Miniature gearmotors for grippers and end effectors

A gripper spends its life at stall. Grip force is set by current, holding is set by whether the screw self-locks, and the stall torque through the ratio decides whether the part survives being held.

Wedge
Industrial instrumentation and automation
Catalog reviewed
2026-08-18
Configurations screened
32 published

A two-jaw parallel gripper drives both jaws from one screw with opposed threads, or from a rack pair, or through a cam. The motor closes the jaws over a free stroke at essentially no load, meets the part, and then the mechanism stops moving while the motor continues to apply torque. Everything interesting happens after motion stops. The same is true of a clamp, a latch, a collet actuator, or a tool changer.

Grip force is a torque problem run backwards. For a screw-driven gripper the total axial reaction is F_total = T·2π·η / L, and each jaw sees half of it if the threads are symmetric. Because torque is proportional to current through the motor's torque constant, grip force is proportional to current: F_total = k_t·I·N·η_gear·2π·η_screw / L. That is the whole control law. The free-stroke phase is a speed problem, not a torque problem: the closing time sets the required output speed as stroke divided by lead divided by time. The holding phase is neither, because a self-locking screw holds at zero current.

Whether the mechanism self-locks, and what the drive does if it does not.

A gripper that has to hold a part with the motor energized is a gripper that is dissipating power continuously at stall, with no rotation to move heat and no back-EMF to limit current. A gripper with a self-locking screw holds at zero current indefinitely and also holds through a power failure, which is usually a safety requirement rather than a convenience. After that, the governing check is the stall torque through the ratio, because a gripper reaches stall on every cycle by design, and if that stall exceeds the gearhead rating then every cycle is an overload.

Continuous, peak, starting and abnormal

Sizing on one of them is how a mechanism gets a part that works on the bench and fails in the field.

Row Continuous Sizes the thermal case Peak Sizes the torque case Starting Sizes the transient Abnormal Sizes the protection
What the mechanism is doing The free-stroke close and open. Torque is bearing and screw drag only, typically well under a millinewton-metre. Speed is what has to be checked here, not torque.The moment of grip, at commanded current. This is the design point and it is a stall point.Opening from a fully gripped, self-locked state. Static friction in the screw is higher than running friction, so the release torque is above the grip torque. A drive sized only on grip force can fail to release.Full command into a rigid part with no current limit. Grip force then goes to whatever the stall torque produces, which on a 10 mm 31:1 configuration is 73.6 N per jaw against a 3 N design force.

Failure modes and misleading specifications

What follows is what actually fails, and which published number sends people the wrong way.

Common failure modes
Failure mode Physical cause What it looks like in the field
Crushed or marked partsGrip force controlled by time or by position rather than by current, so the actual force is whatever the drive develops before the controller decides it is done.Damage that varies with supply voltage and with temperature, because both change the current the motor draws at a given command.
Motor burnout while holdingHolding at stall with a non-self-locking mechanism. No rotation, no cooling airflow, full I²R in the winding.Failures concentrated in units that hold parts for long periods rather than in high-cycle-count units. Winding discoloration with no gear damage.
Gearhead output stage failure early in lifeStall torque through the ratio above the gearhead rating, reached on every grip cycle rather than as a fault.Failures at a consistent cycle count rather than a consistent time. Fatigue markings on the output stage, not a single overload fracture.
Grip force drifts over the machine's warm-upCurrent-based force control without temperature compensation. Magnet flux falls as the motor warms, so the same current produces less torque.First parts of a shift gripped harder than later parts. Force recovers overnight.
Specifications that mislead here
Specification Why it misleads in this application
Continuous output torqueA gripper never runs continuously at torque. It runs free, then stalls. The continuous rating describes a thermal state the mechanism does not visit, and the stall the mechanism does visit is not bounded by it.
Stall torqueHere it is the design ceiling rather than a hazard, but only if a current limit exists. Without one, stall torque is the actual grip force and it is far above any sensible design value.
EfficiencyForward efficiency sets the grip force for a given current. It does not tell you the back-driving behaviour, which is a different number, approximately 2 − 1/η, and which decides whether the gripper holds without power.
Gearhead backlashIrrelevant to a gripper. The jaws close until they meet the part; lost motion is taken up before the part is touched and never appears in the force.

Three decisions carry most of the risk, and each one costs something real.

Spur or planetary reduction

Row Spur Planetary
When it is right Very small envelopes where the axial length of a planetary stack does not fit, and the grip force is low enough that a 24.5 mN·m gearhead rating is not the binding limit.Whenever grip force needs real torque, and wherever the stall torque has to sit inside the gearhead rating with margin.
What it costs Low output rating. On a mechanism that stalls every cycle, a low gearhead rating is a cycle-life problem, not just a peak-torque problem.Length. In the 10 mm class a four-stage planetary adds about 12.8 mm to the motor body.

Brushed coreless or brushless

Row Coreless brushed Brushless
When it is right The gripper self-locks, so the motor is never energized at stall for long, and the duty is a short current pulse per cycle.The mechanism cannot self-lock and force has to be held electrically, or the cycle rate is high enough that brush wear becomes the limit.
What it costs Stall with brushes stationary concentrates current in one commutator segment pair. Repeated stalls at the same rotor position wear the commutator unevenly.Holding torque at zero speed with a brushless motor concentrates current in whichever phases the rotor position selects, which is the same problem in a different place. The published brushless platform is 24.2 mm, so a miniature gripper cannot use it.

