Two genuinely different causes of the same rough, uneven rotation — one is purely magnetic and never goes away even unpowered, the other only exists once the drive starts switching current.
"The motor feels rough at low speed" is one of the most common complaints in motion control — and one of the easiest to misdiagnose, because two completely unrelated phenomena produce almost the same symptom. Cogging torque comes from magnets and steel wanting to snap into alignment, with no electricity involved at all. Torque ripple comes from the electronics driving the motor, and only exists while that drive is switched on. Fixing the wrong one wastes an engineering cycle without touching the actual problem.
Cogging torque is a torque disturbance that comes purely from the magnetic attraction between a permanent-magnet rotor and the toothed (slotted) iron stator around it. As the rotor turns, each magnet passes stator teeth that represent a lower-reluctance — magnetically "easier" — position than the slot gaps between them, and the magnet wants to snap toward that easier alignment. This happens whether or not the motor is energized at all: it's baked into the physical geometry of magnets and teeth, not into how the motor is driven. Torque rippleis a torque disturbance that only appears when the motor is actually energized and being electrically commutated — imperfections in the current waveform, back-EMF shape, or the timing of phase switching mean the torque the drive actually delivers isn't perfectly smooth as it steps or sweeps between phases. A motor with essentially zero cogging torque (slotless, or specifically designed with cogging-minimizing geometry) can still ripple badly if its drive electronics commutate imperfectly.
Cogging torque and torque ripple both show up as the same symptom — uneven, notchy rotation, most noticeable at low speed where it isn't smoothed out by inertia — but they come from different physical mechanisms and need different fixes. Cogging torque is a purely magnetic/geometric effect: it exists because permanent magnets and iron stator teeth attract each other toward low-reluctance alignment, and that attraction doesn't care whether any current is flowing. You can feel it by disconnecting the drive entirely and slowly turning the shaft by hand — if it still notches at regular intervals, that's cogging, and the fix lives in the motor's physical design (skewed magnets or slots, optimized tooth/slot geometry, or a slotless topology that removes the teeth altogether). Torque ripple is an electrical effect: it only exists while the drive is actively energizing and commutating the phases, produced by imperfect current waveforms, non-ideal back-EMF shape, or commutation timing that doesn't hand off torque smoothly between phases. It disappears the instant power is removed, and it's fixed on the drive/control side — better current-waveform shaping, tighter commutation timing, or ripple-compensation control algorithms — not by redesigning the motor's magnets or slots.
False, and it leads directly to fixing the wrong thing. Rough rotation can stem from cogging torque — a purely magnetic/geometric effect present even with the motor completely unpowered, addressed through motor design choices like magnet or slot skewing, optimized tooth geometry, or a slotless design — or from torque ripple, an electrical/commutation effect that only exists once the motor is actively energized and driven, addressed through improved drive and control electronics: better current-waveform control, tighter commutation timing, or torque-ripple-compensation algorithms. These are genuinely different root causes. Redesigning the motor's magnets and slots won't touch a ripple problem that's actually coming from imperfect commutation in the drive, and no amount of drive-side tuning will remove cogging torque, since cogging exists independent of how — or whether — the motor is being driven at all. Correctly diagnosing which phenomenon is actually responsible for a given rough-rotation symptom has to come first, and checking whether the roughness persists with the motor unpowered and turned by hand is one real, practical clue for isolating cogging torque specifically.
Explains why cogging torque and torque ripple both produce the same symptom — rough, uneven motor rotation, especially noticeable at low speed — despite coming from completely different root causes: cogging torque is a purely magnetic/geometric effect present even with the motor fully unpowered, while torque ripple is an electrical/commutation effect that only exists once the motor is actively driven. Correctly telling the two apart is necessary before applying the right fix.
Both phenomena show up as the same felt experience — a notchy, uneven torque delivery that's most obvious at low shaft speed, where inertia isn't smoothing it out. It's easy to lump them together as "motor roughness" and assume any fix targeting roughness in general should help. But cogging torque and torque ripple have almost nothing in common mechanistically. One is a static, passive magnetic interaction that exists in a motor sitting on a bench with no wires connected. The other exists only for as long as current is actively flowing and being switched between phases. Treating them as interchangeable leads directly to applying the wrong fix — motor redesign for what is actually a drive problem, or drive retuning for what is actually a magnet/slot geometry problem.
