General Relativity
Frame Dragging
A spinning mass doesn't just curve spacetime — it grabs it and twists it around with the rotation
Frame dragging is the twisting of spacetime by a rotating mass — the Lense-Thirring effect. Earth drags a gyroscope by 37.2 mas/yr.
- Also calledLense-Thirring effect (1918)
- CauseAngular momentum of a rotating mass
- Earth's drag37.2 mas/yr (Gravity Probe B, 2011)
- GR prediction39.2 mas/yr (gyroscope precession)
- Extreme regimeKerr black hole ergosphere
- Weak-field nameGravitomagnetism
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Definition
Frame dragging is the prediction that a rotating mass drags the surrounding spacetime around with it. Spin a heavy enough body, and the local definition of "not rotating" — the frame set by free-falling gyroscopes and the distant stars — is itself dragged around in the direction of the spin.
In Newtonian gravity, only how much mass there is matters. In general relativity, how that mass moves matters too. A static mass curves spacetime; a rotating mass also twists it. This twist is frame dragging, derived by Josef Lense and Hans Thirring in 1918 from Einstein's field equations, and so it is also called the Lense-Thirring effect.
The cleanest signature: a perfect gyroscope, in free fall and feeling no torque whatsoever, will still slowly precess if it sits near a spinning mass. Nothing is pushing on it. The very frame it is measured against has been swept along.
How it works — gravitomagnetism
The deepest way to see frame dragging is the analogy with electromagnetism. When you linearize Einstein's equations for weak fields and slow motion, they take almost exactly the form of Maxwell's equations:
| Electromagnetism | Gravity (linearized GR) | What it does |
|---|---|---|
| Electric charge q | Mass m | Source of the static field |
| Electric field E | Gravitoelectric field (ordinary gravity g) | Pulls toward the source |
| Electric current I | Mass current (rotating / moving mass) | Source of the rotational field |
| Magnetic field B | Gravitomagnetic field Bg | Acts on moving / spinning test bodies |
| Magnetic dipole (current loop) | Spinning body with angular momentum J | Creates a dipole field pattern |
| Larmor precession of a magnetic moment | Lense-Thirring precession of a gyroscope | Spin axis slowly rotates |
Just as a loop of electric current produces a magnetic dipole field that torques a compass needle, a rotating mass produces a gravitomagnetic dipole field that torques a gyroscope. The precession rate of a gyroscope's spin vector S in this field is
dS/dt = Ω_LT × S
Ω_LT = (G / c²r³) · [ 3(J·r̂)r̂ − J ]
where J is the angular momentum of the central body, r is the gyroscope's position, and the explicit factor of 1/c² is exactly why the effect is so feeble: it is a relativistic correction suppressed by the speed of light squared. The same field also makes orbital planes slowly precess — the nodal Lense-Thirring drag measured by laser-ranging satellites.
Worked example — Gravity Probe B
Gravity Probe B (GP-B) is the cleanest measurement of frame dragging ever made. NASA launched it in 2004 into a 642 km polar orbit, carrying four gyroscopes machined from fused quartz — spheres so perfect that, scaled to the size of Earth, the tallest mountain would be about 2.4 metres high. Each gyro spun in a vacuum, was electrically suspended, and its spin axis was read out by the magnetic field of a superconducting coating (the London moment).
A gyroscope in GP-B's orbit experiences two relativistic precessions, and the experiment was designed so they point in perpendicular directions:
| Effect | Origin | GR prediction | GP-B measurement (2011) | Direction |
|---|---|---|---|---|
| Geodetic (de Sitter) | Curved space of a static Earth, gyro carried around the orbit | 6,606.1 mas/yr | 6,601.8 ± 18.3 mas/yr | Orbital plane (north-south) |
| Frame dragging (Lense-Thirring) | Earth's rotation dragging spacetime | 39.2 mas/yr | 37.2 ± 7.2 mas/yr | East-west (perpendicular) |
The geodetic drift is roughly 170 times larger, so isolating the tiny frame-dragging signal required modelling and removing unexpected electrostatic "patch" torques on the gyros — which is why the final analysis took until 2011, seven years after launch. The headline number: Earth drags a freely orbiting gyroscope by 37.2 milliarcseconds per year. To picture how small that is: 37.2 mas is the angle subtended by a human hair (about 0.1 mm) viewed from roughly 16 kilometres away. And it accumulates at only that rate per year.
Why so small? Plug Earth's numbers into ΩLT. Earth's angular momentum is J ≈ 5.86 × 1033 kg·m²/s, the orbital radius is r ≈ 7.02 × 106 m, and the prefactor G/c² ≈ 7.4 × 10−28 m/kg. The result is a precession of order 10−14 radians per second — which, multiplied out over a year, lands at tens of milliarcseconds. Frame dragging is real, but Earth is a weak, slow source.
