Tuesday, 25 February 2014

WSPR'ing? No, Shouting! AARRRGGHHH!!

Sadly though, its me whos shouting 'AARRRGHH!', not the 10m Wispy! I rigged this up again today, this time with the 9t trifilar toroid fitted. Nothing. Then I noticed one of the diodes was backwards! This sorted out, still nothing.

I can see there is some signal getting out of this beast. The spectrum analyser shows the start of a peak with modulation applied, but its many tens of dB down on the levels measurable from the doubler. This is beginning to drive me bloody mental! I just cannot fathom out why this damn mixer wont work! I isolated it from the following driver stage, just in case a fault there was pulling it down, but still nothing. Yet when I mocked this mixer up on its own, it worked? Why will it not work in circuit?

The only thing I can possibly think is that its a phasing issue, I must have the phasing wrong, but it all looks right to me. Im going to have to ask for help with this again from the G-QRP members.

I know I have started Project TennaLady to be a simplified 2nd version of this circuit, but I hate the idea of abandoning Wispy. There really should be no reason for it not working, and ive pestered Roger G3XBM so often about it, and hes always been very helpful, I do feel a bit like i'd be letting him down if I cant make it work.

On a lighter note, the small amount of other radio done today has been good fun. Most of the day has been taken up with church visits as part of my study of local extant war graves (see my other blog http://selbywargraves.blogspot.co.uk/ if your at all interested), which was intended to be a single church visit but turned into five, followed by a bit of gardening. The bit of gardening led to several hours of cutting and blanching parsnips for the freezer!

What I did manage to do though, was to transfer the PSK-20 onto solar power. I have a small solar panel up on the shack roof, about 6" x 18" and rated in the region of 6W, although it is now rather old and inefficient. This feeds a 12v 7Ahr SLAB. I transferred the power cable of the PSK-20 from the main PSU to the distribution block of this system, and switched on. Odd, both LEDs lit? Ah, take the multimeter measuring the charging current out of circuit! Thats better. With the PSK-20 running on stored solar power, I answered a CQ from IP8AAW, an Italian special event, and worked him on first go.

I should get a bit of work on Wispy done tomorrow morning, but I have to be at work mid-day for a secondment interview, followed by Sams parent/teacher evening, then Karate.

Monday, 24 February 2014

Going 'a shade greener'?

There is a company, located not too far from me, who specialise in photovoltaic installations, and trade under the name 'A Shade Greener' http://ashadegreener.co.uk

I am somewhat curious about this. I have for a long time liked the idea of a solar power installation. In fact, I already have a small solar panel used for charging a 7Ahr SLAB, and using that to run small loads such as my weather satellite receiver, which now is very little used (hmmm, theres a thought - I dont use it because I no longer have the antenna, but I might try it on the discone!)

A shade greeners business model seems good, if a little cheeky. Essentially, what they are offering is a totally free installation. Yes, totally free. So, whats the catch? They say there isnt one, and in many ways thats true. It does seem a pretty risk free plan. Basically, you sign to a 25yr lease. They rent your south facing roofspace for 25yr, during which time, they also maintain the system for free. As the consumer, you get the benefit of the free electricity it generates. All sound too good to be true so far?

Well, ok then, heres the deal. Yes you get free electricity, but only during the time the panels are generating. When the panels are not productive, ie when its nighttime or bad weather/thick cloud, you buy your power from the grid as usual. So the savings you can make depend entirely on how much of the free power you can use, during the day, without going above the generated amount. This is where the company's business plan starts working. Most people are not about during the day when the power is being generated. Also, if they are, its likely that less power is used anyway, as lighting and heating are often not required. So, because the home owner isnt using all the generated power, the company makes its money on the payments from government under the FiTs scheme, which pays for microgeneration to feed into the grid, and it seems pays well.

So, how to get the most from a free installation? Its all down to how much of the free power you can use at the time. The usual technique is to change your habits. Run the dishwasher, hoover, washing machine etc, during the day, and in series, not parallel. Running them all together will go above the generated load and cause you to have to buy from the grid. Using them one after the other can make the most of the available solar power.

So also, running a radio station during the day (reasonably convenient for me) uses free power, unless you go over in the transmission peaks! But what about evening or night operations?

It all comes down then, to storage! I am reasonably lucky in that I can occasionally lay my hands on surplus but still reasonably usable large sealed lead acid batteries, ex of UPS systems. By using the free power from a solar array to charge a bank of large SLABs, that power is then available to run the radio station and any other suitable loads later.

Its unlikely that such a system could be built for a reasonable cost that would run such kit as washing machines, but using a suitable inverter it could run a telly, or perhaps the fridge? Also, my kitchen and bathroom lighting is already 12v LED. By replacing more lighting with LED technologies, battery stored power could be convenient.

I have on order a roll of this stuff
this is flexible LED lighting. 5m long, extendable, and comprising about 300 white LEDs. It can be used by sticking it to a surface, say the underside of a shelf on the radio desk, or on the top of a picture rail. Im going to try this out as a way to replace mains powered lighting in my shack and workshop. I will then look at the use of solar charged batteries for running this.

The company's free system doesnt have many disadvantages, but its not easy for the home owner to actually make much of a saving on their power bill, likewise the lease timeframe is somewhat excessive

A Shade Greener free photovoltaic installation
Pros -
Free Installation
Free maintenance
'Some' free power

Cons -
Savings available to home owner are vastly lower than profit available to company
25yr lease is excessive. 10-15yr would be more acceptable
Array remains company property after lease expiry. Array passing to consumer ownership would be preferable 
No bidirectional energy monitoring. I would want to see what im feeding to grid.
No specs regarding RFI/EMC of installation

That last con is of particular concern. The chargers/inverters etc used would have to be properly filtered, in effect 'silent' at radio frequencies. I would not tolerate ANY increase in noise floor from the system.

Now, I know my roof is suitable, even without them doing a survey (hey, you dont study Environmental Science and Renewable Energy without being able to work these things out). I could buy my own installation, in which case I get all the free power, plus the feed in tarif. But heres the reason this company can do this - I dont have the money for that!

So its something im interested in, and might consider. Of course, by blogging about it, ive effectively given A Shade Greener some free advertising! Perhaps they will read this! If so, then they are welcome to contact me to discuss providing me with a 'showcase' installation, tuned somewhat more towards my requirements. Im sure its unlikely they have an amateur radio station in their installation portfolio, and even more unlikely they have a battery storage and comprehensive LED lighting installation. Maybe they would like to go with my plan as a live demo of what solar can do?

1st QSO with PSK-20

After a while just monitoring this morning, and checking settings (remember to turn that sound card AGC off!) I fired up and answered a CQ. My 3W PSK-31 signal was answered first time by 9A5ZM in Croatia. He reported my IMD at -17dB and S/N of 25dB, solid copy.

