China's Bearing Clusters: A Sourcing Guide

"Made in China" describes a border, not a capability. Chinese bearing production spans an extremely wide quality band, and the band is organised geographically: different clusters grew around different product types, different customers and different industrial histories, and the differences are visible in what each cluster is good at. For a buyer, knowing the map is the difference between sourcing from a factory that has been making your product for thirty years and sourcing from a trading company that has been making it since last Tuesday.

This report covers the four clusters that dominate export supply, what each specialises in, how quality tiers divide inside a single cluster, and the verification steps that turn a supplier claim into evidence. It is written from inside the supply chain rather than from a distance, which is also why it says what a certificate does not prove.

Why geography matters in bearing sourcing

Bearing manufacture is a clustered industry everywhere, not only in China. Grinding and heat treatment need skilled operators, precision machinery and a local supply of bearing steel, and those inputs concentrate together. Once a cluster forms, it develops its own labour pool, its own subcontracting network and its own customer base, and the specialisation deepens.

In China the effect is unusually strong because the industry was built deliberately: state plants were sited for strategic and resource reasons, and after the 1990s the surrounding private supply chains grew up around them. The history of ZWZ in Wafangdian is the clearest example - a single large plant that seeded an entire regional industry of component makers, grinders and assemblers.

The four anchor clusters

Four regions account for the overwhelming majority of exported rolling bearings, and each has a characteristic strength that a buyer can use.

Wafangdian: large bearings and heavy industry

Wafangdian, in Liaoning, grew around China's oldest and largest bearing plant and remains the centre for large and extra-large rolling bearings: spherical roller bearings for heavy machinery, cylindrical and tapered roller bearings for rolling mills, and large-bore specials. The technical difficulty in this cluster is not grinding to a fine tolerance but controlling heat treatment and dimensional stability across sections that can be hundreds of millimetres thick, and the equipment to do it exists here because it was built for the state programmes. For very large bearings the cluster competes on capability rather than price, and the practical marker is outside diameter: producers here routinely handle single bearings above 1 m outside diameter, where the constraint is heat treatment furnace capacity and dimensional stability rather than grinding precision.

Luoyang: precision and wind

Luoyang, in Henan, is the second historical centre, with a research institute and a producer base oriented towards precision and large-diameter product. It is a significant source of wind turbine bearings - pitch, yaw and main shaft positions, with bores commonly ranging from 200 mm to 800 mm - and of precision bearings for machine tools and aerospace-adjacent applications. The engineering depth in this cluster is higher than average, which shows up as a willingness to discuss clearance rationale, preload and inspection criteria rather than only price and delivery.

The Yangtze delta: Wuxi, Changzhou and the general supply base

The Yangtze delta - Jiangsu and Zhejiang, with Wuxi and Changzhou as the core - is where most general-purpose industrial bearings for export are made. Deep groove ball bearings, angular contact bearings, cylindrical roller bearings and mounted units are produced in enormous variety and in export-grade quality, supported by one of the densest industrial supply chains in the world. Our own base in Wuxi sits inside this cluster, and the practical advantage is access to the specialised subcontracting that makes a wide catalogue possible: a factory that grinds rings well may not make cages, and the cluster supplies both.

Cixi and Ningbo: miniature and micro bearings

Cixi, in Zhejiang, and the surrounding Ningbo area specialise in miniature and small bearings - bores from roughly 1 mm to 10 mm, with the smallest instrument series sitting between 2 mm and 8 mm - and in the high-volume consumer of those products: micro motors, power tools, fans, medical devices and precision instruments. This cluster competes on volume, consistency at very small sizes and acoustic performance, because a miniature bearing that is audibly rough is a rejected part. The inspection technology here is oriented towards high-speed automated gauging rather than towards large-bore metrology.

The cluster map at a glance

The cluster names are a sourcing heuristic, not a specification. What they tell you is which product type a region's workforce and machinery are tuned to, which is a genuine and useful signal. What they do not tell you is anything about the individual factory you are about to buy from.

What a cluster does and does not guarantee

A cluster guarantees the availability of capability - skilled grinders, heat treatment capacity, cage suppliers, gauging services. It does not guarantee that any given factory uses it. This distinction is the whole of sourcing risk in China, and it is worth stating plainly.

• Capability exists. The machinery, labour and subcontracting to make an excellent bearing are present in every cluster listed above.
• Capability is not evenly used. A factory can own a competent heat treatment line and still buy in rings from a source nobody has audited.
• Certification is a management system claim, not a product measurement. A quality management certificate says a process exists, not that the specific rings you received meet the drawing.
• Price communicates something. When a quotation is far below the cluster's own material and processing floor, the difference is being taken out of something - usually heat treatment, steel grade, or inspection.
• Consistency is the hard part. Making one good sample is easy in every cluster; making the thousandth identical to the first is what separates the factories worth qualifying.

Quality tiers inside a single cluster

The quality band inside one cluster is wider than the band between clusters, which is why geography alone is never enough.

How to choose and qualify a supplier in practice

Qualification is a short, specific process rather than a long general one. Six steps cover almost all of the risk.

