TheRadioBoard

Previously I had built a moderately complex two-turn shielded loop out of coaxial cable, carefully following detailed instructions about cable lengths, cutting gaps in the shield at 2 places, making pigtails on the coax, etc.

Just now I made a simpler "classic-style" receiving loop: 2.5m of 30mm-wide copper foil tape, stuck onto 60mm-wide barrier tape as a backing and formed into a diamond. I taped a junkbox 365pF cap to the loop ends. At the opposite end of the loop I added a 50cm-circumference stiff copper wire loop and soldered the two ends to an RCA jack, which then connects to a run of coax to the rig. Probably took less than an hour to build.

Initial results are very good. I could null the S8-strength Radio New Zealand signal down into inaudibility, so the nulls are working well. The capacitor allows tuning from about 7 to 21 MHz. The whole design is much simpler and allows more frequency coverage than my previous two-band coax loop.

It's working well enough that I want to make a varactor-tuned version so I don't have to reach over to the antenna every time I change frequency.

Very ingenious!

You should call that the Crime Scene Interceptor, CSI Antenna.

If we assume the cap is fully meshed (365 pF) at 7 MHz, the loop inductance is about 1.4 uH. Reactance at 7 MHz is 62 ohms. At 21 MHz the reactance is 186 ohms. This reactance variation will be reflected into the coupling loop, so you're never going to have a perfect match to 50 ohms.

Based on the Hoop Loop experience, I wonder if you could put the feed line right across the loop, use flat speaker wire for the feeder, and put the cap at the receiver. This would be for listening, of course, where a 1:1 SWR isn't required. That will save you running dc out to varactors.

The Hoop Loop on MW needs a matching transformer because of its low inductance relative to wavelength. I had to transform that inductance up to something much higher, that the radio's internal variable cap could tune. With this setup I don't think you have to do that, because the 365-pF cap can tune the antenna directly.

If you can rustle up some speaker wire at the local radio store, I'd be very interested in knowing how that works. There's room here for discussion related to feeder capacitance and the effects of connecting a balanced antenna to an unbalanced port (you're doing that now, in any event), but this scheme may work and save you some trouble.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Is there anything special about speaker wire for this purpose? What would happen if I used coax instead?

Would the feedline be part of the antenna in this case, possibly reducing the directional properties of the loop?

Speaker wire is like tiny twinlead. It's a balanced feedline. Coax is unbalanced. If you use balanced feeders the feedline shouldn't interfere with the antenna. I've proved that here by moving the feedline around while listening, with no observable effect.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

It doesn't work. At 7 MHz tuning the cap has no effect whatsoever on the received signal (or rather, noise). At 20 MHz and above there seems to be a quieting (not a noise-peaking) effect at one setting of the cap. It's definitely not acting like it's tuning the loop to resonance. I used 15m of speaker wire with about 3m extended and the other 12m in a jumble (not a coil) on the floor.

Time for the varactor idea. I'll reuse the speaker wire as the DC power line.

Got the varactor version done. Works great! Now I have a remotely tunable receiving magnetic loop antenna.

At one end of a 15m two-conductor cable, I have a 12V DC power supply and a 150k pot as a voltage divider. The two conductors of the cable are connected to ground and the pot wiper. At the antenna end of the control cable the positive (wiper) lead is bypassed by .1uF to ground and goes through 100k to the reverse-biased varactor to ground. Junction between the 100k and the varactor goes through .1uF coupling to one side of the loop; the other side of the loop is connected to the control cable ground.

On my first try I mistakenly put the DC decoupling cap at the pot end of the cable instead of the antenna end of the cable. That failed miserably as the entire stray capacitance of the cable was appearing across the varactor capacitance. Putting the bypass cap right at the 100k resistor at the antenna end of the control cable fixed that.

Rats. While the varactor can tune the loop to resonance (I can hear the noise peak), I'm picking up S8-level noise presumably through the unshielded control cable (speaker wire). The noise goes away when I disconnect the varactor and switch back to the using the variable capacitor.

I'm thinking now my varactor scheme with separate DC control lines is unbalancing the loop. In the capacitor configuration, each end of the loop is tied to one end of the capacitor. In the varactor configuration, one end of the loop is tied to the varactor anode (mounted at the loop) which in turn is connected to DC ground through the 15m cable to the power supply at the far end of the cable. The other end of the loop is tied to the varactor cathode (mounted at the loop) through a .1uF cap.

