Info on the Popcorn Superhet RF Amp
Discussion
I have received a lot of emails regarding this RF amp from the popcorn
superhet receiver. This amp was designed using a calculator and as it turns out,
my input and output impedance was off by several ohms from the target 50 ohms input Z and 200 ohms
output Z. Today we have computer programs to calculate the resistor values, so
crunching the math is not a factor any more. This RF amp is preceded by a 50 ohm diplexer and followed by a 200 ohm pad, so it actually matches the
preceding and following stages very well considering that it is clearly depicted to be a popcorn receiver. If I had
to redesign this amp in 2008, I would use software to design a CE amp using "noiseless" Norton
feedback, Still, this was a good sounding receiver at the time.
Probably the weakest section of this receiver is the audio preamp from Q5 on through to the LM386. I would likely never use these
noisy stages in 2008-9. Instead I would use low noise op amps. This web site like my own abilities has grown over time.
In the following section, I have paraphrased some writing from Wes, W7ZOI concerning the Q2 RF amp.
An Analysis by W7ZOI
The input impedance of a common emitter bipolar transistor amplifier is related to the standing current, the emitter degeneration resistance, the transistor beta, the F-t, and the operating frequency. Most of those details
can be well modeled with the hybrid pi model. If you put quite a bit of emitter
degeneration in such an amplifier, the gain drops while the input impedance
increases. However, it is much less dependent upon beta and frequency. The
emitter degeneration is one form of negative feedback. But now we introduce a
second form of negative feedback by applying some of the collector voltage to a
resistor that generates a current that is fed back to the base.
This feedback
tends to reduce gain just as does emitter degeneration. But it also tends to
reduce input impedance. By playing a careful balancing act, you can design an
amplifier that has a 50 Ohm input and output impedance with a reasonable amount
of gain. This particular circuit has a gain of 21.7 dB at 10 MHz and an input
return loss of 14 dB. The output return loss is even better at 18.6 dB. Z-in is
39-j14 while Z-out is 172-j34.
These are calculated values using that hybrid pi model. I have devoted a few pages to a discussion of this feedback amplifier topology in the ARRL 2003 book, Experimental Methods in RF Design. This was co-authored by KK7B and W7PUA. You can get the book on line from the ARRL. The book includes computer programs that will handle the design chores. One deals with bias issues while another does the RF design and analysis.
The RF amplifier under discussion may be found on this web page
