1938: ZWZ and China's Oldest Bearing Factory

In 1938, with war already reshaping the industrial map of East Asia, a small plant was set up at Wafangdian, in what is now a district of Dalian in north-east China, to make rolling bearings. It was China's first industrial bearing factory, and the fact that a national industry of considerable size can be traced back to one modest workshop explains a great deal about how that industry still behaves.

This article follows the line from 1938 to the present: what a bearing factory of that era could actually make, how production was rebuilt after 1949, how a specialised cluster grew around the original site, what that cluster makes today, and what the whole story means when you set out to qualify a supplier. The short version is that a cluster gives you capability, not quality - the two are different things, and they are verified in different ways.

The industrial context of the 1930s

China in the 1930s was industrialising unevenly. Rail, mining and textiles were growing, but the machinery that supported them - and almost every rolling bearing inside that machinery - was imported. A bearing is a small, high-value, precision item, and a country without the ability to make one depends on foreign suppliers for something it can neither easily substitute nor easily repair.

That dependency is why the 1938 plant mattered out of proportion to its size. It was not founded to capture a large market; it was founded to establish that bearings could be made domestically at all. The enabling conditions were thin: bearing-quality steel was scarce, grinding machines were imported, and the gauging needed to hold a few micrometres of roundness was itself a specialist skill.

So the first factory started at the shallow end of the product range, where the tolerances were loosest and the volumes were easiest to justify. That pattern repeats whenever a country builds a bearing industry from nothing, and it is the reason the early output looks so modest beside today's catalogue.

What a bearing factory of that era could make

The honest answer is: much less than a modern plant, and much of it by hand. A 1930s workshop could produce single-row deep groove ball bearings in small and medium bores, grind the raceways on manually controlled machines, sort balls by size using gauge sets, and assemble the rings with a cage pressed by hand. What it could not do reliably was hold close tolerance across a large batch, or make a large bearing at all.

Single-row ball bearings and small bores

The product was the plain deep groove ball bearing. It carries radial load and light axial load, it is the simplest family to grind, and it is the one every motor, fan and bicycle needs. In the 1930s the practical ceiling was set by the size of the grinding and heat-treatment equipment: bearings with an outside diameter beyond roughly 100 mm and a bore beyond about 50 mm were hard to harden evenly and hard to grind true.

Hand assembly and hand gauging

Before automatic gauging, an operator matched rings and balls by feel and by gauge, and the result was a wide spread of internal clearance from one bearing to the next. That spread, not the load rating, was the real limit on quality. Two bearings from the same bench could behave quite differently in the same machine, and the only defence was sorting and matching rather than controlled manufacture.

Early capability. Bores above roughly 50 mm were limited by the hardening and grinding equipment of the period rather than by design intent.

The timeline: from 1938 to a modern industry

The century since 1938 breaks into a handful of clear phases, and the direction of travel in each one is easy to read.

Reconstruction after 1949 and the turn to heavy duty

After 1949 the plant was rebuilt as part of a wider programme of industrial reconstruction, and the emphasis shifted. A country building steel mills, mines, power stations and heavy machinery does not need millions of the smallest ball bearings; it needs a smaller number of very large ones that can carry enormous radial loads at low speed.

That requirement pulled the factory towards large-bore and heavy-duty production, and it changed the manufacturing problem completely. Large bearings cannot simply be assembled by hand into a single product and sold; they are made in small batches, each one tailored to a machine, and the difficulty moves from volume control to the materials, forming and heat treatment of thick sections.

Why large bearings are a different business

A large-bore bearing has different failure physics from a small one. Thick sections harden unevenly, so case depth and core toughness must be managed in three dimensions. Rings are forged and rolled rather than turned from bar. Rolling elements are heavy, so pockets, cages and lubrication are engineered as a system. Above roughly 800 mm bore the bearing is often supplied as a split or special design for one specific machine.

How the Wafangdian cluster formed

The presence of one large, technically demanding plant creates the conditions for many others. A factory of that scale needs steel, forgings, cages, seals, grinding wheels, gauges, heat-treatment services and skilled machinists, and once those exist in one place it becomes cheaper and faster to start another bearing business there than almost anywhere else.

Over the decades that produced a genuine cluster: a dense network of plants, subcontractors and service firms sharing a region, a workforce and a supply base. A buyer should read the word "cluster" literally - it describes geography and shared capability, not a single owner or a single quality standard.

Why geography still matters

Cluster economics are real: shared subcontractors, a deep labour pool and short logistics all reduce cost and lead time. But those advantages apply to the whole region, which means they tell you nothing about the individual plant you are buying from.

What the cluster makes today

Modern production around Wafangdian spans the whole range, but its reputation rests on the large and extra-large end. The heavy plants in and around the district make bearings for steel rolling mills, mining equipment and metallurgy, and increasingly for wind turbine main shafts and gearboxes, where an outside diameter beyond 1 m and a bore beyond 800 mm are routine.

That range is why the cluster appears so often in heavy industry. Our pages on spherical roller bearings and tapered roller bearings list the sizes a heavy plant actually orders, and industries maps the sectors where these bearings do their work.

The current structure of the Chinese bearing industry

China's bearing industry today is not a single pyramid. It is closer to three overlapping tiers, and a buyer deals with all three depending on the application.

• The large heavy-duty tier. A small number of very large plants, often long-established, capable of extra-large bearings, wind components and mill bearings.
• The broad cluster tier. Regionally concentrated groups of hundreds of small and medium plants, strongest in the common ball and roller sizes.
• The specialist and foreign-invested tier. Joint ventures and specialised makers serving automotive, machine-tool and export customers with tighter process control.
• The standards are common to all three. Designations follow ISO and GB/T conventions, so the same base number does not by itself tell you which tier you are dealing with.

Where the three tiers overlap

The overlap is where procurement risk lives. A part number can be produced to very different levels of process control by different plants, so the same designation on a quotation does not imply the same bearing inside.

What the 1938 Start Means for Buyers Today

The lesson from 1938 to now is that capability and quality are two separate questions. The cluster can make a very large bearing at a competitive cost, but only the individual supplier's process control decides whether the bearing meets its stated dimensions, clearance and steel quality. Use a short step-by-step verification before you release an order.

1. Identify the actual maker. A cluster address tells you where a plant sits, not who shaped the rings or heat-treated them.
2. Ask for the boundary dimensions to the standard. Bore, outside diameter and width must match the declared ISO 15 or GB/T 307.1 figures, not an approximate tape measurement.
3. Check the steel and the heat treatment. Steel grade and the hardening route matter more at large sizes than small ones, because thick sections are harder to treat evenly.
4. Verify clearance against the standard range. Internal clearance is controlled by ISO 5753 values, and at large bore the nominal band is wide.
5. Confirm the tolerance class and the inspection record. Ask which class under ISO 492 the supplier works to, and whether dimensional reports travel with the batch.
6. Run a first-article check on one bearing. Measure, mount and evaluate it on your own equipment before the order scales.

Cross-referencing is the practical front end of all this. Our interchange lookup maps designations between makers, the cluster guide explains how the regional supply base is organised, and the ANDGE range shows how a supplier's own catalogue records dimensions and ratings. For the material behind the bearing, see our note on bearing steel supply; for the demand that drives the large-bearing cluster, see offshore wind large bearings and the spherical roller guide.