Lashed up MF antenna but some good DX

Overnight, I left my FRG-100 tuned to 472kHz (474.2kHz to be exact) in USB mode, and the PC on running WSPR. The antenna was a serious bodge job

 

After the failure some time back of the random wire that fed both the FRG-100 and Sams VOLMET 5MHz receiver, all that was left feeding the Hi-Z port of the receiver was about two feet of speaker wire. With no time to fiddle (It was gone 22:00 by this time!) the best I could do was switch the ATU to bypass, can croc-clip the bit of wire to the center pin of the patch lead I use to the antenna switch for connecting up such things as Wispy or my SWL PSK-20.

Some fiddling was needed to get the audio into the PC as well, but I was remarkably surprised to see stations coming in. I left the whole lot running overnight. 13 different stations copied! Including, finally at about 03:00, including Roger G3XBM

Timestamp	Call	MHz	SNR	Drift	Grid	Pwr	Reporter	RGrid	km	az
2014-08-1506:48	PA3ABK/2	0.475739	-8	0	JO21it	0.5	G7MRV	IO93ks	448	301
2014-08-1502:54	GM4SLV	0.475725	-13	0	IP90gg	0.2	G7MRV	IO93ks	723	178
2014-08-1503:20	G8HUH	0.475610	-1	0	IO81mg	1	G7MRV	IO93ks	304	23
2014-08-1422:44	G6AVK	0.475762	-21	0	JO01ho	0.01	G7MRV	IO93ks	268	335
2014-08-1504:56	G3XBM	0.475739	-29	0	JO02dg	0.005	G7MRV	IO93ks	192	331
2014-08-1505:56	G3WCB	0.475654	-22	0	IO91rm	0.2	G7MRV	IO93ks	253	351
2014-08-1501:56	F6HCC	0.475677	-29	0	IN87hs	0.1	G7MRV	IO93ks	686	12
2014-08-1500:40	DK7FC	0.475684	-21	0	JN49ik	1	G7MRV	IO93ks	832	309
2014-08-1422:44	DK2DB	0.475641	-19	0	JN48fw	0.5	G7MRV	IO93ks	854	313
2014-08-1500:30	DJ0ABR	0.475759	-22	0	JN68nt	0.5	G7MRV	IO93ks	1131	305
2014-08-1501:18	DH5RAE	0.475749	-23	0	JN68qv	0.2	G7MRV	IO93ks	1141	304
2014-08-1422:58	DG3LV	0.475785	-25	0	JO53gv	20	G7MRV	IO93ks	765	274
2014-08-1501:44	DC0DX	0.475697	-19	0	JO31lk	2	G7MRV	IO93ks	604	299

New Modes and New Bands

With a few days with time on my hands recently, Ive spent more time playing with the WSJT suite of digital modes, particularly more WSPR on various bands, and a play with the JT6M mode for 6m meteor scatter. As yet, I havent made a full QSO in MS, but have been received a few times, and had part QSOs. Its just a matter of refining my technique, and being more patient.

Playing with WSPR, Ive had no luck on 6m, and very limited success on 10m. I have got a bit further with Wispy though. The photo below shows the experimental parallel transistor amplifier, supposed to be 1W, on the left, and the new LPF on the right

The new 7-pole Low Pass Filter is very effective. Built as an add-on board, I can easily test the levels of harmonics before and after the filter, and it is quite clean post filtering. Ive had far less luck with the parallel transistor amp though. I can see that it is amplifying, as can be observed using the spectrum analyser, but the output is just 10s of mW! Its not even enough to raise the meter!

Why this should be I dont yet know. Perhaps its something to do with the biasing or quiescent current, or maybe its the emitter resistors and emitter bypass capacitors.

Ive also, as can just be seen in the lower right of the above picture, made a bit of progress on the receive chain. The two red components are 470nF capacitors, which due to not having that value I had to rob from Daz's Theremin project! I need another, but dont have one anywhere, I will have to try something like a 220nF in its place and hope its value isnt too critical.

One thing I noticed today about 6m WSPR is the number of stations that are only listening! Today when I was active, out of a slack dozen of stations, less than a third were actually transmitting at all! For a propagation beacon network this is poor. It relies on each station transmitting in order to establish the propagation paths.

