Homebrew magnetic loop antenna with DIY variable capacitor
#Make#RadioRecently, I moved into a new shared flat. Unfortunately, that meant I could no longer install a random wire antenna for amateur radio. For the past few weeks now, I have been looking for an alternative antenna setup to get on the air from my new home. Finally, I have settled on the concept of building a magnetic loop antenna.
This post documents the design of such a magnetic loop antenna for indoor QRP operation as well as its assembly process using only inexpensive parts from the local hardware store.
Note of warning: Magnetic loop antennas tend to reach thousands of volts across the capacitor terminals even at QRP operation. If this post inspires you to build your own loop, please proceed with care and with caution.
Do not touch a magnetic loop antenna during transmission and keep a safe distance.
What is a magnetic loop antenna?
A magnetic loop antenna is basically an LC circuit consisting of the main loop as inductor and a variable capacitor for tuning to the operating frequency. Since the impedance of the LC circuit at resonance will be far lower than the 50 Ω impedance that is common in amateur radio, a smaller loop - called the coupling loop - is placed inside the main loop. Together, the two loops act as a transformer for impedance matching.
The antenna is called a magnetic loop because electromagnetic fields are the combination of a time-varying electric field component and a magnetic field component , which are perpendicular to each other mutually coupled, thus propagating electromagnetic waves through space. Unlike an antenna like a standard electrical dipole, whose near field is dominated by its electric field - the magnetic loop antenna with its primary loop as an inductor, has a near field dominated by its magnetic component. Although it should be mentioned, that in the far field both antennas produce electromagnetic waves consisting of perpendicular and components.
Furthermore, the voltage across the terminals of the variable capacitor can easily exceed a thousand volts, even with a low transmission power of 5 watts. Therefore the capacitor needs to be able to withstand a high voltage. High voltage air variable capacitors - where the plates are separated only by air - can be bought on the internet. But they are not as common as they used to be and quite expensive. However, they are not that complex mechanically. Therefore, I decided to build one myself.
How to build an air variable capacitor?
An air variable capacitor consists of two groups of plates. One group is fixed - called the stator and the other group is placed on a rotating shaft - called the rotor. By slowly rotating the rotor over the stator, the overlapping area between the two groups of plates can be increased or decreased - thus varying the capacitance.
The dielectric strength of air is given as approximately under ideal conditions in textbooks. However, the actual dielectric strength can vary and sharp edges may significantly lower the breakdown voltage. Therefore, as a rule of thumb, I want to have at least one millimeter of distance between the plates for every kilovolt in my design to be on the safe side.
Next, to avoid making an electrical connection to a moving part of the capacitor, such as the rotor, I used two groups of stator plates which are electrically separated and use each group as a terminal. Electrically, this is equivalent to putting two identical capacitors in series, which requires more plates as the total capacitance becomes halved. However at the same time, in a series circuit of two equal ideal capacitors, the voltage at each capacitor equals half the total voltage, thus also lowering the required voltage tolerance for my DIY capacitor.
But how many plates will my capacitor need, and how large should they be? The capacitance required to tune the loop can be calculated with the Magloop Antenna Calculator from Miguel VK3CPU and the formula for the capacitance of a plate capacitor is
Now, while in theory the number of plates can be calculated exactly for a given loop size, overlapping area and distance between the plates, the built capacitor will most likely not match the calculated capacitance at all due to several factors. Twisted plates, separators and terminals may induce additional capacitance.
Nevertheless, you have to start somewhere. The following section contains a calculator for the capacitance of two variable capacitors connected in series based on their number of rotor plates and the plate distance:
In this design, the two capacitors in series will have 6 stator plates each and active gaps between plates. This results in a capacitance of
per capacitor in series. With , this yields a total capacitance of
In the end, I decided to cut 10 semicircular rotor plates from a 0.5 mm thick sheet of aluminum and placed them on threaded rods, separated by hex nuts. Next, the stators were cut as rectangular plates and also separated by nuts. The overlapping area between each stator and rotor plate is about 25 when fully closed. Using the nuts as spacers gives a distance of about 2.5 millimeters between the plates. Furthermore, I cut additional spacers from an aluminum pipe for the threaded rods that do not hold plates, but serve only as stabilizers for the construction. Finally, the top and bottom plates of the capacitor were cut from acrylic glass.
After assembling the capacitor, I cut a little cogwheel from acrylic glass and mounted it on the rotor, to have some more torque when turning the rotor. Then, I calibrated my NanoVNA from 1 to 30 MHz and connected the capacitor to it to get a rough estimate of its actual capacitance.
While I do not trust the results too much due to inaccuracies introduced by the VNA itself and the test leads, I measured the capacitance to range from about 20 pF fully open to 60 pF fully closed. This differs quite a bit from my previously calculated capacitance of about 45 pF fully closed. However, I consider this still a reasonable deviation given the additional capacitance introduced by the threaded rods and spacers, as well as inaccuracies in the construction.
How to assemble the antenna?
With the capacitor finished, building the loop was easy. I got all the parts except the coax for connecting the antenna to my transceiver from the local hardware store. The full bill of materials for the antenna, including the variable capacitor, is as follows:
| Material | Price |
|---|---|
| Aluminum plate 250 x 500 x 0.5 mm | 4.45 € |
| Threaded rod M4 x 1000 mm | 0.80 € |
| 100 M4 hex nuts (3 mm thick) | 2.55 € |
| Acrylic glass plate 250 x 500 x 4 mm | 7.95 € |
| Aluminum beam 25 x 2 mm, 2.6 meters | 8.45 € |
| Aluminum pipe Ø 6 mm, 1 m | 1.80 € |
| 2 cylinder bolts M4 16 mm | 0.66 € |
| 2 wingnuts M4 | 0.44 € |
| 1 meter of H07 V-K 1x10 mm² wire | 1.98 € |
| Total | 29.08 € |
At just under 30 Euros, that seems like a reasonable price to me.
For the construction, first I bent the main loop out of an aluminum beam. The longest available beam was 2.6 meters long. That gives a loop diameter of about 80 centimeters. With a cross section of 25 x 2 mm, bending the beam was easy. In fact, I already bent it at the hardware store as a ring was easier to transport than the long beam. Finally at home, I drilled two 4-millimeter holes at the ends of the beam for connecting the terminals of my variable capacitor.
Next, I took about 60 centimeters of a 10 cable for the coupling loop and connected it to a BNC-to-screw-terminal adapter. Then I placed the coupling loop inside the main loop, turned the capacitor half open and connected the NanoVNA to the coupling loop.
Sweeping from 1 to 30 MHz, I could already see the resonance frequency, but the lowest SWR was still above 3:1. To improve the SWR match, I reduced the length of the coupling loop a bit and used duct tape to secure it tightly to the top of the main loop.
And that’s it - the magnetic loop antenna is finished:
After assembling the two loops and wiring up the capacitor, I used some leftover rope to attach the antenna to the ceiling of my room.
How does the loop perform?
Well, I am satisfied with the antenna’s performance. Turning the variable capacitor from fully open to fully closed tunes the antenna over the range from 16 to 30 MHz. This includes the 10, 12, 15 and 17 meter amateur radio bands.
I may modify the capacitor in the future so it can also tune the 20-meter
band at 14 MHz, but for now, I am happy that I get on the air in the first place.
Look at all those FT8 contacts I made over the span of a few days, using only 5
watts on all four bands.
And that’s despite the fact that we are currently in the declining phase of the
solar cycle and it’s winter: