The small SWL beam (1...30 MHz)

[Andrew (grayhat)]

The image below (full size HERE) contains all the infos needed to build a small loop antenna, fitting even in restricted spaces, which will offer an "unidirectional" lobe like the one from a "beam", the lobe is mantained from a few KHz up to above 30 MHz and allows to "null" out signals from the opposite direction



the diamond shaped loop has a side of 76 cm (2.5ft), the 530 Ohm resistor placed at one of the corners has a dual duty, that is forming the directional lobe and offering a good match, placing a 9:1 BalUn at the bottom corner (feedpoint) the SWR up to 30 MHz will be under 2:1, the antenna gain isn't an issue, since we're seeking for S/N not for gain, but willing some boost, one may add a small preamp, the antenna will work even if placed very low on ground, although simulations show that the optimal height is between 2 meters and 5 meters (6 to 16 ft, 3 meters - about 10ft - in the pic above), at such height the antenna offers a lobe which allows good reception of both NVIS and DX (low angle) signals coming from the main lobe direction (that is the direction opposite from the resistor one), further raising the antenna will lower the lobe and the antenna wil become a DX only one, loosing sensitivity to NVIS signals

To build the antenna it will suffice to form a cross shaped structure using whatever non conductive material (wood, PVC pipes or whatever you have at hand, as long as it isn't conductive) and then use some insulated wire to form the diamond loop, add the resistor at a corner, a 9:1 BalUn at the bottom and the "SWL beam" will be ready for use

For the ones interested in running some simulation, the NEC model file is below

Code: Select all

CM File: DirLoop.nec
CM Small Receiving Loop antenna
CM with unidirectional pattern
CE

' symbols definition
SY freq=7.100               ' test frequenct
SY hght=3                   ' height of base corner from ground
SY side=0.762               ' length of one side
SY diag=(sqr(2)*(side/2))   ' half diagonal
SY wire=0.00250             ' wire radius
SY vres=530                 ' loading resistor value
SY segs=13                  ' number of segment in wires
SY wres=4                   ' wire hosting the resistor
SY sres=1                   ' segment hosting the resistor

' wires geometry
'  ID seg    x0     y0   z0         x1     y1     z1          wire rad
GW  1 segs     0     0  hght      -diag     0   hght+diag       wire
GW  2 segs     0     0  hght       diag     0   hght+diag       wire
GW  3 segs -diag     0  hght+diag     0     0   hght+(diag*2)   wire
GW  4 segs  diag     0  hght+diag     0     0   hght+(diag*2)   wire

' ground parameters
GE  1
GN  2  0  0  0  13  0.005

' wires loading
LD  5  0  0  0  58000000
LD  1 wres  sres sres vres 0

' enable extended kernel for calc
EK

' feedpoint
EX  0  1 1 0 1 0 0

' initial test frequency
FR  0  1  0  0  freq  0

' end
EN

Note: nothing forbids increasing the size of the loop (if you have room), I simulated sizes up to 2m side, but in my opinion it's better keeping it relatively small, at any rate, the model is above, and changing height or side and running a simulation will be easy

[13dka]

Hi Andrew,

This is super exiting! I meant to comment on your cardioid loop thread but my reply would've created a new page and pushed your great post to the previous one. What I meant to comment on that thread is that the "cardioid loop" concept looks particularly promising because it could mitigate or correct the (for my purposes) most adverse property of SMLs - their "independency to height" (sales euphemism for "unfixable cloud warmer").

Now here we have something quite similar, minus the need for a ground rod. How did you come up with that? Is there a downside? This seems to be way easier also because this can be done with the NooElec 9:1?

I hope I find some time to run this through 4NEC2 tomorrow, I'm really curious what the ground conductivity vs. height requirements behavior is. Due to the relatively small size this would also lend itself to stacking 2 of them on a simple fiberglass pole for additional gain and some phase magic takeoff angle control? I have to admit that I don't have any PCB trash in the house to harvest any parts so I need to buy 500 resistors off Amazon before I can start playing with one of them.

P.S. This needs to go on the blog site!

[Andrew (grayhat)]

Hi Andrew,

Hi there, Ollie !

This is super exiting! I meant to comment on your cardioid loop thread but my reply would've created a new page and pushed your great post to the previous one. What I meant to comment on that thread is that the "cardioid loop" concept looks particularly promising because it could mitigate or correct the (for my purposes) most adverse property of SMLs - their "independency to height" (sales euphemism for "unfixable cloud warmer").

