Friday, 22 May 2015

Reverse Beacons, PSKReporter, etc

This month, there's a great article about sporadic E in the RSGB's periodical, RadCom.

I've tried to use the Reverse Beacon Network from time to time, as I have the alternative, PSK Reporter.  But I can't say I've found them very useful, and I have always quickly abandoned them.  Sure, a real-time map of where people's signals are going should be great.

But I think there's a problem for authors and researchers who start to rely on the databases generated by these 'skimming' services.  For one thing, the number of reporting stations is quite small and the geographic spread biased for, largely, reasons of humanity's inability to spread wealth equally.

So, when you don't see spots in some place and then conclude, as many do, that there was some fundamental physical reason for this happening, you run the considerable risk of being led up the garden path to a very wrong conclusion.

This came to prominence when I looked at the discussion about noctilucent clouds potentially being involved in JA-Europe propagation.  Whilst NLC themeselves may not necessarily be the culprit, several other charged 'clouds' associated with them may well be.

However, the discussion in RadCom involved a central German station receiver, and a poor correlation with visual NLC was highlighted.  Unfortunately, whilst central Germany can sometimes see NLC, that location is generally too far south to see them as often as the latitudes, say, of 53-56 degrees north, where they are, from late May to early August, daily or near-daily occurrences.  So, conclusions based on what this station was receiving might be very different to what a RBN or PSK reporter receiver might report back if it were located at a higher latitude. 

This isn't to criticise the article or the author - it's a very thought-provoking piece of work.  But it might inform future analysis, or even promote a greater use of these skimming systems.  For my part, I'm sticking with WSPR, which is just much more user-friendly and intuitive.

Thursday, 21 May 2015

Transatlantic 6m WSPRing

I often wonder, caught up in the excitement of the first 6m seasonal openings, why it is we all get so worked-up about DX that, on any lower band, would be easy using a shopping trolley and counterpoise for an antenna (of course, this has long ago been done!)

Now beaming the USA.  Yes, the sun does occasionally shine up here!

So, I decided this morning to turn my manually-rotated 6m 2 ele quad away from the usual targets in central Europe and, occasionally, northern Africa.  The beam is now pointing squarely at the east coast of the US, which, as you may recall from an earlier post, has extremely good ground gain.  Maybe now there will be some 'proper' 6m DX?

I ran the ARRL HF Terrain Assessment model this lunchtime, entering 50MHz into the equations.  I was astounded to find the output 2-ele quad antenna-plus-ground figure comes out at over +15dBi.  In other words, that 5 Watts I'm sending into the antenna is effectively bumped-up to around 96 Watts!  You can find a calculator for this here.  What's more, that peak gain occurs at just a couple of degrees above the horizon.

Once again, I'm always cautious and cannot say whether this figure is correct, only that I know from operating that the conditions towards the US from here are certainly very good.  Here's the output graph, with blue being the beam heading towards the US, and the red the much less impressive figure beaming due south, towards north Africa, Spain, France and Portugal.

So, in short, if I don't start seeing multi-hop Es WSPR spots across the Atlantic by tea time, I shall want to know why!  Returning to reality, I'll make this summer the transatlantic 6m experiment, and see what, if anything, happens over the long term.  I just hope a few more US WSPR stations come on line soon...

Wednesday, 20 May 2015

Falklands to Anglesey (and vice versa)

15m WSPR was very interesting last evening, as VP8ALJ became the focus of two-way signals between him, the US and Europe, whilst no EU station was hearing the US, or being heard there.  In other words, it was a 'V' formation of signal exchanges between VP8ALJ, the EU and the US.

Nothing very exciting to report, but because sunset at Stanley was just 14 minutes earlier than sunset at my QTH, it was interesting to plot the signals both ways, showing a pretty sharpish drop and then loss of both signals.

Monday, 18 May 2015

Me vs. The Rest of the World - on 15m

As I'm always keen to point out on this blog, amateur radio doesn't need to be expensive.  The station here is very modest, more especially at 15m, where the antenna I've been using for many years is 'just' a vertically-polarised delta loop made of hard-drawn copper.  It's corner-fed with a 4:1 balun at the feed point, which is theoretically non-ideal in terms of the radiation pattern (less rejection at higher angles.)  In practice, I've always found there is no real difference between corner and 1/4 wave-up-a-side feeding, provided you are using the antenna on its cut band.

