The published evidence settles the evaluation method but not per-brand performance: size per-caster load with a safety factor above the naive divide-by-four result, filter wheel material against the duty, match mounting dimensions, and require a named test standard and rating basis in writing.
The sources disagree on the load divisor and the safety-factor magnitude, so the buyer must record the chosen basis in the specification rather than assume a mandatory figure. No source in this set publishes verified per-brand quality, delivery or price data, so this article does not rank brands or quote prices.
Size the per-caster load before comparing any quote
A quote built on total weight divided by four is optimistic. On uneven or worn industrial floors one of four casters is routinely off the ground through frame racking or floor slope, so the remaining casters absorb the full load 1 .
Dividing by three, or by the number of casters minus one, is the procurement-safe assumption; dividing by four leads to early failure 1 , 6 . The published safety-factor ranges vary with duty: 1.3-1.5 for a four-wheel example 9 , 1.0-1.3 normal and 1.5-2.0 rough floors or impact 16 , and 1.3-3.0 for production carts 6 .
For an 800 kg four-wheel cart, dividing by four gives 200 kg per caster; applying a 1.3-1.5 factor raises the selection baseline to 260-300 kg per caster 9 . That is the number to compare against a quoted rating, not the 200 kg figure.
The sources do not establish a single mandatory factor or a governing standard that fixes the divisor, so the buyer should select the factor matching the actual floor condition and impact severity and record it in the specification 1 , 6 , 9 , 16 , 27 . Rating type matters as much as the divisor.
Dynamic rating is the service-life number for rolling applications and is typically 60-75% of static; static rating applies only to stationary equipment such as machine bases and fixed stands 1 . A quote that leads with a static figure is not comparable to one that leads with dynamic.
Check whether the quoted capacity holds at your actual speed and duty
A rating is only valid at its reference speed. Standard polyurethane loses roughly 25% of dynamic capacity per mph above 3 mph, so a caster rated 1,000 lb at 3 mph is about 750 lb at 4 mph and about 560 lb at 5 mph 1 .
Above 5 mph the standard rating table does not apply and high-speed or AGV-grade precision bearings must be specified 1 . Dynamic capacity is established by validation testing that includes speed, maximum obstacle height, floor surface condition, rest and run time, and wheel size, and it is calculated on one less wheel 17 .
A supplier's dynamic test simulates fully loaded dynamic conditions, while a static pressure test applies continuous load above nominal static capacity 18 . Because the rating depends on those test conditions, the buyer should ask each supplier for the reference speed behind any quoted figure and confirm it matches the application.
No source in this set reproduces the ICWM or EN test procedure text, so the exact reference speed, cycle count and pass/fail criteria cannot be verified from these sources 1 , 17 , 18 .
Filter wheel material against the operating environment
Five wheel materials dominate industrial casters: polyurethane, rubber, nylon, phenolic and steel. Each owns a load rating, a temperature band, a floor-protection profile and a chemical resistance, so material is a first filter rather than a final choice 2 .
Polyurethane offers higher load capacity, oil, grease and abrasion resistance and longer wear life, but less cushioning and floor protection than rubber 19 , 20 . Nylon carries high load and impact resistance and resists chemicals and abrasion, but is loud and may damage softer floors; polyurethane is quieter and non-marking 20 .
Phenolic handles modest heat and tops out at a lower capacity than forged steel, which reaches the heaviest capacities 21 . Corrosion and hygiene narrow the choice further.
Where the environment is food acids, sanitizers, hospital washdown, salt air or light marine, 304 stainless handles most of it; 316 is required for sustained salt spray, strong chlorides and some pharmaceutical processes 5 . Hygiene-sensitive applications call for NSF-approved wheels paired with stainless steel casters 10 .
The sources do not provide comparable rolling-resistance or abrasion-life test numbers across materials, so a quantitative material ranking cannot be derived 2 , 19–21 .
Confirm the caster will physically fit before discussing price
A caster that clears the capacity math can still be the wrong part. Collect these dimensions from the existing equipment or the new design, and require the supplier to confirm each against the drawing 11 , 22 , 23 .
- ✓Overall height from mounting surface to floor — all four positions on one cart must share the same overall height, or the cart rocks
- ✓Wheel diameter and wheel width
- ✓Swivel radius — horizontal distance from raceway or kingpin centre to the furthest edge of the wheel
- ✓Top plate length and width, bolt-hole spacing and mounting hole diameter
- ✓Mounting type — plate, stem or bolt hole — and its unique dimensions
- ✓Brake type: a total lock holds wheel and swivel, which is what a cart that must not creep or rotate needs; a top lock holds the wheel only
Put certification and test-basis questions in writing to every shortlisted supplier
A capacity number without a test basis cannot be compared across suppliers, so the rating basis belongs in the RFQ alongside price 25 , 27 . ISO 9001:2015 is the site-level certification observed on caster distributor and manufacturer sites 4 , 12 .
In North America, ANSI MH28.1 defines caster nomenclature and rating method and ICWM test procedures cover dynamic and brake fatigue testing 6 , 13 . In Europe, EN 12532 covers industrial wheels and castors up to 1.1 m/s (about 4 km/h) and its scope includes dynamic, fatigue, performance and load-capacity testing; the EN 12527-12533 series applies 25 , 27 .