Well, at first I was thinking to post this in the "cardioid loop" thread, but then I decided to open a new one, since this antenna overcomes some of the "cardioid loop" limitations, basically it doesn't need a ground wire and, to some degree, it is more independent from height and ground type, while the "cardioid loop" pattern depends heavily from terrain and its height is almost "fixed" due to the need for a ground wire, and then the "cardioid loop" is almost impossible to install on (say) a terrace/balcony that is w/o a good ground connection

Now here we have something quite similar, minus the need for a ground rod. How did you come up with that? Is there a downside? This seems to be way easier also because this can be done with the NooElec 9:1?

I think you remember my little "balcony SWL antenna" experiment, Thomas kindly posted some of my tests (endfed "random", "mini whip", "linear loaded dipole"), and I kept experimenting, my idea was to try some antennas (either commercial or homemade) which could help an apartment dweller putting up a decent listening post; now, the "cardioid loop", needs to be relatively low on ground and needs a good ground connection, and both those factors make it unsuitable for a balcony antenna, plus, even if installed in a yard, it will still be very dependent on the nature of the terrain, so... I kept thinking and after a while I had an idea; if you look at a T2FD antenna placed horizontally



you'll see that the main lobe radiates toward the opposite direction from that where the loading resistor is placed, so I thought "why couldn't this work with a small vertical loop ?" at that point, since I already had a NEC model for the KK5JY "SRL" antenna I edited the model adding a resistor, and after experimenting a bit with height and resistor positioning/value I found a good combo, and the resulting antenna is the one we're discussing

I hope I find some time to run this through 4NEC2 tomorrow, I'm really curious what the ground conductivity vs. height requirements behavior is. Due to the relatively small size this would also lend itself to stacking 2 of them on a simple fiberglass pole for additional gain and some phase magic takeoff angle control? I have to admit that I don't have any PCB trash in the house to harvest any parts so I need to buy 500 resistors off Amazon before I can start playing with one of them.

Truth is... that due to some (hopefully minor) health issues I won't be able to build the loop anytime soon (probably next fall) so at this point I just simulated the antenna using NEC, so... if you want to go on and build a prototype you'll probably be faster than me; as for building, either the commercial NooElec 9:1 Balun V2 or a homemade balun built using a binocular ferrite, like the one used for the SRL/LoG or, with a different winding ratio, the one used for the NCPL, will work, as for the resistor, two of these with a 270 Ohm values, connected in series will fit (and those are 100pcs)

P.S. This needs to go on the blog site!

Well, I'd prefer testing a real thing before going for the blog, but as I wrote, it will take quite some time, if you want to go for a build, we may run a "four hands" experiment, and then ask Thomas if he'd be interested in posting something on the blog...

[Andrew (grayhat)]

as for height effect on this antenna this first image shows how the pattern changes when we change the antenna height



as you can see, raising the antenna we lower the radiation lobe so cutting off more and more the sensitivity to high angle signals while increasing the one to low angle ones, but the directional pattern remains unchanged with height, and although when placed at 9m we start seeing the typical cardioid pattern, the pattern at lower heights offers pretty good directionality for low angle signals; to better compare the patterns the image below shows two 3D renderings of the antenna which may help better understanding the pattern shape and the effect of height



to the left we can see the pattern at 3m height, as you can see the higher gain area of the beam has a vertical angle ranging from low angle to almost vertical, to the right we can see what happens raising the antenna at 9m height, in such a case, the lobe is almost totally concentrated at low/very-low angles while NVIS signals are almost totally "killed"; both configurations are good though, the first one (low) will allow using the antenna for both DX and "domestic" listening while still allowing to null off signals from the opposite direction (to the one at which the antenna is aimed), while the second one will be fine for DX since it will be almost insensitive to local NVIS signals

As a note, this small antenna could also be useful for Hams; putting up a directional beam, say a Yagi, for the "top bands" (160, 80, 40) isn't exactly practical, at least if you don't own a quite large amount of terrain, but using this small directional loop, we'll still be able to enjoy the directionality benefit, at least in RX, so using a different antenna (say a vertical) for transmission and this small loop for reception may allow to "pull out" that faint DX signal we have been seeking for

[13dka]

Thanks!

Truth is... that due to some (hopefully minor) health issues I won't be able to build the loop anytime soon (probably next fall) so at this point I just simulated the antenna using NEC, so...

I hope this is nothing serious and that you'll get better soon!