Cogging torque arises from the magnetic attraction between a permanent-magnet rotor and the iron teeth of a slotted stator. As the rotor turns, each magnet pole seeks the position of lowest magnetic reluctance relative to the surrounding stator teeth — effectively the position where the magnetic circuit is "easiest" — and resists being turned away from it. This produces a torque disturbance that repeats periodically as the rotor rotates, with a period set by the number of stator slots and magnet poles. Critically, this happens with zero current in the stator coils: it is a property of the physical geometry (magnet placement, pole shape, slot shape and count) and exists whether or not the motor is ever electrically driven. It is addressed at the design stage — skewing the rotor magnets or stator slots relative to the shaft axis, choosing a fractional-slot winding, optimizing tooth-tip and slot-opening geometry, or eliminating the effect almost entirely with a slotless (ironless-airgap) stator design.
Torque ripple is a torque disturbance that exists only while the motor is energized and being commutated — that is, while the drive electronics are actively switching current between phases to produce rotation. Its sources are electrical: a trapezoidal or otherwise imperfect back-EMF waveform being driven by current that doesn't match its shape, discretized or non-ideal current control around each commutation event, dead-time effects in the power-switching devices, or simply imperfect timing of when each phase's current is switched in relative to rotor position. A motor engineered to have essentially zero cogging torque — slotless, or with heavily optimized geometry — can still exhibit significant torque ripple if its drive's current control or commutation timing is imperfect, because the ripple is being generated by the electronics, not by the motor's passive magnetic structure. It is addressed on the control/drive side: shaping the commanded current waveform to better match the motor's actual back-EMF profile, tightening commutation timing (particularly around each phase switch point), or applying dedicated torque-ripple-compensation control algorithms that actively counteract the known disturbance pattern.
Because motor design fixes and drive/control fixes solve different problems, misdiagnosing rough rotation wastes engineering effort without resolving it. One genuinely useful diagnostic: disconnect the drive electronics entirely and rotate the shaft by hand. If a periodic notchy resistance is still present with zero current flowing anywhere in the motor, that is cogging torque — a motor-design issue. If the roughness only appears once the drive is powered up and actively commutating, that points to torque ripple — a drive/control issue. In practice, a real system can exhibit both simultaneously, and separating their contributions (through direct torque measurement or motor-only vs. drive-only testing) is often necessary before deciding whether the next design cycle should focus on motor geometry, drive electronics, or both.
Yes. A slotless or cogging-minimized motor design can have near-zero cogging torque by construction, yet still exhibit noticeable torque ripple if the drive's current waveform, commutation timing, or back-EMF matching is imperfect — because ripple is generated by the electrical drive, not by the motor's passive magnetic geometry.
Not entirely with conventional slotted designs, but it can be reduced substantially through skewing (angling the rotor magnets or stator slots along the shaft axis so the low-reluctance positions don't all align at once), optimized slot/tooth geometry, or fractional-slot windings. A slotless (ironless-airgap) stator design removes the stator teeth that cause cogging almost entirely, at some cost to torque density and manufacturing complexity.
Not directly — torque ripple is primarily about how well the current waveform and commutation timing match the motor's actual back-EMF and torque-production characteristics, not simply how much current is flowing. A poorly commutated drive can ripple badly even at modest current, while a well-tuned drive can deliver high current with comparatively little ripple.
At low shaft speed, the rotor has less kinetic energy and the mechanical system's inertia does less to smooth out a periodic torque disturbance, so each cogging "notch" is felt distinctly as uneven motion or speed variation. At higher speeds, the same disturbance still exists but is largely averaged out by the rotating inertia and often becomes audible as noise or vibration rather than felt as uneven rotation.
Yes — permanent-magnet and hybrid stepper motors have the same magnet-to-stator-tooth interaction as other permanent-magnet motors, and it's often called "detent torque" in stepper datasheets. It's the same underlying magnetic-geometry phenomenon as cogging torque in servo and BLDC motors, just under a different name in that context.
Try our Mechatronics Studio
More calculators, simulators, and guides for this discipline.