15m Magloop repair.

Some spare time to attend to a repair of a parallel 4-loop array for the upper HF bands today, prompted by some screen-burningly strong WSPR signals on 12 and 10m.

The loops had been soldered together onto lengths of round copper tube.  I had rushed the build and not made a very good job of it.  My failures were:

(1) Round copper tube doesn't offer much surface area to solder onto another round tube.  It's better to flatten everything and make a better joint.

(2) I hadn't paid enough attention to putting it all together in one go, with all the copper cleaned and soldered at the same time.  The soldering was, predictably, not very good.

(3) I didn't use any additional mechanical securing system.  Bolts help to keep everything stable whilst soldering, and help strength when complete.

Rushed into operation, but not very well made.

 

So, I flattened the end of the tubes, all cleaned to perfection with a rotary tool with abrasive wheel attached, and fixed these to a flattened, cleaned 15mm copper strip on either side, initially using 4mm stainless bolts.  Make sure you don't use 'Nyloc' locking nuts at this stage, because the plastic will melt and contaminate your soldering later on!

All I had to do now was solder it all up.  The key to this kind of job is to use soldering flux and keep the flame on the joint being soldered.  You can do it without flux, but if you remove the flame even momentarily, oxides will form immediately on the surface of the metal, and you won't get a good - or any - joint.  

Soldered and bolted.  Back in operation, soon!

After the whole thing had cooled down, I removed the nuts and replaced them with self-locking 'Nyloc' nuts.  Whole thing now nice and secure, with good electrical joints as well!

Yagi vs. Vertical

Doing the research so you don't have to!

An interesting period of FT8-based RX-only comparisons between my 3-ele monoband LFA at home, 19km inland in the direction of South America, to which the antenna was pointed, and a 1/4 wave vertical at the beach.  Made possible by the very useful 'ALL' file record within WSJT-X and JTDX of all activity.  Both used the same ZLP interface and latest edition of WSJT-X.

 

Results for this band are not the same as lower down the HF bands, due to the Es propagation mode that is dominant.  As you can see, the results are very interesting, with wide variations in the signal levels, and in relative performance.  EA8BFK is included as it is on the beaming path, and a nice, shorter-skip comparison:

EA8BFK receptions, 24MHz FT8.

 

PU5LTI receptions, 24MHz FT8.

PY4EV receptions, 24MHz FT8.

12m at the beach

Well, my wife and daughter wanted a swim in the sea last evening.  So I came along to play some radio on dry land!

I was meant to be comparing reception at the beach with a vertical using FT8, and my 3-element LFA at home.  Unfortunately, simply by checking the ALL.txt file at home, the software stopped recording reception reports!  It seems one has to stop and restart WSJT-X after interrogating the ALL file.

Anyhow, all not lost.  Reception was very strong from Latin America, up in positive SNR territory, even for the most distant stations.  My ~3W even managed to reach LU8ENU, albeit at a meagre -17dB.  At CU2CE, in the Azores, I was a much healthier -2dB.  Of course, with Es propagation, the sea doesn't necessarily provide the same enhancement as lower down on HF, because it depends on the position of the first Es cloud.

 

Unhealthy.

There's a fine line between persistence and simply not getting the message that the propagation was not yet open to Italy.  Just a small part of this man's obsessive, repeated calling over about half an hour this morning on 12m:

Time to try something else, IZ1ERR...

 

Now that's what I call PMSE!

We are less than a month from mid-summer.  And just above the Arctic circle, things are getting very busy.  Es and noctilucent clouds are now pretty much with us again:

Big PMSE formations all day today.

A NLC display, with Comet Neowise, in July 2020.

Digilink Nano - WSPR results.

I've had a chance now to run a fairly limited test with the Digilink Nano, despite the temptations of a very busy upper HF and VHF period, where making 'real' QSOs competes strongly for attention!

The WSPR Challenge result for the day was number 15, which is of course a 'total station' assessment, and does not necessarily reflect the performance of the interface.  Tomorrow, I could be number 8, or 5, or something else; it depends critically on many factors.

 

I had believed at the time of writing this post, that the Nano did not have a soundcard of its own.  The maker of the Nano - it has to be said a very long time after having been first been made aware of this blogpost - today (14/7/21) corrected this belief as follows: 

'DigiLink Nano is an autonomous external USB soundcard and is completely independent from the onboard soundcard. In fact uses the very same chip to do this as the ZLP devices, the TI PCM2912. As such, I'd expect similar noise levels. For full disclosure, I’m HB9ZHK, the creator of DigiLink Nano.'

