Anyone reverse-engineered a commercial actuator module? What did you find?

Harmonic drives, quasi-direct drive, series elastic actuators, motor selection, gearing, and everything that makes joints move.
benjaminsanchez
Posts: 180
Joined: Fri Apr 18, 2025 1:58 am

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by benjaminsanchez »

Small correction on one detail: Backdrivability is the property that lets an external force move a joint without destroying the gearbox or the motor - it's central to both safe human-robot contact and to letting a leg comply naturally when the robot stumbles. Cycloidal drives are a less common but real alternative to harmonic drives - lower cost at high ratios, decent efficiency, but historically bulkier for the same torque density, which is part of why they haven't fully displaced harmonic drives in commercial hips and knees.
kwilliams
Posts: 309
Joined: Sat Sep 21, 2024 1:57 pm

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by kwilliams »

@benjaminsanchez Agreed, and I'd add: Harmonic drives use high reduction ratios (often around 100:1), which gives excellent force density, stiffness, and thermal endurance for continuous-duty lifting - the tradeoff is low backdrivability and some backlash. Reflected inertia is what a limb 'feels like' to the outside world through the gearbox - a high-ratio harmonic drive reflects a lot of the rotor's inertia back to the joint, making the limb feel stiffer and less forgiving on unexpected impacts.
he/him
giulia.roberts4
Posts: 109
Joined: Wed Jun 25, 2025 1:06 pm

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by giulia.roberts4 »

Pretty much this. One thing to add: A lot of the per-joint cost in a modern actuator isn't the motor - it's the combination of a precision gearbox, integrated encoders, torque sensing, and the driver electronics, all of which have to fit inside a housing the size of a fist. Encoder resolution requirements scale with how tightly you need to control low-speed motion - coarse encoders are fine for open-loop swing phases but cause visible stutter during precise placement or fine force control.
"The best actuator is the one that doesn't overheat."
young56
Posts: 123
Joined: Mon Jun 09, 2025 5:26 pm

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by young56 »

Not sure I fully agree here. A lot of the per-joint cost in a modern actuator isn't the motor - it's the combination of a precision gearbox, integrated encoders, torque sensing, and the driver electronics, all of which have to fit inside a housing the size of a fist.
she/her | grad student, biped locomotion
park44
Posts: 156
Joined: Sat Nov 30, 2024 12:03 am

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by park44 »

@young56 Follow-up question though - Harmonic drives use high reduction ratios (often around 100:1), which gives excellent force density, stiffness, and thermal endurance for continuous-duty lifting - the tradeoff is low backdrivability and some backlash. Cycloidal drives are a less common but real alternative to harmonic drives - lower cost at high ratios, decent efficiency, but historically bulkier for the same torque density, which is part of why they haven't fully displaced harmonic drives in commercial hips and knees.
she/her
olga_lind
Posts: 170
Joined: Tue Dec 24, 2024 12:11 pm

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by olga_lind »

@park44 I'd frame this differently. Backdrivability is the property that lets an external force move a joint without destroying the gearbox or the motor - it's central to both safe human-robot contact and to letting a leg comply naturally when the robot stumbles. Totally unrelated but has anyone else noticed how fast component costs are dropping this year.
rtorres
Posts: 54
Joined: Wed Mar 04, 2026 1:31 am

Re: Anyone reverse-engineered a commercial actuator module? What did you find?

Post by rtorres »

@olga_lind Pretty much this. One thing to add: For dynamic walking and recovery, joint control bandwidth in the tens of Hz range is typically necessary to react to a stumble before the center of mass gets too far outside the support polygon.
Opinions my own, not my employer's.
Post Reply