I'm currently running this for higher frequencies over the assumed "dike" ground conditions and the results were initially ...umm... not intuitive:

1. Height flattens the lobe but above 14 MHz, it looks like more than 3-5m can have the opposite effect:

3m:

dirloop_dike_21mhz_3m.jpg (56.52 KiB) Viewed 4119 times

5m:

dirloop_dike_21mhz_5m.jpg (57.82 KiB) Viewed 4119 times

9m:

dirloop_dike_21mhz_9m.jpg (57.22 KiB) Viewed 4119 times

Adding height and changing good ground to saltwater tends to split the main lobe as per usual, but in this particular case not in the horizontal pane (it always remains some kind of cardioid pattern) but the vertical/elevation pattern. Back to 3m there's less splitting, if we compare poor ground conductivity to seawater, the difference in losses (5dB) seems to be negligible, and the takeoff angle difference between the ground extremes is only 15 degrees:

Poor ground:

dirloop_poor_21mhz_3m.jpg (58.05 KiB) Viewed 4119 times

Ocean:

dirloop_seawater_21mhz_3m.jpg (56.12 KiB) Viewed 4119 times

Over salt water,the takeoff angle is only 10°with a height of only 3m, 20dB gain from a preamp is likely mandatory but that would be quite a treat for DX. Curious how this works out in practice but that all seems to be too good to be true - maybe it tends to react pretty unpredictable when the balcony is higher than the 2nd floor or something but it looks like this could do away with everything making DX antennas a hassle, space requirements, height requirements, ground requirements, the only downside being that people may want to use it with some rotor, except if you have a specific target, or something you want in the null zone (some QRM source, Radio Romania International, some ham you don't like...).

[Andrew (grayhat)]

I hope this is nothing serious and that you'll get better soon!

I hope that too, and hope it will be over in a while, but let's move on

1. Height flattens the lobe but above 14 MHz, it looks like more than 3-5m can have the opposite effect:

You're further confirming my observations, to mantain the lobe consistent throughout the whole frequency range, the optimal height for the loop ranges between 2 and 5 meters (more or less), that's why I choose 3 meters for the model, that way the lobe is quite predictable even if changing terrain

Adding height and changing good ground to saltwater tends to split the main lobe as per usual, but in this particular case not in the horizontal pane (it always remains some kind of cardioid pattern) but the vertical/elevation pattern. Back to 3m there's less splitting, if we compare poor ground conductivity to seawater, the difference in losses (5dB) seems to be negligible, and the takeoff angle difference between the ground extremes is only 15 degrees

Exactly so, and this, again, confirms what I noticed; while the other "cardioid loop" pattern and efficiency is heavily dependent from frequency and terrain (height remaining the same), the loop in discussion seems to be pretty unaffected by the type of terrain, sure, it will somewhat alter the lobe, but to a lesser extent than with other types of antennas, and this in my opinion, is a big plus for such a simple/cheap antenna

Over salt water,the takeoff angle is only 10°with a height of only 3m, 20dB gain from a preamp is likely mandatory but that would be quite a treat for DX. Curious how this works out in practice but that all seems to be too good to be true - maybe it tends to react pretty unpredictable when the balcony is higher than the 2nd floor or something but it looks like this could do away with everything making DX antennas a hassle, space requirements, height requirements, ground requirements, the only downside being that people may want to use it with some rotor, except if you have a specific target, or something you want in the null zone (some QRM source, Radio Romania International, some ham you don't like...).

Yes, placing the antenna at an "optimal" height, the terrain will alter the takeoff angle a bit, and over saltwater it will offer a very low angle and a pretty narrow lobe which will help with DX signals, regarding the balcony/terrace, those usually have a concrete floor reinfornced with metal bars (or grid), so don't be fooled by the height above terrain, since the balcony floor will offer a "ground plane"

As for the preamplifier, if cranking up the receiver preamp doesn't suffice, then a small preamp at the feedpoint may be needed to boost the signals, but that would be all, the remainder will be done by the antenna directional lobe the only problem I see is shown in the two pics below

dirloop03.png (50.55 KiB) Viewed 4110 times

dirloop04.jpg (186.4 KiB) Viewed 4110 times

that is, between (about) 79 and 130 MHz the antenna shows positive gain and the lobes are sideways, this means that strong FM signals may be an issue; if that's the case then an FM "bandstop" (notch) filter (e.g. the "Distill FM" or the like) should be placed between the 9:1 BalUn and the preamp (or the coax if an external preamp isn't used AND if the receiver doesn't have decent passband filters at its input), on the other hand, w/o the filter and due to the low angle, the antenna may also be used for FM band DX

[Andrew (grayhat)]

update:

ran some tests and, slightly changing the length of the side it's possible to relieve the FM band "issue", by shortening the side to (about) 65cm or lenghtening it to (about) 80cm the gain in the 70...120 MHz range goes back to negative and may avoid issues with FM "blowtorch" stations w/o adding an FM filter, didn't check the radiation pattern though, work in progress...