So all is well with the rig. It has a 5A fuse in at the moment, which I will change for something more appropriately rated later. But im happy with it and its performance. Im only a little sad that some of the lettering rubbed off during drilling for the mods. I'll find  a nice box to keep it and its cables in, ready for field use.

Sunday, 23 February 2014

PSK-20 working, plus Sudden -2 and a voice keyer

Most of today was taken up with Sams karate grading, so it was very late that I was free to play with any projects. Sam passed and is now 8th Kyu, which meant I had to cook his favorite tea - lasagne. With a good meal and a few bevvies in me, I found a few moments to complete the modifications of the PSK-20 transceiver.

The RF sniffer circuit needed to be as small and as close to the antenna connector as possible. So I decided to build it 'ugly' with all the parts having their leads cropped back as far as possible. The 2k2 resistor is soldered directly to the pin of the BNC socket, and the ground connections are straight to the ground-plane of the PCB.

With the mods done and a power cable made up, it was time to add power!. Always a daunting time, she fired up with no obvious signs of distress, and a healthy green power LED. I had calculated the current limit resistor for the power LED based on a 13.8v supply, which came out as 1k2, so I bumped that up to 1k5 to give a bit of leeway.  I then connected the audio out to the USB sound-card on the laptop. Hmm, nothing? Ah, plug it into the microphone port! Despite it being late evening, several PSK signals began to scroll down the FLDIGI waterfall.

Whilst connecting the antenna patch lead from the rig to the ATU, I had to move a few half finished projects out of the way. One is a Kanga Kits Sudden-2 receiver. This is effectively 'finished' apart from a case and knobs,

but it has provision on-board for an audio filter. After contacting Kanga, I now have the schematic of the filter, and as its pretty simple im sure I have the parts somewhere already, so I intend fitting that before boxing the project up. I already have a large plastic box for this, selected with the intention of it having an internal battery.

Also knocking about on the desk was this little digital voice recorder that cost me about two quid from Hong Kong a good few months back

 This was supposed to have been wired up to the headset system of my mobile installation, so I could call CQ whilst driving and not worry about my throat drying out! But when the rig went faulty this got forgotten about. I really ought to get it finished!

Ive run the PSK-20 now into test load, and also a quick test burst to air. No QSOs as yet, as I want to check the levels etc first and make sure every things aligned properly. In idle mode, with full audio drive the power meter was reading a good steady 3W. As idle mode is a 50% duty cycle, I need to back the audio drive off to a point where an actual data transmission gives 3W, which I think means idle should be about 2W. I'll sort that tomorrow and give it its first real run out on air.

The rather out of focus picture above shows the PSK-20 on receive, with the power LED lit. On Tx, the red LED below is driven by the RF sniffer and lights brightly, flickering slightly with data modulation. I will put the case back on tomorrow. My only concern is that the PA heat sink does get bloody hot. Perhaps the addition of a small cooling fan could be useful here.

Saturday, 22 February 2014

Adding a Tx/modulation indicator to PSK-20

The older version of PSK-20 was somewhat different to mine. For a start, it had a 9-pin D socket to allow control of the PTT by a computers RS232 serial port (RS232 was mentioned in a recent training course at work, and I, rather embarrassingly, was able to real the protocol specs off!). The newer version like mine has 'tone controlled PTT', meaning that the rig detects a control tone on the sound card right channel to control the transmitter, eliminating the need for a serial port, and making connecting up one less cable.

Something however that the older version had, which the newer one doesnt, is a Tx/modulation indicator LED. Now, although the computer will say when its in transmit mode, I rather like having a hardware confirmation! So as previously mentioned I will be adding this Tx LED. The additional circuitry is very simple and will easily fit on the underside of the PCB

The schematic above is the recommended circuit taken from the old PSK-20 manual. Its not a great diagram, as I recreated it in MS Paint, which is frankly rubbish, but it gets across just how easy this mod is.

This will need building as close as possible to the antenna connection. Luckily, the antenna socket is an insulated PCB connecting but panel mounting BNC, which means it has easily accessible pins. The simple RF sampling circuit will fit directly to the antenna connections, I might even be able to build it from SMT parts for the absolute minimum of adverse insertion effects. I dont really know why they didnt keep this circuit in the kit in the newer versions.

Heres what the manual has to say about this circuit under 'Transmitter Operational Adjustment -

If no wattmeter:
Install a T1-3/4 LED at TP2. The 'flat' edge of the
LED body should face to the left. When transmitting,
this diode should glow steadily when idling and will
flicker noticeably brighter when you're typing. Adjust
the master volume slider as needed to achieve this
indication. Note that this diode may be installed
remotely on a front panel to serve as a rough power

Im not exactly sure what a T1-3/4 LED is, but I suspect its a common or garden 5mm LED!

UPDATE - Ive just looked this up and it seems T1-3/4 is just the package style for a typical 5mm LED. So thats that sorted. Im off to bed now as its Sams karate grading in the morning and ive already had three glasses of wine...

DC Power input mod on SWL's PSK-20

 I have mentioned the Small Wonder Labs PSK-20 a few times, and it occurred to me earlier today that I really ought to explain these things a bit more when I mention them, before going on and on about what im doing to them!

Well, the PSK-20 is a small single conversion superhet transceiver from Small Wonder Labs. It is fixed tuned to the digital sub-band on 14MHz for operation of BPSK (binary phase shift keying) mode transmissions. Its a very simple rig in terms of use, you simply connect a 12v power supply, a 50 ohm antenna, and a pair of audio cables from and to a PC sound card. Then, running suitable datamode software such as FLDIGI, HRD, MultiPSK etc etc, this simple set should give about 2W of output, ample for plenty of fun making PSK contacts on 20m. It was supplied as a kit, but sadly by the time I wanted one the owner of the company had retired, and they were no longer available, so I bought mine from someone else, meaning I didnt have the fun of building it.

My only concern with the  transceiver was the lack of any protection on the power input. DC at 12v is fed via a DC barrel connector to the unit, via only a series diode. This means, apart from the Vcc being 0.6v lower than the incoming supply due to the drop across the diode, that the only protection is against reverse polarity. I doubt however, that the diode would really hold up in that situation for very long!

The unit is relatively simple, and so I was more worried about a  fault in the transceiver knackering an expensive power supply! Or, and probably of greater real concern, of causing a huge current draw and melting the power cable, a definite fire risk. As someone who has but one level of OCD, which is an obsession with fire extinguishers, this is what needed sorting.

The solution was, obviously, to fit a fuse holder! This needed a surprising amount of thought, as not only must it be in keeping with the layout of the rest of the rear panel, but I also had to consider the depth of the devices penetration into the housing. Too deep, or in the wrong place, and I run the risk of damage to the transceiver components. The photo shows the new fuse holder fitted in place. This holder had a little locating stud, so as well as drilling a 12mm hole, I had to file a little extra cutout for the locating stud. A small piece of heat shrink tubing covers the end terminal as further assurance against any contact with the board components.