1. Ask for the factory, not the office. The address of the plant that will make your bearing, with the name of the person responsible for quality.
2. Ask which steel and which heat treatment. Through-hardened chromium steel to ISO 683 for standard rings, case-hardening steel where the duty requires, and the name of the heat treatment operation if it is not in-house.
3. Order samples and measure them. Not a visual inspection - measured bore, outside diameter, width, radial runout and clearance against the applicable standard.
4. Check the batch, then check a second batch. Consistency between orders is the property you are actually buying, and it can only be observed over repeat deliveries.
5. Ask for the inspection report with the shipment, not on request. A supplier who produces reports routinely has a process; one who produces them when pressed has a document.
6. Start with a position you can afford to lose, then extend the order to critical duty only after the second batch measures the same as the first.

Measurement is the only real verification

Everything else is a claim. Two families of test cover most of the practical risk.

Dimensional and geometric checks

Bore, outside diameter and width against ISO 15, radial and axial runout against ISO 492 for the declared tolerance class, and internal clearance against ISO 5753 and DIN 620. These are cheap, fast and decisive: a bearing that is out of tolerance on a bench gauge will be out of tolerance in a housing, and no amount of paperwork changes that. Our guide to measuring bearing accuracy covers the practical method.

Metallurgical and heat treatment checks

Hardness testing and microstructure examination detect the failure that dimensional inspection cannot see: a ring that is correctly ground but inadequately hardened, or one whose case depth is insufficient for the load. A through-hardened bearing ring typically measures in the region of 58 to 62 HRC, and a reading below that band on a production sample is the clearest possible signal about the heat treatment behind it. This is the test that most often explains a bearing which measures perfectly and still fails early, and it is usually the step that separates a genuinely qualified supplier from a superficially acceptable one.

Documentation, traceability and export practice

Documentation is not bureaucracy; it is the mechanism by which a claim becomes checkable later. Three documents carry most of the weight.

• The dimensional inspection report tied to the shipment, showing the measured values and the tolerance bands they were compared against.
• The material certificate with a heat number, which allows the steel to be traced back to a melt rather than described generically.
• The packing and marking record, because a bearing that cannot be identified on the shelf cannot be traced to its batch after installation.

Export practice follows from the same principle. Marking, packaging, labelling and palletisation need to be agreed in writing before production, because changing them afterwards is expensive. Incoterms and payment terms should be fixed at the same time as the specification, since the delivered cost is what the decision actually turns on - our reference library collects the standards that apply to inspection and marking.

Freight, ports and lead time

The Yangtze delta cluster ships through Shanghai and Ningbo, the northern cluster through Dalian, and the choice of port affects lead time more than most buyers expect. A factory four hours inland from a port and a factory twenty minutes away produce different transit times for the same production schedule.

• Production lead time depends on whether the size is a stock series or a special, and on whether tooling has to be made. Standard envelopes from a stocked factory ship in days; specials with new tooling take weeks to months.
• Inland transit to port is the component most often omitted from a quotation, and it is the one that varies most between suppliers in the same cluster.
• Consolidation matters for mixed orders. A supplier who can consolidate a multi-item order into one shipment removes cost and handling that a per-item supplier cannot.
• Ocean transit from Shanghai or Ningbo to a European port is commonly in the region of 30 to 40 days, which is usually longer than the production time for a standard envelope.
• Seasonal effects - the roughly 3 week window around the Chinese New Year in particular - affect capacity and transit predictably, and a sourcing plan that ignores them will miss its window.

What changes when the order becomes recurring

Qualifying a supplier for a single order and qualifying one for a recurring programme are different tasks, and conflating them is how a good first delivery turns into a bad third one. A one-off order tests whether a factory can make the part; a recurring programme tests whether it can keep making it.

First order versus repeat order

The first delivery is usually made with attention. It is the second and third that reveal the process, because by then the order is routine and the attention has moved elsewhere. This is why sample approval on its own is a weak qualification, and why the practical test is whether successive batches measure the same as the approved sample without anyone being reminded to check.

• Hold the specification in writing and reference it on every purchase order, including clearance class, tolerance class, closure and any documentation requirement.
• Measure a sample from every batch, not only from the first. Five pieces gauged on arrival is enough to detect a change in process.
• Keep one approved sample sealed, so a later disagreement about whether the part has changed can be settled by measurement rather than by memory.
• Re-qualify after any change in the supply chain, including a change of steel source or of heat treatment subcontractor, because that is where consistency is usually lost.

Where our own supply sits

We are a Wuxi-based supplier inside the Yangtze delta cluster, which determines what we are good at and what we are not. Our strength is breadth within the general industrial families and access to the cluster's subcontracting network, which is how the catalogue reaches 323,474 orderable SKUs across 291,217 unique model numbers and eleven brands. For large-bore heavy-industry product we source from the northern cluster, and for miniature sizes from Zhejiang, because that is where the relevant capability sits rather than where it is convenient.

Every listing in the catalogue carries the same dimensional and rating data regardless of origin, and the ANDGE house brand is specified and documented to the same standard as the named brands. The company profile describes how we qualify suppliers, and the contact page is the fastest route to a specification review.