Now, ZLP has consistently advertised his interfaces as having low noise, and I certainly have no complaints about how reliable and apparently low noise they are. He asserts that his interface has a noise value "better than -93dB", contrasting this with the Tigertronics Signalink noise figure of "-71dB" (ZLP site accessed 07:28UT, 25/05/2021).  I can't say whether this is an accurate statement, but a search online does seem to reveal the Signalink did, and may still have, noise issues that led some to make modifications, which were apparently successful.

The 'D' layer forms early, now it's summer, so a quick and useful test was to look at a few short-haul spots and then switch back to the ZLP interface for a comparison.  I know that, under prolonged testing, I am typically in the top 10 or better; sometimes I get to number 1.  So the ZLP is certainly a quiet circuit.

I looked at a number of stations, and present the results for UR5KHL, where spots before 06:34UT were the Nano, and for the ZLP thereafter.  This had to be a very short test, because propagation might change quickly, especially in the morning:

As you can see, there is no difference between the interfaces.

Another typical result, showing again no significant difference, is as follows:

And then again, for F6KOP, no significant difference:

With this particular set of equipment, there seems to be no significant difference between the ZLP and Digilink Nano interfaces. Note that both sets of computer-to-interface cables had a TDK, HF-rated split ferrite fitted (two turns).

What you do certainly need to look out for with any interface using a USB connection to a computer, whether it is using its own soundcard or the computer's, is RFI conducted along the cables.  Even if you can't hear/see RFI on one band, the chances are it will be there, degrading RX performance.  Use good quality, appropriate-for-frequency ferrites everywhere!

With both interfaces competing at the same price point, and both now having no on-board physical level controls, the question of which interface to choose is perhaps just a matter of personal preference, and maybe the fact that the Nano is considerably smaller and lighter than even the latest ZLP unit. 

 

New digital interface.

Lots of progress with radio-related technology this week, as my latest gadget - the Digilink Nano audio interface - arrived from Switzerland.

This was a 'let's see' purchase, as I had become irritated by ZLP Electronics' fine interfaces no longer featuring on-board in/out volume controls.  So I accepted that this is the way of interfaces in general now, and decided to look at alternatives, hopefully with noise levels similar to, or better than, the ZLP units that I've used heavily every day for so many years.

Small!  The digilink and supplied cables (you can choose cables for your particular brand of radio when buying the interface).  Coin is Peruvian 1 Centimo, 16mm diameter.

 

Despite potential Brexit delays, the Digilink arrived in just over a week.  The price for the unit works out much the same as a ZLP unit, so the question of which to use is down to noise performance and size - the ZLP is much larger and heavier.  Early WSPR testing, which will last for a day or two here, shows promise.  For example, all reception of 9Z4FV this morning so far:

[Update] A longer test undertaken the following day can be found here.

What I can say for now is that the Digilink is pretty easy to install, even on a Raspberry Pi, which has somewhat more involved processes than Windows machines.  After a very quick package download, the Pi was seeing the Digilink immediately.  Connecting to my Windows 10 laptop was entirely 'plug-and-play', with no need for any downloads at all.

Volume control from the Pi is now pretty easy, using simple Windows-like levels software.  The only problem I had was a 'senior moment', which I have quite often with the FT818, in that I had the wrong transmission mode set on the radio, so the rig was keyed, but no signal was appearing.  Once I woke up from my stupidity, the output power level controls on WSJT-X and JTDX worked as expected.

ZLP Datalink (no longer produced), compared to the Digilink Nano.

 

The online support pages are pretty comprehensive, extending not only to installing the Digilink, but also the various software packages we use.  There is personal support in the event of difficulty.

I think the seller has a nice interface, very small, neatly packaged and well-supported on an attractive web site.  It's not especially cheap, but if it has low noise, it's pretty much the going rate for a good audio interface.  Data on the Audio codec is found here.

In fairness to ZLP, though, I can see why he has ditched the on-board level controls now.  Whilst I still much prefer those controls because they are simpler and faster to access than software, level control from the computer is easy, and not quite the irritation I had expected, particularly with the Pi.

Time updates in no-signal areas.

If you like /P work as I do, then it won't be long before you find yourself in a place that has no mobile data signal.  Here in Wales, though coverage is very good, locations such as low down on a beach, behind sand dunes, and in the middle of mountains, inevitably mean no data.

Weak and intermmittent mobile data signals, and time updating, is a problem in many locations.