In light of this, and after reading Roger G3XBM bemoaning the lack of stations active at MF, I have considered today looking at ways to get onto 472kHz. This is an altogether different proposal to most HF/low VHF work. The best I can manage at present is to set up my FRG-100 receiver for 472kHz, and, in the absence of a random wire for that Rx at present, Ive had to take what left of the high-Z input wire (about 2ft of 7/.32 stranded speaker wire!) and using a croc clip lead, patch it to the main doublet antenna via the center pin of the antenna switch port, and set the ATU to bypass. I also have found that the audio feed into the PC is a bad fit, but after much juggling, have managed to feed the audio from the FRG-100 to the PC.

The WSPR software is not at all happy at being fed a 24dB noise floor, and the hum is atrocious, but surprisingly some decent decodes are coming through -

2238 -23 -0.9   0.475749  0 DH5RAE JN68 23
2240 -22 -0.1   0.475684  0 DK7FC JN49 30
2240   2  0.2   0.475739  0 PA3ABK/2 27
2244 -19 -1.1   0.475641  0 DK2DB JN48 27
2244 -14  0.5   0.475653  0 G3WCB IO91 23
2244 -18 -1.5   0.475697  0 DC0DX JO31 33
2244 -12 -0.7   0.475725  1 GM4SLV IP90 23
2244 -21 -1.1   0.475762  0 G6AVK JO01 10

Best DX so far is DH5RAE at 1,141km!

Of course, ive yet to hear Roger!

UPDATE :  Managed to eliminate the hum by swapping from the USB soundcard, which it  seems doesnt like mono jacks, to use the on-board mic input. This has brought the noise floor on WSPR down to a more manageable 17dB. Unfortunately, all the messing about has upset the program, which now is crashing out a bit!

Perseids Meteor Shower Peak Soon - Lets Try JT6M!

Something ive thought about doing but never tried before, is Meteor Scatter. This, just like EME and the various QRP HF modes, is something that with the advent of fast computers, and particularly with Joe Taylor K1JT writing amazing software and developing amazing weak signal modes, is now within the grasp of even the smallest station.

The Perseids meteor shower is due to peak over the next couple of days, with average ZHRs of about 20-30, but with the severe disadvantage for visual observing that theres a full moon, which will wash out all but the brightest bolides. But the full moon doesnt affect VHF!

So, whilst Whispy gently toasts its PA transistor running all night on 10m, Ive downloaded Joe Taylors WSJT software suite, and am listening on 6m for JT6M meteor pings

Although not shown in the screenshot, I have picked up a few pings, along with a horrid pulsing noise which seems to come from the computers mouse, and annoyingly covers around 40kHz of the 6m band, centered right on top of the MS window!

Another thing of course with meteor scatter, is that it will work in daylight, something that is very rare for visual observation of meteors!

I dont yet fully understand the system, or procedures, so will need to see a good few received pings first before I decide to try for a QSO.

10m Wispy Tx running overnight

The Wispy beacon Tx is set up to air via my ATU on the main station antenna, Toshiba laptop providing the audio drive WSPR signal at a 50% Tx cycle.

Its running from a 12v 7Ahr SLAB. As it uses pretty much the same current, and runs at the same temperature, regardless of whether its actually on Tx at the time or not, I am going to put it to 80% Tx overnight, and hope that at some point, perhaps morning grey-line, some propagation occurs that at least gets a spot to somewhere!

Wispy on test!

After finally discovering that I had simply not been driving her hard enough, I have the 10m Wispy beacon transmitter testing to air.

Several problems had to be solved to get this far. First, I found the laptops clock is woeful! I now have Dimension4 running on both the laptop and the desktop to keep the clocks happy. Since doing this, my own decoding of the beacon (via the main station) is almost solid. Second, the main antenna, during high winds this morning, failed. The wire snapped right at the feedpoint, necessitating some quick repair work to get the antenna back in service.

 

The picture shows my own decodes of the beacon. From this, and some other test information I now now that the Wispy Tx takes around 30mins to stabilize during warm-up, after which its drift rate is negligible, but the initial drift during warm-up is about 60Hz. I  also know that the Tx takes a tad below 160mA, even in standby. This is due to the amplifiers always being powered. I might, later, add a Tx relay so the amplifiers can be off and allowed to cool when not in use, but that will add the complication of needing either a PTT input or a tone detector circuit, like in the FT-857D Interface.

