Measuring the Solar Flux Index at home ...

Foundations of Amateur Radio

Recently the Australian Space Weather Forecasting Centre issued an alert for a Coronal Mass Ejection or CME expected to impact Earth within 24 to 36 hours. This was presented within the context of seeing the resulting Aurora, but as a user of the HF radio spectrum, I'm subscribed to their email list, not for the pretty pictures, though I would be delighted to actually see them with my mark one eyeball, I'm on the list for the impact on propagation for my hobby.

As a good citizen I shared the alert with my community both via email and social media and as a result I received some questions and comments. One question was, "What does this mean?", one comment was "it's not going to impact the United States."

My response was to point out that HF propagation and the impact of the Sun is a very deep rabbit hole and encouraged further research by supplying several links, including a very detailed video by Rohde and Schwarz titled "Understanding HF Propagation", very, highly, recommended.

Whilst watching that video I discovered that the Solar Flux Index is measured using a receiver tuned to 2800 MHz or 2.8 GHz. Being in the business of having receivers scattered around my shack, I asked myself if I had something that was able to receive on that frequency. My RTL-SDR dongle doesn't cut it without extra hardware, it tops out at 1.75 GHz. However, my PlutoSDR has a standard frequency range that goes up to 3.8 GHz out - of the box - and with some tweaks can make it to 6 GHz, so well and truly within range.

Now, before I move on, I should mention that an RTL-SDR is a cheap, as-in $20, USB computer accessory that looks like a thumb-drive and is ostensibly built to receive digital television, or DVB-T signals. I've spoken about this previously. It can be used to receive radio frequencies outside the purpose it was built for. The PlutoSDR, or to give its official name the ADALM-PLUTO, on the other hand, something which I've also spoken about, is a single board Linux computer made by some smart people at Analog Devices, specifically for the purposes of learning and experimentation with receiving and transmitting RF. It comes with all manner of documentation and software and to be honest, I'm a little bit in love with mine.

Back to measuring stuff. In this case I'm attempting to measure the power levels of radio frequencies at 2.8 GHz. I know of a simple tool called rtl_power that can measure RF power over time and started investigating if that tool had been hacked to be able to use the PlutoSDR, rather than the RTL-SDR dongle. It might have been, but I've not yet discovered it, however, that in turn led me to several other tools, most of which I'm still investigating.

What it does tell me is that I'm not the first person to tread these paths, much has happened and been documented in the analogue sphere, some has been done using digital I/Q data and a transverter, a device that can multiply radio frequencies to make them appear in a different part of the radio spectrum, but I'm not yet sure if anyone has made a Solar Flux Index device out of a PlutoSDR.

I recalled a wonderful little tool that I've also talked about before, there's a theme here, I'm sure, but the tool, "csdr", written by Andras HA7ILM, which allows you to do all manner of interesting things to a stream of raw data, specifically RF raw data. It has a function called logpower_cf which Andras describes as "useful for drawing power spectrum graphs", which is precisely what I'm looking for.

Armed with that I'm now in the process of building a compiled version using Docker, so I can run csdr on my PlutoSDR and perhaps generate a power spectrum graph for 2.8 GHz. Of course that will now require that I learn how to extract raw data, known as I/Q data from the PlutoSDR command-line, process it through the logpower_cf function, output an image and hopefully show the result as a web-page. At the moment I'm still in the weeds with a Makefile, but that's not unusual.

Needless to say that I'm working on it and the result will no doubt turn up on my github page when it's done. In the most innocent terms possible, how hard can it be?

One takeaway that really needs to be expressed out loud, even if I've hinted at it. All the bits I've talked about here are things I've already been playing with. It wasn't until I came across a salient piece of information about the Solar Flux Index and how it was measured that all the puzzle pieces came together, the PlutoSDR, csdr, rtl-power, logpower_cf, the SFI and a web-server, that I could even imagine this happening.

The point being that even if you have all of the puzzle pieces in your hands, it might still take one missing piece of information for your brain to go "Ah-ha, wow, yes, this makes sense, I can do this."

So, keep collecting puzzle pieces, relevant to your own interests and one day you too will get to this point.

I'm Onno VK6FLAB