 

This is a real problem with the Raspberry Pi, as it has no internal real time clock; it gets its time from the internet every time it is switched on.  If you have no internet, you will be stuck with the time it was last switched off, and unable to update it.  That means you can't decode most digital signals, as the Pi's clock will be unsynchronised with the transmission cycles.

I decided I would get a cheap GNSS dongle off e-bay, which arrived direct from China in the typically admirable time that puts many UK-based suppliers to shame.  It cost me just £6.42, including delivery!

3B+ with a £6 GNSS receiver.

 

With a bit of help from this excellent video and associated document, I had no trouble at all getting a GNSS signal on the Raspberry Pi 3B+ that I previously used with my portable radios.  The time updated nicely, and can be forced with a manual command, too.  

For future reference, in case the video vanishes, here is the code:

#Turn Bluetooth off on Pi

#Install Updates 

sudo apt-get update 

sudo apt-get upgrade  

#see USB devices

lsusb 

 #Install GPS software 

sudo apt -y install gpsd gpsd-clients python-gps 

 #Edit GPS config file 

sudo nano /etc/default/gpsd 

 #Add this to file 

START_DAEMON=”true” 

USBAUTO=”true” 

DEVICES=”/dev/ttyACM0″ 

GPSD_OPTIONS=”-n” 

 #Install chrony 

sudo apt-get install chrony

 #reboot pi 

 #check to see if services are running 

systemctl is-active 

gpsd systemctl is-active chronyd

 #Check GPS output  (cgps -s probably won't in fact)

cgps – s

gpsmon -n 

 #Edit chrony config file 

sudo nano /etc/chrony/chrony.conf 

 #Add this to end of file 

refclock SHM 0 offset 0.5 delay 0.2 refid NMEA

 #Check Chrony Output

 sudo chronyc sources -v 

 # Check chrony output 

sudo chronyc tracking

 #Force time sync 

sudo chronyc makestep 

 

Useful links: Mike Richards G4WNC https://photobyte.org/raspberry-pi-st...

Good morning!

Always nice to switch on before breakfast and get a good DX contact!  The signal levels changed very rapidly, as they often do on the long path - though this could also have been a short path route:

Not bad for ~35W and a vertical delta!  Initial repeat calls from me were due to TX frequency changes on a very crowded band.

Grey line still works.

With a lot of unseasonably strong wind yesterday, I set the rig to RX-only for 14MHz WSPR overnight.

Nice to see a very clear effect of the passing greyline period between 9Z4FV and Wales:

Situation at peak signal(22:30UT) from 9Z4FV, 14MHz. Image: DXAtlas, with permission.

Reception SNR of 9Z4FV's 20W 14MHz signal.

Could be a good Es day...

Wow!  PMSE keeps getting stronger and stronger.  10m WSPR signals are correspondingly strong, too:

5W, 28MHz WSPR signal from OZ7IT 19-20 May 2021.

Summer 6m!

Very good conditions on 6m across the Atlantic yesterday.  Spots to/from my 2-ele quad:

And, once again, the conditions in the mesosphere provide one reason why.  Look at those beautiful waves - the reason that you often hear bursts of 6m signal come through for just a fraction of a second:

Ice!

Judging by the extremely strong reflections seen at 70 degrees north, it seems charged ice/meteoric debris surfaces are now forming in the mesosphere.  It is very close to noctilucent cloud season now - just a few days to go.

Strong reds and oranges indicate ice formation.

Nice 6m SSB with S57A today, using just 6W PEP and a 1/4 wave vertical by the lake!

At IO73ug (Cefni lake, central Anglesey).

 

A friendly Alder Fly comes to visit me!

Stick to shore.

Time to test the Ampro stick-based antenna (price: not the £200 being asked for the commercial version!) at the beach last evening, passing a couple of hours whilst on taxi service for my daughter!

It's simple, and can take good winds.

 

Key to the success of this antenna is the requirement for 1/4 wave radials for the band of interest, and the raising of those radials off the ground - even if that is only a few centimetres.  With the radials on the ground, the matching is way off.  As soon as they leave the ground, matching becomes perfect.

It only takes a small stick to raise the radials enough for a perfect match.

This outing wasn't long enough to get a good WSPR dataset.  All I can say is that, with the tide a long way out, and no seawater-saturated ground beneath the antenna, the stick antenna was nevertheless achieving exactly the same SNR at the DX station, VE6JY, as my full wave delta back home - also a simple but much more complex to deploy antenna (think: elevated feedpoint and ~8m total height) than the Ampro - just ~2.5m tall.

Nice and warm, lying on the coastal grass.