Regimes — from satellites to the ergosphere
The strength of frame dragging spans an enormous range depending on the source's compactness (GM/rc²) and how fast it spins:
| Source | Frame-dragging strength | Observable consequence |
|---|---|---|
| Earth (LAGEOS / LARES orbits) | Tens of mas/yr of nodal drag | Measured by satellite laser ranging to a few percent |
| Earth (Gravity Probe B gyros) | 37.2 mas/yr precession | Confirmed gyroscopic Lense-Thirring, 2011 |
| Sun | Sub-mas/yr near planetary orbits | Tiny contribution to perihelion precession |
| Millisecond pulsar / neutron star | Large near the surface | Affects orbital dynamics in pulsar binaries |
| Accreting Kerr black hole | Tilted disk Lense-Thirring torque | Candidate driver of QPOs in X-ray binaries |
| Maximal Kerr black hole (ergosphere) | Faster than light can resist | Nothing can stay at rest; Penrose energy extraction |
The extreme case is the Kerr black hole, Roy Kerr's 1963 exact solution for a rotating mass. Outside the event horizon of a spinning black hole lies a region called the ergosphere. Inside it, frame dragging is so violent that spacetime is dragged around faster than light could travel against the rotation. The consequence is startling: no object inside the ergosphere can remain at rest relative to the distant stars. Fire your rockets as hard as you like — you are still forced to co-rotate with the black hole. You can leave the ergosphere (you're above the horizon), you simply cannot stand still inside it.
The ergosphere is also where you can steal energy from a black hole via the Penrose process: drop an object in, let it split, send one fragment onto a negative-energy trajectory into the hole, and the other escapes carrying more energy than you put in. The surplus is drained from the black hole's rotational energy — up to about 29% of a maximally spinning Kerr hole's mass-energy is extractable in principle.
Common pitfalls and misconceptions
- "Frame dragging means the gyroscope feels a force." It feels no torque. It is in free fall, following a geodesic. What changes is the local inertial frame itself — the gyro's axis stays "fixed" relative to a frame that is being swept around.
- Confusing it with geodetic precession. Geodetic precession occurs even for a non-spinning central mass; it comes from moving a gyroscope through curved space. Frame dragging is the extra precession that exists only because the central body rotates. For GP-B they were perpendicular, which is how they were disentangled.
- "The ergosphere is the same as the event horizon." No. The ergosphere lies outside the horizon and you can escape it. The horizon is the true point of no return; the ergosphere is the region where standing still is impossible.
- "It's frame dragging that causes Mercury's perihelion shift." Mercury's famous 43 arcsec/century precession is dominated by the Sun's static spacetime curvature (a geodetic-type effect), not the Sun's spin. Solar frame dragging is a vastly smaller correction.
- Treating it as instantaneous "drag" like a viscous fluid. The honey-and-marble picture is a cartoon. Nothing is physically stuck to spacetime; the metric's off-diagonal time-space components (gtφ) encode a rotational tilt of the light cones, not friction.
- Forgetting the 1/c² suppression. Frame dragging is a genuine but tiny effect for ordinary bodies. Expecting Earth-scale gyroscopes to precess visibly is a unit-analysis error; the effect only becomes dramatic for compact, rapidly rotating objects.
Applications and where it shows up
- Testing general relativity. GP-B and the LAGEOS/LARES laser-ranging satellites turned a 1918 paper prediction into a measured number, ruling out alternative gravity theories that predict different gravitomagnetic strengths.
- Black-hole astrophysics. Misaligned accretion disks around spinning black holes are torqued by Lense-Thirring dragging, a leading model for quasi-periodic oscillations (QPOs) seen in X-ray binary light curves.
- Relativistic jets. Frame dragging around a Kerr black hole is central to the Blandford-Znajek mechanism, in which the hole's rotational energy is extracted electromagnetically to power astrophysical jets.
- Gravitational-wave sources. The spin of merging black holes — and the frame dragging it produces — shapes the late-inspiral and ringdown waveforms detected by LIGO and Virgo.
- Precision geodesy and navigation. Gravitomagnetic corrections enter the most precise satellite-ranging and reference-frame models, alongside the better-known geodetic and gravitational time-dilation terms.
Derivation and scaling analysis
Start from the Kerr metric in Boyer-Lindquist coordinates. The single feature responsible for frame dragging is the cross term between time and the azimuthal angle, gtφ. For a slowly spinning body of angular momentum J, far from the source, the metric picks up
ds² ≈ −(1 − 2GM/rc²)c²dt² + … − (4GJ/c²r) sin²θ · dt dφ
A "zero-angular-momentum observer" (ZAMO) — one who simply lets go and falls freely with no orbital angular momentum — is nonetheless dragged with angular velocity
ω(r) = − g_tφ / g_φφ ≈ 2GJ / (c²r³)
This is the rotation rate of spacetime itself at radius r. Two scaling lessons fall straight out:
- It dies as 1/r³. Frame dragging is a dipole-like field — far stronger near the source. Doubling the distance cuts the dragging eightfold.
