Old Rig

[Albert H]

I used to make my exciters thermally separated from the hot parts (the Power Amplifier and the Power Supplies).

One run of rigs were built from three aluminium boes (H.L. Smith ones) assembled into a "U" shape, with the exciter at one end, and the PA (mounted vertically) at the other. The exciter box was attached to the middle box with nylon screws, with a piece of polystyrene ceiling tile sandwiched in there.

Some of them had a handle across the top (between the PA and Exciter boxes) to make them easier to carry. The middle box had the heaviest parts (mains transformer and smoothing capacitors), which helped it to stand upright. Looking back, it was a good way to construct the rigs with the maximum screening between sections.

Coupling between stages was done with RG58 (or smaller) coax. Many or them were retro-fitted with PLL boards, and we built them with several different output power ratings from 50 - 200 Watts.

[Shedbuilt]

Funnily enough, I think the silver mica caps I was talking about, came from the metal trays in HL Smith. Always remember the white haired old feller in there (can’t remember his name though). Also had a number of cases from them (they used to do a few in grey hammerite type finish. Did quite a few with exciter and PA thermally separated too; mounted in separate boxes, with each box screwed to blockboard. Some used DC coupling via chokes; so that the exciter box could be powered via the co-ax connecting the PA and exciter (what you might call phantom powering). Very early days again.

[Shedbuilt]

Here's another blast from the past. I have little doubt that Albert will recognise this design and probably the article too. Obviously not suitable for use on air as shown.

[EFR]

Radio Is My Bomb by Hooligan press!

[jvok]

Lol who needs harmonic filters anyway?

Seems strange to have the short length of coax from pot to mod input but then just normal wire from pot to jack. Also normal wire for RF between the stages

[Shedbuilt]

Radio Is My Bomb by Hooligan press!

Got it in one !

[Shedbuilt]

Lol who needs harmonic filters anyway

Yep. Welcome to 1981 LoL.

[Shedbuilt]

Seems strange to have the short length of coax from pot to mod input but then just normal wire from pot to jack. Also normal wire for RF between the stages

Yep. Probably best not to overthink or overanalyse this one. The oscillator is decent (above average for a VFO). The RF amplifier (final) is fairly standard, but the intermediate stages are a bit questionable.

[Albert H]

Way back, I used an oscillator a bit like that, with a dual-gate FET. As I recall, Alfred D wrote it up and put it in FRM magazine in the Netherlands. If you used the specified parts, and constructed it rigidly and neatly, the stability was pretty good. After having two rigs "take-off" because of RF feedback problems, I abandoned the use of "at frequency" exciters, and used frequency multiplication to get to Band II from a lower frequency source.

With careful construction, I could get remarkably good frequency stability from my oscillators. I used fairly elaborate buffer circuits after the oscillator, to isolate the oscillator stage from the subsequent stages.

However, PLLs guaranteed that rigs were as frequency stable as their reference crystals, with no need to spend hours selecting components to get the oscillator "just right".

[jvok]

Just out of interest whats the advantage of a dual gate fet in a vco? Normally you see them used as mixers with RF and LO fed to each gate but in the oscillator the one gate is just tied to +ve, seems like that would just work same as a normal fet

[Albert H]

The big advantage in my circuit (see below) is because I can use the second gate to optimise the output amplitude of the oscillator. No oscillator will produce exactly the same amplitude all the way across the band, so I use a simple AGC loop to guarantee the same drive level right across the band. The other advantage in this instance is that the input impedance of the device is extremely high and there's very low capacitance. In mass manufacture, it's a good idea to repeat the same component for various functions - "economies of scale" can save a lot of component cost.

VCO+ALC.png

In this circuit, I used an RF transformer for the VCO coil, with its secondary providing a very perfect, low harmonic content sinewave. It went a long way to ensuring that the subsequent stages would have no "loading" effect on the VCO. The "synth" output is a nice low impedance signal to drive into either a PLL IC or a logic-based prescaler. The two op-amps were an NE5532, which has extremely low noise.

Later versions of the above circuit were scaled for half-frequency and an 8V VCO supply, and the accurate output level made driving the diode-based frequency doubler stage much more predictable, which allowed the doubler to be "no-tune", whilst retaining minimal ½f breakthrough.