The SULA antenna

[13dka]

Dug out another clip - this one is kind of special, not only because 21:30 local is kind of late for 15m and the station being 13,300km away on the Falkland Islands yet giving me a very solid signal on the SULA:



That's Bob (VP8LP), the guy who informed BBC's Laurie Margolis about the invasion having taken place on April 2nd, 1982.

https://www.bbc.com/news/av/uk-20857595

It's moments like this that make all the efforts of carrying stuff to places and building Andrew's strange contraptions worthwhile to me. As for the performance of the SULA, you can see on the waterfall that the band is pretty deserted and only signals with flat incidence angles can come in at that time, which is where the SULA shines. I would have certainly received that e.g. with my vertical, but more often than not I get a better SNR from the SULA simply because there is (QRN) noise pickup from only one hemisphere and the highest sensitivity is limited to a range of takeoff angles within that hemisphere.

[rayb]

Well, I finished construction on my Sula, and have it temporarily lashed to a metal fencepost on the lawn. I say temporary because it would never stay vertical with the very high winds that often blow through this rural location. It's just for testing... seeing how it works on the different bands before finding a "permanent" location. At present, I have it decoding wspr "spots" on the 22m experimenters band.

Here's a look at my Sula:

https://www.dropbox.com/s/oq8anmoq564ozfv/sula.jpg?dl=0

I'm using an octagon shape rather than the traditional square. It has the same length of wire as the square loop, so I'm hoping the change in shape isn't a performance issue. At this time, I'm not using a preamp (at the antenna). My rcvr (at the shack) is an Airspy HF+ Discovery with a built in adjustable gain amp. I figured that the upper HF range may suffer a bit with no amp directly at the antenna (cable loss to the shack), but for now, I'm basically interested in 22m and lower in freq so I think the snr would be roughly the same either way.

I'm noting one "minus" with my installation location, not at all the fault of the Sula. My QTH would be an excellent radio location except for one problem... powerline noise. Here's the situation... yes, I can minimize the noise with rotation of the Sula. However, when I do that, the main lobe is then oriented in a direction away from my desired signal sources. I want to "look" mostly East, but the Sula must face West to null the noise. Relocating the antenna to accomplish my goal, so that the antenna is on the other side of the noise, unfortunately is not an option. The antenna, because of "logistics" must remain the general area of the photograph. Murphy strikes again.

Still, I'm happy with the antenna so far and look forward see what it can do compared to wires I previously had up.

[13dka]

Congrats! A few notes though:

1. I have no hard data points what it really does but a metal support is not recommendable as it will very likely mess with the antenna's performance in some way or other. Or is that a wood pole (hard to see on the pic)?

2. The SULA is lacking a lot of sensitivity without a local preamp. The Airspy may do better than average radios (it copes better with e.g. a lossy YouLoop) but it may not make a preamp obsolete since the SULA is very lossy. It's also kind of the other way around: the lower the frequency, the more the SULA needs the preamp.

3. Sorry to hear about the power line between you and your stations of interest! Nearby power lines can be a problem for any antenna (even magnetic) but the SULA is not a magnetic loop and fully susceptible to E-field interference. Depending on how far away they are it can be acceptable on dry days and on higher frequencies but for example, a main high voltage line is spoiling the fun in one of the best imaginable listening posts around here, in the middle of a large natural reserve. Low noise and high sensitivity work together to make its effects audible even though it's several kilometers away.

4. The shape looks cool but did you run that through 4NEC2 to see what it does to the pattern?

[rayb]

Hi 13dka...

