■ FIGURE 2
varactor V1, and capacitors C7 and C8 (C7 is big enough
to swamp out stray capacitance from Q1). Note that Q1
is a common-collector amplifier (i.e., an emitter-follower)
with the tank circuit from base to ground. That might
seem surprising since an emitter-follower has a voltage
gain of less than 1. But for an oscillator, it's power gain
that counts. As long as the amplifier replaces the signal
power lost in resistance, the oscillation keeps on going.
V1 is actually a dual varactor to eliminate the
possibility of forward conduction at the sinewave peaks.
The frequency of oscillation is set by adjusting the DC
voltage on V1 with potentiometer R2 (see the sidebar
on varactors). R4 and C3 form a low-pass filter to prevent
RF from feeding back onto the DC. C5 couples the tank
circuit to the base of Q1, but blocks the DC bias (of R5
and R6) from shorting to ground through L. Capacitors C7
and C8 form an AC voltage divider to provide feedback
at the emitter of Q1 to sustain oscillation. A necessary
condition for oscillation to start is for the ratio (C7 +
C8)/C7 to be sufficiently bigger than 1.
Modulation
Modulation is done by superimposing an audio signal
onto the DC bias applied to V1. The audio comes from an
electret microphone (MIC). Since the electret mic has a
built-in FET amplifier, it supplies a relatively large signal.
The audio is applied to V1 via C2 to block DC. R3 serves
two purposes. First, R3 and C1 form a low-pass filter to
prevent RF from feeding back to the microphone. Second,
R3, R4, and R2 form a voltage divider for the audio.
Increasing R3 decreases the sensitivity of the microphone.
DC to power the mic is fed through R1, which also sets
the gain of the FET amplifier. IC1 provides a regulated five
volts to power the microphone and its built-in amplifier.
For detailed information about
Colpitts oscillators, see these websites:
www.ee.adfa.edu.au or www.rfic.eecs.berkeley.edu
Output Stage
The output of the oscillator is fed through C9 to the
Q2 emitter-follower. The output of Q2 drives the antenna
through C11. The Q2 emitter-follower does two things
for the circuit. First, it ensures that the oscillator is not
loaded down by the impedance at the antenna. Second,
it provides power gain to drive the antenna.
DC Power Decoupling
The transmitter is powered by a standard 9V battery.
As the battery ages, its internal resistance increases and as
current is drawn there is a voltage drop across that resistance. Since the current in Q2 varies with the 100 MHz
signal, you get an RF ripple voltage across the battery. C10
is there to bypass that RF. Also, R7 and C6 form a low-pass filter to make sure that any ripple that C10 didn't
remove is blocked from feeding back along the DC rail.
The Layout
Figure 3 shows the layout of the PCB. Since this was
my first SMD project, I wanted to keep it simple, so the
only surface-mount devices I used were resistors and
capacitors (non-polar devices). To make it easy to tell the
caps from the resistors, all the caps are size 0805 and all
the resistors are size 1206. I used through-hole capacitors
for C4 and C10 because I first built this circuit with all
■ FIGURE 3
September 2008 41
