Writing an NVH Clause for Precision Bearings: ISO 15242 Method, Vibration Grade and Acceptance Limit

Why "ISO 15242 compliant" is not an acceptance criterion

The most expensive mistake in an NVH clause is treating a method standard as a pass/fail limit. ISO 15242 Parts 1–4 define measurement conditions and calibration only.

Part 1 covers fundamentals, Part 2 radial ball bearings with cylindrical bore and outside surface, Part 3 radial spherical and tapered roller bearings, and Part 4 radial cylindrical roller bearings¹⁸.

A supplier can therefore truthfully claim conformance while telling you nothing about whether the bearing is quiet, because conformance only establishes that the measurement was taken correctly^18,23,26. That is why the acceptance limit has to live somewhere else — in a product specification or in your own drawing^18,23.

When you write the clause, name the ISO 15242 part that matches your bearing type and state the limit as a separate line item. If a quotation comes back with only "ISO 15242" on the datasheet, you have a statement about procedure, not a statement about the part you will receive²³.

Structure-borne vibration on a rig, or airborne noise in a room?

The two domains are not interchangeable. An RFQ that does not say which one it wants can be answered with the wrong test entirely.

The bearing-level measurement described in the sources is structure-borne vibration velocity taken on a spindle rig. The inner ring is driven, the outer ring is stationary and preloaded, and a velocity sensor reads the signal off the ring itself^23,26.

Airborne acoustic measurement is a different domain — ISO 3745 defines precision methods for sound power and sound energy in anechoic or hemi-anechoic rooms²². NVH testing in general combines acoustic measurement, structural vibration measurement, and interpretation of the mechanisms behind each symptom.

Engineers distinguish airborne from structure-borne noise as separate categories²². For your RFQ, that means the clause must state which domain applies to your assembly.

If your concern is what the operator hears, you are specifying an airborne measurement and the rig data will not answer it. If your concern is what the bearing transmits into the housing, the rig measurement is the relevant one.

The acoustic conditions, microphone distance, and background-noise correction for the airborne route are not established in these sources, so those have to be agreed with the supplier rather than assumed.

Fixing the test condition and reporting format so two quotes can be compared

A vibration number without its test condition cannot be compared across suppliers. The rig condition described is an inner ring driven at 1,800 rpm with a stationary preloaded outer ring²³ (supplier-reported figure).

The signal is split into three bands reported in Anderons: L from 50 to 300 Hz, M from 300 to 1,800 Hz, and H from 1,800 to 10,000 Hz^23,26.

Because the inner ring is driven by a high-precision spindle and the signal is taken directly off the outer ring, small vibrations can be detected without external influences²⁶. Write the clause so the supplier reports all three bands, not a single figure, and states the rig speed and the sensor location.

The 1,800 rpm figure is the test-rig condition described in these sources, not your machine's operating speed²³ (supplier-reported figure). Do not let a supplier's report be read as a statement about your application.

If you need the limit to hold at your own operating speed, that is a separate requirement to put in the RFQ and to confirm with the supplier.

Precision class is not an NVH specification

A higher ABEC or ISO precision class will not buy you a quieter bearing, and the RFQ should not be written as if it will.

The ABEC scale explicitly does not measure or control noise and vibration levels, lubrication quality, ball grade, radial internal clearance, surface finish, material composition, heat treatment, cage quality, speed ratings, or load capacity — it is a tolerance band specification only²⁸. The same separation applies on the ISO side.

ABEC ratings are mainly used for ball bearings, roller bearings use RBEC grades or ISO precision classes such as P4 or P2, and ABEC 7 does not automatically mean the bearing is faster. Speed also depends on cage design, lubrication, clearance, preload, sealing, material, and temperature²⁹.

For your clause, that means the vibration grade is a separate line from the precision class, and it needs its own defining standard and edition. If a supplier offers to meet your NVH requirement by upgrading the precision class, the offer does not address the requirement.

Choosing a vibration grade designation and rejecting the "Z" ladder

GB/T 32325-2015 defines group V as the basic vibration requirement for deep groove ball bearings, with V1, V2, V3 and V4 covering higher requirements and VF3 and VF4 the tightest. VF3 and VF4 apply only up to a 100 mm outside diameter²⁶.

The "Z" ladder that circulates in supplier quotations is described as shorthand rather than a designation from the current specification. A quotation that says "Z3" or "Z4" without naming a standard and edition is not a grade you can inspect against²⁶.