Prior to any of that, I'd like to play with the biasing and see if I can lower the quiescent current somewhat. I also want to play with the drive and the output filtering to further reduce the harmonics in the output. If I can reduce the spurii going into the amplifier, then I can ensure that as much of the power consumed as possible is being used to generate my wanted 28MHz DSB signal.

To this end, Ive added the little 28MHz bandpass filter to the output of the SBL-1 DBM on the 'TennaLady' mock up, as well as adding the necessary parts to convert the oscillator to 14MHz self-doubling to 28MHz. If I can kill the spurii prior to the amplification chain then so much the better.

Another little beastie im looking at is a 28MHz 1W Power Amplifier, using parallel small signal bipolar transistors. Roger G3XBM has used this approach in the Tx only version of Wispy, plus in his 'tenbox' AM set. With 4x 2N3904 NPN transistors, Roger gets 500mW DSB, 250mW SSB equivalent. So, Im going to try with 8x transistors! Hopefully giving me an equivalent of 500mW SSB.

The photo just shows two transistors fitted, which would give the same output as the single 2N3866 im currently using in Wispy.

Unfortunately, so far my Wispy beacon has not yet been spotted. Ive had her running most of the day, but the propagation on 10m at present is dire. I'll let her run over night, and into the morning. Depending on the wind, I might put the 10m J-pole up and run her to that tomorrow, see if having a resonant antenna and not having to go through an ATU and long feeders makes much difference. At 100mW I think every little helps. At some point however I want to work on the receive side, so if I make decent progress on TennaLady tomorrow, I might try her to air for a bit while I do the Rx chain on Wispy.

WSPR, Window, and Rain

It seems the window line section of my antenna isn't too keen on the endless rain we've had overnight. After I discovered that WSPR had crashed sometime in the small hours, probably when the antivirus decided to tell me it was due to expire, I restarted on 20m, but found not a single person was hearing me!

Playing with the ATU showed it was very difficult to get a decent match anywhere, suggesting the now soggy vegetation near the balun is probably drooping onto the window line section and detuning it. I have managed to get a usable match on 40m, so have popped up there at 2W. The system went straight into transmit with the results as below

 

Now into the receive cycle, lets see where I can get today 

UPDATE: The rain has stopped (mostly) so I managed to get out and take the shears to the mass of Ivy that has, seemingly in the last 12h, utterly overgrown my balun and lower section of window line feeder. With that cleared I can once again get a match on 10m, so have QSY'd up there to see what can be done.

WSPR - A fantastic mode, so long as someone uses it!

Yesterday, I set up the main station to beacon in WSPR mode on 10m. This was sometime just after lunch.

Around 21:30, I retasked the system to 40m. Not one, single, solitary spot on 10m! Neither hearing nor heard. It seems there were less than a bakers dozen of us on 10m on the entire planet! This is simply rediculous. There were nearly 200 stations on 20m, a band whos characteristics mean it really doesnt need any propagation investigations. This is just people who want to do something without actually having any real involvement in it.

On tasking the system to 40m, as soon as I came out of idle the system went into transmit, and was immediately received in Tasmania!

Today, shortly after 15:00, I started beaconing, 10% Tx/Rx cycle, at 2W on 6m, beaming SE. I dont hold out much hope, as theres fewer than ten stations around the world on 6m today! And, although I dont know whether to believe it or not, the WSPRnet site says most of them are running 0%, i.e. Receive only!

FT-857D Digital Interface Mk2 (Rebuild)

Yesterdays successful test of the digital interface for my FT-857D did unfortunately leave me with a rather convoluted collection of bits of circuits all cobbled together. Todays task then was to build the Mk2 interface - in other words, get it all on the one, neat and tidy board, ready to be boxed.

Last night I drew out the schematic, planned a Veroboard layout, and selected a suitable hunk of board. The only minor change was to include a level control potentiometer on the receive path. Not really needed for WSPR, but perhaps useful for other modes.

The picture above is how it all looked at close of play yesterday, not including the opto-isolator. The board to the lower right is the single transistor audio amplifier, that drives the audio detector and PTT switch (right hand side of main board). The modified el-cheapo Chinese USB soundcard hanging off the whole lot to the left.

This is the new board. As an odd off-cut the sizes actually turned out to be pretty good for the job, with the longer, lower section fitting the PTT system nicely. Here the audio isolation transformers, DC switch and opto-isolator are installed, along with the radio side ground connections. 