I guess the long 15m ground wire is probably the problem. So this means the DC power supply for the varactor (+12v and ground) should be right at the antenna so the grounded end of the varactor doesn't need to go through 15m of cable to reach ground potential. This requires a 3-cable conductor: two wires carrying full 12v DC voltage from the antenna to the pot, and 1 return wire carrying the adjusted control voltage from the pot wiper back down to the varactor at the antenna. So everything except the pot would be mounted right at the antenna.

Sound reasonable?

http://home.earthlink.net/~christrask/Indoor%20Loop%20Antenna.pdf

Balance tuning with two varactors, decouple tuning control from rf.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

While there is much of interest with that design, one of my goals was to have the tuning control physically separated from the antenna element itself (to allow optimal placement of the antenna while keeping the tuning control next to the rig).

I'm not sure if the referenced design will work (any better than my design) with the tuning knob separated (by a 15m cable) from the antenna and varactor assembly. I'm starting to wonder whether this can work at all, because any way you slice it you need a 15m-long DC ground wire running to the pot, which would seem to invite noise pickup.

Maybe a motorized variable capacitor would be easier...

I thought that circuit would give you some insights. Here's a suggestion. There are examples of this elsewhere online, mostly used to tuned air-core MW loops, but should work with yours as well.



If you're connecting the feedline directly across the loop (as opposed to using a second, coupling loop -- I can't remember), connect it as you have in the past, but use 0.1-uF coupling caps to isolate dc from the feedline.

The rf chokes should have several times the reactance of the loop to adequately isolate the rf from the dc.

If you're connecting coax directly to the loop, you could dispense with one varactor and get twice the capacitance variation/volt. In that case, use an rf choke to ground one side of the loop for dc and again, use 0.1-uF coupling caps from the loop to the coax, so no dc appears on the feedline.

A broadband balun at the loop and the balanced circuit shown would be preferable if you care about keeping the loop balanced. If you're going to connect a balun, don't forget the dc-isolating coupling caps.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Thanks for the suggestions. I'll give it some thought and revise the design. For clarity, here's what I've got now. It's pretty minimalist.

Compared to your suggestion I'm basically lacking a choke and an additional decoupling cap. It's hard for me to believe that an additional cap and choke will be able to banish the S8-level noise riding on the control line, but perhaps I just need some more faith in (and experience with) the choking ability of chokes.

It's rf noise, right? So why wouldn't rf chokes and low-impedance bypass caps reduce it? The noise, being superimposed on the control voltage, could be modulating the varactor. Better decoupling and balancing the tuning will help, as will that grounded centertap between the varactors. On the Hoop Loop matching transformer, adding a center-tapped winding and grounding the center tap, recommended by a SiLabs engineer, dropped signal pickup by the transformer and feed line by about 30 dB.

Radiated noise getting onto the control wire is common-mode. That the loop is picking it up and radiating it into the pickup loop tells me the system is not balanced. You may never get rid of all noise, but running a noise-receiving control wire to a motor-driven mechanical cap is still going to put re-radiated noise in proximity to the antenna.

There are other options: 1. Move to a quieter location. 2. Find and eliminate noise sources.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Something very simple that you could try is to insert a second 100k resistor in the line running from C1 to the anode of the varactor. that will help decouple the ground side of the control line from your antenna.

Doing some experiments this morning with the loop and the variable capacitor (no varactor or control wire) revealed that moving the coax around changed the noise. I understand that this is not supposed to happen, right? (The coax is about 10m long mostly in a bunch on the floor.) This would mean my loop is unbalanced, which needs to be solved first.

I'm getting good nulls 180 degrees separated, which is supposed to be a good sign with regard to balance, but it's apparently not the whole story.

Is achieving balance typically difficult with receiving loop antennas, requiring exacting physical balance and measurements down to the mm level? Or is there some easy fix? I just made a reasonably symmetric main loop and a reasonably symmetric coupling loop (though the coupling loop is slightly lopsided at the bottom due to the shape of the RCA jack it's connected to).

One suspicious thing is my use of an aluminum tripod which may be disturbing the balance.


OK, maybe a balun would be a good start. Time for more study...