I was in the shack, messing about with ferrite rods trying to replicate Roger G3XBM’s success using a ferrite rod as a transmit antenna. Meanwhile I was monitoring 14.236MHz, the 20m digital voice calling frequency.
Some audio came through the PC speakers. I didn’t catch what it was, but the callsign OZ1BXN appeared below the DV application waterfall. I waited, and shortly Chris called again. We completed a QSO with 100% solid copy each side. I gave Chris 5 by 8, which was the actual K3 S meter reading, and Chris gave me 5 and 9. Chris said he was running 200W peak power. I had the K3 in DATA mode with the power level set to 25W which is the recommended level to use with a 100W transceiver. Whether that results in an average or a peak power of 25W I have no idea.
Although Chris’s signal was solid copy with no dropouts the audio was rather boxy and I found it more difficult to read than a typical SSB signal. This is probably a consequence of the fact that Codec2 has been designed to use a 1.4kHz bandwidth rather than the 2.8kHz required by SSB. I can understand the thinking behind this decision but I agree with David G8JGO who commented on one of my earlier DV posts that it is a pity the codec wasn’t designed to give better than SSB fidelity within an SSB bandwidth. I think that would make the adoption of digital voice more compelling. I can see many people, particularly skeptics who can’t see the point of digital voice at all, being turned away by the audio quality. Has DV shot itself in the foot?
Having said that, Chris reported that I had very good audio. I had tweaked the equalisation built into FreeDV to give some treble emphasis and bass cut, so perhaps some adjustments are possible.