance (achieved by a loading inductance), and the inner loop must be low impdeance.
CM 
CM The model has initially been adjusted for a unidirectional null at 1.8 MHz in free space.
CE
SY h=1.0
SY r=6.508e-4
SY c1=2.51e-10
SY c2pf=8.441943
SY c2=c2pf*1e-12
SY rl=205.253
GW	1	7	-0.5	0	h	0.5	0	h	r	'Wire 1, 9 segments, halve wavelength long.
GW	2	7	0.5	0	h	0.5	0	h+1	r
GW	3	7	0.5	0	h+1	-0.5	0	h+1	r
GW	4	7	-0.5	0	h+1	-0.5	0	h	r
GW	10	9	-0.642857	0	h-0.142857	0.642857	0	h-0.142857	r
GW	11	9	0.642857	0	h-0.142857	0.642857	0	h+1.142857	r
GW	12	9	0.642857	0	h+1.142857	-0.642857	0	h+1.142857	r
GW	13	9	-0.642857	0	h+1.142857	-0.642857	0	h-0.142857	r
GE	0
LD	0	2	4	4	0	0	c1
LD	0	11	5	5	0	0	c2
LD	0	13	5	5	rl	400e-6	0
LD	5	0	0	0	9.9E+99
LD	0	4	4	4	0	25e-6	0
GN	-1
EK
EX	6	4	4	0	1	0	0	'Voltage source (1+j0) at wire 1 segment 5.
FR	0	0	0	0	1.8	0
EN

[Andrew (grayhat)]

Thank you, ran the model and it behaves as expected, I just changed the "GN" card to put the antenna over real ground and not in free space ("GN 2 0 0 0 13 0.005") the results are pretty good, but the impedance is around 2 Ohm which means that, to couple the antenna to a receiver we'll need to use a preamp/buffer designed to accept such low impedance and offering an output impedance around 50 (or 75) Ohms, and this adds another degree on complexity, not that it's bad, but sincerely I prefer simpler stuff in the "SULA" case, the same resistor which gives the cardioid pattern also serves to give a wideband impedance around 450 Ohms, which just using a 9:1 can be easily matched to a coax feeder or to a standard LNA; a thing which would be interesting to test, either with the SULA or with the CTL is a tuned stage like the one described here

http://www.arrl.org/files/file/QEX_Next ... Steber.pdf

such a unit would allow to "tune in" the desired frequency and coupled with the unidirectional characteristics of the loops should greatly improve them

just as a note, if you want to try a circular loop design, the model below is a vanilla circular loop model

Code: Select all

CM --------------------------------------------------------
CM File: loop_antenna.nec
CM --------------------------------------------------------
CE

' symbols
SY frq=7.100              ' test frequency
SY ray=0.00250            ' wire radius
SY rad=0.78               ' loop radius
SY hgh=3                  ' height from ground
SY seg=3                  ' segmentation

