Installing & tuning the nested loops
Tests have shown that at least 200 Watts can be transmitted in cw and ssb-pep without noticeable tvi. However most rigs that are nominally rated at 100 Watts can only achieve 40 Watts continuously e.g. in fm or fsk. Hence a 3 or 4 amp dc rated 2-core cable would normally be adequate for the construction of the nested loops.
Determine the velocity factor for the cable and calculate the distance along the cable where the two cores have to be shorted together for the particular frequency band. Continue laying out the cable and allow 15cm at the end to make it easy to fold back and tie in place in case adjustments are required.
Roughly install all of the loops for the bands you are interested in before starting to tune them up because there will be some slight interaction between them.
It is good practice to make a choke balun at the connection to the antennas (Ref.21) by using the 50 Ω coaxial cable to form a 6 turn coil with a diameter of 10cm (Note:This should be increased to 7 turns for the 10MHz antenna). This produces an inductance of approximately 6 μH which has a reactance of about 500 Ω at 14 MHz, corresponding to 10x the input impedance of the antenna. This helps to achieve consistent swr measurements by reducing any stray RF by making it at least 10x harder for it to flow down the outside of the co-ax rather than onto the antenna loops.
Tune the loops starting with the lowest frequency band. This is done by altering the effective lengths of the loops by tying back more of less of the cable at the remote ends, just as you would do for a normal dipole. If a loop has to be increased in length do not overlap the ends because the capacitance between them will make any adjustments very delicate. It is better to increase the overall size of the loop. It should be possible to achieve an swr of between 1.0 and 1.5 on all of the bands at resonance without an atu. If not, it may be because there is a bad connection at the terminal block, or because there is wiring or plumbing in the vicinity loading up the antenna loop which will need to be detuned.
Its a good idea to place a smoke alarm in the attic just to be on the safe side, we have not had any problems with their rf immunity.
Checking for parasitic conductors in the attic
If you are unlucky the propagation pattern may be distorted by some local conductor. This can show up as an inability to get the swr below 1.5 at resonance on one or more of the bands. It may also be revealed by noticeable differences in the field strengths outside the property at equal distances ether side of the centre line of the nested loops. The problem could be confirmed by modelling the situation using an NEC based program.
Detuning parasitic wiring and plumbing
If the culprit is the electric wiring it will most likely be supplying the lighting for the rooms immediately below. A standard electrical lighting circuit would be typically fed from a 5Amp or 10Amp MCB on the distribution panel. The top floor lighting sockets would be connected up radially in a daisy chain fashion forming an open loop. The cable can be detuned by clipping on broadband EMI ferrite split/snap-on-cores (Ref.22). Generally there is not a great deal of difference between equivalent sized parts produced by different manufacturers and performance details are given in Ref.23. A typical ferrite such as HEM3018 from Maplins has a 10mm hole diameter, it adds an inductance of 0.32 μH and a resistance of approximately 50 Ω at ≥14MHz in series with the cable. For maximum effect the ferrite should be clipped on 1/4 of a wavelength from the end of the cable, and then every 1/2 wavelength apart at the problem frequency. Experiments with a cable 5.3m long which was resonant at approximately 14MHz showed that the loading effects were nullified by the addition of 3 HEM3018 ferrites clipped on at the middle.
It is not practical to clip ferrite to a parasitic conductor if it is a copper pipe of more than 12mm in diameter. However experiments have shown that it is possible to detune the problem pipe but using metal hose clips to connect two wires to form a rectangular loop 0.3 x 0.3m closed by 1μF capacitor to form a resonant loop. Experiments were made with a parasitic conductor resonant at 14.15 MHz placed 1m in front of and parallel with a receiving dipole also tuned for 14.15 MHz. The connection of the loop to the middle of the parasitic conductor raised the resonance of the conductor by 320kHz and restored 70% of the gain lost due to the presence of the conductor.
Tackling EMC issues
The compactness of the double folded dipole makes it much less likely for the house wiring in the attic to match its shape and couple strongly with the antenna loops. The actions taken to detect and detune any parasitic effects will also minimise the rf energy picked up by the house wiring.
The next stage is to follow the ‘first steps’ (Ref.24) by testing each loop in cw and then ssb at increasing pep levels to see if there is any break-through into your domestic electronic equipment e.g. radio, hi-fi, television, computer, internet link, security light/alarm, telephone.
- Where your transmissions are affecting your other equipment, try to resolve the problem through application of the relevant filters.
- Look for the obvious sources of the problem, what is acting as an antenna to pick up the signals? Very often this will be long loudspeaker leads to audio systems (fit ferrite toroids near the audio amplifier output), the coaxial down-lead to the television (try a braid-breaker filter), or the mains cable to the affected equipment – again, try a toroid at the point that the mains cable enters the equipment.
- Breakthrough onto telephones can be tricky (Ref.25). Again, ferrite toroids fitted to the telephone cable near the handset can help. But some telephones are much more susceptible to breakthrough than others. It may be helpful to consider having a very basic handset (with minimum amount of electronics) to confirm the source of the problem. Disconnect all telephones, modems etc from your internal telephone network, and connect the simple phone. Confirm that the breakthrough is cleared – if not, fit toroids to that handset until it is. Then progressively add back telephones on the network, one at a time, checking each time that the breakthrough is cleared. Remember that one telephone that is susceptible to RF can add back interference on to the whole network. When one handset added causes breakthrough to return, try toroids, but be prepared to discard the offending handset if all else fails. Ultimately you should have a complete telephone network free from breakthrough.
- Intruder alarms (& lights) – It is possible to do tests without setting off the alarm. The red ‘walk test’ LED normally flashes if the Passive Infra-Red sensor (PIR) is triggered by movement or RF. Find out how much power is needed on each band to make the LED flash. Nearly all PIRs use pulse counting so for an adequate test the RF carrier must be pulsed on and off by transmitting a few seconds of Morse. If you prove that the PIRs are RF triggered you will need to contact the installer. The most effective solution is to replace all PIRs with a more RF immune model (Ref.26).
Ref.21 – http://www.hamuniverse.com/testingswr.html
Ref.22 – Laird Tecchnologies : Broadband EMI ferrite split/snap-on cores in plastic cases.
Ref.23 – http://www.arvelo.net/N2MZZ/graphics/A Hamsguide-to-RFI-ferrrtes-baulns-audio-interfacing-k9yc.pdf
Ref.24 – http://www.rsgb.org/main/technical/emc/i-am-causing-breakthrough/
Ref.25 – http://www.rsgb.org/main/files/2012/11/EMC05-Final.pdf – ‘Radio Transmitters and Telephones, Information for Telephone Users.’
Ref.26 – http://www.rsgb.org/main/files/2012/11/EMC03-final.pdf – ‘Dealing with alarm emc problems.’