As the PSK-20 has no power switch, I decided to add one at the same time. The connection to the fuse holder for the incoming power comes from the reverse polarity protection diode. The cathode leg of this was cut and the wire removed, the holder and switch then being wired in series.

The two other issues with the transceiver design were the lack of any indications of when the unit is powered, and when it is transmitting. To solve this, a pair of 5mm LEDs have been fitted onto the front panel

Keeping things simple, ive gone for red and green. The green LED will be the power indicator, and will be lit whenever the rig is switched on. The red LED will indicate Tx, and is a bit more complex! The drive for this will come from the addition of an RF power sniffer circuit, which will sense the RF power at the output of the LPF, just prior to the antenna connector. Its a simple diode detector circuit. The addition of this will also act as a rudimentary VSWR indicator, although im likely still to make use of a proper VSWR bridge in line with the antenna.

This is the whole thing back in its case, but without the lid. I was a bit concerned about the LEDs being too close, but there is in fact plenty of space. The LEDs need their current limiting resistors fitting, and then wiring up to the board. Im almost ready to try this on air, but I also have the power cable to make up before that. The audio cables have been tested. That reminds me, I need to organise a container to keep audio patch leads in! Ive got one now for USB cables, that sits under the PC desk, so all those cables are now tidy!

Friday, 21 February 2014

Small Wonder Labs PSK-20

The two 3.5mm jack leads have arrived, so I can now set up to try out the Small Wonder Labs PSK-20 14MHz PSK Transceiver. Ive waited this long to save dwindling down my meager stock of jack plugs!

Ive metered out the DC plug and socket, and all looks good there. However, Im not happy about directly connecting this little beauty to my PSU, which is capable to over 20A, so intend making a small modification tomorrow and fitting a fuseholder to the connector panel of the unit.

 So the cabinet has been dismantled to remove the panel ready for marking and drilling. Theres plenty of space to install a fuse holder on the connector panel. The 'front' panel itself is completely blank.

The SWL PSK-20 is internally a handsome bit of kit, sadly now out of production. All being well, I will have it on air tomorrow evening. I still need to give the soldering a visual inspection, just in case, as it were, although ive no cause to think there are any likely problems

Ive also ordered another DVB-T dongle, identical to the other I have on order. This is because I have an idea or two regarding doppler RDF, but might try a few other ideas out as well. If nothing else, the two mini-magmounts i'll end up with im hoping to modify to complete my TDOA (Time Difference Of Arrival) direction finder. This will just be a few PIN diodes and resistors.

No more from me for now. Ive been awake since 16:00 yesterday, first to do a night shift, but then I hung about and joined my colleague Dave Preston and four of the Field Support Team girls, for a visit to ITV studios and the set of the long standing soap Emmerdale

The photo above is on the set of the world famous Woolpack Inn, with myself in the role of customer. And yes, it is real beer! These sets are at ITVs Kirkstall Road studios in Leeds, where we also had a smashing dinner of fish and chips in the canteen, before visiting the 'external' sets, which comprise a custom built entire village, carefully built well away from prying eyes in the Yorkshire countryside.

So im now rather tired, and am going to bed.

Wednesday, 19 February 2014

The slow boat comes in!

I didnt get much radio done at all today, instead I decided to get some sleep in advance of night-shifts. I did, however, set up for 10% WSPR beaconing on 6m. The result of which, I found when I woke up, was a grand total of no spots whatsoever! Oh well, i'll stick to HF for WSPR until the conditions improve.

Several more items arrived today from China. These included the 10x 10k potentiometers and 5x SPST toggle switches, both packs cost less, including shipping, than just one item would have cost from a UK seller! I do try and support British business, but its a no brainer to me to buy parts for experimental purposes where I can get them at the best value. I'll support British business by buying local veg, meat and cheese, and      local beer (oh, and whilst there still in the UK - Scottish single malts!), or for components, from sources such as rallies, and club sales.

Also today came the 'windows 7 compatible' USB micro Bluetooth dongle. This cost a whopping 99 new pence, so I really wasnt holding out much hope. But Windows 7 saw and installed it no problem. So far so good then, I then tried to use it. It was at first a bit flakey, but then settled down. Bluetooth is very easy to use, once you know how, and it took me a while to work out what to set on both the PC and the external device (in this case my mobile phone) to be able to transfer files, but that learning curve mounted and a few files were successfully transferred each way. So, for less than a quid, im fairly happy with that. It should help Julie transfer pictures from her phone to the PC.

I'm still waiting on the 100k potentiometers, the audio jack cables and the SDR DVB-T receiver dongle. Hopefully these will arrive soon.

The mixer balun/doubler tuned circuit core for Wispy has been rewound as a 9t trifilar winding, and is ready to be re-installed. I feel somehow happier with this than with Rogers original 9+2+2 toroid, as I know all the turns are in the right place. Ive already swapped the 1N34s for 1N4148s, and will see just what the output is like. I have  a feeling something is amiss with the driver amp following the modulator, which might be why im not getting much output. I'll also improve the LPF with a 7-pole version.

All this will probably have to wait until saturday. Although I will be off shift by friday morning, im on a visit to Yorkshire Television and the Emmerdale studios, so im likely to be rather knackered after!

Tuesday, 18 February 2014

Possible 50MHz WSPR operation tomorrow

Ive decided to re-jig the station for more control over my 6m activity, with the idea of potentially putting WSPR on 6m tomorrow.

The big problem with 6m is lack of antenna monitoring. (gawd im starting to talk like at work!) so I have moved the Diamond SWR bridge out of the antenna feed line from my 2m station up to the halo, and will patch this into the 6m beam feeder line tomorrow morning when im back from hospital. I will be able to accurately set the Tx power on 50MHz then, although running to test load will still be a manual patch.

This will leave the 2m antenna unmonitored and the 2m station unprotected, but its not often used at the moment, so I can always put the Bird 43 in line if needed.

Now the storms are behind us, hopefully i'll get a spot or two on 6m!

Waiting for the slow boat from China

Several projects have effectively ground to a halt at the moment, as Im awaiting the delivery of a number of parts, the majority of which are on the slow boat from China, or, at the very least, on the low priority air mail from Hong Kong!

One that has arrived is the 1mA panel meter, but the swift delivery of that probably has more to do with the depot its come from being in Swindon! The wait for parts has delayed my live testing of the Small Wonder Labs PSK-20, as the necessary audio cables are on the shipment from the Far East, this being because at 99p each I simply cant make them myself for that price!

Little has been done radio wise other than the sound-card mod these last couple of days, and little will be done over the next either. Yesterday I sacrificed time in the workshop for a boys day with my eldest son Sam, and enjoyed the Lego Movie in 4D followed by pizza! Today I have been 'dataminer'ed' to death, and indoctrinated into the dark arts of 'site manager' at work. Tomorrow, after another hospital appointment, I may get a bit more done on one project or another, before I have to go to work and keep the UKs wonderfully educational television working for another night. On the plus side - tomorrow night is company sponsored pizza night!

