Re: HF Magnetic Loop Antenna - prototype under test

David Cutter

Some interesting thoughts there, Chris. 
I made a couple of loops and the best one was far too cumbersome being made of 1mm aluminium sheet 8ft long and I learnt stuff on the way. 
The best one I've seen which tickled my technical bits was a "trombone" system out of QST using copper pipes and a lining of PTFE between, so this was a linear capacitor.  The annoying thing is how slow they are to tune.
I read a discussion amongst erudite professionals in regard to the use of PTFE sheets between the blades/vanes of a cap and the conclusion was that the insulation needed to be intimately bonded to the surfaces to avoid any air gaps across which arcs could develop.  That meant that the vanes would be dipped, I presume.  The current withstand was governed by cooling and thickness and I've seen some very thick beasts out of Russian marine transmitters!
Good luck

----- Original Message -----
Sent: Friday, January 13, 2017 11:29 AM
Subject: [CrossCountryWireless] HF Magnetic Loop Antenna - prototype under test

We've been making the HF Active Loop Antenna for a few years now and whenever we attend radio rallies and hamfests or mention it to other radio amateurs I'm always asked "When are you going to make a transmit version?".

Over Christmas and New Year I've been laid low with shingles so I've taken the opportunity of enforced idleness to research and design a magnetic loop antenna for HF.

The first thing to note is that there is an awful lot of incorrect and misguided information on the internet about magnetic loop antennas. Since I built the prototype I can understand how that comes about as the resistive losses have to be kept extremely low and it's possible with a small amount of loss (say 0.1 ohm) to have an antenna that appears to work and match correctly but has high RF losses giving misleading results.

The first prototype uses a 1m diameter loop of S400 (LMR400 equivalent in the UK) with high quality N type male connectors. These fasten to two N type female connectors mounted on one of the polycarbonate boxes used for our existing antennas.

This allowed me to test various feed arrangements and a range of variable capacitors.

The first capacitor was an Oren Elliot 384 pF single gang variable. This gave very poor results on transmit but appeared to tune and peak perfectly on receive. A DC check showed 0.1 ohms across the capacitor frame and sllding contact to the rotor. This was enough to lose a lot of power on transmit.

The second capacitor was a new old stock Jackson Brothers 365 pF dual gang capacitor. The ends of the loop were connected to the stators allowing the rotor to tune without any RF current flowing through the sliding contact. This was efficient but the maximum power that could be applied was 25W before the capacitor flashed over. I'd like to manufacture an antenna that could take 100W as least.

The supply of the JB capacitors is limited not being manufactured anymore and the Oren Elliot range of capacitors are extremely expensive once they have been shipped to the UK and had tax applied.

At this point it's worth noting that the magnetic loop antenna is a resonant tuned circuit and the RF voltages across the capacitor and the RF current flowing through the loop should be very high if the losses are low. With 100W on 14 MHz the RF voltage across the capacitor should be 2579 V rms and the RF current in the loop 16.6A. Serious stuff!

There are very few variable capacitor manufacturers left so the choices are very limited and expensive.

At this stage the only option was to look at making a capacitor ourselves.

The first prototype is a compression type capacitor using PTFE sheet as a dielectric. This should be capable of withstanding the high RF voltages and current.

The first test of the PTFE capacitor worked well. The tuning bandwidth is sharper than the air spaced capacitors indicating that the losses are lower.

Applying RF power to it there was a dramatic flashover at 80W. After stripping it down the flashover was not across the capacitor but from one of the mounting screws on the N type sockets. The arc had jumped across the air gap to the capacitor plate.

That's the progress so far. The flashover indicates that the RF losses in the antenna are very low allowing high RF voltages. The RF matching works well giving a VSWR of 1:1 at resonance.

Before it goes into production it needs work on the safety aspects insulating the N type connectors and replacing the mounting screws with polycarbonate ones for example.


Chris G4HYG

Join to automatically receive all group messages.