[Andrew (grayhat)]

Another note, I forgot to write that the antenna SWR curve "flatness" is also dependent on the diameter of the conductor, in the model I left the diameter to 5mm (0.00250m radius) as for the Matt Roberts (KK5JY) "SRL" antenna, lowering the wire diameter to 2.5mm will cause the SWR to get above 2:1 around 25 MHz, not a big issue by the way, just a note, and if someone is curious here are the SWR curves for conductors with 2.5mm, 5mm and 10mm diameter

[Andrew (grayhat)]

as for pattern, here are the radiation patterns of the loop placed at 3m above ground on average terrain on the "classic" bands (160, 80, 40, 20, 15, 10 meters)



as you can see, the vertical aperture becomes narrower (and lower) when frequency raises, but at the same time, the horizontal aperture becomes progressively wider and the backside "null" becomes deeper, all in all considering the directivity at low frequencies, where a beam antenna would be impractically big, the small directional loop should offer pretty nice performances, now... the confirmation may only come from "real world" tests

[Andrew (grayhat)]

Some updates on the "experiment", leaving the other parameters almost unchanged (side 76cm, height of bottom corner from ground 3m, the wire forming the loop has a diameter of 2.5mm, the resistor has a value of 530 Ohm, the feedpoint hosts a 9:1 transformer) I moved the feedpoint of the loop as shown below



that is, moving it to the opposite (side) corner from the one hosting the resistor, then I ran some NEC simulations and the results are shown in this image



with such a config, the small loop shows a more directional pattern, and both the horizontal and vertical aperture are much narrower, so improving the overall directivity of the antenna, and at the same time, giving a true "cardioid" pattern on almost all the frequencies (from 1 to 30MHz); as for the matching and gain, the image below shows the graphs generated by running a sweep between 1 and 31MHz



it's easy to see that the matching, after the 9:1 transformer at the feedpoint is very good, and while the gain is low, it's in line with the original Matt Roberts "SRL", plus, given the good match, it will allow to use whatever preamp at the feedpoint and not just some preamp designed for a loop antenna (a 20dB preamp will boost signals enough - assuming the receiver preamp won't suffice - or a remotely tuned preamp like this one, would further improve the antenna performances), I believe that this design is the most promising one, the only drawback is the slightly more complex coax routing; assuming to have a cross shaped support structure (built using NON conductive material) for the loop, the coax will run from the feedpoint (balun, at a corner) along one half of the horizontal arm and then, from there, down along the vertical support, not a big issue, but worth mentioning; also, to avoid that the coax (braid) could distort the loop pattern, it may be advisable to put some snap on ferrites over the coax running horizontally from the antenna feedpoint, that should take care of the issue (assuming my reasoning is correct)

Now, the only "problem" will be building a prototype, due to an issue with my wrist, I don't think I will be able to go on and build the antenna soon, so if someone would like to volunteer and try a build, it would be very appreciated

[edit]

forgot, the NEC model for the modified "beam loop" is below

Code: Select all

CM File: DirLoop.nec
CM Small Receiving Loop antenna
CM with unidirectional pattern
CE

' symbols definition
SY freq=7.100               ' test frequenct
SY hght=3                   ' height of base corner from ground
SY side=0.762               ' length of one side
SY diag=(sqr(2)*(side/2))   ' half diagonal
SY wire=0.00125             ' wire radius
SY vres=530                 ' loading resistor value
SY segs=13                  ' number of segment in wires
SY wfed=1                   ' feedpoint wire
SY sfed=segs                   ' feedpoint segment
SY wres=4                   ' wire hosting the resistor
SY sres=1                   ' segment hosting the resistor

' wires geometry
'  ID seg    x0     y0   z0         x1     y1     z1          wire rad
GW  1 segs     0     0  hght      -diag     0   hght+diag       wire
GW  2 segs     0     0  hght       diag     0   hght+diag       wire
GW  3 segs -diag     0  hght+diag     0     0   hght+(diag*2)   wire
GW  4 segs  diag     0  hght+diag     0     0   hght+(diag*2)   wire

' ground parameters
GE  1
GN  2  0  0  0  13  0.005

' wires loading
LD  5  0  0  0  58000000
LD  1 wres  sres sres vres 0


' enable extended kernel for calc
EK

' feedpoint
EX  0  wfed sfed 0 1 0 0

' initial test frequency
FR  0  1  0  0  freq  0

' end
EN