As you can see from the infrared photo below, the delta, on the right and propped-up by a pole off an old chimney, is hardly in the most isolated situation possible, being within about 5m each of a 12m beam and tower, a 2-ele 6m quad, and a 20m vertical!

As soon as I put up this delta one rainy Christmas holiday morning, it became clear my experience of radio had changed dramatically in comparison to the earlier, sloping long wire from which the delta had been made.  The DX was now easy, whereas it was more hit and miss with the LW.

The 15m delta is visible at right, apex supported off a pole on an old chimney.  Hardly an ideal environment!

Recently, I haven't been on 15m much, but conditions there were quite good this weekend, so I ran some WSPR transmissions towards the end of the day.  The time range in question spans 18:52UT to 19:08UT on 17/05/15, so well before sunset at any point in the UK, so the greyline enhancement would not have occurred yet for any of the stations in question.

What happened next was remarkable - even for a station based on an old copper mine ridge, 320 feet up in the air. 

The receiving station was K9AN.  By any standards, K9AN runs a good quality station from what you can see from Google Streetview, is a pretty clear, rural location where the band is clear of most interference.  Anyone who's a regular on WSPR will know K9AN is a reliable, reference station in the US.

Now, my 5W from a delta loop with a base 2m off the ground, and an apex at roughly 6m, was hitting K9AN at +2dB.  How were others doing?

Predictably, my friend Ken Franklin, G3JKF, was the second strongest signal to K9AN at the time, coming in at -13dB, also from 5W to a very -carefully researched and maintained magnetic, 3-loop array which is very often equal in performance to full-sized antennas.  So, a difference there of 15dB.

When I looked at others, G4KYA was at -21dB (5W), G8VDQ at -23 (5W), GM4SFW at -20 (1W) and G3TXA at -22 (1W).  These are enormous differences in signal strength compared to mine, even allowing for the output differences of the 1-Watters.  I should also point out my spot was repeatedly as strong, at +1dB, a few minutes earlier.  It's a shame I didn't start earlier in the day, actually.

Now, there is no question that this QTH benefits from (a) lack of man-made development (b) ground conditions that are, from the point of view of how many other stations are so situated, unique, and (c) a modest elevation that nevertheless puts it above the rest of the landscape, with clear sea views to the Americas.

According to the ARRL HF Terrain Assessment software, the model I've carefully prepared of the landscape towards north America yields a antenna-plus-ground gain figure of +8dBi at an elevation of 17degrees.  Between 2 and 5 degrees, it's about +6dBi.  Note that, because the software only deals with horizontal antennas, I've been forced to input a dipole at half a wave up.

Clearly, this model does show a very good gain figure for a dipole, reaching values more indicative of a 2 element beam at very low angles of radiation, and a 3 element beam at the still-low elevations around 12-15 degrees.

What it doesn't do is explain the enormous differences in received WSPR signals by K9AN.  One might be tempted to look at 'patchy' propagation conditions, where a cloud of ionisation has fortuitously aligned itself between me and K9AN.  But, because of the spread of other stations spanning from the south east of England to the north of Scotland, and that those signals were of very comparable strength at K9AN, this explanation seems to fall flat on its face.

How software models a dipole at half a wave above ground, at 21MHz.

Now, how about we model a vertically polarised delta at the real height, with real (or as best as can be estimated) ground conditions input into the mix?  Here's what we get:

With real ground conditions: +4.5dBi gain at 14 degrees elevation.
With perfect ground conditions: +7.63dBi gain at 0 degrees elevation
Free space: +2.71 dBi.

Does this really tell us anything?  Well, let's assume that, based on what we know about the QTH here, that my signal of +2dB, coming from an undeveloped, clear site with sea horizon views, is close to what K9AN should receive at 5W, at that time of day.  This must mean that other stations are, for reasons of their environment, probably above all else, losing a fantastical amount of energy to their surroundings. 

This is why, of course, much cleverer radio people than me made a point of buying houses on hillsides wherever possible; the late Les Moxon, G6XN, was one of them, and often cited SSB contacts with VK using milliWatts of power from simple verticals when working from such locations.