Monday, 17 February 2014

El Cheapo soundcard - Modified

 A quick test of the 99p soundcard showed 2.2v DC on each headphone channel, which was easily solved using a small milling bit on my Dremel to remove the offending PCB trace.

A pair of 100uF 16v electrolytics were then soldered into place. As it turns out, on this board the silkscreening was correct, the + markers being in the right place.

The connections were probed again after modification with the card attached to my laptop. All seemed well so an audio feed was arranged from the FRG-100 tuned to the PSK sub-band of 20m, and FLDIGI started and configured.

The photo on the below shows FLDIGI running. Lots of nice clean decodes coming through. The level is a little high (the computers sound mixer is set to 0) but not a problem here.

I then switched to WSPR reception. The indicated audio noise level was too high, at 43dB, but I remembered what was needed to set up the other USB sound-card for WSPR, and found and switched off the cards AGC. Levels are now in the 20-21dB region. Im getting a lot of very nice WSPR decodes, including plenty of spots from across the pond.

Whats required now is to add the audio out feed and check performance on transmit. This is more awkward as I dont currently have a cable with the right inline socket to extend the Tx audio feed path.

No more radio play for now, its a boys day for me and Sam today and pictures and pizza await!

Sunday, 16 February 2014

El Cheapo Chinese USB Sound

The 99p USB soundcard arrived yesterday from China. My other USB soundcard, the one im running WSPR with, was from a UK seller and cost a whopping £1.49. So, are there any differences?

Oh, yes, indeed there are! The first difference, is that the plastic case 'feels' a bit more flimsy. Only a bit, and it has a few bits where the molding machine wasnt quite as clean and accurate as it should have been.

But the real obvious difference is internal. I had read up about this possibility with these cheap units, and its not in anyway a big problem for me, but if you bought one of these and happened not to be an engineer, it could cause you some grief. Take a look at the second photo. You can see, close to the two sockets, two pairs of solder pads. These are bridged by a PCB track. But what should be there, are the two DC blocking electrolytic coupling capacitors!

Without those capacitors, the audio lines have a DC bias on them. This might not be a problem for, say, an electret microphone, but its probably none too clever for your headphones or most other things you might connect, including in my case radios!

So it seems the difference in price is accounted for by the manufacturer saving a few pennies on a pair of electrolytics. This isnt a problem for me, as it was always my intention with this one to separate it from its case, and use it as a built in soundcard for a 'universal' digimode interface. I will be removing the connectors, wiring the soundcard PCB direct to the isolation transformers and level controls. The USB plug will also be removed and a panel mount USB socket added, along with line filtering. I will put suitable capacitors on at the same time.

I have connected this up to my laptop, and the drivers installed fine, and the computer sees it just fine. I will test the performance later.

Wednesday, 12 February 2014

Meters, OCXOs and strange goings on in the workshop

Mark G0MGX has returned Wispy to me, and now while im sat here in windy rainy Yorkshire, he's off working in A71 land! Im very grateful to him for the tests done on Wispy. I now know where to start on improving it and getting it working properly, and I also know to beware of the odd behavior of test equipment at times!

This occasional odd behavior showed itself again this afternoon. The photo below shows my mock-up 28MHz oscillator, and my frequency counter

So, doesnt seem much wrong there, does there? But look again. Look at the connections to the oscillator...

Apart from the scope probe, there are no connections! It ISNT POWERED! This lasted while I took the photos, and checked about the bench to see if I could fathom out what was happening. Then The counter suddenly went to zero, as it should have been! How very odd!

I put this bizarre occurrence behind me and cracked on with the tasks in hand. I didnt want to spend too long out in the workshop as the weather was really bad. The first thing was to test my various panel meters and determine their FSD currents. Having ordered a 1mA panel meter, I knew of course that some of them would prove to be 1mA, and I had a suspicion that the biggest and best for use as a meter for very low power levels, where I want a nice big movement, would be the ones to do so.

I rigged the variable PSU to 5v, and with a 100k linear taper pot in line with the multimeter set on its 2mA scale, proceeded to turn each meter up to full scale deflection (FSD) and read the required current.

Three of my best and biggest meters, all dB scale, and two of them being very big movement and anti-parallax mirrored, all proved to be 1mA! The trick now will be finding or building a box for the 50mW power meter that one of these beasts will fit! Were talking 3inch high and 4 inch wide meters!

My final task was discovering how to connect up a GFS-203A 10MHz OCXO (I think I referred to these at TCXOs last time). I tried to work the connections out from an old RC-690 processor board but the only one I could be certain of was the output, as this went to a coax connection. I was fairly confident though that I knew which was Ground, as it also connected to the devices metal case. But what voltage? A bit of research led me to the idea that 5V might be enough, and so I rigged it up to try

With a 5V supply, this little wonder fired up and put out an (excuse the pun) rock steady signal at 9.99994MHz. Just 60Hz off. A tweek of the adjustment trimmer and I found it can be adjusted +/- about 200Hz. Adjusted properly, it just sat there spot on 10.00000MHz.

Unfortunately, this means my idea of multiplying it up to 50MHz for WSPR is out, as there isnt the frequency swing needed. So i'll have to think of another way to get 50.294MHz.

Although still waiting for most other things ive ordered, my items from G-QRP club sales have arrived. I can now try a frequency doubler oscillator for 10m WSPR from 14.060MHz (which I know works) to feed as LO for TennaLady.

Ive rerigged the station back to HF + 6m, which means for now 6m WSPR is out. This is due to in this configuration the 50MHz port of the rig goes direct to the beam, without any metering, so I cannot set Tx powers for WSPR on 6m at the moment.

Anyway, thats enough radio for now. Ive ironing to do. Both mine and Sams Gi's need a press before tonights Karate session.

UPDATE - Odd frequency counter indication SOLVED!

It never occurred to me  at the time. I was sat here thinking, shall I turn off the 10m WSPR beacon? All while I was in the workshop, this has been running - on 28.126MHz! The workshop, and the counter, are about 16ft underneath the antenna...

Tuesday, 11 February 2014

6m WSPR - NO success

I allowed my 6m WSPR beacon to run from about 06:15 until 20:15 today, at a 15% Tx cycle, running 33dBm. Roger G3XBM also ran his beacon at 20% 33dBm, plus a few others down south popped up now and again. But unfortunately, not a dickiebird was heard by any of us!

I cant run 6m WSPR again for a while, as I need to put the station back to HF use. But it started me thinking about the possibility of a permanent 50MHz QRP WSPR beacon.

If I follow a DSB architecture, then the circuits are fairly simple. If I aim initially for Tx only then even easier. One of my SBL-1 DBMs will serve as modulator, a simple audio buffer amp ahead, and followed by a small chain of linear amplifiers up to, say, a few watts, and an efficient low pass filter.

