Skip to content 
A castor with a tight swivel head is harder to push and may have a fit problem. Buyers should isolate each interface before approving the product.
When a castor swivel feels too tight, check the swivel head, wheel and axle bush, fork-leg clearance, brake release, mounting, floor, and load separately. Compare unloaded and loaded conditions, then record the test method and approval limit before accepting a production lot.
The case below shows why these checks need to be separated.
A customer reported that a heavy-duty braked castor with a wheel fitted was almost impossible to swivel. Earlier feedback on the same housing had also shown that fork-leg width and screw passage could affect assembly. The case evidence therefore separates the dimensional fit issue from the later swivel concern. A revised ball-race arrangement was reported to pass an 800 kg load test, but customer sample approval was not recorded. The lesson for buyers is to complete a functional check rather than rely on one result.
The wheel and axle interface sits beside the swivel housing, so both parts need checking.
Before measuring resistance, define normal movement and then identify what may change it.
How Should a Castor Swivel?
A castor should turn smoothly around its vertical axis while the wheel stays supported, aligned, and clear of unwanted contact.
The swivel head uses a race and balls to guide movement. That function is different from the wheel’s rolling function. A correct load rating does not prove free swivelling.
The wheel must remain free to roll, while the swivel head must turn without binding. These functions should be checked separately.
ISO 22878 describes separate checks for castor performance, including swivel play and swivel resistance. Buyers should therefore define the intended movement for the complete assembly, not judge a loose housing by hand alone.[1]
This leads to the next question: what can change castor swivel movement?
What Affects Castor Swivel Movement?
A tight swivel is a symptom, not a diagnosis. Resistance may come from the swivel race, wheel, axle bush, fork legs, mounting, brake, load, or floor.
Check one condition at a time. Preload, damage, contamination, poor seating, wheel interference, uneven mounting, brake contact, and load position can all change the force needed to start turning.
Add the wheel, brake, mounting, and load in stages, changing one condition at a time.
The case involved a wheel-fitted heavy-duty braked castor. Earlier feedback on the same housing concerned fork-leg width and screw passage. These observations should guide separate assembly checks, not be treated as proof of one confirmed root cause.
This leads to the next question: how can wheel and housing fit affect the result?
How Can Wheel and Housing Fit Affect Swivelling?
A castor may swivel freely on a bench but tighten after the wheel and hardware are fitted.
The case report records an earlier assembly problem in which the fork legs were too wide for the screw to pass. It also records the need to validate different wheel and axle-bush combinations. This shows why the housing, wheel, and hardware should be checked as one assembly.
Use the intended wheel and axle bush, not a convenient substitute. Check fork-leg spacing, wheel width, hub, axle, bush, screw engagement, and housing alignment together. A correct wheel diameter alone is not enough.
ISO 22878 states that relevant measurements use the wheel and axle bush assembled as during the test, while fork width is maintained and swivel movement is not impaired.[1]
This leads to the next question: how should buyers test the complete castor?
How Should Buyers Test a Castor With and Without a Wheel?
Use a staged check to show when resistance enters the assembly:
- Isolate the swivel head unloaded.
- Fit the intended wheel, axle, and bush; release the brake.
- Repeat the check under the agreed working load.
- Record the surface, load, direction, measurement point, duration, and sample.
The test method and the acceptance limit are different things. ISO 22878 provides test methods and apparatus. The buyer still needs a product- and application-specific condition for acceptance. BSI also separates resistance measurement, dimensional tolerances, static loading, acceptance, and test conditions.[2]
This leads to the final question: why is a load result alone not enough?
Why Is a Load Test Alone Not Enough?
A load result shows capacity under one condition, not free swivelling in the full application.
The reported 800 kg result concerned a revised ball-race arrangement, but it did not show whether the intended wheel, axle bush, brake, mounting, floor, and load position allowed smooth swivel movement.
The approval record should answer these questions:
- Was the swivel checked unloaded, loaded, or both?
- Was the intended wheel and axle bush fitted?
- Was the brake fully released?
- Was resistance measured, or was the result only a static load observation?
- What were the surface, direction, duration, and sample identity?
- Which result is the reference for later production samples?
Customer sample approval was not recorded in this case. That does not make the load result useless. It means the result remains evidence for the stated load test, not a confirmed resolution of the broader swivel concern.
Buyers should approve the complete castor against written functional conditions, not against a load number alone.
References
- International Organization for Standardization, ISO 22878:2004, Castors and wheels — Test methods and apparatus.
- BSI Knowledge, Castors and wheels — Test methods and apparatus.
