A pendulum clock counts swings. That can work only if one swing lasts as nearly as possible as long as the next. Christiaan Huygens realised the first successful pendulum clock in 1656; in 1673 his Horologium oscillatorium presented its construction and his geometric theory of the pendulum. The drawing on this page shows how closely theory and clockwork were linked from the beginning.
Why does a pendulum swing almost uniformly?
For an ideal pendulum moving through a small angle, its period is approximated by T = 2π √(L/g). It depends on its length L and local gravitational acceleration g, but not on the mass of the bob. Within this approximation the pendulum is isochronous: small and slightly larger swings take the same time.
The word “approximation” matters. A real pendulum follows a circular arc. As the amplitude grows, its period becomes longer. The first correction is proportional to the square of the angle. The difference is small at a few degrees, but a clock accumulates it over tens of thousands of swings each day.
What disturbs the beat?
Air resistance and friction drain energy from the pendulum. Without a supply, its arc shrinks until it stops. The escapement therefore gives it a small impulse on each swing while also counting the beat for the wheel train. This intervention preserves the oscillation, but it is not invisible: the strength, timing and geometry of the impulse affect the motion. Precision comes from a controlled interaction, not from a freely swinging ideal.
Nor does the length remain constant. NIST gives 11.5 micrometres per metre and degree Celsius as a typical coefficient of thermal expansion for steel. An uncompensated steel pendulum lengthens as it gets warmer, so the clock runs more slowly. For a seconds pendulum about one metre long, a ten-degree rise produces an idealised loss of roughly five seconds per day. Temperature-compensated pendulums were developed to reduce precisely this effect.
What did Huygens solve?
Huygens knew that a circular arc is not perfectly isochronous. His cycloidal cheeks constrained the suspension thread to a path whose period is theoretically independent of amplitude. Later anchor escapements made smaller arcs possible and reduced circular error in a more practical way. Yet air, material, suspension and escapement remained real parts of the clock.
The pendulum's strength therefore does not lie in perfection. It makes its deviations calculable, observable and adjustable. A good pendulum clock keeps time not because nothing disturbs it, but because clockmakers learned to address each disturbance in turn.
Sources
Christiaan Huygens, Horologium oscillatorium (1673), digitised by ETH Library Zurich
MIT OpenCourseWare, Classical Mechanics, Appendix 24A: Higher-Order Corrections to the Period for Larger Amplitudes
Allan R. Willms, Petko M. Kitanov and William F. Langford, Huygens' clocks revisited, Royal Society Open Science (2017)
National Institute of Standards and Technology, Engineering Metrology Toolbox: Temperature and thermal expansion
Science Museum Group, Early pendulum clock by Salomon Coster, c. 1657