When you write the clause, name the grade and the standard that defines it, and check that your part falls inside the grade's scope. For the V groups that means confirming the diameter series and outside diameter range, and for VF3/VF4 the 100 mm outside-diameter ceiling²⁶.

Then ask for the bench report that shows the measured values against that grade. The numeric values behind the V groups are not published in these sources, so the numbers themselves have to come from the supplier's report rather than from the standard text you cite.

Setting the operating point the limit must hold at

Internal clearance interacts with noise and vibration, so the clearance call-out and the stated operating condition belong in the same clause as the limit. A C3 bearing has larger initial clearance and can run slightly noisier or with more vibration than a standard-clearance bearing when cold or lightly loaded.

It improves once the temperature rises and clearance converges to its working value³⁰. If the noise stays clearly noticeable, the source suggests checking whether the clearance is too large, the fit is incorrect, or lubrication is at fault³⁰.

For the RFQ, state the internal clearance group and the operating temperature and load condition the limit must hold at. That lets the supplier flag a mismatch rather than quote against the wrong condition.

Housing fit also enters the picture: for applications requiring low-noise rotation with an easily displaceable outer ring, an H6 housing bore tolerance class is listed¹⁹.

Whether the NVH limit applies at your operating speed or only at the standard test speed is not settled in these sources, so put that question to the supplier explicitly.

Commercial conditions: catalog item or special?

Before the RFQ goes out, decide whether the graded part is a catalog item or a special, because the commercial terms differ. Custom-engineered bearings frequently require minimum order quantities of 500 to 1,000 units and lead times of 12 to 16 weeks, while standard catalog bearings are available immediately³⁴.

MOQ for a clearance-specific part varies by type, specification, and whether customization is needed, with standard and custom parts on different conditions³². A supplier-evaluation framework in the sources proposes MOQ below 5,000 units and lead time below 60 days as commercial-term thresholds, but that is a framework threshold rather than observed market data³⁵.

Ask the supplier to state whether the vibration-graded part is a catalog item or a special, and to quote the MOQ and lead time for the graded lot specifically. No price premium for vibration-graded versus standard bearings is given in these sources, nor tooling or gauging cost for a custom test fixture.

Those figures have to come from the quotation rather than from a benchmark.

Qualification evidence to demand with the quote

The evidence package is what converts the clause from a wish into an inspection. Ask for these with the quotation, and treat the numeric thresholds and grade values as supplier-supplied rather than standard-supplied, because the sources do not publish them.

• ✓Bench report per lot, traceable to the lot number, showing L, M and H band values
• ✓The defining standard and edition for any vibration grade quoted, including the "Z" ladder if offered
• ✓Confirmation that the part falls within the grade's scope, including the outside-diameter limit for the tightest grades
• ✓Quality certification status: ISO 9001 as a baseline, with IATF 16949 preferred where the framework applies
• ✓Manufacturing capability evidence for the noise-relevant process, such as automated grinding and Z3/V3 testing where the supplier claims it
• ✓Whether the supplier offers customization of cage material, ball material, internal clearance, shields, lubrication, contact angle, preload, or seals for the precision part

Where the sources disagree or fall silent on figures

These are the numbers a buyer is most likely to want in the RFQ and that the sources do not settle. Treat each one as a question for the supplier, not a value to write in from a single vendor-tier page.

What the sources do not establish

• No source gives pass/fail vibration thresholds per bearing size class or per grade.
• No source describes DIN 620-7, JIS B 1519, or ANSI-ABMA equivalents to ISO 15242.
• No source gives SKF Q/C/V, NSK V, or NTN N grade definitions or values.
• No source specifies ISO 3744, ISO 11201, or ISO 11202 conditions, microphone distance, or background-noise correction.
• No source states whether noise is measured on the bearing alone or in a reference housing or motor.
• No source specifies 100% vibration screening versus AQL sampling, or the test instrument and its calibration traceability.
• No source specifies whether a per-lot vibration test report or a certificate of conformance is issued.
• No source gives ISO 3290 ball grades (G5, G10) or raceway waviness limits.
• No source states whether the NVH limit applies at the customer's operating speed or only at the standard test speed.
• No source gives a price premium specifically for vibration-graded versus standard bearings, or tooling and gauging cost for a custom test fixture.
• No source addresses ISO 17025 calibration of the vibration rig, AS9100 certification, or a sample vibration test report or PPAP-style submission for a specific part number.