@ XXL, Just a thought?

It might be worth checking the enamelled coils for proper connection?

I have had trouble with those before! A nice trick i have learned is to burn the enamel off with a lighter, Then scrape off the residue with a blade or abrasive paper!

I also check the connections once soldered with a multimeter to verify!

I'm no RF enginner so lets get that out the way!

So quick look over Alberts circuit, the transformer looks like its trifilar wound but looks wrong to me as the 2 x 6T are connected to the + rail not the collector side of the 4427???

Anyway I made the circuit up to the 100n (dc block). Dont know if because I've built it on copper clad and there are some long leads with RF getting into the connections but something is not right plus I'm getting a second modulated carrier a couple of meg up.

But am getting some output, not alot mind!

I cant comment on Alberts design? I guess he can give explain it?

The only significant spurs i get (@55Mhz) are one @ 27.5Mhz and the other @ 110Mhz which can normally be reduced by keeping the oscillator 'Symetrical' and adding a low pass filter!

What are you using to check the spectrum by the way?

The trifilar coil is just a convenient way of doing both the collector choke and the output impedance commercial. I found that with the output after the filter was a shade over 1 Watt with the "standard" NRG dual transistor oscillator driving it when running at 15V. My push-pull oscillator has three turns in each of the six coils, and is run from a 10.1 Volt regulator (a 78L05 on top of a 5V1 zener). The output stage is running at 14V1 for exactly 1 Watt and I have an LM317 variable regulator to set the supply voltage.

The PLL is a diode-programmed 74HC4040, a diode-programmed 74HC4060 with a 4 MHz crystal for the reference, and a phase comparator and lock detector built from a 4013 and a 4011. The loop filter uses a LM358 dual op-amp.

The whole little rig is in a small diecast box, and the purity at 1 Watt is excellent. It's tuneable from around 46.5 MHz up to 58 MHz, giving roughly the same power out at any frequency it covers. If I was building this years ago, I would have used 65p trimmers in the output filter to tweak for maximum power / best purity, but the trimmers these days trimmers would cost almost as much as the rest of the rig!

Ofcourse Albert is correct! There are endless possibilities!

It all depends on your goal? Mine was to make a simple 'Oldskool Pirate' link which was tuned manually via the front panel.

In the past i saw some excellent designs, Some similar to the BW VFO (Without PLL) which had bi-filiar transformers for the output stage!

The BLF244 was often present too!

Just to add!

I loved that the early B1 links had a power setting for the final amplifier which could be set for location and distance!

F*cking Genius!!

I hope to add this to my project soon!