The big issue of course at 50MHz will be frequency stability. How to generate 50.293MHz, in a way that will remain accurate? Well, I thought, I could start at 10MHz, pulled a little, and use a multiplier to extract the 5th harmonic. But how to ensure stability? Ah, perhaps a good use for one of these -

This  is the 10MHz TCXO from an old Marconi RC-690 public safety radio. I have at least two of these, although one is still attached to the rest of the RC-690! This, followed by a tuned buffer amp at the 5th harmonic (50MHz) should give me the accurate and stable LO signal needed. The only problem is I dont know the pin-out of this beastie! Luckily the other one is still on the RC-690s control board, so I should be able to trace the connections out. 

Whether I will be able to pull one of these far enough I dont know yet. IF not, then a more complex oscillator chain might be called for.

Monday, 10 February 2014

6m WSPR Ready

This evening I have rigged up to try WSPR on 50MHz. As usual Im set for 33dBm (2W), this time the antenna though is my rather low 3 element yagi, on a bearing of about 140 degrees (roughly SSE). This wasnt easy, as its essential to accurately set the output power, but I dont usually have a meter in line on 6m, so ive had to disconnect the HF port feed to the ATU, swap that with the 6m port, and transfer the 6m beam feeder to the unused Coax 2 port on the ATU. Im running at my usual 10% Tx cycle.

I will run a few test cycles tonight just to check alls in order, then if all is well run the beacon during the day tomorrow.

All together, accurately

One of the problems I have with all this construction, and in particular with the QRP side of things, is measurement.

I have for instance just built a little QRP power meter that should read about 100mW up to 2W, BUT, I have no way to calibrate it that I know is accurate! I tested it on my HF rig, at what I think was a couple of watts, but I only have the meter in the ATU to go by, and Im not sure thats accurate! So for even lower powers, making a meter is tricky as I dont know the power out of anything is accurate enough to calibrate it against. Swings and roundabouts here. I cant check the levels out of oscillators and small amps etc without some calibrated measuring kit, and I cant calibrate it without a known power signal source.

The 28MHz 3rd overtone oscillator I built however should be giving 7dBm (5mW), so its theoretically possible that I could use that as a known source, IF I can prove its output level. Im hoping I can measure it on the spectrum analyser. From that I should be able to build a QRPp power meter and calibrate it.

Ive learnt a fair bit about the pullability of crystal oscillators recently, in an attempt to find the best way of putting TennaLady onto the PSK and WSPR 10m frequencies. It looks like the best course of action will be to rubber a 14.060MHz crystal up a notch, and make the oscillator self doubling.

Not much I can do until the parts arrive now.

Ive also ordered a bag of snap-on ferrites to put onto the LED lighting circuits in the bathroom and kitchen to try and kill the DAB RFI. I dont think I have enough cable available to wrap around the cores, so will just have to hope I can fit enough on to kill the RFI as single turns.

Ive also bit the bullet and ordered a copy of 'Experimental Methods in RF Design', spending the very last of my hobby funds. Oh, well, another call-out payment due end of the month! Ive already, from a preview online, selected one of its designs to build! So long as I have a 1mA panel meter amongst my junk, I should have a 50mW RF power meter ready in a day or two, that should read down to at least the 5mW expected from the oscillator mentioned above.

Friday, 7 February 2014

Project TennaLady

So this is a new project, in effect a Wispy Mk II. The designated name comes from the fact that A) Its built for WSPR operation (and you'd only whisper about wearing TennaLady's wouldnt you?) on the 10m band (Tenna, geddit?), and is hopefully a more elegant solution (hence Lady),

that, and B) because it uses hermetically sealed Double Balanced Mixers, so I can avoid all the 'pissing about' with matching diodes and trifilar windings!

The Local Oscillator will use the 3rd overtone circuit built today, IF I can pull the crystal to 28.1246MHz. Unfortunately, overtone crystals dont like being pulled, so this might be a no go. If thats the case, I have a 28.060MHz fundamental crystal on order, and a 14.060MHz crystal. I KNOW the 14MHz will pull enough that when doubled it will work (thats what Wispy uses), but thats an extra transistor then! Maybe the 28MHz fundamental crystal will pull far enough, 65kHz to shift!

Its 18:40, its dark, and my 10m 500mW beacon is still being heard on the US east coast! And im still hearing five other US beacons!

One key aspect of Project TennaLady, somewhat in keeping with the nature of its namesake, will be shielding! I found today that the spectrum analyser was easily picking up the oscillator just by the proximity of the probe, so shielding the oscillator is essential to prevent LO leakage to beyond the mixer.

If I cant pull the crystal as far as the WSPR allocation, I will try for 28.120MHz, the PSK sub-band, thats just 60kHz to pull!

UPDATE - I tried taking a few turns off L1 on the oscillator, and it doesnt seem possible to pull the crystal more than a bit over 10kHz. The best I could get was 28.073MHz, before it jumped to 5th overtone. I'll just have to see what can be done with the 28MHz fundamental crystal.

Day of mistakes, but some successes

First it was the 3rd overtone crystal oscillator that I was trying to make work without giving it any power!, then, wondering why no one was hearing me on 10m WSPR, I realised I was still switched to voice operation, not data!

So both those are now working how they should, and the 500mW WSPR is being received on the US east coast.

Success however, with the proof of concept oscillator/mixer - The 2955B has decided to play again, and feeding the mixer RF port with a signal at 28.060MHz, a kHz or so from the oscillator frequency, I can see audio on the oscilloscope! So it works on receive as well. I'll need to filter it and feed it to an audio amp before I can measure the sensitivity. I suspect an RF preamp will be needed ahead of it for receive.

Also successful was the removal by heat gun of the relays and transformers from the modem cards. The 32kHz crystals also came off ok, but none of the buzzers survived! After some hard work and leverage the USB-B socket is also off the old BT router board, the rest is in the bin!

I will order the fundamental mode crystals and some more MeSquares from club sales later.

The plug for Sams tablet charger has arrived (although Toms extra memory hasnt yet) so time to do a repair job.

28MHz oscillator and mixer Proof Of Concept

Due to the odd troubles im having with Wispy, I decided today I would make use of one of my stock of SBL-1 0-500MHz diode ring double balanced mixers, and make a 'Proof Of Concept' mock up of a 28MHz DSB generator/receive mixer. As the SBL-1 is double balanced and entirely self contained, it means unlike Wispy theres no toroids to wind, and no diodes to match, so the isolation and carrier balance should be much better.

After a false start with a Colpitts oscillator design, I went with this circuit designed by the respected Wez Hayward W7ZOI, and blatently pinched from AA3SJ's website

Now this is designed for 32MHz, so I used the various online calculators provided by Amidon etc, and recalculated L1 for 28MHz. This proved to be 19 turns, which just fit on the same specified T37-6 core.