' circular loop geometry
GW   1 seg 0 rad*sin(0)   hgh+rad*cos(0)   0 rad*sin(10)  hgh+rad*cos(10)  ray
GW   2 seg 0 rad*sin(10)  hgh+rad*cos(10)  0 rad*sin(20)  hgh+rad*cos(20)  ray
GW   3 seg 0 rad*sin(20)  hgh+rad*cos(20)  0 rad*sin(30)  hgh+rad*cos(30)  ray
GW   4 seg 0 rad*sin(30)  hgh+rad*cos(30)  0 rad*sin(40)  hgh+rad*cos(40)  ray
GW   5 seg 0 rad*sin(40)  hgh+rad*cos(40)  0 rad*sin(50)  hgh+rad*cos(50)  ray
GW   6 seg 0 rad*sin(50)  hgh+rad*cos(50)  0 rad*sin(60)  hgh+rad*cos(60)  ray
GW   7 seg 0 rad*sin(60)  hgh+rad*cos(60)  0 rad*sin(70)  hgh+rad*cos(70)  ray
GW   8 seg 0 rad*sin(70)  hgh+rad*cos(70)  0 rad*sin(80)  hgh+rad*cos(80)  ray
GW   9 seg 0 rad*sin(80)  hgh+rad*cos(80)  0 rad*sin(90)  hgh+rad*cos(90)  ray
GW  10 seg 0 rad*sin(90)  hgh+rad*cos(90)  0 rad*sin(100) hgh+rad*cos(100) ray
GW  11 seg 0 rad*sin(100) hgh+rad*cos(100) 0 rad*sin(110) hgh+rad*cos(110) ray
GW  12 seg 0 rad*sin(110) hgh+rad*cos(110) 0 rad*sin(120) hgh+rad*cos(120) ray
GW  13 seg 0 rad*sin(120) hgh+rad*cos(120) 0 rad*sin(130) hgh+rad*cos(130) ray
GW  14 seg 0 rad*sin(130) hgh+rad*cos(130) 0 rad*sin(140) hgh+rad*cos(140) ray
GW  15 seg 0 rad*sin(140) hgh+rad*cos(140) 0 rad*sin(150) hgh+rad*cos(150) ray
GW  16 seg 0 rad*sin(150) hgh+rad*cos(150) 0 rad*sin(160) hgh+rad*cos(160) ray
GW  17 seg 0 rad*sin(160) hgh+rad*cos(160) 0 rad*sin(170) hgh+rad*cos(170) ray
GW  18 seg 0 rad*sin(170) hgh+rad*cos(170) 0 rad*sin(180) hgh+rad*cos(180) ray
GW  19 seg 0 rad*sin(180) hgh+rad*cos(180) 0 rad*sin(190) hgh+rad*cos(190) ray
GW  20 seg 0 rad*sin(190) hgh+rad*cos(190) 0 rad*sin(200) hgh+rad*cos(200) ray
GW  21 seg 0 rad*sin(200) hgh+rad*cos(200) 0 rad*sin(210) hgh+rad*cos(210) ray
GW  22 seg 0 rad*sin(210) hgh+rad*cos(210) 0 rad*sin(220) hgh+rad*cos(220) ray
GW  23 seg 0 rad*sin(220) hgh+rad*cos(220) 0 rad*sin(230) hgh+rad*cos(230) ray
GW  24 seg 0 rad*sin(230) hgh+rad*cos(230) 0 rad*sin(240) hgh+rad*cos(240) ray
GW  25 seg 0 rad*sin(240) hgh+rad*cos(240) 0 rad*sin(250) hgh+rad*cos(250) ray
GW  26 seg 0 rad*sin(250) hgh+rad*cos(250) 0 rad*sin(260) hgh+rad*cos(260) ray
GW  27 seg 0 rad*sin(260) hgh+rad*cos(260) 0 rad*sin(270) hgh+rad*cos(270) ray
GW  28 seg 0 rad*sin(270) hgh+rad*cos(270) 0 rad*sin(280) hgh+rad*cos(280) ray
GW  29 seg 0 rad*sin(280) hgh+rad*cos(280) 0 rad*sin(290) hgh+rad*cos(290) ray
GW  30 seg 0 rad*sin(290) hgh+rad*cos(290) 0 rad*sin(300) hgh+rad*cos(300) ray
GW  31 seg 0 rad*sin(300) hgh+rad*cos(300) 0 rad*sin(310) hgh+rad*cos(310) ray
GW  32 seg 0 rad*sin(310) hgh+rad*cos(310) 0 rad*sin(320) hgh+rad*cos(320) ray
GW  33 seg 0 rad*sin(320) hgh+rad*cos(320) 0 rad*sin(330) hgh+rad*cos(330) ray
GW  34 seg 0 rad*sin(330) hgh+rad*cos(330) 0 rad*sin(340) hgh+rad*cos(340) ray
GW  35 seg 0 rad*sin(340) hgh+rad*cos(340) 0 rad*sin(350) hgh+rad*cos(350) ray
GW  36 seg 0 rad*sin(350) hgh+rad*cos(350) 0 rad*sin(360) hgh+rad*cos(360) ray

' end of geometry
GE  1

' ground parameters
GN  2  0  0  0  13  0.005

' wire loading and resistor
LD  7  0  0  0  2.1  ray        ' insulation
LD  5  0  0  0  58000000        ' copper

' enable extended kernel
EK 

' feedpoint
EX 0 19  1  0  1.0 0.0

' frequency
FR 0 1 0 0 frq 1

' end of model
EN
CM --------------------------------------------------------
CM File: loop_antenna.nec
CM --------------------------------------------------------
CE

' symbols
SY frq=1.800              ' test frequency
SY ray=0.00250            ' wire radius
SY rad=0.78               ' loop radius
SY hgh=3                  ' height from ground
SY seg=3                  ' segmentation