How grip force is set and held

Row Current limit, self-locking screw Current hold, non-locking mechanism Force sensor on the jaw
When it is right The default. Grip force is set by current, then the screw holds it with the motor de-energized.The mechanism must be able to release instantly, or the screw lead is too coarse to self-lock.Grip force tolerance is tight, or the part is fragile enough that a factor-of-two error is unacceptable.
What it costs Force accuracy is only as good as the torque constant and the screw efficiency, both of which vary. Release needs more torque than grip did.Continuous dissipation at stall for as long as the part is held, and the part drops on power loss.A sensor, its wiring through a moving joint, and a calibration. It measures the quantity you actually care about, which the current does not.

Worked example

Size a miniature two-jaw gripper for 3 N per jaw in a 10 mm envelope, and check what it does without a current limit.

Grip force
3 N per jaw, 6 N total reaction
Jaw stroke
3 mm per jaw
Closing time
1.2 s
Screw lead
0.5 mm
Screw nominal diameter
3 mm
Calculation
Step Expression Result
Screw torque at gripT = 6 N × 0.0005 m ÷ (2π × 0.25)1.91 mN·m
Closing speed3 mm ÷ 0.5 mm per rev ÷ 1.2 s6 revolutions in 1.2 s, 300 rpm
Screw lead angleλ = arctan(0.5 mm ÷ (π × 3 mm))3.04°
Friction angleφ = arctan(0.15)8.53°, so λ < φ and the screw is statically self-locking
Back-driving efficiency of the geartrainη_back ≈ 2 − 1 ÷ 0.660.48, so the geartrain does not self-lock and the screw has to
Output torque per amp2.343 mN·m/A × 31.12 × 0.6648.1 mN·m per amp
Grip force per amp0.0481 N·m × 2π × 0.25 ÷ 0.0005 m ÷ 2 jaws75.6 N per jaw per amp
Current for 3 N per jaw3 N ÷ 75.6 N/A39.7 mA of torque-producing current, 4 percent of the 973 mA stall current
Output torque at a hard stall2.28 mN·m × 31.12 × 0.6646.8 mN·m, 1.91 times the 24.5 mN·m gearhead rating
Grip force at a hard stall0.0468 N·m × 2π × 0.25 ÷ 0.0005 m ÷ 2 jaws73.6 N per jaw, 24.5 times the design force

Result

MM-C1017-P00311-060A resolves it, and it resolves the stall case as well as the running case, which is rare. It is a 10 mm coreless brushed configuration, 31.12:1 planetary, 6 V, rated 366.3 rpm at the output against the 300 rpm closing demand, 4.4 mN·m continuous against the 1.91 mN·m grip torque, and 29.8 mm long. What makes it the right choice is that its stall torque through the ratio is 24.9 mN·m against a 24.5 mN·m gearhead rating, a factor of 1.01. It is one of only four published configurations where a hard stall stays essentially at the gearhead rating rather than above it, which is exactly the property a mechanism that stalls every cycle needs. The 3 V alternative in the same 10 mm class, MM-C1025-P00311-030B, has more torque headroom at 10.1 mN·m continuous but stalls at 1.91 times its gearhead rating. On a gripper that is the wrong trade. Grip force still has to be set by a current limit near 40 mA, because at full stall the same drive applies 73.6 N per jaw.

What would change it

A coarser screw lead raises the closing speed for the same output speed but reduces the self-locking margin. At a 1 mm lead on the same 3 mm screw the lead angle rises to 6.06°, still below the 8.53° friction angle but with much less margin, and a lubricated or worn thread with a friction coefficient of 0.10 would no longer lock. Higher grip force scales the torque linearly and would push past the 4.4 mN·m continuous rating above about 7 N per jaw, at which point the 10 mm class runs out and the choice moves to 12 or 13 mm. If the gripper cannot self-lock, the entire holding case changes and the selection becomes a thermal problem at stall, which none of the published continuous ratings describes.

A short qualification plan

Each step names the measurement, not the intention.

  1. Measure grip force against current

    Force gauge at the jaw across the commanded current range, on at least five units, at both temperature extremes. The slope is the real force constant and it will not equal the calculated one.

  2. Verify self-locking on the real thread

    Applied opening force at the jaw with the motor de-energized and the leads open, then repeated with the leads shorted, then repeated after the lubrication the assembly will actually ship with.

  3. Measure release torque

    Current required to open from a fully gripped state, which is higher than the grip current and is the number that sizes the drive if it is higher still than the closing case.

  4. Cycle at the commanded limit

    Full-life cycling with a real part in the jaws, then teardown of the output stage and the commutator. Repeated stalls at the same rotor position are what to look for.

  5. Prove the power-loss case

    Grip retained through a supply interruption with a real part, at the minimum grip force setting.

Send the gripper geometry for an application review

Send the grip force, stroke, closing time and screw or cam geometry. You get back the current that produces the force, the self-locking check, and the stall case for each candidate.

Send the gripper geometry for an application review