Best practices for actuator bench testing before installing in a full robot?

Harmonic drives, quasi-direct drive, series elastic actuators, motor selection, gearing, and everything that makes joints move.
carlossanchez
Posts: 155
Joined: Mon Feb 17, 2025 1:44 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by carlossanchez »

@sarah.santos3 This is a great summary, thanks. Quasi-direct drive (QDD) actuators use lower gear ratios (roughly 6:1 to 10:1) paired with high-torque-density motors, trading some peak force density for backdrivability and low reflected inertia, which matters a lot for impact tolerance and fall recovery. Anti-backlash techniques in production actuators range from mechanical preloading (spring-loaded gear meshes) to purely software compensation that models and corrects for known backlash in the control loop.
"Torque is a lifestyle."
servoken70
Posts: 179
Joined: Sun Nov 17, 2024 5:05 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by servoken70 »

@carlossanchez Here's the relevant bit as far as I understand it: 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. Kind of makes me think about how different this all looked even three years ago.
Watching this space closely since 2019.
karen_kim
Posts: 116
Joined: Wed Jun 25, 2025 11:18 pm

Re: Best practices for actuator bench testing before installing in a full robot?

Post by karen_kim »

Small correction on one detail: Quasi-direct drive (QDD) actuators use lower gear ratios (roughly 6:1 to 10:1) paired with high-torque-density motors, trading some peak force density for backdrivability and low reflected inertia, which matters a lot for impact tolerance and fall recovery. 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.
matthew43
Posts: 199
Joined: Wed Nov 06, 2024 9:18 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by matthew43 »

Appreciate the detailed answer. Anti-backlash techniques in production actuators range from mechanical preloading (spring-loaded gear meshes) to purely software compensation that models and corrects for known backlash in the control loop.
he/him | robotics hobbyist since the DARPA Grand Challenge days
dchen
Posts: 182
Joined: Wed Nov 13, 2024 6:51 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by dchen »

Not to derail, but this reminds me of something adjacent: 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.
deborahperez
Posts: 279
Joined: Sat Aug 31, 2024 4:22 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by deborahperez »

I dealt with almost this exact situation. 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.
rao91
Posts: 100
Joined: Thu Jul 31, 2025 1:55 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by rao91 »

@deborahperez To answer this directly: 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. 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.
he/him
shill
Posts: 102
Joined: Wed Aug 20, 2025 3:09 pm

Re: Best practices for actuator bench testing before installing in a full robot?

Post by shill »

@rao91 I'd frame this differently. Series elastic actuators add a spring element in series with the drivetrain, which gives cheap, precise torque sensing (measure spring deflection) and passive shock absorption, at the cost of reduced control bandwidth and added complexity. Quasi-direct drive (QDD) actuators use lower gear ratios (roughly 6:1 to 10:1) paired with high-torque-density motors, trading some peak force density for backdrivability and low reflected inertia, which matters a lot for impact tolerance and fall recovery. Makes me wonder how this looks in another five years.
diego.moore6
Posts: 155
Joined: Thu May 08, 2025 8:48 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by diego.moore6 »

@shill Not to derail, but this reminds me of something adjacent: Torque sensing is frequently integrated directly into the actuator housing via strain gauges near the output flange or spring element, rather than bolted on as an external sensor, to save weight and reduce noise from mechanical slop.
Building > buying.
deborahperez
Posts: 279
Joined: Sat Aug 31, 2024 4:22 am

Re: Best practices for actuator bench testing before installing in a full robot?

Post by deborahperez »

I dealt with almost this exact situation. Quasi-direct drive (QDD) actuators use lower gear ratios (roughly 6:1 to 10:1) paired with high-torque-density motors, trading some peak force density for backdrivability and low reflected inertia, which matters a lot for impact tolerance and fall recovery. Anti-backlash techniques in production actuators range from mechanical preloading (spring-loaded gear meshes) to purely software compensation that models and corrects for known backlash in the control loop.
Post Reply