CCW Active Loop Ethernet cable.

Sean - G4UCJ

Hi Chris,

Thanks for the info about the Ethernet cable. I ended up buying a 20m exterior quality CAT-6 UTP cable, which seems to be working fine. As I can't do a before and after comparison, I will have to assume all is well (both LEDS are lit in the base unit and plenty of signals are being heard. I don't know if the extra 'bandwidth' of CAT-6 makes any positive difference or could be a hindrance. The only reason I went for CAT6 is that it was actually cheaper than the equivalent CAT-5 cable (plus it would arrive a few days faster, ever the impatient!).


Now I asked this question once before but I think I asked it in the wrong way as the answer wasn't what I expected, so I'll ask it in a slightly different way: I am going to be using the CCW Active loop (the triangular one), together with my Wellbrook ALA1530 through the MFJ1026 noise canceller/phaser with the hope of making a steerable phased array for use on the lower bands, and to see if I can lower the noise floor by targeting the major noise sources on each band. The Wellbrook is on a rotator and about 4m above sloping ground. The CCW loop is on top of my mast, about 6m above flat, but slightly higher, ground (so about 4-5m higher than the Wellbrook). The CCW is not on a rotator, but can be turned by hand if really necessary.


My question is this: Do the Wellbrook and CCW loops behave in the same manner when nulling or peaking? That is, the open loop (like looking at a bicycle wheel), as you look at it, point to the signal null. The edges of the loop (like the rim on a bike wheel) 'point' to the signal maxima? An answer to that would be really handy so that I can position the loop accordingly. I haven't been able to do any null/peak tests with the loop as yet.


To go back to the Ethernet cable, in answer to Simon, yes I have ferrite(s) on pretty much every cable plus two mains chokes (well one choke, and one mains filter). It makes a big difference to the noise floor, but of course it can't get rid of VDSL/PLT type noises - 30m is really tough going here and is the worst affected band, followed by part of 80m.

73, Sean



From: [] On Behalf Of Chris Moulding
Sent: Wednesday, July 15, 2020 10:59 AM
Subject: Re: [CrossCountryWireless] CCW Active Loop ethernet cable.


Hi Sean,

The first antenna we made that used ethernet cable as a feeder was the HF Active Antenna.

This was designed for use with the RSGB Noise measurement program some years ago so I put a lot of time and effort into finding what was the best ethernet cable to use as we wanted to reduce any local noise pick-up on the cable to avoid corrupting the noise measurements.

To give some information to those who have never used ethernet cable as a RF feeder the cable has eight wires arranged in four twisted pairs. Each twisted pair has a slightly different twist rate to minimise crosstalk between the pairs.

The characteristic impedance of each pair is 110 ohms.

It is very important that each pair is correctly terminated at each end or if it is an unused pair shorted out.

In the active antennas we use one pair for the RF connection correctly terminated at 110 ohms. Two pairs are used for power and effectively shorted out for RF by decoupling capacitors at each end.

The unused pair is shorted out at each end. This was eventually used as an additional RF feeder for the Dual Polarisation Active Antenna.

When the cable is correctly terminated like this the best cable to use was found to be UTP (unshielded twisted pair). Adding shielding in the STP (shielded twisted pair) added extra capacitance and increased attenuation in longer feeder runs. Using UTP the antenna was tested working with a 3 dB loss on 10m with the full 305m drum of cable.

The measured crosstalk between the RF pairs in the Dual Polarisation antenna measured greater than 50 dB down (the cable spec is 40 dB).

As is is a correctly terminated balanced feeder it doesn't need extra ferrite or coils of cable to act as common mode filters.

To prove this I made a 15m long test cable between the prototype antenna head unit and base unit and erected the feeder as an inverted vee antenna to try and pick up the huge signal at our location on 10100.8 kHz from the German weather station DDK9. When the active antenna was powered down the received signal picked up by the elevated feeder was 40 dB  down on the signal received by the active antenna.

In answer to your last question ethernet cable will only radiate or pick up unwanted RF only if it is incorrectly terminated say one of the crimp connections is not secure.



Chris Moulding

It depends on the RF frequency received by the loop.

At LF and HF where the diameter of the loop is less than 0.2 wavelengths it works as you describe where the null point is where you look through the loop.

All small (< 0.2 wavelength) loops such as loops made with our Loop Antenna Amplifier, Wellbrook etc work this way. I hesitate to include the MLA-30 in this list as there is evidence from my own tests on one sample that it picks a lot of signal up from the coax so the the loop nulls are filled by random signal and noise pick-up from the coax cable.

When the loop diameter approaches 1 wavelength the loop radiation pattern changes to that of a resonant quad loop where the nulls appear off the edges of the loop and maximum signal is the direction looking through the loop.

In the case of the 3m wire supplied with the Loop Antenna Amplifier this occurs around 100 MHz. Changing the circumference of the loop allows this to be tuned so say 2.5m will resonate on VHF airband and 2m will resonate on the 2m amateur band. The latest Loop Antenna Amplifier + works up to 150 MHz to take advantage of this.

Sorry for the longer reply than you expected, in RF engineering there are very few yes or no answers!




To add to the question, what differences/improvements would cascading the turns achieve rather than 1 turn with all the wires paralleled ?

Just something I have mulled over in my mind while looking at my Ethernet loop form the kitchen window ;-))


Chris Moulding

With small (< 0.2 wavelength) aperiodic loops the efficiency increases when the inductance is reduced.

This is why for 1 turn loops constructors go for copper or aluminium tube rather than wire.

Increasing the number of turns increases the inductance reducing efficiency in untuned loops.

Where multiple turn loops work is where the loop is used as the inductor in a tuned circuit and is tuned to resonance by a variable capacitor. This gives a very efficient receiving antenna but only at one frequency.

By the way the very first antenna I ever made was a simple frame antenna for medium wave using about 5 turns of wire on a wooden frame resonated with a 365 pF capacitor liberated from a scrap receiver. Putting my little medium wave "tranny" inside the loop and tuning to resonance probably sparked my career in radio engineering!



Sean - G4UCJ

Hi Chris,
Thanks for the additional information, that is very useful as I do use the loop on the VHF bands as well as HF. Looking at the orientation, I think I may get better results on 4m and Band II by turning it 90 degrees. I've mounted it about 12" above my 5 element 2m beam, so there may be some interaction between the two antennas, but I haven't noticed anything odd so far.

I also made MW loops, back when I was a schoolboy. Usually, 9 turns on a 1m square, or 15 turns on a 30cm frame. All tuned by a 500pf air-spaced capacitor. To stretch the range I added switchable parallel fixed capacitors. I did at one point add further inductance in the form of a switchable ferrite rod/Litz wire coil (as found in the old transistor radios back then). I experimented a great deal back in the early 80's. The largest loop I made was about 5 feet per side - I did have trouble keeping the tension on the wires as the frame was a bit wobbly, due to less than ideal woodworking skills! It worked very well indeed but in the end it was just too big to fit in the room and be turnable. Ahh those were the days of little QRM (and no PC's either)! 73, Sean G4UCJ