REE DRM @ 9630 kHz, 18 Aug 2012

Around 1910 UTC someone must have tripped over the transmitter power cable.

At 1938 UTC, transmission resumed. Man, the noise floor at 31 m is atrocious. -100 dBm? The co-channel interference doesn't help any.

The Disco Palace DRM @ 17875 kHz, 18 Aug 2012

I think signal was strongest I've ever seen from this station, with non-pilot subcarrier hitting -75 dBm.

What's interesting is, there is clearly some energy bleed into the region beyond its 10 kHz bandwidth. In the past I chalked this up to intermodulation in the antenna, but I'm starting to reconsider. There are plenty of strong double-sideband stations (all of the 31 m band) that don't exhibit this behaviour.

Here it's easier to see that fades start from outside the signal bandwidth and cross through from one side to the other over time. Intermod in the receiver chain would mean that fades outside the bandwidth start at the same time as fades inside.

What is probably happening is the transmitted spectrum mask is not 55 dB down, but more like 40. Usually this is still below the noise floor, but if subcarrier power exceeds -80 dBm the "skirt" becomes noticeable.

Also, check out this 9800 kHz Vatican DRM broadcast where, upon review, the signal mask is only about 35 dB down.

The Disco Palace DRM @ 17875 kHz, 12 Aug 2012

Forgot to add this last week. Audio stream was present, but the song title service was still broken. Last broadcast it was Rapture by Blondie, and this broadcast was the same.

The Disco Palace DRM @ 17875 kHz, 11 Aug 2012

Decent reception, but no audio. Usually if there's even the existence of an audio stream there will be a spectrum around DC at about -88 dB. There isn't even that.

That song title has been stuck for at least 10 minutes too.

RTTY notes on 20 m

On many occasions when listening on the amateur radio bands I rarely hear a complete exchange on digital modes.

Case in point, around 21:30 UTC K4ADR (Florida USA) comes in on 14088 kHz with -85 dBm signal power, and makes the following contacts:

• SP4EOO (Poland)
• SP5ECC (Poland)
• RW0LIF/6 (Russia)
• AD2Q (USA)
• OZ5NJ (Denmark)
• DG4NAT (Germany)
• IZ1UJM (Italy)
• SQ7EQB (Poland)

And I only know because K4ADR copied back those, except for DG4NAT whose reply I did hear, and maybe a IZ0PMV that was later on another frequency.

Maybe I'm just not on beam and all those operators have directional antennas pointed towards Florida. Or maybe I need even more gain than I already have, but that's not going to happen any time soon.

The Disco Palace DRM @ 17875 kHz, 6 Aug 2012

Looking pretty good, but I had to expand the demod bandwidth by 200 Hz because the centre was drifting.

I forgot to show the time, so you'll have to take my word for it, but it was around the same time. The center continued to drift, and by the end it had drifted about 191 Hz.

This isn't a big deal; 10.5 kHz demod bandwidth should be sufficient and Dream successfully tracks the drift. But starting out off by 138 Hz and then drifting about 50 Hz over the rest of the transmission doesn't scream stable mixer.

TDP and RNZI DRM off-center, 4/5 Aug 2012

This afternoon was pretty good for The Disco Palace until about 20:30 UTC when a deep and prolonged fade struck, and audio remained choppy until 20:50 UTC, wherein it finished strong.

Anyway, earlier I noticed that the right edge of the signal seemed pressed right up against the edge of the demod sent to Dream. It didn't really cause a degradation in Dream's reception, but I think it's normal to have the signal roll off contained in the signal bandwidth passed to the decode chain; it's what transition bands are for.

So I shifted the demod frequency up. Closer inspection revealed that the suppressed carrier was about 176 Hz higher than expected.

I also saw a roughly 150 Hz shift in RNZI tonight, on 17675 kHz.

But RNZI on 11675 kHz is pretty dead on. Unfortunately decode is spotty. I've had better signal strength on 25 m before.

Noise spectrum around 9330 kHz, 21:15 UTC 4 Aug 2012

Usually I don't loiter around 31 m in the afternoon, especially now that there is no DRM from Sackville on 9800 kHz, but during a quick scan I came across this weird noise structure. Can't imagine it's some electronic interference as I couldn't find any harmonics.

While I'm here, some comments about WBCQ out of Monticello Maine. From the waterfall, it looks like it's trying to do upper sideband with carrier. But there's still a visible LSB, about -15 dB relative to the USB. Unless it's suppressed by like, at least another 10 dB (-25 dB total), I don't see a point other than to troll anyone who doesn't have synchronous / product detection.

Then again, most of the content comes across as a giant troll, but it pays the bills so whatever.

Also, the carrier is not even on 9330 kHz. It's usually about 33 Hz higher, but it can sometimes be below target. And lastly, it experiences complete loss of audio for several seconds at a time, as if a resettable fuse is constantly tripping. Run more bible thumping anti-science/anti-government/anti-anything programs and get better broadcasting equipment. Or don't.

RNZI DRM @ 17675 kHz, 4 Aug 2012

Last night I thought to check the signal periodically over the course of two hours. At the beginning there was just noise, and then the faintest indications of a DRM signal on the waterfall, and from there it just slowly rose out of the noise until what you see here. Propagation conditions were really good.

But the other interesting thing was how even though the signal power didn't change much, I'd still incur periods of frequent frame loss due to a fall in SNR.

In the demod spectrum on the right, the nearby noise power is about -115 dBm in the second image, versus -120 dBm in the first. It did translate neatly to a 5 dB difference in SNR reported by Dream, which is all the difference between no frame loss to bursts of frame loss. The funny thing was the transition was sudden, like a switch, and it was being toggled every six some odd minutes.

I tracked it down to the air conditioner. So the A/C raises the noise floor by 5 dB across at least the 16 m band.