2008-09-28
Resonant Coaxial Feedline Choke
This is a tool I've been meaning to make for a while. For HF antenna experiments it is often useful to be able to selectively choke the feedline of common-mode currents. While you can use a ferrite bead choke or some arbitrary coil of coax to provide a bit of impedance, a resonant choke is tunable so you can make measurements with it still inline but detuned and then quickly tune it up for comparison.
The coil form is simply a plastic pill bottle about 50 mm in diameter. Onto this 9 turns of RG-58 coax are wound giving an inductance of about 3.5 uH. The coax passes out the bottom of the bottle and is terminated in BNCs. Inside the bottle the braid of the coax is exposed at each end of the coil and connected across a polyvaricon installed in the lid of the bottle. Liquid electrical tape (like RTV) is used to seal up the coax again and bottle penetrations.
The lid is calibrated in MHz, performed by dipping the coil or tuning it to resonance near the output of my signal generator while monitoring the voltage across the BNC plugs on the CRO with a suitably small cap in series. The device tunes 5.7 MHz to beyond 16 MHz using the full 210 pF maximum capacitance of the polyvaricon's combined gangs. The distributed capacitance of the coil plus internal wiring capacitance is around 12 pF limiting operation to around 20 MHz tops.
Usage
It covers 20, 30 and 40 metres (it was designed for 40 metres in particular), and 17 metres too but the tuning gets fiddly up there. The general idea is scalable to any frequency. Tune-up is simple, place a clamp-on RF ammeter over the coax and tune for a dip in current. Alternatively you may pre-tune with a dipper, but it is easier to tune in-place where the extra capacitances of the feedline are attached. Pre-tuning will generally need to be tweaked up (less capacitance) because of the extra capacitance between the feedline ends.
It is fairly common practice to use self-resonance without additional lumped capacitance at VHF and UHF to choke feedlines, but it seems few people actually try for a resonant breaker, just adding "a few turns" in the feedline. With a dipper it is easy to fiddle with a roughly dimensioned coil of coax until you get resonance where you want it, then tighten zip-ties to hold the coil in place. Q is generally low at higher frequencies using the coil's self-resonance so tuning is not too touchy, but it must be done with care to get the best results.
The resonating of the coax braid like this DOES NOT effect currents flowing inside the coax. (I confirmed this experimentally to the limits of my return loss bridge.) Only currents trying to flow on the outside of the coax experience the impedance of the tank so formed. The Q of the tank is not especially good, the coil being close-wound of coax braid, but it offers at least several tens of kilo-Ohms to braid currents at resonance and is broad enough to not need much retuning in narrow bands. It is possible to construct a similar device for balanced lines, using isolated dual capacitors across bifilar mutually wound inductors fed by the feedline.
2 comments.