Satellite magnetorquers turn a spacecraft by creating a magnetic dipole that interacts with an external magnetic field, often Earth’s. Current flowing through coils produces the dipole; the interaction supplies a torque, or turning effect. No propellant needs to be expelled for that control action, although the electrical system must supply power.
The central limitation is geometric: at any instant, magnetic torque is perpendicular to the local field. Even three coils cannot command arbitrary torque along that field direction. Effective control therefore depends on the orbit, the changing field, and the spacecraft’s other control equipment.
A controllable magnetic dipole
A coil carrying electric current acts as a magnetic dipole, with a magnitude and direction. Reversing the current reverses the dipole. Changing the current changes its strength within the device’s operating limits.
OpenStax’s current-loop treatment connects the dipole moment to the current, loop area, and number of turns. It also distinguishes a turning effect from a net force: a current loop in a uniform field can experience torque even when its total force is zero.
The spacecraft analogy is closer to an electrically controlled compass than to an engine blowing gas. Unlike a loose compass needle, the coil is attached to the vehicle. The magnetic interaction can therefore change the vehicle’s orientation.
NASA’s small-spacecraft survey, section 5.2.3, distinguishes air coils from rods wound around ferromagnetic cores. Both generate a dipole from current. A core can strengthen the magnetic response, but residual magnetism is one of the tradeoffs, particularly near sensitive magnetic instruments.
These are actuators: they apply a control action. They do not independently tell a spacecraft which way it is pointing. Sensors and a control calculation must supply that information.
The angle can remove the torque entirely
The ideal dipole relation is τ = μ × B, where μ is magnetic dipole moment and B is the external field. The magnitude is μB sin θ, with θ the angle between those vectors.
Imagine a hypothetical torquer producing 0.2 ampere-square-meters in a local field of 30 microteslas. If its dipole is perpendicular to the field, the torque magnitude is:
0.2 A·m² × 0.000030 T = 0.000006 N·m, or 6 micronewton-meters.
At a 30-degree angle, sin 30° is one-half, so the same dipole and field produce 3 micronewton-meters. Aligned with the field, the ideal torque is zero. The coil can still be powered while producing no turning effect in that alignment.
These are calculated examples, not measurements of a particular satellite or a specification for Earth’s field everywhere. They show why a current setting alone cannot specify torque: the external field and the angle are part of the answer.
Why three axes do not remove every limitation
Mounting coils along three spacecraft axes lets a controller synthesize different dipole directions. But every resulting torque still lies perpendicular to the same local field. There is no choice of dipole that produces an instantaneous torque parallel to it.
As a satellite moves through orbit, the field direction and strength generally change relative to the spacecraft. A controller can use that evolving geometry over time. This is different from having unrestricted three-axis control at every moment.
NASA’s survey notes that some coarse magnetic pointing approaches rely on field variability over multiple orbits. That is not a guarantee of the rapid, steady pointing required by every imaging mission. Suitability depends on the mission’s actual tolerances and geometry.
Where magnetic control is useful
The NASA-hosted ADCS primer by A. Nguyen and A. Zufall identifies detumbling after release, coarse attitude control, and reaction-wheel momentum dumping as common uses. Those jobs have different timing and precision requirements.
For wheel unloading, the distinction between internal and external torque matters. A reaction wheel exchanges angular momentum with the spacecraft body. A magnetorquer interacts with the surrounding field, allowing the control system to change the spacecraft-and-wheel system’s stored angular momentum. The existing reaction-wheel explanation follows that internal exchange; magnetic unloading is a complementary action.
A star tracker can help determine orientation, but knowing the pointing error does not make every correction physically available. Sensors, algorithms, and actuators perform separate parts of the job.
Why this is not a universal deep-space engine
A magnetorquer needs a sufficiently strong external field in its operating environment. NASA cautions that use beyond low Earth orbit or in interplanetary space requires investigation of the actual field available. “There is magnetism somewhere in space” is not enough to establish useful control authority.
The same caution applies to fuel-free claims. This mechanism is intended to control orientation; it is not a general substitute for the propulsion required to change a spacecraft’s orbit. Electrical power, thermal behavior, magnetic interference, and control precision still constrain the design.
The useful question is therefore not whether a satellite can turn without expelling gas. It can, under appropriate conditions. The engineering question is whether the local magnetic field supplies the torque directions, strength, and timing that this particular spacecraft needs.
AI-generated illustration for Galileo Whispers of a generic small satellite with an illustrative view of magnetic coils. Internal layout is simplified; no magnetic field, measured torque or named flight hardware is shown.
