Another Look at the BFO from The Progressive Communications Receiver from QST for November 1981

The Progressive Communications Receiver from November 1981 written by Wes, W7ZOI and John, K5IRK provides 2 of my all-time favorite analog superhet circuits: The BFO and IF amp/AGC system exalt fabulous design and function. Later, Wes -- along with Jeff, WA7MLH seemingly evolved this IF system into the hybrid cascode version presented in QST for December 2007. 

I've built 4 versions of the HYCAS IF circuit, but, I won't lie -- I prefer the IF system from the Progressive Receiver -- & I collected dual gate MOSFETs to build it back in the day. Fantastica.

BFO

I sought a 9 MHz Beat Frequency Oscillator to switch a a diode ring product detector in a new, popcorn 9 MHz IF superheterodyne receiver.   I built a couple standard BFO designs, but didn't feel happy.
To get a 50 ohm output impedance with low distortion -- plus an available power output close to 7 dBm takes a lot of buffer stage current,  filtration & effort; at least for me.  On the other hand, to make a low distortion BFO for 0 to -10 dBm available power output seems easy.

Then I remembered the BFO from the Progressive Receiver. Just 1 transistor gets it done. Popcorn! The output AC voltage is in the order of 2.3 volts peak to peak -- and the harmonic distortion is relatively low; especially for a bipolar transistor amplifier since an oscillator with a BJT amp tends to clip quicker (or more abruptly) than a JFET amp oscillator.

Above — My take on the Progressive Receiver BFO. Although an excellent design, I have never exactly copied any circuit in my life. e.g. 'I am an experimenter'.  I removed the DC from the secondary, bumped up the base bias resistors to reduce current loss through the base bias network to ground -- and found that a 35 to 5 turns ratio gave slightly better output amplitude + distortion performance @ 9 MHz.

The output circuit is part of the feedback loop so a delicate dance ensues when tweaking this interesting design. Further, a pi low-pass filter was added (918 nH inductor wound on a T37-6 dust powder toroid). Additionally, using math or software + measurement, a builder could adapt this pi filter to perform impedance transformation to boost output return loss.

The output was buffered/padded with a 5 dB pad to give an output available power of 6.26 dBm. I tuned the oscillator output with a 68 pF plus a 2.2 pF in parallel after removing and measuring the variable trimmer capacitor ( 70 pF ) after peaking the output in my DSO.

Above — The raw output of the BFO into a 50 Ω  input impedance DSO.  The BNC output port was connected to the DSO with a 50 Ω  coaxial cable.

Above — A bench 6 dB 50 Ω attenuator pad was placed between the BFO output port and the 50 Ω coax going to the DSO.

Above —  The output of the circuit shown in the schematic. Likely, the low pass filter is not needed, however, I demand a 50% duty cycle from my BFO circuit to switch a diode ring mixer. You might argue that the raw, padded output is good enough. This is a genius design after all.

In BFOs circuits with 2 - 3 transistors getting a clean signal output usually takes >= 10 mA between the stages -- or a 5th order low pass filter to mop up the mess at the end. This circuit gets it done with a current draw of ~ 7.7 mA and minimal filtration. I'm happy.

FILTER DESIGN using Web tools

From site data analysis -- just over 50% of my readers do NOT use the Windows OS. They use Mac, Linux, Android and a myriad of other operating systems. Many hardcore security-biased builders choose not to use Windows emulators -- and thus cannot run Windows-based software.  I recently got a question from a Norwegian reader who uses Mac -- How do I design my filters?

I've built up a collection of useful web tools that improve my design workflow. These mostly run java script in your web browser. For filter design, I recommend the LC design system of Marki microwave. Click here for the link  

For example, here is a screen shot of the filter I used in the BFO:

Above — My filter design by Marki Microwave's tool. I manipulated the cutoff frequency to get standard value caps ( 330 pF ) and used the plot ( unshown) to check that S11 at 9 MHz was OK. The Marki microwave site, tutorials and products prove excellent. 

Thanks and best!!

Above — This week snow drops bloomed in the garden. In the foreground the cat mint is budding. Spring emanates.