So, I mocked this circuit up, minus at first the L2 tap and the 5dB pad. And of course I couldnt get it to work!

So why wouldnt it work? Its a W7ZOI design, so I'd every confidence in it. Perhaps something was missing? I decided to check the volts on the base of Q1 - nothing? Hmmm, what about the collector - nothing?! Glance towards the PSU - BUGGER! Turn the flippin' power supply back on! Ah, look at that! 28MHz on the frequency counter!

At this point I came back in and made a cuppa, and had a miniroll, and sent an apology to the G-QRP club forum for being such a muppet!

Back at the bench, I had a brief play with trying to pull the crystal, as the output im getting is a tad over the 28.070MHz it should be. No success there. Not to worry, this is a proof of concept, the actual frequency doesnt matter at this time. So I added the 3t tap to the toroid, which now causes the oscillator to be a bit more picky over the setting of the trimmer capacitor! Too far one way and the frequency first jumps down a dozen kHz, then oscillation stops, too far the other way and it starts operating at 5th overtone on 56MHz!

I then added the 5dB pad, and the SBL-1 DBM

It was my intention at this point to feed it a 28MHz signal and see if I could get audio out (i.e. a bare bones DC receiver) but the 2955B has decided to pack-up again! So instead, I connected the spectrum analyser to the SBL-1's RF port, and fed the IF port with the two-tone audio oscillator. The result was what seems to me to be a DSB signal. As my spec. an. doesnt go low enough in spread to show the sidebands, I have to just hope thats the case.

I did however, notice that the analyser easily picked up the oscillator with the probe near the circuit. This is probably why I had so much trouble with Wispy. So the next version of this will have the oscillator screened.

But, the concept was indeed proven. A simple crystal oscillator and an SBL-1 can make a very easy DSB generator for WSPR or PSK use, at least on Tx. I'd like to know how sensitive it is on receive.

One problem though is the pullability of the crystal, which is pretty much none. For a converter this would be fine, but here I need to shift the frequency a bit, so a fundamental crystal is needed. Club sales have 28.070MHz fundamentals, so I will order one and another 14.070MHz crystal, and see if I can get a decent pull from the first, if not I will have to go with a doubler and start at 14MHz.

40m WSPR beaconing over night showed some good spots across the pond. Ive switched to 10m again today, again at 500mW, but propagation seems very poor and ive yet to be spotted, and am only hearing a few European stations so far.

Thursday, 6 February 2014

10m 500mW WSPR success

Ran up on 10m for WSPR this afternoon for a few hours, just as dusk was approaching. This time, to reasonably mimic what I want Wispy to achieve, I set up for just half a watt of output power (lord knows what the ERP is!). This was instantly spotted by a pair of stations on the US east coast! I dont know if the conditions were particularly good on 10m or not, but I was impressed by that result.

Im now running at 2W on 40m again overnight. Initially, I didnt seem to be getting many decodes, until I thought I hadnt sync'd the clock. This got me thinking - how often does Windows 7 look for an NTP (Network Time Protocol) sync? This turns out to be every 60,400s. A week! A bloody WEEK??

A quick Google (other search engines are available) and I found a guide to changing the NTP poll period. Rather than give you a simple 'how often would you like to sync?' option, you have to modify the registry. So my clock is now set to poll for NTP sync every hour. Since setting the clock, the decodes have been rolling in.

Mark G0MGX has received Wispy and run her up on his test kit. It seems I have no test equipment issues, but Wispy does indeed have a few problems, not least of which is that the low pass filter isnt cutting off early enough, so letting most of the 42MHz harmonic through. So I now know what to work on. The output power is also wayyyy low, so some work on the output level of the mixer and perhaps the driver is in order.

Ive just opened up an old Marconi RC-690 radio to start dismantling. It amazes me that so few people will strip old kit for parts. The 690, although its often a pig to remove bits from, is a VHF/UHF  Public Safety rig and furnishes some fabulous swag. As well as the common or garden parts, theres three UHF rated power transistors (these beasts gave about 30W), a 10MHz precision crystal oscillator, an SBL-1 500MHz diode ring mixer, a lovely little DPDT RF relay, and a fair bit of RG-178 miniature coax to connect it all up with! I'll start pulling bits out tomorrow. Even the case is solid die cast aluminium!

The 40m WSPR beacon has just reported that its heard LU8EX in Argentina!

Wednesday, 5 February 2014

40m WSPR beacon now QRT

I shut down the 40m WSPR beacon at exactly 22:00 after a successful test this evening. The USB sound card is behaving and working quite well. Both transmit and receive were doing good throughout the evening since switching over to the new soundcard, so I decided to call it a day at that point. I just had a brief nosey at what was going down on 40m PSK before closing the shack.

Ive now been awake 30 3/4 hours, including working a night shift, a hospital visit and a karate training session, and think I really ought to stop playing radio and get some kip.

So of course, before shutting everything down I popped into the workshop to take those modem cards and the new 3.5mm stereo jack sockets and put them in stock, and to move the 4m transverter out of its box and onto the bench ready for some more work tomorrow. I had noticed a musty smell and discovered that all the recent heavy rain had seeped in in one corner and the bag for the Larkspur mast kit was damp, thats now been moved and I will run the heater tomorrow to help dry everything out.

Whilst in there I took the meter to the potentiometer on my mocked up level control. Hmmm, its a 100k pot and it reads - 100k. Checked the wiper action and wherever on the track I had the wiper the value came out about right. So not the pot then. I then tested the jack plug... bugger, open circuit to the center pin! This isnt the first of these ive had that with, this particular batch had a few duffers in it. It seems for whatever reason the center pin isnt properly crimped onto its solder tag. I dont know if its salvageble, perhaps the center pin can be soldered to directly. So I will make up another plug for it tomorrow and retry the maximum value setting exercise. I now have a rough maximum level to aim for, a mere 10mV, and the design of a suitable 40dB 'T' attenuator to make a fixed maximum control. I need to do this exercise on the soundcard mic input as well, to nudge the noise level down a bit closer to the ideal of 0dB. I dont think that will need as much attenuation though!

I also need to dig out a big heatsink and plan a board layout for a build of ON6MU's 7W HF linear amplifier design. This is a 10dB gain linear that I intend driving with a 500mW input, or theres about, from the 10m Wispy. Once Mark has had it on his test bench I should know just what power I have available as drive for the linear, and hence my maximum power. David G4HMC has very kindly offered me a VN66AF power MOSFET, which is the active device. I dont wish to go too high, as Wispy is a DSB transmitter, and the lower sideband will become a nuisance to others at more than a few watts. I'd like an ideal of 5W SSB equivalent, but will see what I can drive the linear to. IF I can hit 3.5W equivalent (7W DSB) then that will do fine.

And now, children...

Boiiiiinngggg, said Zebedee, time for bed.

