How much heat does the power electronics itself add versus the actuators?

[karen.chen3]

@scott21 This is exactly the kind of context I was looking for. Fast charging accelerates capacity fade over repeated cycles, so fleet operators generally have to choose between minimizing downtime (fast charging) and maximizing pack lifespan (slower charging or swap-based approaches) rather than getting both for free.

[deborahperez]

New to this, so forgive me if this is obvious - DC-DC conversion losses across all the individual actuator drivers add up across a whole robot - it's a less glamorous efficiency question than battery chemistry, but power electronics efficiency meaningfully affects real-world runtime too. Solid-state battery claims from platforms like XPeng's IRON, GAC's GoMate, and EngineAI's T800 are genuinely promising on paper for energy density and safety margins, but independent, large-scale field validation of those runtime claims is still fairly limited as of 2026 - it's real progress, not yet fully proven at scale.

[thomas65]

@deborahperez I see it a little differently. Best-in-class lithium-ion cells used in humanoids are currently landing around 280-300 Wh/kg, which is respectable but still leaves battery mass as one of the largest single contributors to total robot weight.

[ramirez77]

Sorry if this is a basic question, but Thermal margin in a densely packed humanoid chassis is often the real limiting factor on sustained performance, not raw motor power - actuators get thermally throttled well before they'd hit their absolute torque limits, especially during repeated high-load cycles like continuous lifting.

[carlossanchez]

This lines up with my experience. Battery placement (torso-centered vs backpack vs distributed through the limbs) is a real tradeoff between center-of-mass/balance considerations and thermal/cooling access - a torso-centered pack helps balance but is harder to cool than a more exposed backpack placement. Makes me wonder how this looks in another five years.

[novikova63]

@carlossanchez Not sure I fully agree here. Solid-state battery claims from platforms like XPeng's IRON, GAC's GoMate, and EngineAI's T800 are genuinely promising on paper for energy density and safety margins, but independent, large-scale field validation of those runtime claims is still fairly limited as of 2026 - it's real progress, not yet fully proven at scale.

[williams84]

@novikova63 Slightly off-topic, but related: Best-in-class lithium-ion cells used in humanoids are currently landing around 280-300 Wh/kg, which is respectable but still leaves battery mass as one of the largest single contributors to total robot weight.

[greta78]

@williams84 Ran into exactly this myself. Battery placement (torso-centered vs backpack vs distributed through the limbs) is a real tradeoff between center-of-mass/balance considerations and thermal/cooling access - a torso-centered pack helps balance but is harder to cool than a more exposed backpack placement.

[young56]

@greta78 Follow-up question though - The average humanoid in 2026 carries under 2.5 kWh of battery capacity, with real-world runtimes clustering between two and four hours depending on how dynamic the workload is - static, low-motion tasks stretch runtime much further than continuous walking or lifting. Idle/standing power draw is often surprisingly close to a meaningful fraction of active walking power draw once you account for onboard compute, sensors, and balance-holding torque - 'doing nothing' still costs real energy on a humanoid.

[karen.chen3]

@young56 Minor factual note: Hot-swappable battery packs solve the runtime bottleneck for continuous operations (like a 24/7 warehouse shift) without needing a much bigger, heavier pack, but they add mechanical complexity, a failure-prone connector interface, and logistics overhead for managing spare packs. Solid-state battery claims from platforms like XPeng's IRON, GAC's GoMate, and EngineAI's T800 are genuinely promising on paper for energy density and safety margins, but independent, large-scale field validation of those runtime claims is still fairly limited as of 2026 - it's real progress, not yet fully proven at scale.