Revisiting a Familiar Circuit
As his final look back, for the time being at least, Lee Aldridge G4EJB revisits a preselector from the G3RJV stable.
Lee Aldridge G4EJB
So there I was, still looking through all of the work that George Dobbs G3RJV had put together over all those years of Carrying on the Practical Way when I stumbled on his take of a circuit familiar to me – one that I had already built.
About two and a half years ago, while wrestling with my Howes 20m CW transceiver, I thought a modest tuned RF amplifier might help liven up the simple Direct Conversion (DC) receiver as well as improve broadcast station rejection. I’d found an HF preselector circuit on the G-QRP Sprat DVD (Winter 1995) in Steve Ortmayer G4RAW’s Novice section. It was a groundedgate amplifier with very few parts and at that time that suited my very limited resources.
So what is a preselector? George describes that in his PW CoTPW March 2007 article: “The tuned circuits in the preselector allow only a very small portion of the RF spectrum to reach the receiver and attenuate other (unwanted) frequencies. Reducing the out-of-band signals increases the signal-tonoise ratio and reduces undesired effects such as cross modulation and broadcast signal breakthrough”.
George’s circuit is shown in Fig. 1 with photo as Fig. 2.
Well, I did build the preselector and it worked well in front of the Howes receiver without any detrimental effects I could hear. It did improve sensitivity and reduce the occasional HF broadcast breakthrough.
My Board
My board was similar to George’s example in his article with two minor differences, I’d used the 2N3819 FET and a 12V supply (so I had replaced the 100Ω drain resistor with 510Ω – though probably not that critical). But there was one major difference. I’d built it on copper clad stripboard! Some constructors would be horrified about building RF circuitry on stripboard, others have no such qualms. (At that time I hadn’t ventured on to other ways of building).
The preselector was to be fitted inside my transceiver without the dual 365pF variable capacitor (Fig. 1 C1a and C1b) as it would only be required for 20m. (At that time I started to call it an RF amplifier as the Howes DC receiver didn’t have one). Each section of the variable capacitor was to be replaced with a fixed capacitor and 10-40pF ceramic trimmer in parallel. I had no way of measuring capacitance back then and could only guess the capacitance value by the amount of moving vane meshed when the preselector was peaked on 20m.
Sadly I destroyed the board attempting to modify it and built another, again on stripboard using most of the parts from the original. The RF amplifier trimmers were peaked by assuming if they peaked twice, there was a good chance the tuned circuit was resonant at that point – well that was the theory.
It seemed to perform well and was fitted into the transceiver with antenna changeover relay and in/out of circuit relay. The RF amplifier and the Howes radio lived happily ever after – well, for about two years or so. Then one day, I switched on the transceiver and signals seemed very weak, switched out the RF amplifier and signals improved. Off came the lid and I found one of the ceramic trimmers had gone open circuit – yes, I’d had this happen in the past (we called it drying out then) and knew it just needed a few little turns backwards and forwards to remake the riveted connection on the underside – see photo, Fig. 3. Sure enough, the trimmer started to work again.
Why Revisit?
So why the revisit now? George’s article had sown a few seeds with me. A little line he mentioned and can also be found in the Doug DeMaw W1FB article he quoted... “keep the gate lead as short as possible” ... ... I knew my board was not laid out that well and I also have to admit one of the trimmers didn’t seem to peak twice as expounded in my theory.
With George’s article now as my guide, my plan was to rebuild the board using a variation of the Manhattan style of building on a small copper clad board with as many of the components as I could salvage again from the original.
That plan came to a halt almost immediately because close inspection of the RF amplifier revealed the epoxy-resin fairies had for some reason glued down most of the components. Just don’t ask me what possessed me to epoxy resin so much of the board at that time.
Again, George’s advice would have made a huge difference: “My usual method is to prepare on the main windings and cover the windings with a layer of bee’s wax to hold the turns in place. Bee’s wax can easily be melted with the soldering iron tip. The wax quickly burns off leaving the tip ready for fresh solder.”
With the salvage plan seemingly in tatters, the Arkwright streak took over and I decided it seemed tragic to throw away a ‘working’ board when a few enhancements might be worthy of George’s insights – particularly if money wasn’t involved.
One of the first concerns I had with my board was that a number of copper strips were not grounded so using old desoldering braid, I built up a much bigger grounded area on the board, hopefully to reduce RF resistance, earth paths and stray capacitance. One upshot was the connection of the FET gate was now much shorter to ground and another was better grounding of the tuned circuits as George also detailed, Fig. 4.
A final touch was a little screening between input and output (it also meant I picked up one last floating piece of track and earthed it), Fig. 5. I then cleaned the board of flux and checked for shorts.
Before I refitted the board, I checked the current drawn by the circuit to ensure I hadn’t got carried away with the earthing on the board. I measured 5.8mA – that’ll do. The board was re-fitted to the solid copper wire support and re-wired into the radio.
With the radio powered up, the RF amplifier was to be peaked in the CW section of the 20m band as it’s a CW transceiver, and checked for operation throughout the band.
The input trimmer still didn’t peak twice so, a little playing with the value of the fixed capacitor in parallel with the trimmer was required. Why the difference between the input and output capacitors? Well, my best guesses are differences in the two coils and maybe capacitance effects of the board. I’m fairly sure I would have wound the main windings with the same number of turns but with the other winding of similar gauge, it was very difficult to check this time around. Anyway, once sorted, the RF amplifier really seemed to perform better than previously and the Howes transceiver receiver input tuning (Preselector) peaked more sharply than before. Over a number of days, there seemed to be no sign of broadcast breakthrough. To be honest, I was really pleased with the improvements. Thank you again, George.
This may be my final tribute to George’s work for now but it has been a privilege to tread the path that he laid for us. I hope I’ve been able to demonstrate his knowledge and experience has helped me improve my very modest amateur radio station and may inspire you to have a go (or even do better). The PW Carrying on the Practical Way CD is well worthy of your time and money.
And a final quote from a G3RJV article that I think sums up his approach to his work: “If you can’t explain it simply, you don’t understand it well enough.” (Albert Einstein, 1879 – 1955)
The Carrying on the Practical Way CD is available from the Radio Enthusiast website at: