Sanhua Intelligent Controls (SZSE: 002050): From HVAC to Tesla's Optimus Joints
Sanhua built its business on refrigeration valves and electric-vehicle thermal management — components nobody outside the industry thinks about. Then it became the name most persistently linked, credibly but never officially confirmed, to Tesla's Optimus actuator supply, and robotics revenue grew 320% year over year almost overnight.
Who they are
Sanhua Intelligent Controls is a global leader in HVAC controls and thermal management components, with roots in refrigeration valves and a much larger, more recent business in electric vehicle thermal management systems — the plumbing that keeps EV batteries and motors at the right temperature. That core business is genuinely large and profitable in its own right, generating H1 2025 revenue of RMB 16.26 billion and net profit of RMB 2.11 billion, up 39.3% year over year, almost entirely unrelated to robotics.
The company operates 13 factories overseas across Poland, Thailand, Vietnam and Mexico, with overseas revenue accounting for over 45% of the total between 2017 and 2024 — a genuinely global manufacturing footprint built for HVAC and EV thermal management long before robotics became a strategic priority, which now doubles as ready-made capacity for robotics component production, including a Thailand plant specifically integrated into the company's robotics strategy.
In January 2024, Sanhua announced a 3.8 billion yuan investment in Hangzhou to build a dedicated R&D and production base for robot electromechanical actuators and domain controllers — a serious capital commitment made well before any confirmed large-scale customer order, in the same pattern seen across much of this stage.
What they do
Sanhua's robotics products centre on linear and rotary actuator assemblies — complete Tier-1 units rather than raw components — and the company is widely reported, though never officially confirmed by Tesla, as the exclusive Tier-1 supplier of these assemblies for Tesla's Optimus program. In October 2025, Chinese media reported Tesla had placed a $685 million order for linear actuators, with deliveries beginning Q1 2026 at a volume market analysts estimated could support at least 180,000 Optimus units.
Beyond the disputed Tesla figure, Sanhua has disclosed real operational detail: in the first half of 2025 it engaged with customers across R&D, trial production, iteration, and sample delivery of a full range of robotics products, and the company's Thailand facility has been integrated into its humanoid robot manufacturing strategy alongside peers Xinjian Transmission, Bete Technology, Tuopu Group and Xusheng Group, all approved by Thailand's Board of Investment to build structural frames, joints and control systems for humanoid robots.
How they make money
The core HVAC and automotive thermal management business remains, by a wide margin, where Sanhua's revenue and profit come from today. H1 2025 group revenue of RMB 16.26 billion is overwhelmingly non-robotics, and analyst commentary is candid that this business will continue to "carry the load" for the foreseeable future while robotics scales from a much smaller base.
Sell-side estimates on the robotics opportunity vary widely, reflecting genuine uncertainty about volume: one set of industry estimates projects Sanhua's robotics business could contribute RMB 3.5 billion in revenue by 2026 if Tesla produces 50,000 Optimus units with Sanhua holding a 70% share at roughly RMB 50,000 per actuator assembly — implying perhaps RMB 350 million of net profit at a 10% margin. Longer-range, more speculative estimates run to RMB 33.8 billion of robotics revenue by 2030 if Optimus scales to 2.25 million units with Sanhua maintaining a 50% share, a scenario Goldman Sachs separately projected could see robotics reach 25% of Sanhua's total revenue and 18% of net profit by that year.
Those 2030 numbers are aggressive extrapolations built on Tesla achieving a production scale that has not yet materialised, and should be read as illustrative rather than forecasts. The disclosed, audited number that matters most today is the 320% year-over-year robotics revenue growth in H1 2025 — real, but off a small base.
Where it sits in the value chain
Sanhua sits in Stage 3 — Actuators, screws & joint modules, alongside Schaeffler, THK, Hyundai Mobis, HL Mando and Robotis. Its specific niche — complete linear and rotary actuator assemblies rather than raw components — puts it in direct competition with the integrated-module strategy other Stage 3 names are pursuing, but Sanhua's advantage is a manufacturing cost base built over decades in the famously thin-margin HVAC and automotive parts industries.
Whether that cost advantage translates into humanoid robot design wins the way it has in automotive thermal management is the central question for Sanhua's robotics thesis. Chinese peers like Tuopu Group (rotary actuators) and Green Harmonic (harmonic reducers) are pursuing similar strategies simultaneously, suggesting the whole Chinese automotive-supplier ecosystem is converging on humanoid actuators as a natural adjacency.
The bigger trend
Sanhua is the clearest single illustration of the pattern running through most of Stage 3: an automotive-scale manufacturer with existing cost discipline and global factory footprint deciding that a robot joint is, mechanically, a familiar problem worth solving with existing capabilities rather than new ones. The 320% robotics revenue growth is real evidence that customers are already buying, even without Sanhua confirming exactly who.
The unresolved tension is between growth and disclosure. Sanhua's denial of the specific $685 million Tesla figure, combined with genuinely fast disclosed robotics revenue growth, leaves outside observers unable to size the actual customer concentration or order economics with precision — a gap that will likely persist until either Tesla or Sanhua chooses to disclose more, which neither company has strong incentive to do while the relationship remains commercially sensitive.
What to watch
- Continued robotics revenue growth rate disclosure in subsequent quarterly and annual reports.
- Any confirmation, from either Tesla or Sanhua, of actual actuator order volumes or customer names.
- Utilisation and expansion of the Hangzhou robotics R&D and production base.
- Thailand facility ramp for humanoid robot component manufacturing.
- Whether robotics revenue as a share of total company revenue approaches the more aggressive 2030 sell-side projections or tracks a more modest path.
Related companies
🇨🇳Fellow Chinese component maker also reportedly supplying Tesla, at the reducer end of the joint.
🇰🇷The Korean equivalent bet: an automotive supplier converting cost discipline into robot actuator wins.
🇩🇪Western counterpart with a comparably large, explicitly disclosed actuator supply commitment.
FAQ
Did Tesla really place a $685 million order with Sanhua?
Chinese media reported this in October 2025, but Sanhua issued a formal clarification stating the rumor was untrue and that no material events requiring disclosure had occurred. The company has not confirmed a Tesla relationship at that specific figure, though its disclosed 320% year-over-year growth in robotics revenue for H1 2025 suggests some significant customer demand exists.
What does Sanhua actually make for robots?
Complete linear and rotary actuator assemblies — the joints that convert electrical signals into a robot's physical movement — rather than raw components sold separately. This positions Sanhua as a systems integrator within the actuator stage, similar in structure to what Hyundai Mobis and Schaeffler are building.
Is robotics Sanhua's main business?
No. Sanhua's core business is HVAC controls and electric vehicle thermal management, which generated the vast majority of its RMB 16.26 billion of H1 2025 revenue. Robotics is a fast-growing but still small part of the business, disclosed as growing 320% year over year in that same period.
How large could Sanhua's robotics business become?
Estimates vary widely by analyst and assumption. One projection sees roughly RMB 3.5 billion of robotics revenue by 2026 under conservative Tesla volume assumptions, while more aggressive 2030 scenarios from firms like Goldman Sachs suggest robotics could reach 25% of total revenue if humanoid production scales into the millions of units. These are projections, not disclosed results.