Speaker cable as RF transmission line

Speaker cable and similar Figure-8 twin in various sizes is popular as RF transmission line, particularly with QRP operators, especially those operating portable in the field (eg SOTA activators). Indeed, SOTA seems to have triggered a revival in the use of these cables. With the large number of online recommendations, one could be forgiven for thinking that a dipole and feed line formed entirely from #24 speaker wire is the antenna of choice.

With that in mind, this article looks at the performance of Jaycar WB1702 speaker cable, 14×0.14mm conductors (#24) with PVC insulation, cost $0.50/m.

A 15m length was cut from the roll, inspected for defects, then a set of measurements of the input impedance of the section from 1 to 40MHz was made with S/C and O/C termination using a Rigexpert AA-600. This is sufficient information to calculate the frequency dependent Zo, vf, and loss.

Zo is nominally 130Ω, nominal vf is 0.66.

Clip 128

The model MLL=2.02e-5f^0.5+4.2e-9f dB/m (f in Hz) was derived from the measurements. The graph above shows the measured data, the total loss model and its conductor and dielectric components. It is unusual for quality RF transmission lines to have such high dielectric loss at HF, this is the result of the PVC insulation.

The loss model allows calculation of loss under mismatched conditions. Let us consider three half waves of this transmission line at 14MHz (a little over 20m in length) feeding a low half wave dipole of feed point impedance 60+j0Ω. Contrary to popular belief, the input impedance is not 60+j0Ω, but 88+j0Ω, enough to activate VSWR protection in some transceivers, thus reducing output power, but in the absence of that, still likely to achieve less than rated power to the load. The loss in the feed line under those conditions is 3.3dB, just over 50% of the power into the line is converted to heat.

By contrast, the same load fed with three half waves of RG6 ($0.35/m) would have less than a quarter of the loss and input Z would be 62+j0Ω, better suited to the transceiver. Not as easy to pack into a back pack, but probably double the EIRP in this scenario.