Bearing Life Calculator

The L10 Bearing Life Formula

Rolling element bearing life is statistical by nature. The L10 rating life is the number of operating hours that 90% of a large population of identical bearings will survive under a given load at a given speed. It is defined by ISO 281 as:

The exponent p reflects the fatigue mechanism of the contact geometry. Ball bearings have point contact (p = 3), while roller bearings have line contact (p = 10/3 ≈ 3.33), making them more sensitive to overload but generally capable of carrying higher radial loads.

Life Adjustment Factors

The basic L10 life assumes ideal operating conditions: good lubrication, clean environment, and normal internal clearances. ISO 281 introduced life adjustment factors to account for real-world conditions:

• a₁ (reliability factor) — adjusts for reliability levels other than 90%. For 95% reliability, a₁ = 0.62; for 99%, a₁ = 0.21.
• a₂₃ (material/lubrication factor) — accounts for lubrication quality (viscosity ratio κ), contamination, and material cleanliness. Under ideal conditions a₂₃ can reach 3-5×; in contaminated environments it drops below 1.

The adjusted life is L_na = a₁ × a₂₃ × L10h. In practice, maintaining adequate lubrication viscosity relative to speed and load (the κ ratio) is the single most impactful factor within an engineer's control. Many premature bearing failures trace directly to under-lubrication or contamination rather than overloading.

The L50 life (median life) is approximately 5 times the L10 life, meaning the average bearing in a population lasts about 5 times longer than the rated life before failure.

Worked Example

A deep groove ball bearing 6206 has a dynamic load rating C = 19,500 N. It operates at 1,200 RPM under a radial load of 2,000 N with no significant axial load (P = Fr = 2,000 N).

More Worked Examples

Example 2 — Roller bearing on a conveyor drive shaft: A cylindrical roller bearing NU308 has C = 81,000 N. It runs at 350 RPM under an 18,000 N radial load from belt tension and shaft weight. With p = 10/3 for line contact: L10h = (10&sup6;/(60 × 350)) × (81,000/18,000)^3.33 = 47.6 × (4.5)^3.33 = 47.6 × 121 = 5,760 hours, or about 8 months of 24/7 operation. To reach 40,000 hours (general continuous-duty target), the load must drop to about 8,500 kN or the bearing must upsize to NU312 (C ≈ 156 kN).

Example 3 — Pump bearing showing the load-cubed sensitivity: A pump with deep-groove ball 6310 (C = 62,000 N) at 2,900 RPM was designed for P = 8,000 N → L10h = (10&sup6;/174,000) × (7.75)³ = 5.75 × 465 = 2,670 hours. A field measurement shows actual P = 12,000 N due to shaft misalignment. New L10h = 5.75 × (5.17)³ = 5.75 × 138 = 794 hours. The 50% overload cuts life to 30% of design — laser alignment on startup would restore the original life and pay back in months.

Example 4 — Wind turbine main bearing (ultra-long life): A spherical roller bearing 240/600 has C = 9,900 kN. Load is dominated by rotor weight and wind thrust, combined equivalent P = 800 kN. Speed is only 15 RPM (gearbox upstream). With p = 10/3: L10h = (10&sup6;/(60 × 15)) × (9,900/800)^3.33 = 1,111 × (12.375)^3.33 = 1,111 × 3,430 = 3.8 million hours ≈ 430 years. This seems absurd until you factor in real-world a23 < 0.1 from grease ageing, micro-pitting from water ingress, and start-stop cycles — actual field life is 15 to 20 years, showing that catalogue L10 is only the starting point.

Example 5 — Reliability trade-off for medical equipment: An MRI gantry drive uses ball bearing 6205 (C = 14 kN) at 60 RPM under P = 2.5 kN. L10h = (10&sup6;/3,600) × (5.6)³ = 277 × 175.6 = 48,700 hours. At 99% reliability (required for life-critical medical equipment), a1 = 0.21 → L1h = 10,200 hours ≈ 1.2 years of continuous use. Even a single bearing failure stops the scanner and delays diagnoses, so the design specifies preventive replacement every 12 months regardless of condition.