- It is suppressed by 1/c². The factor G/c² ≈ 7.4 × 10−28 m/kg is what makes Earth's drag a mere 37.2 mas/yr while a Kerr black hole's ergosphere drags faster than light can fight.
The ergosphere boundary (the "static limit") is where this dragging becomes total — where gtt changes sign and a stationary observer would need to move faster than light to keep up. For a Kerr hole of mass M and spin parameter a = J/Mc, it sits at
r_ergo(θ) = GM/c² + √( (GM/c²)² − a² cos²θ )
At the poles (θ = 0) it touches the event horizon; at the equator it bulges out to the full Schwarzschild radius 2GM/c². Between that outer surface and the horizon is the region where standing still is not allowed — the deepest, most physical consequence of a rotating mass dragging spacetime along for the ride.
Frequently asked questions
What is frame dragging?
Frame dragging is the prediction of general relativity that a rotating mass twists the surrounding spacetime around with it, like a spinning ball wrapped in honey. The local inertial frames — the definition of "not rotating" set by free-falling gyroscopes and distant stars — are themselves dragged around in the direction of the spin. It was derived by Josef Lense and Hans Thirring in 1918, which is why it's also called the Lense-Thirring effect.
How did Gravity Probe B measure frame dragging?
Gravity Probe B, launched by NASA in 2004, carried four fused-quartz gyroscopes — the most perfect spheres ever manufactured — in a polar orbit around Earth. It watched the gyro spin axes drift against a guide star (IM Pegasi). After separating the much larger geodetic effect (6,601.8 mas/yr) from the tiny frame-dragging signal, the 2011 final result was a frame-dragging precession of 37.2 milliarcseconds per year, matching general relativity's prediction of 39.2 mas/yr within the experiment's uncertainty.
What is an ergosphere?
The ergosphere is a region surrounding a rotating (Kerr) black hole, lying outside the event horizon, where frame dragging is so extreme that spacetime itself is dragged faster than light can travel against it. Inside the ergosphere no object can remain at rest relative to distant stars — everything is forced to co-rotate with the black hole, no matter how powerful its rocket. You can still escape the ergosphere because you are above the event horizon; you simply cannot stand still.
Is frame dragging the same as gravitomagnetism?
In the weak-field limit they are essentially the same thing. Einstein's equations, when linearized, look almost exactly like Maxwell's equations: mass-energy plays the role of charge, and mass currents (moving or rotating mass) produce a "gravitomagnetic" field analogous to the magnetic field from electric current. Frame dragging is what this gravitomagnetic field does to gyroscopes and orbits, so "gravitomagnetism" is the engineering-friendly way of describing weak-field frame dragging.
Why is frame dragging so small for Earth?
Frame dragging scales with the body's angular momentum J divided by distance cubed, and is suppressed by a factor of G/c². Earth spins slowly and isn't very massive on cosmic scales, so its gyroscopic drag is only about 37.2 milliarcseconds per year — roughly the angular width of a human hair seen from 16 kilometres away. The effect only becomes dramatic near compact, fast-spinning objects like neutron stars and Kerr black holes, where J is enormous and r is tiny.
Can you extract energy from frame dragging?
Yes — this is the Penrose process. If an object entering the ergosphere splits in two, one fragment can fall into the black hole carrying negative energy (as measured from infinity) while the other escapes with more energy than the original object had. The extra energy comes from the black hole's rotational energy, slowing its spin. Up to about 29% of a maximally rotating Kerr black hole's total mass-energy is theoretically extractable this way.
How is frame dragging different from geodetic precession?
Both make a gyroscope precess, but for different reasons. Geodetic precession (de Sitter precession) comes from moving a gyroscope through the curved spacetime of a static mass — it happens even if the central body is not spinning, and for Gravity Probe B it was about 6,601.8 mas/yr. Frame dragging (Lense-Thirring) is an additional, much smaller precession that exists only because the central body rotates — for Gravity Probe B it was 37.2 mas/yr, perpendicular to the geodetic drift, which is why the two could be separated.
Has frame dragging been measured anywhere other than Earth?
Yes. The LAGEOS and LARES satellites measured the Lense-Thirring drag of their orbital planes around Earth using laser ranging, reaching a few-percent agreement with general relativity. Frame dragging is also inferred near accreting black holes, where the misalignment between the black hole's spin and the accretion disk causes Lense-Thirring torquing that may drive quasi-periodic oscillations seen in X-ray binaries.