Thanks for your reply.
1. Yes, it is supported by a wood pole. Sorry I didn't supply a better photo.
2. Ok on the Sula needing the amp at the source. I'll try it. My thoughts were that as the frequency was lowered, the coax loss to the shack would be less, so as to make up for the decreasing "pickup" of the Sula.
3. Understood that the Sula is still susceptible to the E-field. Since I posted earlier this day, that's becoming more apparent after further testing. As you say, that doesn't necessarily mean the mag loops would do any better. Every location has its own particular circumstances as far as noise is concerned, and there just isn't a "one size fits all" solution.
4. No, I haven't run the octagon through a simulation. I haven't yet learned how to use the software.

rayb

[Andrew (grayhat)]

Late to the party

Ok, here we go, I quickly built a NEC model for a round loop which should be pretty similar to your octagonal, the results of the simulation are below, notice that to recover the cardioid I had to change the resistor value to (around) 642 Ohm, also notice that the horizontal pattern is skewed because I made a little mistake in the model, but that doesn't change the pattern, as for the preamp, I inserted some infos about where to put it, also, the choke at bottom is a W2DU (Maxwell) type, that is a bunch of snapon ferrites over the coax, the bottom position is the best one looking at the currents distribution in the model

sula_round.jpg (164.8 KiB) Viewed 32296 times

HTH

P.S. right click the image and select "open in new tab" to see it full size

[rayb]

Thanks for taking the time to model this configuration Andrew.

I did notice yesterday after a few actual field tests, that the resulting pattern seemed to have a null that wasn't all that deep. The noise level didn't decrease much at the null. Perhaps your results from the modeling are pointing this out. I'll try again by raising the resistance from the 530 to approximately 640-650 ohms or so, then see what the real world pattern looks like in the field. I'll also have to look into some sort of choking the coax as you mention.

rayb

[Andrew (grayhat)]

the sula size, resistor and height have been optimized for the "diamond" shape, so changing shape or size will require remodeling and optimizing the antenna, that said, yes, try raising the resistor value and also lay the coax so that the section coming down from feedpoint is long about 1 full "diameter" before curving toward the support pole, as for the resistor 640 Ohm should fit your config

[rayb]

So the resistor changeout to approx. 640 ohms worked. The antenna seems to be doing a good job. I re-adjusted the coax run from the balun, as per your instructions. I found I needed to replace my 9:1 balun. The balun I had was a binocular with a 43 mix. With two turns to the coax, there really wasn't enough inductance to satisfy the "4-5 times" impedance rule when attaching to the coax. I found a BN-73-202 with two turns a better situation when down on the 160m band.

One interesting point to mention. The noise heading (direction) that produces a null at 20m is not the same noise heading at 40m. Obviously, that means that I would need a rotor if I wished to minimize/null noise as I moved in freq across the HF bands. Be nice if that wasn't so, but that's just the "luck of the draw" I guess. I'm sure I'm not the only one that has run across that situation.

[Andrew (grayhat)]

As for the noise direction, I think the noise comes from different sources, regarding aiming the antenna, the SULA is directional, so it needs to be aimed, and then I suggest adding a preamp at the antenna

Regarding the transformer, the turns ratio is the square root of the transformation ratio, so willing a 9:1 we calculate sqrt(9) and obtain 3, so a transformer with 6 turns to the antenna and 2 turns to the coax will give us a 9:1 ratio, then the number of turns also depends from the permeability of the chosen core material, as a general rule of thumb, pick a material allowing to use the lower number of turns, the #73 is a better choice for low frequencies, willing to use a #43 you'd probably need (did not make core calcs, just "guessing" !) 9 turns and 3 turns for the 9:1

Also, each time the two ends of the wire enter from one side and exit to the other, counts one full turn, not "a half one", just to be clear

HTH

[rayb]

Not sure Andrew if the noise is different sources, just no way to know. Whatever the source, I'll need a rotor if I intend on making the Sula practical for me. I now have a preamp attached.

If you feel the 43 is a better choice over the 73 for hf, I'll have to go with increasing the coax side number of turns so as to have the required inductance needed for the low end of HF. Most likely, the three turns will work. I'll measure it with an LCR meter to make sure. Of course, that will make the other side 9 turns.