 Overall, though, the stick wasn't anywhere near as good as the delta, as this map shows:

Over on a few bursts of FT4 - at about 3W - the stick is certainly not a dummy load:

And, of course, with any portable antenna, you get the distinct advantage of fresh air and wonderful views.  Even a little nibble on the very salty but refreshing sea kale, now in full bloom:

On the horizon is the lighthouse on the remote 'Skerries' - a name revealing of the area's Norse heritage.

Sea kale.  Delicious!

Below: 6m operation, the following afternoon, cut short due to oncoming heavy rain!  Initial assessment is that the Rapsberry Pi causes RFI at 50MHz, so will need some additional ferrites on cables.  I was getting to the Azores at -10dB on FT8, whereas my 2-ele quad was only managing -17dB at the time.  The radials are significantly less than 1/4 wave, being only about 1m each.

Hurrah! 6m Transatlantic!

Last evening, I received 6m signals until just before 00:30UT, the action restarting a few hours later.  Quite amazing!

Hello, Canada!

With my antenna - 'just' a 2 ele quad - pointing south to catch most of the signals today (it was mostly north and NE yesterday), I received a signal from VO1HP, at 3410km.  The appearance of strong and prolonged PMSE will not be entirely unrelated to this:

Usual response to the VO1 signal: run outside; get the ladder up; try to turn the antenna; forgot to undo the safety rope!  Undo that; align the beam; run back inside.

VO1HP was still there!  I sent a reply and got him first try, albeit a bit weak from me, although I came up to -6dB a few minutes later:

From this year onwards, I think I may well start concentrating on the 6m band, as well as 2m.  Some big changes may happen at the station fairly soon.  More about that in a future post...

Going simple.

In this month's Practical Wireless, there's a nice review by G3UGF of what is actually a very simple new product: the Ampro portable kit.

It's nothing more than a collection of different band sticks, which fit onto a vehicle-type mount, clamped to a solid earth rod.  The idea being that it's all simply a stick-it-in-the-ground, cheap antenna that might also be effective.  

Carrying sticks is easy, but the stainless element at the top is very springy, and becomes very irritating when walking any distance.  But that's a minor point, really.

It's certainly a great way to sell more Ampro sticks, which are actually very well-made; I have several for the car already.  

But the price!  How much for eight sticks, a clamp and a copper rod?  £238, reduced in some outlets to 'just' £199.99!  I suspect most of us will buy only the sticks we really want (80m is a bit wishful thinking for good performance at this size, for example), make our own ground rod, like I did (it's not ideal, copper tube is too soft at this kind of length, unless you have soft ground or sand), and buy or make the mount separately. 

Very simple, but you do need radials. 

I never thought of making exactly this type of antenna, but I did come close some years ago, with a manpack.

So, I went to get my manpack from a remote shack in the fields that I have; it's been unused for a couple of years now.  There was a drama underway, as some cows had escaped into the main road!  After all that was sorted out, I made a simple copper tube mount with a short 90 degree leg to attach the type of clamp I already had.

Unlike my manpack, proper matching (at 14MHz) was not possible without the addition of two 1/4 wave radials.  With the radials, I got down below 1.09:1. 

Great stealth antenna, with very good performance.  WSPR tests at 14MHz.

Excellent match for much of the 14MHz band (scale is too wide - my fault!)

How did it do on 200mW WSPR?  Surprisingly well, despite horrendous HF conditions in a G3 storm.  I can't remember what antenna the other guy has, but he is a good UK station (reduced duration relative to the other comparisons - he didn't keep going as long!):

Here's another - again no idea what antenna, and the other guy is using 500mW, against my 200mW (40h of testing):

And, finally, another 200mW:200mW comparison (40h duration), at which point I think we can safely conclude that the Ampro stick-and-ground rod system leaves nothing to be desired:

Distances are only one element of performance.  Also important to consider relative signal strengths.  Here's one comparison across all distances.  Antenna is described as follows: "TX antenna is 1/2 wave end fed wire against a short counterpoise with a tuned matching transformer on T82-2 core."

Essentially identical performance against G7CKX across all receiver distances.

And at DX distances, beyond 5010km:

Only 1.9dB down, on average, against the end-fed.

 

This antenna is a bit of a compromise, obviously.  It works well for its size, simplicty and pocket-money cost of building or buying.  It's surprisingly stable in a good wind, too.  Even if you have to carry, say, three whips along with you, it's always going to be a lot less hassle than tuning a magnetic loop.  Once your whips and radials are set-up properly, you just stick and go!