HUMANOID ROBOTICS • DEXTEROUS HANDS • MOTOR SELECTION

From Coreless BLDC to Outer-Rotor BLDC: Solving the Hidden Thermal Problem in Dexterous-Hand Joints

A practical engineering case study on why a motor that looks excellent on a datasheet can still struggle when a robotic hand must grip, hold and repeat the same task all day.

The short version: the original coreless BLDC solution was compact and smooth, but repeated gripping exposed a thermal weakness around long holding and near-stall operation. We redesigned the joint around a custom outer-rotor BLDC architecture, then treated motor, reducer and drive control as one thermal-mechanical system.

Figure 1. Simplified topology comparison for a dexterous-hand actuator.

SEO publishing notes

Suggested title tag	Coreless vs Outer-Rotor BLDC Motors for Dexterous Robot Hands | Panda Electrics
Meta description	See how a dexterous-hand joint moved from a coreless BLDC motor to a custom outer-rotor BLDC design to improve holding torque, thermal robustness and system reliability.
Suggested URL slug	/coreless-vs-outer-rotor-bldc-dexterous-hand-motor/
Primary keyword	dexterous hand motor
Secondary keywords	robot hand joint motor; coreless BLDC motor; outer-rotor BLDC motor; humanoid robot actuator; stall torque thermal design

The motor worked. The hand still had a problem.

The first version of the joint used a coreless brushless DC motor. On paper, it was an easy choice to defend: compact diameter, low inertia, smooth rotation and good controllability. During normal motion, the joint behaved well. It opened, closed and followed commands without drama.

The trouble appeared in a less glamorous part of the duty cycle: holding. A dexterous hand does not spend all of its time spinning. It closes around an object, builds grip force, stops moving and then keeps producing torque. Electrically, that can look uncomfortably close to a prolonged stall. Current remains high while speed — and therefore self-cooling — falls toward zero.

That was the customer’s real pain point. The question was no longer “Can the motor reach the required speed?” It became “Can the joint keep the required grip force without accumulating heat faster than the system can remove it?”

This distinction sounds small. In practice it changes the whole motor-selection process. The source design brief describes the same failure mode: extended gripping and posture-holding can build heat in a compact coreless motor, accelerating insulation ageing and raising the risk of thermal shutdown or winding damage.

Why coreless BLDC motors are attractive — and why we did not simply blame the motor

Coreless motors are not a poor technology. Quite the opposite. Their low rotor inertia, smooth behavior and lack of cogging make them very strong candidates for compact precision motion. FAULHABER, for example, highlights low inertia and smooth, cogging-free operation as key advantages of ironless winding technology. The same manufacturer also notes that cooling the self-supporting winding is a design challenge — exactly the point that matters when current stays high for an extended period.

So the engineering decision was not “coreless is bad, outer-rotor is good.” That would be too simplistic. The right statement is more specific: the original motor topology and thermal path were not well matched to this customer’s grip-and-hold duty cycle.

The duty cycle mattered more than the no-load speed

· Fast finger motion required responsive acceleration and smooth low-speed control.

· Gripping required a short burst of higher torque.

· Holding required meaningful torque at little or no speed — the thermally difficult part.

· Repeated open/close cycles added thermal accumulation rather than giving the motor unlimited time to cool.

· The motor lived inside a dense hand mechanism beside gears, sensors and structural parts, so local temperature affected more than the winding alone.

Figure 2. The practical design question is the temperature trajectory during holding, not only the motor’s free-running performance.

Our redesign: move to a custom outer-rotor BLDC architecture

We changed the motor concept rather than continuing to increase the current limit on the original design. The new joint used a custom outer-rotor BLDC motor. That gave us a different set of design levers: a larger effective air-gap radius for torque production, a stationary stator that could be tied more deliberately into the housing, and more freedom to use the surrounding structure as part of the thermal path.

This is not just a marketing claim. Recent research on high-torque-density motors for humanoid robotics describes external-rotor architectures as a way to increase torque through a larger air-gap diameter and better space utilization. The broader motor-design literature makes the same mechanical point: torque benefits from a larger effective radius, although the final result still depends on magnetic loading, copper loading, geometry and cooling.

What we changed at system level

The outer-rotor motor was only one piece of the solution. We also treated the reducer, current limit and housing as part of the actuator design. That mattered because a motor can survive a short torque spike and still overheat if the controller allows the same current indefinitely.

· We selected the motor around the actual holding torque, not only peak motion torque.

· We designed the stator-to-housing interface as a thermal path instead of treating the housing as packaging.

· We matched the motor constant and gear ratio so the joint could generate useful grip force without living near the electrical limit.

· We used current limiting and thermal protection as part of normal control behavior, not as a last-resort fault response.

· We validated the complete joint under repeated gripping and holding cycles rather than qualifying the bare motor on a bench.

Figure 3. A motor-selection workflow built around the real joint duty cycle.

What improved after the change

The biggest improvement was not a dramatic increase in free-running speed. It was margin. The redesigned actuator had more room between normal gripping current and the point where thermal protection became the dominant concern. That made the joint easier to tune and more tolerant of real-world variation: different objects, longer holds, repeated cycles and small differences between assemblies.

The source project presentation records a substantial improvement in stall tolerance, torque density, heat dissipation and impact life for the custom outer-rotor concept. Those figures are useful as project evidence, but they should be published as formal product claims only when the test method, sample size, ambient temperature, current limit and pass/fail criteria are documented. The numbers below are therefore presented as project-source data rather than universal motor-topology guarantees.

Figure 4. Selected comparison data from the supplied project presentation.

