Remotely tuning a magnetic loop antenna

Magnetic loop antennas are great. They are relative small in size compared to the frequency, they are pretty easy to construct and parts are generally available. However tuning the antenna, is a bit tricky.

In 2016 I decided to make my own loop antenna. Tuning is done using a variable capacitor. These can be obtained from ebay for example. Search for “Vacuum variable capacitor”. This page is not intended as a guide on how to build your loop antenna. There are many good pages found on internet which already do this. However here a few picture to give an impression how I build it and especially how I tune the antenna!

Variable capacitor

Protecting the capacitor and motor from the weather, is a plastic shield. Now 4 years later, all parts are still in a good condition.

For matching the coax onto the antenna I use a gamma match. A plastic box is mounted on the top of the antenna, it consists of a balun, and a relay to choose between the 2 gamma matches. The shorter one is used for the 40m band, the larger is used for 80 and 160m.

The tuning capacitor is turned using a stepper motor.

To main problem I faced was controlling the stepper motor I mounted onto the capacitor. I decided to build my own controller. Here a short video clip showing how the stepper motor is controlling the capacitor. The controller is inside the house seen in the background.

Tuning the capacitor remotely

Mechanically this is pretty simple. A few parts connect the stepper motor to the capacitor. A rubber belt is used to obtain electrical separation.

Controller in the shack

In the shack I have a small controller, that easily and accurately turns the capacitor. In addition to different speeds, it also has 4 memory buttons. So it is easy to “move bands” by just going to a memory.

Updated design of the controller: LT100, Nov 2020

After using this controller, and gain enough experience, it’s time for an update.

This is the list of improvements:

  1. Design for manufacturability
  2. It should have additional switches to control the gamma match and or other equipment
  3. Improved speed selection
  4. Improved visibility of where the capacitor is currently set to
  5. Better connectivity, to have at least USB
    1. Fully support remote operations
    2. Be able to update software
  6. Easily controllable from the front panel
  7. Wider voltage input to support long wires (10 ….. 24V)

Hereby the first ideas


There will be 4 memory buttons. Whenever the antenna has been tuned to a frequency, long press any of the memories to store the current position and Switch settings. Above the memories there will be a led bar, indicating the position.

Tune & Speed controls:

The tune knob is a Rotary Encoder, with 30 Detents per full rotation which is used to tune the capacitor.There are 3 speed settings, above is a picture of the idea, it has a button and 3 led indicators.

Each speed can be configured, for example: 1 detent / 1 step on the motor, 1 detent / 10 steps, 1 detent / 100 step.


Switch selection consists of 2 buttons and 2 indicators. These switches are 2 relays which can be controlled. The relays are rated at 2A. These can be used to control antenna switches or anything else really. The position of the switches is stored at memory locations as well.

Stepper motor

One stepper motor is supported, the current range will be between 500mA …. 2A. Whenever the motor is not in used, it will be physically disconnected via relays. Also the windings will be shorted. This will have the following benefits:

  1. No RFI into the antenna
  2. Shorted windings act like a “brake” it will be hard for the motor to turn.

Max current and max speed can be set with potentiometers on the back.

Remote station support

This controller should fully support remote operations. All functions can be configured and controlled via a USB interface. In addition current and voltage monitoring is implemented as well.

Current status 02 Nov 2020

Design files are done, and prototypes are ordered. Prototypes are expected to be delivered in December. After that verification and software implementation will take place.

If you have any questions / comments, or if you’re interested to buy, please contact me !

Update 27th Nov 2020

Prototype is coming along nicely, PCB for the frontpanel is done:

However comments suggest that the LT100 should not only have USB. With USB a separate program, running an a computer is needed to remotely control the LT100.
As remote control is an important aspect of this design, it should have an internet connection. I therefor decided the following:

  • Update the prototype to include an ethernet connection, no WiFi, reasoning:
    • Ethernet is more reliable than WiFi
    • Easier to setup compared to WiFi -> better customer experience
    • Typical use case for the LT100 is in a fixed, already wired setup -> no need for WiFi
  • To add ethernet capabilities, the following technical changes need to be made:
    • Update the MCU from a ATSAM D20 to a D50 series, the new MCU has:
      • A bit more compute power and more RAM
      • More Flash to host the webpages
      • Dual Flash to support firmware update via the webpage
    • MAC & PHY chip
    • Ethernet connector with magnetics

2 Dec 2020

Prototype of the LT100E with Ethernet almost complete. Overall the same concept, and these are the main changes:

  • Some layout changes to fit everything on the board
  • MCU change from the original SAMD21 to SAMD51.
    • 1MB of memory instead of 256KB
    • dual Flash, required for over the air updates
    • Flash protection unit (MPU) to prevent accidental overwrite of the bootloader code
  • MAC & PHY chip and an ethernet port with integrated magnetics added
  • USB port from type B to micro to save space
  • Different relays which have better availability
  • Small change in the power supply configuration, now both 5V and 3V3 available to power the 5V relays
  • Smaller connectors on the back to save space
  • security chip ATECC608B added to support secure boot

Secure boot

As this device will be facing the internet, there are a few security concerns. The risks analysis is as follows:

  • running malicious code:
    • Somebody manages to upload malicious code and the device becomes part of a botnet.
      The risk is low, but the impact is high. To mitigate this risk the LT100E must support “secure boot”.
      Secure boot prevents running malicious code. The software itself will be open source, and if anybody wants to run their own code, that is fine, and possible. To do this, you physically need to open the box and connect a hardware programmer.
  • Password of the web interface is known by others.
    • This is probable, if the same password is used on all devices. The impact is low however. Just annoying.
    • To prevent this, each device will need it’s own unique password, or a forced password change at boot is required

11 Dec 2020

Very short update, the LT100E (with Ethernet) board has been ordered, expected delivery end of December.

Received the boards of the LT100 (no Ethernet), and started verification and software implementation.
One component was incorrect and I had to replace this. The exact root cause is still unknown, currently pending with the
manufacturer. Front panel seems to work correctly, the software logic is nearly done for that. Next up will be the motor driver logic.

20 Dec 2020

First prototype is now working and I am currently using this, great to work the bands using the loop antenna 🙂
Still some pretty hard work ahead but this feels good!

Finished first prototype

30 Dec 2020

Main board

PCB manufacturer informed that the board is ready for shipment. Quite exiting! The last few weeks have been a bit slow due to the holiday season. However I did find time to study the mac & phy chip and got that to work on a development board. And I am now fairly confident that this will work on the LT100E as well. Also started to experiment with the security chip, currently awaiting a development board to help proceed with this.

Demonstration Video

12 JAN 2020

Boards arrived last week. And this week has been used lost on countless software issues in getting the TCPIP stack to run properly on the board. Still some dark days ahead writing software unfortunately…