As usual, I'd be interested to hear of any considered input into this.  My enduring frustration is being unable to properly assess the ground conditions; according to the ARRL, the highest values for ground dielectric - much higher than seawater - have been found at highly-mineralised places - such as copper mines!  It seems they got their measurements right, anyway!


Just to keep things balanced, here's the ARRL HF Terrain Assessment for my QTH, but to the south, where, at about 42km, lie the Llyn Peninsula (southern Snowdonia) mountains, only a couple of thousand feet high.  Even so, and even at that distance where they are only about a little finger's width above the horizon, they have a marked effect on low angle signals, taking them into negative territory for a while.

Although this gentleman's talk is a little long, it does neverthless get to some important points, and worth watching from about 23 minutes on.

So, it's not all good news here.  Maybe this effect of hills explains why so many "kiloWatters" exist in California, surrounded by vast mountains to the east?  I'm too lazy to build a terrain model for that!

Distant hills bring considerable gain penalties at low angles...

Saturday, 16 May 2015

The Last Helix...

About 30 years ago, James Miller, G3RUH, started fabricating helix antennas in kit form for the amateur market.

Somehow, by a stroke of unusually good luck for me, I came across James' web site a couple of weeks ago, which was last updated at the end of 2014.

The helix in its 1980s heyday - together with the copper element of kit 170 - the last one - ready for assembly. 

After a few e-mails, I found James still had the kit for a 16-turn antenna.  Being over 3m long, there was no way that would work at this hurricane-infested QTH.  So, the 9-turn kit arrived yesterday, in 3 packages, leaving me with a fair bit of careful assembly to get on with one rainy day.

So, a big thanks to James for selling me Number 170 - The Last Helix.

Thursday, 14 May 2015

6m Up and Running

Well, there have been very few spots on 6m WSPR last few days, but today is showing more solid Es formations.  Later in the day, full openings on SSB were running for lengthy periods; I managed 12 QSOs into central and eastern Europe over just a few minutes at lunchtime today.

The Es season is OPEN!

Tuesday, 12 May 2015

'Seeing' Sporadic E Clouds.

The 6m sporadic E season is pretty much upon us, with good hints of short-lived Es clouds over the past few days.

Whilst I watch the WSPR screen carefully, it's often seen that signals sometimes follow strange deviations by Doppler shift.  Most times, this is down to reflections from passing aircraft.  But not always.

If you've ever made an SSB QSO on 6m, or followed WSPR traces during the peak of the season, you will know that Es propagation can come and go in seconds, but also last for a few hours.  Oftentimes, the signal comes and goes in gentle wave-like patterns.  Overall, the Es paths are quite complex, even if the underlying mechanism is relatively simple.

Polar mesospheric summer echoes at 46.5MHz - loose links with NLC and Es.

Now, the ionisation that leads to Es is at a height of 90-120km.  This isn't visible, other than by radio reflections.  Noctilucent clouds, on the other hand - which also occur in the summer months (but do not appear at all in winter, unlike Es) - are readily visible from mid-high latitudes (~40-65 N and S.)  Like Es, NLC also drift east to west in the summer.  NLC occur somewhat lower in the atmosphere, at about 80-90km.

We also see reflections at 46.5MHz from polar mesospheric summer echoes - PMSE - at the same height as NLC.  PMSE is caused by ionisaton, and NLC form around charged, metallic meteor debris.  Clearly, all these phenomena are linked to a greater or lesser degree.

Noctilucent cloud over the Irish Sea, 2009.

Whilst NLC are not always linked to Es occurrence, there is a loose correspondence.  The point of this post is to highlight the likely similarity in the form of Es clouds with the visible NLC.  From there, we can start to appreciate how these strange changes in propagation conditions occur, and what kinds of structures cause them.  No doubt Es clouds are modulated pretty much in the same way as NLC - dominated by the breaking of gravity waves propagating up from the lower atmosphere. 

The best thing to do is watch a video of NLC, and think about how propagation characteristics in Es might be understood in terms of the kinds of structure and reflection chracteristics evident in NLC.  Here's a good example, showing plenty of evidence of breaking atmospheric waves later in the sequence.  Remember that almost all the apparent change in cloud brightness and extent is due to changes in the sun-cloud-observer angle.

lapse 2014June19 20 from John Rowlands on Vimeo.