Early Xmas present arrives!

At long last, after a whole week in the post, my SDRPlay RSP1a arrived this morning!

First thing to do is laugh at the several comments on Eham reviews about the box being 'cheap plastic'.  Well, both those statements are probably true, but that doesn't make it a weak or poor enclosure.  It's just a circuit enclosure, with what seems to be perfectly good RF shielding (I think it has foil shielding inside, but if that's not enough for you, it's hardly difficult to add some more!)

My new RSP1a, with grounding clip!

The SDRUno software is also very easy to understand without instruction and nicely thought out.  Ehammers, of course, criticise this as well! 

Connection of the unit to my computer posed no problem or difficulty at all - you just plug it in and you're off straight away!

The only slight problem I had was getting the sound to route properly from Virtual Cable into WSJT-X.   Also, I usually have my main rig running as CAT-controlled, so for non-CAT use (though the RSP1a can be configured to run as a CAT unit), you have to select 'none' in WSJT-X CAT control.  This has worked properly only very recently, but wouldn't work at all today.  Was this due to the recent WSJT-X update, or something I didn't understand?

It turns out that I had to connect the main Yaesu rig up, run WSJT-X select 'none' for CAT, then switch WSJT-X off.  Then connect the SDR and start WSJT-X again.  Problem solved!

On the audio side, a bit of online digging revealed a useful video that suggested once I had selected Virtual Cable as the audio input stream into SDRUno, you had to press 'Stop' on the receiver, and then 'Play' again.  This solved the problem as promised.

I did notice some very big frequency drifts at the start, but these soon settled down to nice straight WSPR lines.  A couple of software selections probably helped that along, too.  Sensitivity and decode rate seems to be very good at the moment, as this shot of the previous 30 minutes at around 17:20UT today shows (global spots plot of those hearing FR4RX, i.e. just me and Antarctica!)

And on 30m, the RSP1a is doing similarly well this morning (30 minute slot of global VK3DQL spots, back from 09:57UT):

We'll see how well this unit brings me in on the WSPR listening challenge over coming days, though I suspect that will have to wait until any constests this weekend have fallen silent.  One thought is whether the laptop, which now has to act as a fully-featured SDR receiver as well as run WSJT-X and Virtual Cable, will be caused to yield fewer decodes as a result of heavier processing demand.

WSPR Listening - top slot!

Well, I'm not sure yet what part the new, possibly improved WSPR software played in this, but I managed to get to the joint top slot in yesterday's listening ranks!

This is even more remarkable, given that I was off 14MHz and on 50MHz meteor scatter (very little activity, no QSOs) for over an hour and a half last night!

The new, 'improved' WSPR!

Yesterday, I responded to a comment by my helpful friend, PE4BAS, about his understanding that WSPR sensitivity had been improved under the latest WSJT-X general release software.

I thought Bas was mistaken, as I had read somewhere that, other than FT8 and MSK, all other modes were unchanged.

But, it seems that I was the one who was mistaken.  There are a couple of references in some of the new release documentation that clearly say that WSPR sensitivity has been improved.  My error, such as it was, was to believe the following within the 'Quick Start Guide':

"Operating modes other than FT8 and MSK144 are unchanged from v1.9.1"

Well, if we choose to believe it has improved, that is very welcome.  I have been using JTDX for serious WSPR work for a couple of years now.  It seems to me to have been better than the old versions of WSJT-X.  If you could even remotely see a straight, non-aurora affected signal, then it would be decoded.  

Unfortunately, I can't run side-by-side comparisons of the two softwares, so I can only go on long experience and give a subjective view.  It does seem to me that the new WSJT-X might be equal to, and perhaps slightly better than JTDX, though it is far too early to draw any definite conclusions.

What is curious, though, is the claim made as follows by the WSJT-X team in the WSJT-X User Guide:

"The protocol is effective at signal-to-noise ratios as low as –31 dB in a 2500 Hz bandwidth."

Well, I've seen decodes at -31dB for years with earlier software.  I'm pretty certain I've even seen a few down to -33dB, though they are not common.  But -31dB is a daily occurrence.  If -31dB SNR/2500Hz really is the limit of sensitivity, I was doing well in reaching it several times yesterday - and the claim of improved sensitivity might be misplaced.  Either that, or the documentation has fallen out of synchronisation with the software!

One indication that WSJT-X might be better now than before is that I didn't listen for practically the whole daytime yesterday on 14MHz WSPR, yet I still ended up being number 18 in the daily listening challenge!  I can usually end up in the top 4 stations if I listen for the full 24 hours and the geomagnetic field is quieter (wilder conditions then strongly favour lower latitude stations).

