The Sagnac effect is the difference in propagation time between two signals traveling in opposite directions around a closed path when the apparatus rotates. For light, this time difference produces a measurable phase difference when the beams are recombined, shifting their interference pattern. Named after Georges Sagnac, who demonstrated the effect optically in 1913, it provides a way to measure rotation without observing external landmarks. It underlies optical gyroscopes and must be accounted for in precision timing and satellite navigation.
Physical principle
A typical Sagnac interferometer splits a light beam into two beams that traverse the same closed route in opposite directions, using mirrors or an optical fiber. The beams then return to a common detector.
If the apparatus is stationary in an inertial reference frame and both directions have otherwise identical propagation conditions, the travel times are equal. If the apparatus rotates, the detector moves during the beams’ transit. In an inertial description, the beam traveling in the direction of rotation must catch up with the moving detector, while the oppositely traveling beam meets it sooner.
This explanation does not require light to have different speeds in opposite directions in an inertial frame. Both beams propagate at the same vacuum speed of light, ; their return events differ because the receiving apparatus moves.
The resulting time difference can be measured with pulses, but optical instruments commonly detect it through interferometry. Even when the time difference is extremely small, it can correspond to an observable fraction of an optical cycle.
Leading-order equations
For a closed optical path rotating rigidly with angular velocity , with apparatus speeds much smaller than , the travel-time difference is
Here is the oriented area vector enclosed by the path. Its direction is normal to a planar loop, with its sign fixed by a chosen circulation convention. Reversing the rotation or exchanging the labels assigned to the two beams reverses the sign of .
For a planar loop of area , whose normal makes an angle with the rotation axis,
Thus, a planar loop senses the component of rotation perpendicular to its plane. It has no leading-order response to rotation about an axis lying in that plane.
For monochromatic light of vacuum wavelength , the corresponding phase difference is
For identical turns of a fiber coil, the effective vector area is approximately . More generally, it is the sum of the oriented areas of the individual turns. Increasing enclosed area, adding turns, or using a shorter wavelength increases the phase response.
These expressions assume a stable path and the usual low-speed operating regime. Moving media, time-varying paths, or strongly dispersive resonators require a more detailed propagation model.
Circular-path derivation
Consider an ideal circular vacuum path of radius , rotating about its center at angular velocity . In an inertial frame centered on the apparatus, the detector has tangential speed . The two return times are
Their difference is
where . For , this reduces to .
These are coordinate times in the central inertial frame. The rotating detector measures proper time, giving
The distinction is negligible for most practical instruments but matters in an exact relativistic treatment.
Relativity and synchronization
The Sagnac effect is consistent with special relativity. A rotating apparatus is not a single inertial reference frame, so conclusions about synchronized clocks in a globally inertial system cannot be transferred unchanged to a rotating ring.
In rotating coordinates, the coordinate speed of light around the loop can differ between the two directions, although every local observer measuring light in vacuum obtains . This distinction between local measurements and global coordinate descriptions is essential.
The effect also exposes a limitation of clock synchronization around a rotating closed circuit. Applying local Einstein synchronization successively around the ring does not return a globally consistent time assignment without a time discontinuity or an equivalent coordinate convention. The observed return-time difference, however, does not depend on which synchronization convention is chosen: both return events can be compared at the same detector.
General relativity provides a broader framework for propagation in rotating coordinates and gravitational fields, but spacetime curvature is not required for the ordinary Sagnac effect. It occurs in flat spacetime. It should therefore not be identified with frame dragging, which is a gravitational effect associated with rotating mass.
Sagnac originally interpreted his experiment as support for a stationary luminiferous ether. That interpretation is not required by the observation, which is fully explained within relativity.
Historical development
Sagnac’s 1913 experiment used a rotating optical apparatus in which counterpropagating beams followed a closed path defined by mirrors. Rotation displaced the interference fringes, demonstrating the expected dependence on rotation and enclosed area.
In 1925, Albert A. Michelson, Henry G. Gale, and Fred Pearson reported a large terrestrial interferometer experiment that detected the effect associated with Earth’s rotation. Unlike Sagnac’s apparatus, the installation did not need to be mechanically spun relative to the ground: its attachment to the rotating Earth supplied the rotation.
Later developments in lasers, low-loss optical fiber, and optical detection transformed the effect from a laboratory demonstration into a practical sensing principle.
Principal instruments and applications
Fiber-optic gyroscopes
A fiber-optic gyroscope sends light in opposite directions through a coil of optical fiber. The beams are recombined, and their Sagnac phase difference is used to infer angular velocity.
In the standard case of a dielectric fiber moving together with the apparatus, the leading rotation-induced phase shift is independent of the fiber’s refractive index. Simply replacing with the light speed in the fiber in the vacuum formula gives an incorrect result. A consistent treatment must account for propagation in the moving dielectric.
Many instruments use broadband light sources, phase modulation, and feedback to extract a stable rotation signal. Because the sensing element has no spinning rotor, these devices can be compact and mechanically robust.
Ring-laser gyroscopes
A ring-laser gyroscope supports laser oscillation in both directions around an optical cavity. Rotation produces a difference between the two resonant frequencies. For a simple ring operating in the low-speed limit,
where is the cavity perimeter.
The resulting beat frequency measures rotation. Large ring lasers can monitor small variations in Earth’s rotation and rotational ground motion. Backscattering between the two beams can couple their oscillations and produce lock-in, reducing sensitivity near zero rotation unless mitigated.
Navigation and timing
Optical gyroscopes are components of inertial navigation systems used in aircraft, ships, spacecraft, and other vehicles. Several differently oriented sensors provide the components of angular velocity needed to track orientation.
In a global navigation satellite system, including GPS, Earth rotates while a signal travels from a satellite to a receiver. Calculations performed in Earth-fixed rotating coordinates therefore require a Sagnac correction. Omitting it can produce positioning errors of tens of metres.
Sagnac corrections also enter precision clock comparisons and time transfer over long terrestrial links. Their form depends on the signal path and coordinate system used.
Matter-wave rotation sensors
Rotation-dependent interference is not restricted to electromagnetic waves. Quantum mechanics associates a phase with particle propagation, allowing atom, neutron, and other matter-wave interferometers to sense rotation.
For an ideal nonrelativistic matter-wave interferometer with particle mass and oriented enclosed area , the rotation phase is commonly written
where is the reduced Planck constant. In practical atom interferometers, the effective area is determined by the particle trajectories and the sequence of beam-splitting interactions.
Measurement limitations
The ideal Sagnac signal depends on rotation and geometry, but real instruments also respond to imperfections that can create or obscure a counterpropagating phase difference.
Important error sources include:
- Thermal changes: expansion changes the effective area, while temperature variations during propagation can produce unequal phase histories.
- Polarization and birefringence: unequal evolution of the two beams’ polarization can introduce phase errors.
- Backscattering and reflections: unwanted coupling can generate spurious interference or frequency locking.
- Optical nonlinearities: unequal beam intensities can cause different intensity-dependent phase shifts.
- Mechanical deformation: vibration and strain alter path geometry and optical properties.
A common-path design suppresses many disturbances because the two beams traverse nearly the same components. It does not cancel disturbances that affect the beams differently or change between their passage times. Practical rotation measurements therefore require calibration, environmental control, and an assessment of measurement uncertainty.