Coreless BLDC vs outer-rotor BLDC for a dexterous-hand joint

Design question	Coreless BLDC	Outer-rotor BLDC	Engineering takeaway
Low inertia / fast response	Excellent	Good to very good	Coreless often has an advantage when minimum inertia is the top priority.
Cogging-free smoothness	Excellent	Depends on slot/pole and magnetic design	An outer-rotor motor must be optimized carefully for low-speed force control.
Torque leverage within diameter	Good	Potentially strong due to larger air-gap radius	Useful when the joint needs more torque without simply increasing current.
Holding / near-stall thermal design	Can be challenging in compact packages	Can be engineered with direct stator-to-housing heat paths	Thermal architecture matters as much as electromagnetic performance.
Integration with reducer	Common and compact	Can support integrated layouts	Choose topology around the full actuator, not the motor alone.
Best fit	Ultra-light, highly dynamic micro-motion	Grip-and-hold joints needing torque and thermal margin	There is no universal winner; the duty cycle decides.

A practical selection rule for humanoid hands

If a finger joint spends most of its time moving lightly and must minimize inertia above everything else, a coreless motor can still be the better answer. We would not replace it just because an outer-rotor option exists.

If the joint repeatedly produces substantial torque at very low speed — especially if it must hold a grasp for seconds at a time — then the selection process should start with thermal behavior. In that case, an outer-rotor BLDC can be a strong architecture because it gives the designer more torque radius and more freedom to connect the stator thermally to the structure.

For fingertip joints where every gram matters, a miniature coreless solution may still make sense. For palm, thumb-base, wrist or other joints that carry larger static loads, the outer-rotor route becomes more attractive. This mixed architecture is also consistent with the supplied project recommendation, which keeps small coreless motors as a useful option for the lightest joints while using outer-rotor motors where holding load is more demanding.

The part that is often missed: the driver must be designed with the motor

Changing the motor without changing the current-control strategy would have left part of the problem untouched. A dexterous-hand actuator needs a driver that understands the difference between acceleration torque, contact torque and long-duration holding torque.

Control features we normally consider

· Fast current-loop control for repeatable torque response.

· Dynamic current limiting based on motor temperature or a validated thermal model.

· Separate peak-current and continuous-current limits.

· Locked-rotor and over-temperature protection that does not destroy normal gripping behavior.

· Encoder or Hall feedback appropriate to the required position and force-control bandwidth.

· A clear definition of what happens after an overload: hold, reduce force, release or retry.

The supplied design document makes the same system-level recommendation: combine the motor change with thermal-model-based current limiting and treat motor, gearbox and control as one integrated solution.

How we would validate the next dexterous-hand motor

A five-minute no-load run tells us very little about whether a hand actuator will survive its real job. Our preferred validation is intentionally less comfortable. We reproduce the load profile that caused the problem in the first place.

1. Measure the real joint torque. Use the finger geometry and contact force to calculate motor-side and output-side torque, including reducer efficiency.

2. Define the hold event. Specify current, torque, duration, ambient temperature and the allowed winding or housing temperature.

3. Repeat the cycle. The worst case is often accumulated heat over many grasps, not one isolated event.

4. Instrument the joint. Measure winding or estimated winding temperature, housing temperature, driver current and output torque.

5. Test abnormal conditions. Include jammed fingers, unexpected contact, repeated reversals and blocked motion.

6. Freeze the current limits only after the thermal test. The software limits should come from evidence, not a convenient round number.

FAQ: choosing a motor for a dexterous robotic hand

Is an outer-rotor BLDC always better than a coreless BLDC?

No. Coreless motors can be excellent for very low inertia, smooth control and highly dynamic micro-motion. Outer-rotor designs become attractive when the joint needs more torque leverage and a more deliberate thermal path during long holding events.

Can I compare motors only by rated power?

Not safely. For a robotic hand, torque at the joint, reducer ratio, current limit, hold duration, thermal resistance and available cooling surface often matter more than a single wattage number.

What causes a hand motor to overheat during gripping?

The joint may produce high torque while rotating very slowly or not at all. Copper loss continues, but airflow and speed-related cooling are minimal. If heat cannot leave the winding quickly enough, temperature climbs.

Do I still need a gearbox with an outer-rotor motor?

Usually, yes. The gearbox is selected together with the motor to reach the required joint torque, speed and backdrivability. In some designs the motor and reducer can be integrated more tightly.

What information is needed for a custom motor recommendation?

Joint torque, speed, maximum diameter and length, supply voltage, peak and continuous current limits, grip/hold duration, duty cycle, reducer ratio, encoder requirement, ambient temperature and target life are the most useful starting points.

Developing a humanoid hand or compact robotic joint? Send us your torque, speed, voltage, size limit and duty cycle. Panda Electrics can evaluate the motor topology, gearbox and control requirements together and propose a prototype path for your OEM application. hp@pandaelectrics.com | pandaelectrics.com

References and engineering notes

· Project source: “Humanoid Robot Dexterous Hand Joint Motor Selection Solution.” Supplied by the user. It describes the original coreless-motor thermal/stall issue, the outer-rotor redesign and internal comparative data.

· FAULHABER. “DC-Micromotors – Technical Information” and “A Heart of Copper.” These sources describe ironless/coreless winding benefits including low inertia, smooth motion and the thermal-management challenge of self-supporting windings.

· Biomimetics (2026), “Comparative Analysis of High-Torque-Density Permanent Magnet Motors Having Similar Slot and Pole Numbers for Humanoid Robot Applications.” The paper discusses external-rotor structures for humanoid actuators and the torque benefit of increased air-gap diameter.

· Important publishing note: exact stall-duration, torque-density, heat-dissipation and life figures in the source project presentation should be retained only if Panda Electrics can document the corresponding test conditions and measurement method. Motor topology alone does not guarantee those values.