Plug 'n' Play-up

My previous post stated that I was getting no 'heard' spots on WSPRnet, but early this morning, that became a false statement. I heard a couple of US stations, but only just. Having been up all night, and then having a hospital appointment ah half ten, by the time I got home there was no point going to bed, so I took a look at the beacon.

After a fair bit of playing and head scratching, I realised it was a noise level problem. But fiddling with the soundcard settings didnt help. After a bit of fiddling, including being told by WSPR that my 'output device was invalid' (eh? I wasnt even touching that!) I found that the soundcard was set for microphone input (ok) but WSPR was set for Line In (ah!). I tried swapping to use the line in socket but that didnt help, so rejigged everything to actually be on the microphone input! Suddenly, with the mic gain at 0, I had an almost decent noise level, just hovering on the limits of the yellow section around 19-21dB. Suddenly my ears were syringed! Spots started to bang in and fill up the waterfall.

I tried my mocked up level controller, but it seems to have no effect! Im not sure what can possibly be wrong with  what is essentially a potentiometer with a plug and a socket attached, although maybe the choice of 100k is a bit too high? I'll test it later if im not too tired after Karate.

Some of the various parts ive ordered recently came today. Amongst them are the stereo jack sockets and the modem cards for dimantling

These are simple PCI cards and have a surprising number of salvagable bits on them. As well as the transformers they were bought for, theres a 47W class relay, a buzzer, a little crystal, probably 32kHz, some 1N4000 class diodes, and a smattering of capacitors. Plus the antistatic bags they came in will be handy!

The extra 32GB of microSD memory for Sams Android tablet came as well. £16 that cost, thats 50p a GB! I still think of myself as reasonably young, but I remember when memory was a £ a MB! And it came in a black rectangle with sharp pointy legs the size of half a breakaway biscuit! This memory for Sam is the size of my little fingernail, and most of thats the package!

Of greater consequence to my WSPR efforts, one of the items that came was the first of my USB sound cards. At £1.50 including postage, it was worth the gamble (the other one, coming from Hong Kong, cost 99p inc shipping, and is most likely identical!). Plugging it into a spare USB port on the back of the PC, the green LED lit up, well, thats a good start!

Instantly, Windows 7 saw it and started looking for drivers. Within a minute, it said it was ready. Ok then, so far so good, lets see how it actually performs.

A bit of changing in the WSPR software to make it look at the USB mic input (still on the on-board PC sound for Tx) and a few minutes playing with the Windows volume control panel, renaming the device so I can tell the difference between it and the on-board sound, and I put WSPR back to active and waited...

... MASSIVE noise level! +40dB! Luckily, I had anticipated this. The mic gain was set to zero, but I had noticed an AGC control was ticked. I unticked it, and the noise level dropped back to 20dB, just in the yellow. Ive had the WSPR software running on this now for 25mins, and theres plenty of received stations. So, its performing at least equally as well as the PCs on-board sound as far as receive goes. The 20dB I think is due to the radios output being just a tad too high.

Next step then, is to try the transmit side and see how this little beasties headphone output performs.

I have found a simple little 28MHz linear amplifier circuit to build, which should give me a bit of a kick on 10m WSPR when I get Wispy working. Its about 10dB gain, so with the Wispy giving its designed 200mW it should give a nice cool running 2W DSB signal, for an SSB equivalent 1W of USB WSPR. If I can find how to ramp the Wispy output up a bit to 500-1000mW DSB, then I can happily run a bit more power. I just need to get hold of a VN66AF MOSFET before I can start building, as its built around this device.

Something I noticed last night whilst updating this blog, was that my posts seem to be attracting at least one reader! It seems there possibly is someone out there who has so little else to do they are prepared to read my mindless ramblings! Whoever you are, feel free to comment, even if just to say hi!

UPDATE! - Found a suitable moment to put WSPR to idle, then set-up ready to test the USB sound cards headphone channel (Tx). Everything ready, clicked Test - No output. OK, stop being thick, and this time remember to change the plug to the new soundcard! That done, fired up for test into dummy load, and carefully set for 2W output. Bit tricky, as I think the level control pots track is damaged, either that or its a log pot! (quite plausible). With the level set to 2W, I set WSPR back to receive, and went off to put the kettle on. A few minutes later I returned and was curious why the receive noise level was 0dB!? Oh, try switching back from test load to antenna! This I did, but not before i'd heard EI6KH on the test load!!!

Now running the beacon entirely via the new USB £1.50 (inc postage) sound card. I wont decide if its good until ive seen a dozen or so transmit cycles out, but already ive five hearing and two heard by spots, and thats just one transmit cycle and a bit of fiddling so far!

40m WSPR overnight

My 2W WSPR beacon has been running overnight on 40m. It doesnt seem to have been very successful this time, with long periods without being spotted. Just as the greyline hits the UK, a few spots from the USA have started to come in, but they are rather sparse.

What doesnt help, is that for some reason I am not being shown as hearing anyone else! I dont yet know why this might be, perhaps ive an antenna or receiver problem? But I suspect its a network problem, perhaps my set-up has got inadvertently set to not upload spots. It did crash and have to be restarted by Julie, but she sent me a screenshot that did seem to show all was in order. I'll be able to see whats amiss in a couple of hours time.

UPDATE -  I am now Hearing KC9NBV, but no one else, and no one is hearing me! Something not quite right methinks.

Tuesday, 4 February 2014

Careless WSPRs

After a decent lie in this morning, I finally dragged myself out of bed at about 11 and started with my list of jobs. I know what your thinking - 11??? but to be fair im on nights! Anyways, since my jobs list would keep me mostly away from the wireless for the afternoon, I set up for WSPR on 20m, then cracked on with dismantling the drains...

After spending quite a while unblocking the drain from the washing machine, for which I found the only thing I had flexible enough to get in there was my 6m whip antenna, I got around to another task on the list - testing the dual band mini-magmount antenna. Well, with the MFJ-259 attached, I put the antenna on a conveniently located groundplane and swept it, resonant at 135MHz. This seems bad, but my convenient groundplane was the top of the fridge! So, I zipped up the neck of my jumper and went and tried it on an actual car. The results were very good. Apart from an odd spot near the drivers door where the 2m SWR came out as 1.7:1, almost everywhere else on the roof was 1.5:1 or better, often 1.1:1. This doesnt mean the antenna is efficient, but it does present a usable match. When I next get time i'll connect one of my little dual band handies to it and try to fire a few repeaters, see just what it will work. You may recall that the original magmount base for this had water ingress in the coax, something that the person I got it from roundly denied when I berated him for it this evening!

By this time it was starting to get late and 20m was dying. After picking Sam up from school, I retuned for 40m WSPR operation. Again running just 2W, and getting a few good spots from the US and Canada, plus Isreal, as well as Europe.