Further on, the Tx 10:1 audio divider, computer side grounds, and the audio rectifier diode doubler are installed. What can't be seen, is the continuous run of track breaks under the transformers and opto, forming a complete separation of the two signal sides.

And, finally, the completed board, in use. The USB sound card is affixed onto the board with hot melt glue. This is during final testing. Im pleased to say all the testing went well, with good receive spots from around Europe, and some decent DX 'heard by' spots into Isreal and the USA, all at about mid-day here in the UK on 20m.

The schematic is shown below. There is nothing new, or original in this design, and it borrows much from KK7UG's interface. The audio activated PTT circuit is a mish-mash of several designs found around and about the internet.

With a couple of simple mods for serial port controlled PTT etc, this could easily be made into a 'universal' digital modes interface.

The final two tasks are to replace the sound-cards USB plug with a wired USB connector, and to put the whole thing into a case.

Mobile WSPR - Finally!

Well this took some flippin' doing! The audio isolation needed to make an effective interface was really quite simple, the trouble came about when trying to use the radios own Digital VOX circuits to control the transmitter. It would work, so long as the audio input was high enough, but that level was always far greater than the power output I wanted, and always driving up into the ALC action region! No matter what settings I used, adjusting the soundcard level, the interface level, the VOX level, or the Digital Gain control on the rig, I could not get the 1-2W I wanted AND have VOX PTT!

So, I had to work on a hardware solution.  This seemed simple - both channels of the soundcard carry the same audio signal at around 4Vpk-pk, it should have been a simple job to rectify this and drive a switching transistor. Er, no. The output of the diode doubler used only generated 0.3V, no where near enough to switch the PTT transistor.

So, some head scratching, and internet searching later, I added a single transistor audio amp to the detector circuit. Yeah! It switches!

Only trouble was, I had to use the 5V supply from the USB soundcard to power the amp, so I now no longer had the isolation between the radio and the computer! The photo above shows the interface in this state of development! I now had to add yet another little bit of veroboard, this time with a 4N25 optoisolator and a resistor on it.

That done, I got out to the car, connected everything up - and it wouldn't Tx! Arrgh! I'd forgotten to put the ground connection from the opto to the rig on! Back to the workshop, telling Julie that 'This is definately the last trip!' and then back to the car, and finally, a working audio driven PTT line!

So, the time now being around 19:00, and with a fully recharged laptop battery ready to give me its whole whopping 20mins capacity, I set up the Tx level to about 3-4W, sync'd the clock with an internet time server, and started WSPR, with one Rx period followed immediately by a Tx period. First spot, both heard and heard by, was HS0ZKM Thailand!

So what remains now is to rebuild the interface with everything, including the USB soundcard and the opto-isolator, all on the one compact board!

Show Cards and Yaesu WSPR

After a slight delay of just about 11 months, I finally took delivery of the QSL cards for GB2LDS this week. They are all now written, along with a few remainders from the year before, and ready to be sent off to the bureau. Thats around 150 cards.

Ive taken the time to check QRZ.com for QSL status, so as not to waste a card on someone who doesnt want one anyway, or who only accepts them direct but who hasnt sent one direct to me. If you havent sent a direct card in 11 months then clearly your not going to send one!

With just a month to go before GB0SML, the final preparations are underway. The main antenna now has actual wire on it, and just needs a trial on site to trim the elements. The digimode interface for the FT-857D is made and due to be tested today, im just waiting for the laptop to charge up!


Well, after giving the lappy a brief session on the charger, it lasted long enough for me to establish a working receive path. With the car parked on the drive, and just the mobile 20m whip, good Rx spots were straight away coming in!

 

Now to get Tx working! This is less straightforward, as it requires some fiddling of levels and power settings. With the Tx PWR setting turned right down to 5W, and the DIG VOX setting right up to 100 (not greatest but will do for now!) ive managed to get the radio to trigger to Tx from the WSPR software. Using FM mode and a dummy load and power meter, Ive set it first to ensure the meter was calibrated at 5W, and then set the audio drive level to give 1W output.

This is all with the DIG GAIN level set to 50, and without reference to the ALC meter! I will probably let a bit more drive into the radio, and adjust the DIG GAIN to set the actual modulation level. However, for now, flat battery stops play! The lappy is back on charge.