OK. I'm going to post a simple PLL and modulator circuit and a PCB layout. It can also be built on Veroboard! It will give you 100mW on any MW frequency. This is enough for house and garden, and to get a feel for the way it works. We're going to use a simple "chopper" modulator, and a couple of cheap transistors for the output stage. I'll post a 100mW in, 5W out amplifier to go with it shortly. The whole thing can be assembled into one box - with careful layout.

It's designed for a dual-rail supply (+/- 12V), because this simplifies the biasing in the amplifiers. The exciter (100mW) part has just two moulded chokes, but the amplifier stage will require some coil winding - making a pair of RF transformers (one in and one out) and the coils for the output lowpass filter.

I'll also provide details for a basic audio processor which will give the right frequency response, and provide compression and limiting to keep the average mod level high (to get over the atmospheric and other noise), but without allowing over-modulation which causes distortion and interference.

If you shop around for the parts, you should be able to build the whole thing for £30. I plan to make PCBs available through OSHPark or another of the Far-Eastern PCB companies. You'll need a multimeter, a medium wave radio and (possibly) a frequency counter to do the basic alignment.

The circuit is an adaption of one that was made available in the 'States in the 90s. A company over there called Panaxis made kits available. Their version wasn't fully frequency agile, but got a lot of people into basic AM operation.

More to follow shortly!

Sounds very interesting, particularly in veroboard form. Keen to hear more.

Be interested in trying this especially if there is a PCB too.

"couple of cheap transistors for the output stage" 2n3904 as I have hundreds

The amp would be a nice addition, IRF510 - cheap and cheerful device!

Soldering iron at the ready

Watching with interest. My soldering iron might also be at the ready!

OK. Just tidying the documentation. I've got a prototype here running 5W carrier, 20W peak. Sounds good too.

Please don't forget about this! I need something to do for the next few weeks of lockdown.

OK. Here's a crude MW job. The carrier power can be around 8 Watts, and the peak will be around 30 Watts with a 30V supply. It's capable of just under 100% mod, and has a very basic audio limiter preventing over-modulation. There's no audio filtering - it's best fed with a lowpass filter to prevent adjacent channel interference.
I've had one of these (built on a mixture of Veroboard and point-to-point) working into an aerial at my place in Croatia for some time. It covers a good few km in each direction in the daytime, and sounds pretty good on the air.

The transformer is wound on a Type 43 ferrite ring of about ⅞" diameter. Mine had 18 turns on the primary and 34 turns on the secondary. The ratio will change with supply voltage. Don't go above 30V supply or the TDA2030 will go bang. You can use reduced supply voltage (mine runs on about 27V) and get a bit less power out. Mine gives about 7W carrier and peaks close to 30W (it's set up to be slightly asymmetrical). The output coils are calculated for 50Ω impedance at your chosen frequency. They're wound on Type 2 (T50-2) toroid cores.
The output capacitors have to be fairly high voltage types. The input and output ones have 50Ω reactance at the output frequency, and the two in the middle give a total Zc of 25Ω.

The 100µH chokes can be as high as 500µH each, but be sure that they can handle the current!

The CMOS chips MUST be ordinary 4000-series. If you want to use the faster 74HC or HCT parts (if you want to go up into the shortwave frequencies) the 7812 will have to be replaced with a 7805. You'll also need to change the crystal to either 5.120MHz or 10.240MHz for 10kHz steps.

Thanks Albert, Food for thought, will have to see whats in the junk box!

Nice work Albert!

Forgive me if this is a silly question? The TDA2030 is quite a substantial audio amplifier, Why do MW rigs need such a high level of MOD drive?

Is it an impedance issue? As we all know the TDA2030 is designed for 8 Ohm loads.

Looking at your diagram? The Drain of the IRF540 is driven by the TDA2030 via the transformer?

The actual carrier frequency enters at the Gate of the IRF540?

I think thats correct?

Haha!! Sorry, I think i have just answered my own question!

Forgot to mention - the glass diode is a 1N4148 (or 914). The bigger black diodes are 1N4001.

The reason for using an audio amplifier with a bit of "grunt" is so there's plenty of current available for the PA, without the amplifier having to struggle. I have used the TDA2003 for the same use, but it's easy to kill it if you're working into a poor match.

I usually build this sort of thing on top of a big heatsink. The one I have running at the moment is on top of a 0.45C/W heatsink, and has a case made from pieces of thick double-sided PCB material, with the "case" (actually just a cover) screwed to the heatsink with self-tapping screws. The LEDs poke through holes in the case, and the audio input (a phono socket), the output socket (an SO239), and the power input (a 5mm power connector) are all screwed through the case. I spray-painted the case with cobalt-blue car acrylic, and it almost looks professional!

The part that gets hottest is the audio chip. I mount it through a rectangular cut-out in the board with a mica insulator and a nylon screw, so that the tab is isolated from the metalwork, to prevent hum loops.

These are only little rigs (as MW rigs go), but matched into a reasonable aerial and a good earth, you'll cover a fair-sized area at little cost.

Finally - I include a 6.5kHz low-pass filter inside the box (built on a bit of Veroboard), using a quad op-amp. I include some mid-boost, so it doesn't sound like it's coming through a sock (so it's unlike Caroline).

The one I have running is now on two "leisure" batteries in series, for about 27.5V supply. I have an old battery charger connected across them all the time, and the rig stays on the air when the power goes off (as it does quite often over there).

Hi,

At the risk of my crappy eyesight making me look silly - is not the 100nf cap on pin 16 of the 40103, feeding the dip bank, actually shorted across on that pcb layout?

Stu

I think it should be between pin 16 and ground.

Premier-Carousel wrote: ↑Tue Jan 26, 2021 12:31 pm Hi,

At the risk of my crappy eyesight making me look silly - is not the 100nf cap on pin 16 of the 40103, feeding the dip bank, actually shorted across on that pcb layout?

Stu

Stu - you're right. My mistake. I'll correct it and upload.

I've also got a "flea-power" job that uses a dual supply, and has some interesting properties. I also have a 5 Watt / 20 Watt peak amplifier for it that's really easy to build. I'll put it all up here shortly.