Saturday, April 17, 2010

VHF NFD. The Presentation.

They say a picture paints a thousand words. Well here are two pictures that are the result of literally thousands of man hours over several years.

Peter G0VVE and Andy G7FWE clearly capture the mood at the award ceremony held before the RSGB AGM in Bedford this morning.







See you all on the hill at 15:00L on Friday 2nd July.
Or on the air from 14:00Z on Saturday 3rd July.








Sunday, April 11, 2010

CARC Spring Surplus Equipment Sale

31st March 2010

Well, we had a reasonable turnout for the Spring Surplus Equipment Sale, and there was quite a wide range of equipment to go under the hammer – signal generators, a nice HP scope, a Dymar 2-tone audio generator, frequency counters, an ERA Morse Reader, a built Howse twin receiver kit, Morse keys, drawers of components, computer stuff and other items too numerous to list. It was also nice to welcome friends from other clubs.

The redoubtable auctioneer and hammer-wielder was Stewart G3YSX, and with Richard G4ANN looking after the financial and admin side of things the evening went without a hitch. And not to forget Ted G4TTY who kept the teas and coffees coming throughout the evening.



Finally, whilst CARC takes as standard a 10 per cent commission on sales we would also like to say a “thank you” to those sellers who donated all the proceeds of their sales to the Club funds – it all helps in these difficult times!

John G3VLH

Monday, April 5, 2010

Mini-Whip with CAT5

I have previously reported on the excellent mini-whip design produced by Roelof Bakker (PA0RDT) (http://carconline.blogspot.com/2009/05/pa0rdt-active-antenna-continued.html).

Recent work on the application of the design to VLF reception has enabled the use of CAT5 (Ethernet) cable in place of a coax feed.



As Roelof said in a note to the RSGB LF reflector : "Note that the RF-isolating transformers in the signal path are not correct. I have used MCL TT 1-6 parts, rated for 0.004 - 300 MHz at the -3 dB points. The common mode chokes are not optimum, as I used the parts I had on hand. Type 75 or 77 ferrite should be better."

Referring to work on 9KHz he continues "Michael used a PERSEUS SDR, which has a noise figure of about 25 dB at 9 kHz. This is due to the fact that there is also a RF-isolating transformer in the signal path.

"It is amazing that with this not optimized set up it was possible to have good reception over a 72 km path. The actual capture area of the antenna is 30 x 40 mm single sided copper clad PCB."

"I have the parts for an optical link active E-field antenna, as suggested by Stefan. Due to other more pressing matters, I have not yet put them to good use."

It looks therefore as if we can expect further developments of this simple but very effective design.

Joze (S52AB) reports on a CAT5 mini-whip http://s52ab.jalbum.net/mw-test-2/ in which he shows a comparison between reports on a CAT5 fed mini-whip and (presumably) the same mini-whip fed with coax. My visual impression from looking at the received signals is that the CAT5 has a higher loss, although it is not clear at what frequency these measurements were taken. For those interested in LF work, Jose's main site (http://lea.hamradio.si/~s52ab/) is worth looking at.

- Stewart/G3YSX

Saturday, April 3, 2010

Triax 2 Lug to 3 Lug Interconnection

At a rally last year I purchased a Keithley 616 Electrometer for a very reasonable price. Unfortunately it had a serious problem as can be seen from carefully looking at the connector in the picture below.



Investigation's on the web through up this Wikipedia article (http://en.wikipedia.org/wiki/BNC_connector) that explained all. As this says:

"Triaxial (http://en.wikipedia.org/wiki/Triaxial_cable) (also known as triax) connectors are a variant on BNC which carry both a signal and guard as well as ground conductor. These are used in sensitive electronic measurement systems, particularly of Keithley manufacture. Early ones were designed with just an extra inner conductor, but later tri-axial connectors also include a three-lug arrangement to rule out an accidental forced mating with a BNC connector. Adaptors exist to allow some interconnection possibilities between tri-ax and BNC connectors."



Unfortunately the plug shown above is very expensive, and even if I had found one I would have then had to find some cable and make the connector, which if making an ordinary BNC up is anything to go by would not have been easy, and the adapters mentioned in the wikipedia article are rarer than hens teeth. I must have searched every new and surplus connector tray at the Friedrichafen show last year, and a good few English rallies and found nothing that would work.

I was at the point of giving up and soldering some wires to the outer contacts (I thought of replacing the connector with a standard BNC, but the internal construction of the Keithley 616 made this a very difficult and hazardous to the instrument) when a friend to whom I am most grateful gave me a triax to BNC adapter.

