Page 88 - Build Your Own Transistor Radios a Hobbyists Guide to High-Performance and Low-Powered Radio Circuits

In terms of the emitter follower amplifiers of Figures 5-3 and 5-4, oscillation occurs

when the emitter of Ql and/or Q2 is loaded into a capacitor, which is always the
case because there is at least capacitance on the board from the emitter to
collector of Ql and Q2, as well as at the base (to ground) of Q3. The capacitance
loading at the emitter of Ql or Q2 causes a phase shift in the minus direction (e.g.,
phase lag or negative phase shift), but the internal capacitance inherent across the
base and emitter of Ql and or Q2 forms a high-pass filter with the inductor of Tl or
T2. Recall that when the variable capacitor VCl or VC2 is tuned to the top of the

AM band, there is very little capacitance across the inductor of Tl or T2. The base
emitter
capacitance of Ql and or Q2 then forms a two-pole high-pass (and high-Q) filter
with amplitude gain and positive phase shift. When the positive phase shift cancels

the negative phase shift with amplitude gain, oscillation occurs. In a later chapter it
will be shown that this principle of introducing a phase lag via a capacitance load at
the emitter and a series LC filter via an inductor connected to the input base lead of
an emitter follower plus the base emitter capacitance will form an oscillator. Note
that the condition for oscillation in a system requires a 0- or 360-degree phase shift
with an amplitude gain of greater than 1 from input to output.
So in the quest to further reduce oscillations, a third design was done with success.

Now let us turn to Figure 5-5.
The L1 inductor and VCl capaCitor tank circuit is connected to the differential pair
amplifier circuit with Q1 and Q2. An emitter follower Q3 provides a gain of about 1
after the differential pair amplifier with load impedance drive into power detector

Q4.
The reason there is no oscillation is that Q1, which looks like an emitter follower,
has a gain of only one-half (e.g., 0.5 at its emitter). And the emitter of Ql is
connected via capaCitor C3 to the emitter of Q2. C3 is chosen to be large in
capacitance and is considered to be an AC short circuit at the RF frequencies. But
"looking" into the impedance of Q2's emitter is resistive, not capacitive. Therefore,

there is no appreciable phase lag or negative phase shift at the emitter of Q1. Thus
Ql's emitter has a gain of one-half and no phase lag, which then enables a
condition for Q1 to not oscillate.
In this radio, a 250-~H ferrite antenna coil is used with a 365-pF variable capacitor.
It is worth noting that the input resistance of Q1 is not as high as the emitter

follower circuits of Figures 5-2 and 5-3. However, it is high enough to maintain
good selectivity from the tank circuit L1 and VC of Figure 5-5.
This radio works fine at 3 volts, with a current drain of less than 200 ~A.
A thought did arise-how about using complimentary metal oxide semiconductor
(CM05) gates such as hex inverter gates for the RF and audio-frequency amplifiers

for a TRF radio? It turns out that when the inverter gates are biased to about
one-half Voo for a 74HC04 chip, both output transistors of the inverter gates are