' circular loop geometry
GW   1 seg 0 rad*sin(0)   hgh+rad*cos(0)   0 rad*sin(10)  hgh+rad*cos(10)  ray
GW   2 seg 0 rad*sin(10)  hgh+rad*cos(10)  0 rad*sin(20)  hgh+rad*cos(20)  ray
GW   3 seg 0 rad*sin(20)  hgh+rad*cos(20)  0 rad*sin(30)  hgh+rad*cos(30)  ray
GW   4 seg 0 rad*sin(30)  hgh+rad*cos(30)  0 rad*sin(40)  hgh+rad*cos(40)  ray
GW   5 seg 0 rad*sin(40)  hgh+rad*cos(40)  0 rad*sin(50)  hgh+rad*cos(50)  ray
GW   6 seg 0 rad*sin(50)  hgh+rad*cos(50)  0 rad*sin(60)  hgh+rad*cos(60)  ray
GW   7 seg 0 rad*sin(60)  hgh+rad*cos(60)  0 rad*sin(70)  hgh+rad*cos(70)  ray
GW   8 seg 0 rad*sin(70)  hgh+rad*cos(70)  0 rad*sin(80)  hgh+rad*cos(80)  ray
GW   9 seg 0 rad*sin(80)  hgh+rad*cos(80)  0 rad*sin(90)  hgh+rad*cos(90)  ray
GW  10 seg 0 rad*sin(90)  hgh+rad*cos(90)  0 rad*sin(100) hgh+rad*cos(100) ray
GW  11 seg 0 rad*sin(100) hgh+rad*cos(100) 0 rad*sin(110) hgh+rad*cos(110) ray
GW  12 seg 0 rad*sin(110) hgh+rad*cos(110) 0 rad*sin(120) hgh+rad*cos(120) ray
GW  13 seg 0 rad*sin(120) hgh+rad*cos(120) 0 rad*sin(130) hgh+rad*cos(130) ray
GW  14 seg 0 rad*sin(130) hgh+rad*cos(130) 0 rad*sin(140) hgh+rad*cos(140) ray
GW  15 seg 0 rad*sin(140) hgh+rad*cos(140) 0 rad*sin(150) hgh+rad*cos(150) ray
GW  16 seg 0 rad*sin(150) hgh+rad*cos(150) 0 rad*sin(160) hgh+rad*cos(160) ray
GW  17 seg 0 rad*sin(160) hgh+rad*cos(160) 0 rad*sin(170) hgh+rad*cos(170) ray
GW  18 seg 0 rad*sin(170) hgh+rad*cos(170) 0 rad*sin(180) hgh+rad*cos(180) ray
GW  19 seg 0 rad*sin(180) hgh+rad*cos(180) 0 rad*sin(190) hgh+rad*cos(190) ray
GW  20 seg 0 rad*sin(190) hgh+rad*cos(190) 0 rad*sin(200) hgh+rad*cos(200) ray
GW  21 seg 0 rad*sin(200) hgh+rad*cos(200) 0 rad*sin(210) hgh+rad*cos(210) ray
GW  22 seg 0 rad*sin(210) hgh+rad*cos(210) 0 rad*sin(220) hgh+rad*cos(220) ray
GW  23 seg 0 rad*sin(220) hgh+rad*cos(220) 0 rad*sin(230) hgh+rad*cos(230) ray
GW  24 seg 0 rad*sin(230) hgh+rad*cos(230) 0 rad*sin(240) hgh+rad*cos(240) ray
GW  25 seg 0 rad*sin(240) hgh+rad*cos(240) 0 rad*sin(250) hgh+rad*cos(250) ray
GW  26 seg 0 rad*sin(250) hgh+rad*cos(250) 0 rad*sin(260) hgh+rad*cos(260) ray
GW  27 seg 0 rad*sin(260) hgh+rad*cos(260) 0 rad*sin(270) hgh+rad*cos(270) ray
GW  28 seg 0 rad*sin(270) hgh+rad*cos(270) 0 rad*sin(280) hgh+rad*cos(280) ray
GW  29 seg 0 rad*sin(280) hgh+rad*cos(280) 0 rad*sin(290) hgh+rad*cos(290) ray
GW  30 seg 0 rad*sin(290) hgh+rad*cos(290) 0 rad*sin(300) hgh+rad*cos(300) ray
GW  31 seg 0 rad*sin(300) hgh+rad*cos(300) 0 rad*sin(310) hgh+rad*cos(310) ray
GW  32 seg 0 rad*sin(310) hgh+rad*cos(310) 0 rad*sin(320) hgh+rad*cos(320) ray
GW  33 seg 0 rad*sin(320) hgh+rad*cos(320) 0 rad*sin(330) hgh+rad*cos(330) ray
GW  34 seg 0 rad*sin(330) hgh+rad*cos(330) 0 rad*sin(340) hgh+rad*cos(340) ray
GW  35 seg 0 rad*sin(340) hgh+rad*cos(340) 0 rad*sin(350) hgh+rad*cos(350) ray
GW  36 seg 0 rad*sin(350) hgh+rad*cos(350) 0 rad*sin(360) hgh+rad*cos(360) ray

