Constraints
Outer diameter capped by adjacent link geometry and cable bend radius.
Technical case studies
Detailed engineering examples showing how actuator decisions changed once interfaces, preload, and build variation were examined.
These studies are structured as engineering reviews rather than marketing summaries. Each one now opens into a subsystem-specific detail page with a downloadable review summary for internal team circulation.
Study index
Open a subsystem review to inspect the original constraint set, intervention logic, installed outcome, and downloadable summary.
Failure mode
Installed lost motion increased after heat soak and repeated service removal because support compliance and datum location changed the effective backlash window.
Design response
The review moved the output datum to the bearing support, revised housing stiffness near the gearbox flange, and separated mesh error from support-stack contributors.
Summary brief
A compact elbow joint gained installed backlash after heat soak and repeated service removal because support compliance and datum location altered the effective mesh window.
| Review dimension | Observed result |
|---|---|
| Primary failure mode | Backlash growth after thermal cycle and service reassembly |
| Critical change | Datum relocation plus housing-stiffness revision |
| Manufacturing effect | Reduced shim sensitivity and fewer rework loops during rebuild |
Filterable case-study matrix
Compare the same symptom across gearbox, bearing, encoder, and housing decisions.
The matrix is intended for engineering triage. Start with the observed failure mode, then reduce by subsystem to isolate where the installed behavior was actually created before opening the subsystem detail page.
Backlash driftGearbox
Arm joint retrofit
Installed lost motion increased after heat soak and repeated service removal because support compliance and datum location changed the effective backlash window.
Outcome / resolution
The review moved the output datum to the bearing support, revised housing stiffness near the gearbox flange, and separated mesh error from support-stack contributors.
Bearing wearBearing assembly
Mobile traction module
Repeated curb-strike reversals and drag variation overloaded the bearing support path because local housing flex amplified radial load instead of distributing it.
Outcome / resolution
Support span and wall distribution around the bearing pocket were revised, and incoming plus end-of-line inspection gates were added for seat geometry and drag signature.
Preload lossHousing
Humanoid knee subsystem
Competing bandwidth and foot-strike requirements pushed the assembly toward preload migration and technician-dependent builds because support order and clamp sequence were not yet repeatable.
Outcome / resolution
The assembly sequence was reordered around bearing support and gearbox pilot engagement, and a verification set was defined for backlash, drag, torsional wind-up, and thermal rise.
Thermal driftEncoder
Integrated sensing axis
Reported angle drift originated in the encoder mounting stack rather than the control loop because the cover-mounted sensing reference moved relative to the bearing-supported output under heat.
Outcome / resolution
The encoder reference was re-tied to the bearing-supported structure, spacer growth was reviewed, and thermal drift was separated from backlash and servo tuning effects.
Cross-study comparison
The recurring pattern is that system behavior changed at the installed interfaces, not at the nominal catalog values.
Backlash shifted when datums, clamp sequence, and support stiffness changed after installation.
Shock survival improved only after the true secondary load path was modeled and constrained.
Pilot-run repeatability depended on inspection gates and assembly order as much as geometry.
Thermal and sensing drift repeatedly traced back to where the measurement or support datum was actually located.
Start an engineering review