Common Pitfalls

• Using C0 instead of C. The basic static load rating (C0) is for stationary bearings — the dynamic load rating (C) applies to rotating bearings. C is roughly 0.6 to 1.0 × C0 for ball bearings, and using the wrong value gives life estimates that are 3 to 10× off.
• Ignoring axial load in combined loading. The equivalent load P = X·Fr + Y·Fa depends on ratios from the bearing catalogue. For a deep-groove ball bearing with Fa/Fr > 0.35 (approximately), axial load dominates and Y can be 1.5 to 2.0 — ignoring it gives optimistic life.
• Forgetting lubricant viscosity drops with temperature. The life adjustment factor a23 collapses when the actual viscosity falls below the required value at operating temperature. A bearing designed for ISO VG 68 grease at 40°C may run at 80°C in service, where the viscosity is half, cutting a23 to 0.4.
• Applying L10 to fleet planning literally. L10 says 10% of a population fails before the rated life. In a single bearing, there's still a 10% chance of failure before L10 — use L1 (a1 = 0.21) for individual critical units, not L10.
• Overlooking contamination. A single 75 µm particle in rolling contact can initiate a spall that propagates to failure within days. Sealing integrity and filtration quality (ISO 4406 cleanliness codes) often matter more than bearing selection itself.
• Static overload during installation. Dropping a press-fit bearing, forcing it over a shaft with hammer blows, or heating beyond 120°C can permanently dent the races. Always use proper induction heaters and hydraulic presses, and check runout after installation.
• Ignoring misalignment. Deep-groove bearings tolerate only 2 to 10 minutes of misalignment; beyond this, life drops exponentially. If misalignment is unavoidable, specify self-aligning ball or spherical roller bearings rather than over-designing the deep-groove bearing.

Frequently Asked Questions

Why is bearing life sensitive to the cube of load? Rolling contact fatigue follows Hertzian stress theory — contact stress scales with load to the 1/3 power, and fatigue life scales with stress to the 9th power (approximately). The net effect is that life scales inversely with load cubed. Halving the load gives 8 times the life; doubling the load gives one-eighth the life.

What is the difference between L10, L50, and MTBF? L10 is the 10th percentile life (90% survive). L50 is the median (50% survive), typically 5× L10. MTBF (mean time between failures) in reliability engineering is the average time to failure for a repaired system — for bearings it approximates L50 for a single unit, but for fleet-level planning L10 is more useful because it bounds the early-failure tail.

How do I include preload from a clamped joint in the equivalent load? Axial preload from a bolted housing or matched-pair arrangement adds directly to the axial component Fa. For back-to-back or face-to-face duplex pairs, the preload is specified by the bearing designation (e.g., DB/DF light, medium, or heavy preload) and must be included in life calculations per the catalogue tables.

Should I design to the catalogue L10 or de-rate further? Catalogue L10 represents laboratory conditions. Real installations see shock loads, misalignment, lubricant degradation, and contamination that the L10 formula cannot capture. Industry practice is to apply a safety factor of 2 to 4× on catalogue L10 for the required service life — so a 20,000-hour target means specifying a bearing with 40,000 to 80,000 hour L10.

When does grease need relubrication? Grease life is separate from bearing life. For ball bearings in clean conditions, the relubrication interval in hours is roughly tf = K · (14,000,000 / (n · √dm)) - 4 · d, where K depends on grease quality and dm is mean bearing diameter. For a 6206 at 1,800 RPM, this yields about 5,000 hours between relubes.

Related Calculators

• Gear Ratio Calculator — design gear trains; bearings absorb the radial and axial loads from gear meshes.
• Bolt Torque Calculator — preload bearing housing bolts correctly to avoid soft-foot distortion.
• Spring Design Calculator — design springs used in preloaded bearing arrangements.
• Thermal Expansion Calculator — compute shaft and housing growth that affects bearing fit and internal clearance.
• Motor Starting Calculator — evaluate starting torque peaks that drive bearing load spectra.
• All Mechanical Calculators — browse the complete mechanical engineering toolkit.

Disclaimer

This calculator is provided for educational and informational purposes only. Bearing selection requires reference to manufacturer catalogues and consideration of application-specific factors that this simplified calculator cannot capture. While we strive for accuracy, users should verify all calculations independently. We are not responsible for any errors, omissions, or damages arising from the use of this calculator.