Friday, 7 December 2012

Software Defined Radio (SDR) - First Steps

I have often seen adverts in the amateur radio press for Software Defined Radios (SDR) and wondered why they are so expensive. During a discussion at my radio club (Cray Valley -, I learned that a TV/Radio Dongle receiver can provide the hardware for SDR. I found a suitable dongle (RTL2832 plus either R820 or E4000 chip set) on Ebay; ten UK Pounds from Hong Kong so, being the last of the big spenders, I went ahead and ordered. While I was waiting for it to arrive, I downloaded a copy of SDR Sharp  (SDR#) and the installation instructions ( ).
There are some hardware things that I needed to do while waiting for the dongle. The dongle has a tiny MCX antenna plug. Fortunately Maplin sells an adaptor cable to standard TV socket (part N59LN) and I had already made a TV-BNC cable for another project, so this was not a problem. These dongles are very vulnerable to high levels of RF, so if you transmit while receiving on the dongle it will fry. The solution is to solder a couple of reversed diodes into the adaptor cable to limit the maximum input voltage.

The dongle arrived. The package also contained a remote control (only useful if you want to use the dongle for its intended purpose, but being radio amateurs we don’t do that do we?), a whip antenna (which works surprisingly well) and a software disk. Do not install this software as this will conflict with SDR#. The only tricky bit was changing the dongle driver from the Windows default to RTL 2832. To do this, you need a piece of software called Zadig (link on the SDR# site) and I also had to temporarily disable Heuristic protection in Norton which took violent exception to what I was trying to do.

All set, and Norton back to full protection, I plugged in the dongle and fired up the software. Operation was quite intuitive, but I started with the FM broadcast band – all was well and all the expected stations showed up on the display any gave me some strong, predictable signals to set up and learn how to use the software.

The dongle, with the E4000 tuner, covers from about 24MHz up to 1.7GHz, so it is a broadband device covering all the amateur bands from 12m up to 23cm. That’s the good news; the bad news is that there is no front-end filtering. Also these dongles mix the RF signal down to a few tens of kilohertz, to input to computer sound cards, so spurious signals are a real problem, particularly on the lower bands. For example I get a strong ‘Classic FM’ signal in the middle of the 10m band. It is possible to manage these phantom signals by carefully choosing the ‘Centre’ frequency and adjusting the RF gain, which can be accessed through the ‘Configuration’ menu, but my next move will be to build some filters to clean up the input signals.

I have also experimented with receiving  some digital modes using this SDR receiver. It is possible to patch the microphone socket to the speaker socket on most PCs so the digital modes can be decoded with standard software. There are a couple of drawbacks: you can’t hear the audio signal and the signal gets converted from digital to audio then back to digital again. There is a better solution – a piece of software called Virtual Audio Cable ( ), which does what the name suggests, but without the Digital/Audio conversions. With VAC, it is possible to ‘split’ the system so the audio output can be heard, recorded or shared with other programmes. I have successfully decoded RTTY, PSK-31 etc.

My Dongle SDR has several advantages: it is very small and everything is controlled from the keyboard so easy to configure. For example, filter bandwidth can be adjusted very precisely and easily.  The waterfall display, which can be adjusted, allows large segments of a band to be viewed at once and you can see when stations come and go then click on the ones you want to listen to. Combining the SDR with the VAC software provides a very flexible setup. For example, I can operate digi-modes while listening to my favourite music on the same computer. 

This SDR radio will never be as good as a full communications receiver, but it was definitely worth a tenner and I hope its performance will be even better with some front-end filtering. It has opened lots of opportunities for experimentation that were not available to me before and after this initial experiment, I may be tempted to invest in an SDR transceiver, but that is for another day.

No comments: