Foundations of Amateur Radio

The other day I went looking for a software defined radio or SDR for HF. This happened because all such devices on my desk are rated at higher frequencies and I've still not managed to fix the broken SMA board connector on the transverter I purchased over a year and a half ago.

In case you're wondering, the design has two SMA connectors attached at either end of a printed circuit board, also known as a PCB. The board slides into a metal case and both connectors are tightened to either side of the case, which causes the problem when the circuit board is slightly shorter than the case and the nuts pull the connector apart, causing the device to fail.

Replacing the SMA board connectors would be relatively simple, but they appear hard to come by and the micro SMA connectors that a friend purchased to help, changed the task into finding adaptors, which I've not managed to solve yet.

I'm detailing this all for a purpose, trust me.

Anyway, the hunt for an SDR for HF lead me to a project called "Radioberry". It's a design by Johan PA3GSB which is designed to be a so-called "hat" for a Raspberry Pi. Think of it as an expansion card to create functionality, in this case a radio capable of transmitting and receiving on HF, covering 0 to 30 MHz, perfect for my current needs.

The design uses a Raspberry Pi computer to power and control the board, including programming the on-board FPGA, accessing the actual data and sharing that with the user, either via a touch screen, or using USB, Ethernet, Bluetooth or Wi-Fi. The board itself has two external connectors, one for transmit, one for receive and when you combine it with the Pi, fits neatly into a box which you could 3D print. Amplifier and band filters are left as an exercise to the enterprising amateur, though there is an amplifier design on the github repository. If you're curious, it's based on the work by the Hermes Lite 2 group.

Johan specifically doesn't sell this device, instead you can choose to buy it from other enterprising individuals, or better still, build your own. Over the last few years I've started noticing several people in the so-called maker community, people, who a lot like radio amateurs, build stuff for fun, using online printed circuit board services.

If you're unfamiliar with the concept, you can design a schematic, layout a PCB, have it manufactured and optionally even built and sent to you. To get an idea of what this might look like, I picked a random online supplier, uploaded the specifications for a Radioberry and costed the whole thing. Suffice to say that the biggest charge is the $50 set-up fee.

Any enterprising engineer would have punched the "Buy Now" button and be done with it, but in some things I'm pretty cautious, so I haven't, yet. I don't know enough about the design or schematic to know how it works, to troubleshoot it, to fix any potential issues, or even to know what kinds of issues there might be, even if they're obvious to anyone with electronics experience.

To make it clear, my electronics experience is rudimentary at best. I'm comfortable with block diagrams, understand the basic principles behind most passive elements, but if you're going to get into trace length and signal timing, I'm not anywhere even remotely qualified to troubleshoot, let alone spot problems. That's not to say that I am stopping before I start, the opposite is true.

I'm using this as an experience to gently get my feet wet.

Back to the apparently too detailed explanation of the transverter. Joining the dots you can probably guess where I'm going with this. Given the access to countless documented transverter designs, I feel comfortable enough to work on a design, construct a PCB and have it manufactured. At the rate I'm going, that should get a solution before I can find a PCB edge-mounted SMA connector, well, at least that's my excuse. I'm also eyeing off this same process to build a logging volt meter, since the Internet seems to believe that I should pay hundreds of dollars for a volt meter and an I/O port, even if the chip inside costs all of $6.

Oh, the transverter I purchased a year and a half ago costs three times as much as having five of them built on demand, so there's that.

For all my life I've been a firm believer in software. I've also been on a computer driven manufacturing journey for a couple of years, still in the process of commissioning my new toys, much to the merriment of some of my fellow amateurs and the idea that I can have a circuit design built and shipped to my door just makes me tingle with anticipation.

If you're already ahead of me on this journey, please don't hesitate to point at any potholes on the road and if you're following along, if you break it, you get to keep both parts.

I'm Onno VK6FLAB