readily available on-line from houses specializing in
discontinued stock. I like it because the feedback loop is
closed internally by a carefully designed driving amplifier
with appropriate input capacity, impedance, and
amplitude limiting for working with a quartz crystal. This
avoids overdriving the crystal, promotes good frequency
stability, and significantly simplifies your design task.
The PLL: Oscillators Enter a
Time was, for every really stable oscillator frequency
required in a design, a circuit designer constructed a
separate crystal oscillator, or at best, contrived to switch
one of a set of quartz crystals into an accommodating
positive feedback circuit. The 1950s and 1960s were the
heyday of amateur radio and there was a large market
for cut crystals in the authorized amateur frequency
bands, as well as in commercial and military ranges. Both
myself and one of the technical editors of this magazine
“cut their electronic teeth” on ham radio during this era.
With the advent of the phase-locked loop (PLL) in
integrated circuit form, frequency synthesis and
synchronization took a great leap forward. Designers were
able to free themselves from the tyranny of multiple
crystals in order to realize a range of frequencies in a
single instrument or communications device. Additional
applications for PLLs include AM/FM detection, accurate
motor control, and noise generation.
As we see in Figure 4, a PLL uses feedback to hold
the output of an otherwise wandering voltage-controlled
oscillator (VCO) in phase with a reference oscillator. Since
the PLL can lock the phase of a wide range of variant sine
wave oscillator frequencies in step with a single reference,
a broad spectrum of stabilized frequencies can be
produced as an output. And this stability rivals that of the
reference oscillator itself. Motorola offers an integrated
circuit phase-locked loop in the form of the high speed
CMOS MC74HC4046A1. Looking at this part’s datasheet
can be intimidating. Designing with PLLs in production
circuits is reserved to those with the courage of the
thoroughly uninformed. But you can have a lot of fun
playing with them!
Stability and response of the PLL feedback loop are
major challenges; I recommend Reference 2 as a guide,
where more acute minds provide a detailed analysis.
Computing the response of a PLL involves some rather
advanced mathematics and should be confirmed by
measurements. As in amplifier and filter design, the
transient response of the PLL (to a sudden change in VCO
frequency) will tell you quite a bit about stability, and
as previously discussed in this series, there is simply no
substitute for comprehensive and careful testing.
FIGURE 3. An integrated
crystal oscillator driver.
FIGURE 4. A block diagram of a Phase-Locked Loop (PLL).
involves jumping on, or in. Since crystal oscillator design is
a non-fatal exercise, you should throw yourself in the pool
and have fun. Unless you are an analog gunslinger, stay
close to actual circuit designs given in the particular
crystal manufacturer’s datasheets when you attempt “cut
and fit” departures for your own designs.
Alternatively, if you want to design crystal oscillators
for production rather than as a hobby, you should
embrace established circuit forms, master the appropriate
mathematics, and test your designs all the way to failure.
A comprehensive discussion of crystal and ceramic
oscillators would require many more pages than contained
in this fine periodical. In any case, I hope I’ve pointed out
the descriptive essentials, the caveats, and some of the
potholes along your oscillator journey. NV
A small kit (with a printed circuit board and all parts for the crystal
oscillator shown in Figure 2) is available from the author for
$14.95, with $5 for shipping and handling. Contact the author at
firstname.lastname@example.org or 619-795-1920.
(1) High-Speed CMOS Data. Motorola, DL129/D, Rev 6. May
Learning to ride a bicycle or to swim ultimately
(2) Radio Frequency Electronics. Jon B. Hagen, Ph.D. Chapter
14, pages 128–140. Cambridge University Press, 1996. ISBN
0-521-55356-3. Dr. Hagen is the Director of the National
Astronomy and Ionosphere Center at Cornell University.
May 2008 75