Mark G0MGX has kindly offered to give my 10m Wispy transceivers Tx chain the once over. His test kit is somewhat more modern and advanced than mine, so he will check to see that the mixer is indeed producing clean DSB, that the carrier suppression is adequate, and also check to see if the PA is performing and what output power its providing. Ive also asked if he'd check the harmonic suppression is acceptable. So Wispy is boxed and addressed ready to send tomorrow. I have plenty of other projects to work on whilst its away, the 4m transverter for one! Or the ARDF 80m beacons ATU, Or the Cobwebb, Or the 4m Moxon...

Just before leaving for work, I put a big notice on the computer saying not to turn it off, and instructed Julie that I was running the 40m WSPR beacon and to leave the PC on when she went to bed. Of course she forgot, turned the PC off, then back on again and emailed me to ask what to do to put things back on. So I talked her through it by email, all seemed well and good. Until a little while later, after she'd gone to bed, when the map on WSPRnet went blank with the notice 'no activity'. I started to think what could have gone wrong. Tx failure? No, theres no received spots. Not uploading? No, theres no 'heard by' spots either... later it dawned on me. I'd asked her to ensure the 'Idle' box was ticked. Which she did. Of course, I should have said UNticked! So, the radio and PC are sat all night, calmly running the WSPR software, in idle mode, just waiting to be told to start working!  Bugger.

Sunday, 2 February 2014

80m digimodes - PSK31 and WSPR

As part of a plan to try out WSPR on 2m soon, I tonight reinstalled the WSPR software, and have started it up, just in receive mode, on 80m. The reason for choosing 80m is that the radio and ATU were already set up there for PSK31, so it was just a quick turn of the big knob.

Just before downloading WSPR, I was watching the waterfall on FLDIGI, and spotted GB6WLB, the SOS Radio Week station from Walton-on-the-Naze RNLI station. As a very big fan of the RNLI's work (and having been one of the people who selected their Swissphone pagers for them!) I jumped on and worked them. I still havent quite got the hang of the macro buttons on the software!

With WSPR now downloaded and installed, I proceeded to try and get it working. After a few minutes head scratching I remembered that unlike most data modes, the waterfall is not dynamic, and I have to await the end of a decode period before it updates! This done, I could start trying to set the receive level. Strangely, no matter what mic gain I use on the soundcard, the indicator goes back to -30dB a few moments later. But, it is decoding successfully.

Theres me, smack bang in the middle of the UK. Once Im happy that alls well decoding, I might be tempted to set the radio to about 2w and Tx as well, see where I get.

The ultimate idea is to try WSPR on 2m using me trusty old Yaesu FT-290mk1. This is a venerable old bugger, but whether its frequency will be stable enough I dont know yet. I need to make up a mic lead to allow me to feed the PC audio into it before I can do much else. Also, a bit of a worry is that my halo antenna on the apex of the roof is fed by a long run of RG-58, not exactly the lowest loss feeder! I do have some unmeasured long lengths of RG-213, so I might ponder swapping the feeder at some point.

On the subject of antennas, yesterday at work during my break I made use of the Baseband desks toolkit, and fitted a crimp on BNC plug to the RG-174 coax of that little magmount that came with the DVB-T2 PCTV dongle, the element of which I had already swapped for the 2m/70cm dual band antenna from the old mini magmount I got off Richard. In case there was any impedance matching network to sort out, tonight I dismantled the old duff magmount to see what was inside

as can easily be seen, theres no matching network inside to worry about. What is worrying, is that the braid was just clamped between the magnet and the casing, hardly good VHF practice! But it does suggest that the rebuilt version should work ok, I will test it tomorrow, always assuming of course its not raining.

For now, im going to sit back, enjoy my well deserved two biscuits and glass of Aberlour, and watch the spots come in

'All in One' Digimode Interface design

Whilst working out the design for my next digimode interface, which will be used with the Small Wonder Labs PSK-20, I looked online at problems and mods for these dirt cheap USB sound cards. One thing that seems to be a trouble with some of them, is a large DC offset on the headphone port. This it seems is due to tight arsed manufacturers saving a penny per unit (or probably per 100 units) by not including blocking capacitors in the audio lines, but instead just having the pads bridged.

So, knowing that one of the first things Im going to do with these sound cards is take them out their case and check them, it occurred to me that I could leave one out of its case, remove its USB plug and the jack sockets, and wire it directly to the interface isolation transformers and level controllers. By including the Tone Controlled PTT circuit as well, this would create an interface that requires just plugging into a computers USB port, and connecting up the radio.

The USB sound cards are fairly simple inside, everything being done by one VLSI chip

But a problem with including it in the box with the rest is that I'd need a USB socket, not plug. Luckily, somewhere in the loft I have a defunct BT Voyager 210 router, which has a USB B socket that I can salvage

Theres not many other bits on it that are likely to be of much use, as it used to run very hot, and I wouldnt trust many of its parts. Perhaps the DC socket and the reset switch could come in handy.

Had a better look at the HF antenna socket on my car roof today. I think the problem is a bit of tarnish on the threads, what I need is to give it a good go over with a wire brush!

DVB-T dongles and old modems

Ive ordered a DVB-T dongle to try and use as an SDR receiver instead of the PCTV one, which doesnt seem to want to work for anything. Having read various reviews, and a couple of articles in Radcom, ive opted for one based on the RTL2832U decoder IC, and the R820T tuner. This combo seems to be touted as having a frequency range a bit lower than the E4000, but is easier to get. Im not too bothered about coverage above about 1.5GHz anyway, so the extra coverage in the lower spectrum is preferred,

 It might take a while to get here from Hong Kong though. I'll probably replace the plastic case with a metal box as suggested in Radcom, along with the power line filtering and a proper coax connector. I have a few small die cast aluminium boxes spare, or it might be another Altoids job! In conjunction with the free download SDRSharp software, Im hoping to put this to use monitoring the UHF military airband.

Various parts are on order for building another digimodes interface. The central part of these is the isolation transformer. Unfortunately, these new are rather expensive and hard to come by. Luckily, what are cheap these days are 56k PCI modem cards, which have on them, along with a few other useful parts, 1:1 600 ohm isolation transformers!

The photo shows a typical example, the transformer can clearly be seen. There is also a DPDT relay, some electrolytics, and a little buzzer! plus various SMT bits which are generally not worth the effort, but sometimes amongst those are SMT LM386 audio amps. The big difficulty is removing the transformers without breaking the very delicate wiring. Cutting the board away is usually the best idea until only easily unsoldered bits are left.

The various jack sockets and potentiometers I need have also been ordered from China. I know people will say to 'buy British' but when I can get ten parts from China with free shipping for the same price as one or two from a UK supplier, knowing full well that the UK suppliers parts are made in China, then why should I be out of pocket?

Tests of the Alinco DX-70 and the digimode interfaces level control last night showed that the same overdrive power reduction occurs on both high and low power settings, so it is almost certainly an excess audio level problem. Monitoring 80m last night showed a rather full band of PSK signals, but I was generally too tired to make any QSOs.