' end of geometry
GE  1

' ground parameters
GN  2  0  0  0  13  0.005

' wire loading and resistor
LD  7  0  0  0  2.1  ray        ' insulation
LD  5  0  0  0  58000000        ' copper

' enable extended kernel
EK 

' feedpoint
EX 0 19  1  0  1.0 0.0

' frequency
FR 0 1 0 0 frq 1

' end of model
EN

which could be easily modified (e.g. adding a second, smaller size loop inside); what else... oh, yes, please give a spin to this modification to your CTL model

Code: Select all

CM Model of Coplanar Twin Loop antenna at 1.8 MHz.
CM Created by qrp-gaijin@yahoo.com (http://qrp-gaijin.blogspot.com).
CM 
CM Adjust model as follows:
CM 
CM 1. Adjust C2 (outer loop capacitance) to be very large (e.g. 1000 pF), so that the outer loop is far away from resonance.
CM 2. Use 4nec2 optimizer to adjust C1 (inner loop capacitance) to resonate the inner loop (X=0 is the optimizer criterion).
CM 3. Use 4nec2 optimizer to adjust C2 (inner loop capacitance) for maximum F/B ratio.
CM 4. Use 4nec2 optimizer to adjust RL (loading resistance of outer loop) for maximum F/B ratio.
CM 5. Repeat step 3 and 4 until no further improvement is possible.
CM 
CM For use at a different frequency f:
CM 
CM 1. Estimate the loading inductance required to resonate the outer loop at frequency f with C2 at 10 pF. This ensures the outer loop is high-impedance. Add this load to the outer loop, opposite of the capacitor.
CM 
CM 2. Estimate the loading inductance required to resonate the inner loop at frequency f with C1 at 300 pF. This ensures the inner loop is low-impedance. Add this load to the inner loop, opposite of the capacitor.
CM 
CM 3. Detune C1. Place the feedpoint on the outer loop, opposite of C2. Optimize C2 for resonance such that X=0. If this is not possible, then the loading inductance for the outer loop is wrong and needs to be adjusted.
CM 
CM 4. Detune C2. Place the feedpoint on the inner loop, opposite of C1. Optimize C1 for resonance such that X=0. If this is not possible, then the loading inductance for the inner loop is wrong and needs to be adjusted.
CM 
CM 5. Set C2 back to its resonant value. Optimize C2 for best F/B ratio.
CM 
CM 6. Optimize loading resistance RL for best F/B ratio.
CM 
CM When properly adjusted, the model shows a unidirectional null, both in free space, and above ground.
CM 
CM Note that the outer loop must be high impedance (achieved by a loading inductance), and the inner loop must be low impdeance.
CM 
CM The model has initially been adjusted for a unidirectional null at 1.8 MHz in free space.
CE

SY h=1.0
SY r=6.508e-4
SY c1=2.51e-10
SY c2pf=8.441943
SY c2=c2pf*1e-12
SY rl=205.253

SY coax=0.00250
SY cspc=(coax*3)


GW	1	7	-0.5	0	h	0.5	0	h	r	'Wire 1, 9 segments, halve wavelength long.
GW	2	7	0.5	0	h	0.5	0	h+1	r
GW	3	7	0.5	0	h+1	-0.5	0	h+1	r
GW	4	7	-0.5	0	h+1	-0.5	0	h	r
GW	10	9	-0.642857	0	h-0.142857	0.642857	0	h-0.142857	r
GW	11	9	0.642857	0	h-0.142857	0.642857	0	h+1.142857	r
GW	12	9	0.642857	0	h+1.142857	-0.642857	0	h+1.142857	r
GW	13	9	-0.642857	0	h+1.142857	-0.642857	0	h-0.142857	r

GW  14 9 0 cspc 0 0 cspc h+0.5 coax
GW  15 9 0 cspc h+0.5 -0.5 cspc h+0.5 coax

GE	1

LD	0	2	4	4	0	0	c1
LD	0	11	5	5	0	0	c2
LD	0	13	5	5	rl	400e-6	0
LD	5	0	0	0	9.9E+99
LD	0	4	4	4	0	25e-6	0

GN  2  0  0  0  13  0.005

EK

EX	6	4	4	0	1	0	0	'Voltage source (1+j0) at wire 1 segment 5.

FR	0	0	0	0	1.8	0

EN

i added a couple wires (tags 14 and 15) and two symbols, I think you'll understand what they're for; give it a try and tell me what you think !