Come join me on a trip back in time when SSB transceivers were typically all vacuum tubes. Names like the KWM-2, Swan SW-120 or the Hallicrafters SR-160 or maybe a National NCX-5 were the rigs on the operating table. Soon we had the Hybrid rigs where all of the lower level stages were solid state but the Driver and Final were tubes (valves for our UK friends). The Sideband Engineers SBE-33 in 1963 was perhaps the first Hybrid rig.

Fast forward to 2022 when I decided to build a Hybrid rig only in reverse where all of the lower level stages are vacuum tubes and the final transmit RF Stages are solid state. But it wasn't until late 2024 I built the Solid State stages and now have a working SSB Transceiver. I now have settled on a single band and that is 40M. The current power output is 6 watts and so semi qualifies for QRP. In a brief two week period I worked eight states including Ohio which is my best DX from southern California. I do have a 40M Delta Loop for an antenna so that helps.

There is a long history with this rig as it actually took shape back in the 1970's when I wanted to build a vacuum tube SSB transceiver. I gave a go at it, but was not successful (read it sucked). I removed all of the components but kept the base plate which is now in this latest version. The original chassis contained most of the same tubes but had a 6BA6 for the Carrier Oscillator and not sure either a 6AQ5 or 6EB8 for the audio stage. So this current radio has a 50 year history right down to the Chassis Base Plate.

There are a couple of innovations in the current build including the Arduino and Si5351 -- no drift and spot on frequency calibration.

A technical issue was that I wanted to eliminate the typical exciter tune, plate and load controls. In the older designs these same controls did double duty as they not only were a part of transmitter tuning but the receiver front end was loosely coupled to these same controls and that provided the front end selectivity. So if you peaked those controls on Receive -- they were close to the same values needed to provide transmit resonance. Thus I opted for broad band networks where there would be a simple tune the operating frequency and automatically you were in the Band Pass filter range.

The technical issue was how to couple to the output of the 12BE6 Transmit Mixer. Enter the Magical FT-82-43 Ferrite Core. If you wind 73 turns on this core its inductance is about 2.5 millihenry. The Al value of the core is 471. This is now a RF choke in the Plate lead to the 12BE6 and for an output winding we use 2 turns. The transform is turns ratio squared so 73^2 = 5329 and the 2^2 = 4. 5329/4 = 1332.25. Now if we look at the 50 Ohm input of the Band Pass Filter that transforms to the plate like 66K. The plate resistance of the 12BE6 is like 400K and the impedance of the 2.5Mhy choke is 113K at 7 MHz. This all seems to work OK.

On the air signal reports are quite excellent and several times now I have reported to me that the radio sounds great but you can tell its old technology. Then it hit me! They are listening to a radio with a 2.4 kHz bandwidth and most of the stations now use SDR where 3 kHz is the norm so while the signal is not pinched and I did spend lots of time adjusting the carrier frequency -- it is still less than 3 kHz!

Based on the success so far, I am thinking of taking the plunge and replacing the solid-state RF section with two tubes (valves). One is a 12BY7 which is sort like the standard meat and potatoes choice for a Driver and for the Final a 12AQ5 run in CLass AB1 and good for about 10 watts. (Maybe a pair for 20 watts.)

 

 

At this juncture it is OK to ask why would you build this radio? Well you could answer that I got bored building all Solid State radios (somewhat true), but I think it was to try how to mate the Tubes section with the solid state part, Along the way I did learn a bit about matching to tubes with a broad band Band Pass Filter and how to cure the hum issue. But I also think that it is because this is not something often not seen done today.

I have a few curious observations from this adventure and that includes using the 6JH8 in lieu of the 7360 for the Balanced Modulator -- it works well in this application. I have lots of tubes in the junk box left over from working on many tube type boat anchors over the years. Thus it was relatively cheap to build from stuff I had. BUT -- the price of tubes has taken a U Turn. Tube manufacturing is not on a scale as it once was making tubes more scarce.

I think that there are but a few places in the USA that actually have a vacuum tube production line and those are for specialty tubes. Off shore you find production facilities in Russia and China -- but low production runs raised the cost. Thus the pre-owned sellers can command higher prices today for tubes. What could be had for say $5 now is $10. When I had thoughts of using tubes in the final my mind turned to a pair of 6BQ5's as I once had a set in an earlier tube adventure. Well shocker a pair can fetch $60. It also seems that the tube type audiophiles are a huge consumer of the ham favorites like 12AX7s, 6L6s, 6V6s and so that has also impacted prices.

 

Never Say Never!

Given my success with the Solid State RF Section, I now decided to shift over to a Vacuum Tube RF Section and employ broad band no tuning on the stages. The two tubes of choice were the 12DQ7 (like a 12BY7) which was used in the SBE-33 and SBE-34. For a Final, I picked the 12 volt version of the 6GE5. Thos who may have had a Heathkit hw-22 or one of the other monoband rigs will realize a pair can develop 100 watts output. I am using but a single tube and shooting for 20 watts PEP output.

The 12GE5 can be operated at low Plate Voltages (275VDC) and has a plate dissipation of 17 watts. For a Screen Supply, I will be using 185VDC and needs about - 22 Volts of Bias. All of these voltages are present in the current Hybrid rig. The 12DQ7 will be grid block operated and the 12GE5 will have the screen supply removed during receive.

I have in work a band pass coupler that will be used as the output stage for the 12DQ7 feeding the input of the 12GE5. It will give a flat respoonce between 7.0 and 7.3 MHz.

 

 

 

 

 

From a cost standpoint today, building a tube rig is a money pit. You could easily spend $200 to build this and that does not even include a power supply. Thus the solid state radio delivering the same performance with an IRF510 is about less than half the cost of the Hybrid and no HV power supply needed. But the real bonus -- I was able to pull it off and this rig is another unique oddity from N6QW.