Unfortunately the adapter was a three lug adapter and as indicated in the wikipedia article it would not mate with the two pin socket (deliberately so). After a lot of measuring and staring at the connectors we concluded that the ONLY difference between the two and three pin varieties was the positioning of the lugs and that whilst not perfect a one lug socket would work well enough (at least compared to the current situation of no mating connector). So out came the Dremel, off came the lug and on went the adapter.





Now in case you are wondering, I do have a reason to want to be able to measure the astonishingly high resistance values that this instrument will measure, but more of that at a later date.

Stewart/G3YSX


Footnote to readers, the reason why CARC shows the full URL hyperlinked to the URL, rather than the more conventional article text to URL linking is to make it easier for our editors to produce the monthly pdf version of the newsletter.

Friday, April 2, 2010

A Loop Experiment

On 15th March I set out for a business meeting in the US and decided that as an experiment I would leave my 40m coax loop that we made at the club connected to my FT817 receiving WSPR. The receiver ran almost continuously for the 14 days that I was away. I say almost continuously, I have a bug in the computer that keeps resetting the the audio port from line-in (which is the audio input port I am using) to mic in (which interestingly still receives the stronger signal due to audio cross talk between the ports) and I needed to remotely log on from time to time to correct this. Because I could not be sure that I would have good internet connectivity whilst I was away I did not attempt to rin the system as a remote transmitter.

Please excuse the mess in the shack, but the loop in use is shown below.



This rather saggy loop is in my loft shack which is constructed with aluminum foil covered plasterboard, although as it is not bonded together I am nit sure what effect it has. It was oriented in a plane exactly East-West.

Over the 14 days on 40m (which is not the most popular WSPR band) I logged 120 unique call signs from 19 countries. To get some assessment of the range groupings I grouped the number of unique contacts into range buckets 1.414 time the previous max range as shown below:

0 to 50Km 2
50 to 121Km 2
121 to 221Km 6
221 to 362Km 9
362 to 562Km 19
562 to 845Km 23
845 to 1245Km 26
1245 to 1811Km 10
1811 to 2611Km 6
3742 to 5342Km 1
5342 to 7605Km 8
7605 to 10805Km 1
10805 to 15330Km 1

The contacts in the range 0 to 500Km were most probably NVIS which is about 38 out of the 120 contacts meaning that about 1/3 were NVIS and 2/3 were contacts that required F layer skip.

I then took a look at number of unique contacts by country, and did a by eye estimate of the baring from my QTH. The results I got were as follows:

Country # Range(Km)
Australia 1 60
Austria 3 111
Belgium 3 102
Brazil 1 220
Denmark 1 47
England 14
Finland 3 45
France 6 160
Germany 35 70 to 110
Italy 9 130
Netherlands 7 76
Norway 1 28
Poland 5 80
Scotland 4 330
Spain 4 190
Sweden 2 47
Ukraine 7 84
USA 3 290

I did not do bearing for UK because they were spread over such a wide arc that I did not feel that the results would contribute much to understand how the antenna behaved.

I then classified the results into front and back (315 deg to 45 deg plus 135 deg to 225 deg) and side (45 deg to 135 deg and 225 deg to 315 deg). The result was that 16 unique reports were made on the front back and 81 off the side. Even if I discount all of the NVIS contacts the majority of contacts were still off the sides. This distribution is heavily skewed by the distribution of round the world stations, but the results would suggest that the small tuned loop probably receives off the side, or omni-directionally rather than off the face which I believe is common wisdom.

Just one final observation I took a look at the powers (in Watts) that various stations that I received were advertising.

Power #
0.05W 1 (Germany)
0.1W 1 (Netherlands)
0.5W 5
1W 12
2W 12
5W 64
10W 8
20W 1 (Italy)
100W 1 (Spain)

I leave you to draw your own conclusions.

It will be interesting to see how this compares to other antennas that I have available and to get a feel for their patterns. I will run some other tests and report what I find.

73

Stewart/G3YSX

Thursday, April 1, 2010

DON'T SHOUT BUT "WSPR"

For our talk on 24th February, Walter Blanchard, G3JKV introduced members to the Weak Signal Propagation Reporter system. This uses a clever suite of software which encodes the message to be transmitted, and monitors and decodes other WSPR stations that you can hear, during precisely timed periods with a 2 second cycle time. The monitored signals are decoded and uploaded to a web site which publishes the results in tabular form for further analysing and mapping. In effect it is a world-wide beacon network additional to the traditional beacons (-which are not comprehensibly monitored and tend to have larger power than "weak")

This new digital system is already used by 2 or 3 CARC members (including G3YSX, G4FYY and 2E0MZB) and I hope they will correct and expand on my notes taken during the talk.

In WSPR "weak" means output from the TSCVR of 5 or less Watts and the software can decode signals received with Signal to Noise (S/N) ratio as low as -28dB. This is about 500 times "better" at decoding than the human ear listening to voice on SSB. If like me you find dBs hard to appreciate the WSPR website provides tables of power in Watts and dB equivalents.

The software has been written by Joe Taylor K1JT who is a physics professor at Princeton University, New Jersey, and who with his colleague Russell Hulse won the 1993 Nobel Prize in Physics. Professor Taylor's group continues to explore problems in astrophysics and gravitational physics by means of radio-wavelength studies of pulsars. Among recent highlights are the discovery of many new pulsars, including millisecond and binary pulsars.



In operation the PC generates the encoded signal and switches the radio to transmit at precisely the right time i.e. at the being of each even minute. The PC needs to know precisely the correct time and this is obtained from a time server on the web e.g. dimension 4 at http://www.thinkman.com. The last 6 seconds of the two minute cycle are silent. The PC can run Linux, Windows etc. and a processor speed of at least 1.5GHz is suggested by K1JT, although Walter suggested that 500MHz is satisfactory.

Interconnection
Interconnect between the PC and TSCVR is identical to that used for PSK , RTTY etc.

The connection of the PC serial port and the radios Push to Talk (PTT) is often made to part of the accessory port of the TSCVR. This can include optical isolation and 1A power transistor to handle the current needed by some TSCVRs to switch to Transmit. To carry the audio between the PC sound card and the TSCVR an isolating transformer will prevent the PC's dc voltage on the sound card reaching the TSCVR and vice versa.

There is a wealth of designs and information about PC to TSVCR connection techniques at provided by W5BBR (Google says that there should be a www.w5bbr.com/soundbd.html site, but this did not seem to be up when the editor looked for it. However the material can also be found at http://www.cqham.ru/soundint.htm). The latest version of the software allows interconnection by CAT system, which additionally ensures the frequency set in software matches the TSCVR setting.

The Encoded Signal
The transmitted signal contains information using Frequency Shift keying (FSK) at the low shift value of only 6 Hertz(Hz). The encoded transmission includes; sender's callsign, Tx location and Power level in dB. As the shift is so small it can not be heard by the average listener and the transmitted signal sounds like a continuous 1500 Hz tone. When listening typically there is ?beat? occurring between signals and "woof-woof" sound is heard. The signal is contained in a 200Hz Pass Band.

WSPR signals heard are decoded by the PC and can then be upload to the web at www.wsprnet.org. There are millions of monitored signals on the site and data can be selected for further analysis. To register with the site you need your Maidenhead locator http://www.levinecentral.com/ham/grid_square.php will give latitude and longitude plus Maidenhead locator directly from your callsign using data from QRZ.com.

Having downloaded the WSPR software, to get started try the 30m band as this has many monitored calls listed in the webpage, where distances and azimuth angles are tabulated. Once monitoring is working and your PC is decoding some of the results you can see on the web, you can consider transmitting. Walter drew our attention to the following points

*Correct UTC time (from internet atomic clock)
*Correct dial and transmit frequency remembering that changing freq on rig does not alter the data set on the PC
*Start with Transmit to Receive ratio set at 20%
*Choose the correct frequency band on the menu bar
*Check Rx noise ? needs to be 1dB plus and minus 10 dB tolerance
*Select USB mode
*Use TSCVR with frequency stability within 2Hz at 28MHz or you may need to add the manufacturer's "high stability Oscillator" cost about £50.

What is it good for?
Well there is all the data to analyse in any way you wish. The effectiveness of aerial types, locations, mounting etc. and TSCVR type and power can be monitored by adding switching to change the equipment automatically during the 6 second silent period.

Map of monitored calls



More information
Try www.g4ilo.com/wspr.html This is a clear comprehensive explanation and I suggest is essential reading. Also search for wspr in www.youtube.com to see wspr in pictures!

Further Links

Steve Nichols G0KYA, RSGB Propagation Studies Committee www.qsl.net/g0kya
Walter Blanchard email wb@g3jkv.co.uk

By Ted Aston G7OBF