Stamped Steel vs Machined Brass vs Polymer Bearing Cages: What the Evidence Settles on Speed, Noise and Temperature

Rule out cage materials on temperature and lubricant first

Start with the hottest sustained condition and the lubricant chemistry, because that combination eliminates cage materials faster than any other input. Polymer cage life is not a single number: it collapses with temperature and falls further with aggressive lubricants, so the temperature limit must be read together with the lubricant, not on its own⁹²⁴.

For PA66-GF25, life expectation with mild lubricants at 240 °C is 10 hours maximum, while below 100 °C life is more than 1,000 times higher⁹. With aggressive lubricants such as gearbox oils, worst-case cage life is a factor of 10 lower than with mild lubricants⁹.

The mechanism behind that collapse is stiffness loss: above 120 °C the elastic modulus of PA66 with 25% glass fibre decays to less than 50% of its room-temperature value¹².

One supplier reports a typical continuous limit of about 120 °C for polyamide cages and about 250 °C for PEEK, and states that brass or PEEK cages are necessary for high-temperature applications²³.

Another supplier states that where continuous operating temperature reaches about 150 °C with aggressive lubrication, PA66 cages are undesirable and metal cages remain the best option, and that PEEK can match metal at high temperature but its cost is prohibitive²⁴.

If your application sits above the polymer limit or uses an aggressive lubricant, the cage decision is between stamped steel and machined brass, and the polymer route is closed unless PEEK is justified.

If you are still inside the polymer window, confirm the actual continuous temperature and the lubricant chemistry with the manufacturer before treating PA66 as acceptable.

Decide whether speed forces a move off stamped steel

Two separate speed rules apply, and they are not the same rule. The first is a design-change threshold stated for riveted stamped steel: above 500,000 dN, a switch from riveted stamped steel construction to riveted machined land-riding designs is stated as required¹⁶.

No equivalent threshold is given for machined brass, so the evidence does not set a brass ceiling you can design to¹⁶. The second rule is a consultation trigger that applies regardless of cage type: when bearing rotational speed exceeds 80% of the catalog specification, the manufacturer should be consulted¹⁴.

A worked case shows how quickly that trigger arrives. An NU1996MC3 cylindrical roller bearing with a machined brass cage has a speed-limit factor of 550,000 mm/min, and an intended 800 rpm sits at 82% of the resulting speed limit, which the source says requires further investigation⁸.

If your dN is above 500,000, plan for a machined land-riding design rather than a riveted stamped one¹⁶. If your operating speed is above 80% of the catalog figure, treat the cage choice as provisional and put the case to the manufacturer¹⁴.

Weigh the noise benefit of a polymer cage, knowing it is qualitative

Two independent publishers state that polymer cages damp noise or run quietly relative to metal cages, but neither gives a sound pressure level, so you can choose on direction of effect and not on a decibel figure¹⁰²³.

One supplier states that injection moulding of polymer cages enables designs offering improvements in roller guidance, damping of bearing noise levels and bearing running temperature¹⁰. Another states that polyamide cages are lightweight, quiet and dominant in many applications, with a typical continuous temperature limit of about 120 °C²³.

The mechanism the sources describe is geometric rather than material: injection moulding lets the pocket geometry be optimised for rolling element guidance, which is where the damping and running-temperature improvements are said to come from¹⁰.

If noise is a specification item for your application, ask the manufacturer for cage-induced vibration or sound data under your speed and load, because the sources here do not provide it¹⁰²³.

Weigh the cost premium and tooling exposure of each cage type

Pressed steel is the cost baseline, and the reported premiums for polyamide and brass are the first cost input to weigh²⁵. The second input is tooling: stamped and moulded cages of unique pitch diameters, rolling element complements and configurations are often unrealistic due to high tooling costs, regardless of suitability for the application⁵.

That matters most in short-run production, where the cage can become the deciding factor on bearing type selection rather than a detail chosen after the bearing is fixed⁵.

One supplier argues that for most applications the cage cost is a small fraction of the total bearing cost, and that specifying the correct cage matters more than minimising cage cost²⁵.

Use that as the supplier's position, not as a rule: if your run is short and your pitch diameter is non-standard, the tooling exposure is the number to check first.

Verify the cage designation and availability per supplier

Cage designation codes are manufacturer-specific and do not transfer directly between suppliers, so interchangeability has to be confirmed per manufacturer¹⁶. A machined brass cage is M at SKF and FY at Koyo, and Koyo splits material and guidance into two designation positions where SKF fuses them into one letter¹⁶.

For cylindrical roller bearings, design and cage codes such as M (machined brass cage) and W/WS (pressed steel cage) appear in the bearing number itself, so the code you read is part of the part identity rather than a free-standing attribute³².

Before you release a drawing or a PO, decode the cage suffix against the specific manufacturer's designation system and confirm that the cage type you selected is actually offered in your bearing size and series¹⁶³².

Check the application against the conditions the evidence covers

The sources settle temperature direction, lubricant aggressiveness, a speed threshold and cost direction, but they do not settle peak temperatures, noise levels, load capacity or per-size availability. Run these checks before releasing the cage choice to the supplier.

• ✓Confirm the continuous operating temperature and the lubricant chemistry against the polymer limit before accepting a PA66 cage.
• ✓If the lubricant is aggressive (gearbox oils, ammonia), check the specific chemical compatibility limit rather than the generic temperature figure.
• ✓If dN is above 500,000, confirm a machined land-riding design is available in your bearing size and series.
• ✓If operating speed is above 80% of the catalog rating, put the case to the manufacturer before releasing the design.
• ✓Decode the cage suffix against the specific manufacturer's designation system before ordering.
• ✓Ask the manufacturer for cage-induced vibration or sound data if noise is a specification item.

Where the sources disagree, and what to verify

These two disagreements matter because each one sits on a decision boundary: the first sets when you must move off stamped steel, and the second sets whether PEEK is a viable high-temperature polymer. Do not pick a side from this page; put both questions to the manufacturer with your actual bearing size, speed and lubricant.

What the sources do not establish

• No source gives limiting or reference speed values for stamped steel, machined brass and polymer cages in the same bearing size and lubrication condition.
• No source gives short-term peak temperature limits for any of the three cage materials.
• No source gives sound pressure level or cage-induced vibration data for the three cage types under comparable speed and load.
• No source gives swelling or hydrolysis data for the three cage materials under a specified medium.
• No source gives MOQ or lead time per cage type in a specific bearing size.
• No source gives cage load capacity or fatigue strength values under dynamic loading for the three cage types.
• No ISO or ABMA cage-selection standard is cited by any source.
• No source establishes availability or interchangeability of all three cage types in a specified bearing size and series.