I hope that the arrival of my SDRPlay unit - still stuck in the chaotic Christmas post - might help, through the DX challenge, to show if the new WSJT-X reallyis a bit better, and that I can climb a little higher in the rankings!

In the meantime, this table from this lunchtime, looking back over the past few hours, suggests that, whatever the software capability, my station's doing very well - one of only three stations hearing VK4TMZ, and the one hearing him (if only just) the strongest:

Seems to be repeatable, in that I'm the only European station hearing ZS1LCD over the past hour or so:

New FT8 QSOs

Well, after a bit of a strop yesterday about FT8 activity being disrupted by a new version being released by Joe Taylor and friends, I have now settled down!

I managed to install the new WSJT-X on both the laptop and Raspberry Pi 3B+ with no difficulty (other than some strange message about 'user activity', which seems to have no relevance to my operating).

As generally agreed, most people clearly haven't heard of the new release, because about 4/5 or more of FT8 signals are still old-school, and not being decoded either way.

Still, the new mode is obviously going - eventually - to become the norm, so I am sticking with it until more people make the switch.

In the meantime, I was very happy to get a call from an incredibly strong VK7BBB (sadly, another operator with no station information on QRZ.com) on 17m this morning on the new version FT8:

Overnight WSPR report, 11-12/12/2018

Some moderate disturbances in the field locally early last evening, followed by a much weaker disturbance after midnight.  The disturbance saw a northerly swing in the Z component, which doesn't generally produce DX spots, which indeed was the case overnight.

Image: Tromso Geophysical Laboratory.

Mid-evening aurora forecast extending into northern UK skies.  Image: NOAA.

There was plenty of shorter-skip activity at 14MHz WSPR until around 23:30UT, and from then on, a couple of distant G stations, all courtesy of the enhanced auroral activity that took the southern limits of auroral visibility as far as Ireland and Wales. Sadly, it was completely cloudy here, so I couldn't groundtruth the forecast.

At 01:16UT, I heard OH5XO once, at -28dB SNR/2500Hz, and an SM station around 45 minutes later, followed by a series of spots of LA6GH between 03:22UT and 03:58UT, all around -26dB SNR/2500Hz.  The LA spots were probably related to the restoration of the field to quieter conditions.

And that was it until advancing-dawn propagation at 06:22UT, when I heard a couple of G stations, followed very shortly thereafter by HZ1SK, presumably all coming in on very low angle paths.

Reluctant advice

FT8 coding has changed.  We have all been told to change to the new WSJT-X version, which is not backwards-compatible with the older versions for FT8 and MSK (so you won't see decodes from those using the old code).

New version is meant to be the global standard as of December 10, 2018.  At the time of writing, only about 8% of FT8 signals on my waterfall are new code versions.

http://physics.princeton.edu/pulsar/K1JT/wsjtx.html

Solar cycle forecast update: hedging in the extreme?

I'm alwaysinterested in matters relating to the solar-terrestrial environment and, like many hams, keep an active eye on how that environment is changing.

At the moment, we've reached what is almost certainly the lowest point in the trough of activity that is solar minimum.  How long this minimum will last is anybody's guess.

But can anyone make a prediction on the solar cycle that is something better than a guess?  I'm not sure there is.

The key point of information for most of us is NOAA.  I've noticed that it is either extremely slow or reluctant (or both) to issue regular updates on how the solar cycle is going.  It can be years between one forecast update and another.

OK, this is not an easy science.  It is trying to grapple with a semi-chaotic and complex system that always can catch us unaware.

But let's look at what NOAA has done in recently updating its solar cycle progression pages:

Now, I remember once being a key witness in a low-level criminal Court case.  The defence asked me whether it was possible the defendant did not hear me tell him he had hit a house with his lorry.  After a quick mental shuffle, I replied: 'I would have to concede anything is possible, but it is unlikely'.  The defence sat down, irritated.

And so it is I find these very odd forecasts.  The nice red trend lines predict that solar minimum is going to continue - and see ever-lower activity - until the end of 2023!  That seems preposterous to anyone following the Sun in detail.   Since at least early 2017, new cycle sunspots have been clearly seen, and more have been seen since.  So, like in the Court case, whilst the outcome is possible, it is unlikely.

It seems to me that NOAA are simply moving the size and position of the goalposts, yielding no useful forecast at all.

Here's my prediction: we are now in the final stage of solar minimum and, by mid-2019, we will see clear signals that the trend is upwards in activity.  Without warning or explanation, at around the same time, NOAA will simply change their red trend line.