Better than Nothing.

Anything on the telly?


Spying on the neighbours.

March 2012:  First comes the latest TV project. I don't know why I keep improving and developing these things, as there is little to see and the interference from local routers and other wi-fi gadgets is depressing. Still, it keeps me out of the pub.




This one uses a Panorama MMDS downconverter (2300 to 2500 MHz in/460 to 660MHz out), the output being 'tuned across' by a standard UHF TV tuner. This produces an IF of 30MHz which is demodulated with an NE564 PLL chip. A bit of amplification of the resultant video signal and it's ready to go to the little telly. Controls are: Tuning, video level, IF gain and video inversion. Another job done cheap and for no good reason!

The scene on the TV is of someone's back garden about 100 metres down the road. Looks like a bit of DIY going on.








It's an SWR meter


Jan 2013:  My recent experiments on the 472 KHz band showed that I needed an extra SWR meter. Borrowing the one that is permanently in line with my Topband system was a pain, so something needed to be done.  After much rooting about on the 'net, I settled on this circuit:



I used a 1 inch diameter ferrite toroid with a length of mini co-ax through the middle (remembering to earth only one end of the braid!) and the rest of it was pretty simple. To calibrate it, I ran enough power through it to give FSD in the FWD position into a 50 ohm load, switched to reverse and adjusted the 1k preset for zero. Then I checked that 100 ohms and 25 ohms gave equal readings and did the same with 200 ohms and 12 ohms. All seemed OK.

After almost demolishing the shack in my quest for a more sensitive meter, I've finally found a 100uA one which will do nicely. All I need now is a box.




Here's the PCB waiting for a nice enclosure. Those rather scruffy-looking sampling resistors have been tidied up now.

Although I wanted this only for the 472 KHz band, it seems to work well up to 30 MHz.













And here is the finished article. The nice fit of the meter in the box is pure chance, the meter face hiding some holes from a previous project.

It takes just under one watt for FSD at max sensitivity.













24 March 2013: Here's a little something that perked me up on this cold Sunday afternoon. Its a do-it-yourself copy of an HP slide-rule thing for converting return loss to VSWR.  Of course there are plenty of look-up tables available, but making one of these is fun.  A friend alerted me to it:


You can slide it around on the net, but making one up from card and using it 'hands on' is pretty nifty. I downloaded both parts as 'save picture' and printed them out onto ordinary paper. Then I stuck them onto card to make them more rigid and cut out the windows. A back part was needed, but this was just a blank copy of the front and the two halves were stuck together with spacers along the edges to leave room for the inner bit to slide freely.





Here it is. I like it because it's an authentic copy of an original HP one, and it slides an'all!











VFO's have come a long way.


July 2013: Remember struggling to keep that VFO stable? Those 'Tempa-trimmers' and 'synthetic rock' circuits? To be fair, they could reward you with a very stable oscillator, but usually only over a limited range and rarely at VHF. Then in the late sixties and early seventies we began to see the emergence of the PLL synthesiser with unambiguous digital readout and memory facilities. Wow!

I'm sure many of us have constructed a few of these, usually having to resort to pre-scalers and mix-down systems for anything remotely HF. And unless everything was carefully optimised, decoupled and screened, we could end up radiating frequencies that were never meant to be radiated!

Fascinating as they are, I no longer 'do' PLL synths. They can be a great plaything and the satisfaction of hearing one plop onto frequency when you apply the volts is difficult to deny, but ......

My first experience with ready-to-go VFOs came with the si570 160 MHz 'digital VFO' from SDR Kits. It took a couple of hours to build and delivered stable signals from about 3 MHz to 160 Mhz with digital readout, IF offsets pretty much everything you could ask for. They also did one that went up to 800 MHz so I had to have one of those too.

Of course you can never have enough knobs to twiddle, so when I saw an unbuilt, unwanted Cumbria Designs si570 kit going cheap at the Stockport Rally, I had to buy it. I also came away with an interesting 6 MHz DDS kit.

These all have one thing in common - they use cheap-as-chips rotary encoders for the tuning. OK they do the job, but they are clicky and the smaller steps fall between the physical detents, which makes the tuning feel odd. As I'd recently scrapped my old faithful (but lately fault-ridden) TS780, I had access to the glorious shaft encoder from that.





Will you take a look at that. No messing - a full-fat, properly-engineered optical encoder, designed with real men in mind. Once I'd discarded the original, rather odd electronics...


















.. and replaced it with something that made sense, it worked a treat on the little DDS board. There are 500 pulses per revolution, so with the DDS set to 10 Hz per step it provides a silky-smooth 5 KHz per turn.





















This is the Cumbria Designs si570 VFO. It covers 7 MHz to 250 MHz and has offsets, selectable steps, band limits, you-name-it. Another well finished kit from Cumbria.

That's the supplied shaft encoder.


















The DDS kit built up as intended. The encoder is built onto the main pcb in this one. It covers almost DC to 6 MHz without fuss, and tuning steps available are from 1Hz up to 10KHz. It's not as clean as the Si570 though, see below.















I was very surprised when I saw how clutter-free the si570 VFO  was (below). This is a 5MHz span centred on 7 MHz with (in yellow) the VFO off and (in purple) the VFO on. Nothing to see all the way down to -75dBc.






 And below is the little DDS in action. A few crawlers popping up to around -65dBc, but for non-critical applications it would be just fine.

Eagle-eyed readers may have noticed I'm now using a RIGOL analyser instead of the Signal Hound.....