Ask each supplier:
- ✓Which standard supports the rating — ANSI ICWM, EN 12532 or another named method?
- ✓Is the quoted value static or dynamic, and at what reference speed does it apply?
- ✓Which wheel and bearing configuration was tested?
- ✓Was the complete castor rated, or only the wheel?
- ✓Does temperature reduce the rating, and by how much?
- ✓Can the supplier provide the test method document, not just a number on a spec sheet?
Build the shortlist by segment and channel
The brand landscape spans a consolidated global leader, European precision and heavy-duty specialists, and cost-effective Asian producers. Colson Group is described as the largest caster and wheel manufacturer in the world, operating in 14 countries with nine global units and owning Colson, Albion, Shepherd, Jarvis, MedCaster, Pemco, Faultless and Bassick 35 .
Named market players also include Tente International, Blickle, TAKIGEN, Payson Casters, Hamilton, TELLURE, samsongcaster, CEBORA, ER Wagner, Flywheel Metalwork, Uchimura Caster, RWM Casters, Darcor, ZONWE Holding Group and Qingdao Shinhee 36 . Channel structure is mixed.
Caster Central is a Colson Group authorized distributor 12 ; MINSEN and Dancy sell direct from factory 30 , 33 ; Algood manufactures in Toronto 29 . General industrial distributors stock breadth across thousands of categories, while caster-specialist distributors stock depth in caster configurations with documented spec sheets and traceable load test methods 6 .
The sources do not provide market share by brand, nor verified quality or delivery performance data per brand, so the shortlist should be built on the evaluation method above rather than on brand reputation alone 6 , 12 , 29 , 30 , 33 , 35 , 36 .
Structure the commercial request around total cost, not sticker price
Total cost of ownership is the amortized purchase cost plus the labour to replace worn or failed units plus the downtime those failures cause, measured across the fleet's service life. On multi-shift fleets, labour and downtime usually exceed the purchase price, so the lowest-priced caster is rarely the lowest-cost caster 7 .
A lower-grade wheel and bearing wear faster and seize more often, and each unplanned failure can stop a cart or a line 7 .
The RFQ should therefore request unit price at the actual volume, volume price breaks, MOQ for the specific configuration, lead time on the required quantity, and whether the rating is per caster or per set 32 .
MOQ can be negotiated by covering raw component costs upfront, paying a higher per-unit price for smaller batches, or providing long-term demand forecasts 31 . Payment terms observed in this set include NET 30 for qualified accounts 6 and T/T 30% deposit with balance against B/L copy, with L/C available for larger orders 33 .
No source provides an on-time delivery rate, verified volume-ramp evidence, actual unit prices, tooling or mould costs, or price-validity periods, so those remain supplier claims to be verified in writing 6 , 12 , 30–33 .
What the sources do not establish
- No numeric published test evidence: dynamic load or endurance cycle counts, impact or drop test results, brake holding force values, salt-spray or corrosion hours, and temperature-cycling results are absent from all sources.
- No IATF 16949, ISO 14001, FDA or EU food-contact, ATEX, or RoHS/REACH certification evidence for any named manufacturer.
- No actual unit prices, tooling or mould costs for custom or private-label wheels, or price-validity periods.
- No verified on-time delivery rate or documented volume-ramp evidence for any supplier.
- No market share data by brand; the player list names companies without shares.
- No dimensional interchangeability standard or tolerance specification; only the dimension set to measure.
- No rolling-resistance or abrasion-life test numbers enabling quantitative material comparison.
- No brake holding force or brake fatigue test data, despite brake fatigue testing being named in ANSI ICWM scope.
- No evidence on whether named brands sell direct or through distributors beyond the examples given (Colson via Caster Central; MINSEN and Dancy direct).
Sources · 28
- 1casterhq.comIndustry publication2026-08
- 2casterhq.comIndustry publication2026-09
- 4casterhq.comIndustry publication2026-06
- 5casterhq.comIndustry publication
- 6casterhq.comIndustry publication2026-04
- 7casterhq.comIndustry publication2026-06
- 9ytopcaster.comManufacturer technical documentation2026-09
- 10algood-casters.comManufacturer technical documentation2026-09
- 11grainger.comIndustry publication2023-12
- 12castercentral.comIndustry publication2025-10
- 13nextdaycasters.comIndustry publication2024-03
- 16bullcaster.comManufacturer technical documentation2026-09
- 17algood-casters.comManufacturer technical documentation2026-01
- 18haioncaster.comUnclassified source2026-06
- 19consolidatedcaster.comManufacturer technical documentation2026-08
- 20bullcaster.comManufacturer technical documentation2024-01
- 21qualitycastersupply.comUnclassified source2026-08
- 22hywproducts.comManufacturer technical documentation2026-09
- 23ytopcaster.comManufacturer technical documentation2024-01
- 25conveyorparts.euIndustry publication2026-09
- 27ironaxis-supply.comUnclassified source2026-09
- 29algood-casters.comManufacturer technical documentation2026-04
- 30kpminsencasters.comIndustry peer technical page2026-08
- 31precintl.comUnclassified source2026-09
- 32skt1m.comUnclassified source2026-09
- 33dancycastor.comUnclassified source
- 35colsoncasters.comManufacturer technical documentation2026-07
- 36verifiedmarketreports.comUnclassified source2026-05
Technical references cited for verifiability — not supplier recommendations. Browse the research library.