A Step-by-Step Guide on How to Size a Bearing: 5 Key Factors for 2025

Abstract

The process of how to size a bearing is a foundational exercise in mechanical design, profoundly influencing the reliability, efficiency, and longevity of rotating machinery. An improperly sized bearing frequently precipitates premature failure, leading to unscheduled downtime, costly repairs, and potential safety hazards. This document examines the multifaceted procedure of bearing selection, moving beyond simplistic catalog lookups to a more holistic engineering analysis. It investigates the principal factors governing this process, including the magnitude and direction of applied loads, rotational speed requirements, desired operational life, and prevailing environmental conditions. The discourse synthesizes two primary calculation methodologies: the static load capacity, which assesses a bearing's resistance to permanent deformation under stationary loads, and the dynamic load capacity, which predicts the fatigue life under rotating conditions. By meticulously evaluating these parameters, an engineer can specify a bearing that not only fits physically but also performs reliably throughout its intended service life, thereby upholding the functional integrity of the entire mechanical system.

Key Takeaways

• Analyze both static and dynamic loads to determine the correct bearing capacity.
• Calculate the required L10 bearing life based on the specific application's reliability needs.
• Consider speed, temperature, and contamination as paramount operational factors.
• Learning how to size a bearing correctly prevents premature equipment failures.
• Verify that the chosen bearing's physical dimensions and fits are compatible with the design.
• Select appropriate seals and lubrication to protect the bearing and extend its life.
• Account for shock loads and vibrations that can drastically reduce bearing longevity.

Table of Contents

• The Foundational Importance of Correct Bearing Sizing
• Factor 1: A Deep Analysis of Loads and Forces
• Factor 2: The Demands of Speed and Rotation
• Factor 3: Calculating Bearing Life and Ensuring Reliability
• Factor 4: Navigating Environmental and Operational Realities
• Factor 5: Accommodating Physical Constraints and Assembly
• A Worked Example: Sizing a Bearing for a Conveyor Pulley
• Frequently Asked Questions (FAQ)
• Conclusione
• References

The Foundational Importance of Correct Bearing Sizing

To embark on the task of designing or maintaining machinery is to engage in a dialogue with the laws of physics. Every rotating shaft, every spinning wheel, every moving component exists in a state of controlled motion, and at the heart of this control lies the humble bearing. It is an element of profound consequence. To ask how to size a bearing is not merely to ask about dimensions; it is to ask how a machine will live, how it will endure stress, and how it will ultimately fulfill its purpose. The selection process is an exercise in foresight, a projection of a machine's future life under operational duress.

Beyond Rotation: The Role of a Bearing in a Machine's Ecosystem

Imagine a complex ecosystem, where each organism has a specific function that contributes to the health of the whole. A machine is no different. The bearing is not an isolated component; it is a lynchpin, a node in a network of interconnected parts. Its primary function is to permit constrained relative motion between two parts, typically rotation or linear movement, while reducing friction. But its responsibilities extend far beyond that. It must support and locate the shaft and other components, holding them in their correct position relative to one another with immense precision.

Think of a large industrial gearbox. The gears must mesh with microscopic accuracy to transmit power efficiently and quietly. It is the bearing's job to hold the gear shafts in that exact position, resisting the powerful forces that are trying to push them apart. If the bearing deflects too much under load, the gears will misalign, leading to noise, excessive wear, and eventual tooth fracture. The bearing, in this sense, is the silent guardian of the machine's internal geometry. Its proper sizing ensures that this geometry remains stable, allowing the entire system to function as its designer intended.

The Cost of Miscalculation: A Cascade of Failure

What happens when this guardian falters? An undersized bearing, chosen without a full appreciation for the loads and speeds it will face, is a seed of destruction planted within the machine. The initial failure may be subtle: a slight increase in vibration, a minor rise in operating temperature. These are the early tremors before the earthquake. As the bearing's internal raceways and rolling elements begin to fatigue and degrade, the precision is lost. The shaft may begin to move erratically, a condition known as runout.

This loss of precision initiates a cascade of failure that can propagate throughout the machine. In our gearbox example, the shaft runout causes the gear meshing to become harsh and uneven. The gear teeth themselves begin to wear rapidly. The vibrations generated by the failing bearing and complaining gears travel through the structure, potentially loosening other components or causing fatigue cracks in the machine housing. The lubricant becomes contaminated with microscopic steel particles from the spalling bearing, turning from a life-preserving fluid into an abrasive slurry that accelerates wear in all other lubricated components. The final result is catastrophic failure, an event far more costly and damaging than the initial bearing problem. Learning how to size a bearing correctly is therefore not just a technical task; it is an act of mechanical stewardship, a commitment to the reliability and safety of the entire system.

Factor 1: A Deep Analysis of Loads and Forces

The first and most foundational question in our inquiry into how to size a bearing concerns the forces it must withstand. A bearing lives in a world of constant physical stress. These stresses, or loads, are the primary antagonist in the story of a bearing's life. Understanding their nature, their magnitude, and their direction is the first step toward selecting a component that can endure them. To ignore this step is to send a soldier into battle without knowing the strength of the enemy.

Understanding the Language of Forces: Radial, Axial, and Moment Loads

Forces acting on a bearing are not monolithic; they have distinct characters. We can categorize them into three main types:

• Radial Load: Imagine the force of gravity acting on a heavy conveyor belt pulley supported by a shaft. That force pushes directly downwards on the shaft, and the bearing must push directly upwards to support it. This force, acting perpendicular to the shaft's axis of rotation, is a radial load. It is like the pressure you feel on your feet when you are standing still.
• Axial Load (or Thrust Load): Now, think of a propeller on a ship. As it spins, it pushes water backward, and the water pushes the propeller shaft forward. This force, acting parallel to the shaft's axis of rotation, is an axial or thrust load. It is a pushing or pulling force along the length of the shaft.
• Moment Load: This is a more complex character. A moment load is a twisting force that tries to tilt the bearing. It often arises from forces that are applied at a distance from the bearing's center. While some bearings can handle moments, many are not designed for them, and these loads must often be resolved into radial and axial components acting on a pair of bearings.

In the real world, a bearing rarely sees only one of these loads in its pure form. Most applications involve a combined load, a mixture of radial and axial forces acting simultaneously. A helical gear, for instance, transmits torque through angled teeth, creating both a radial force that tries to push the shafts apart and an axial force that tries to push one shaft along its length. The process of how to size a bearing must account for this complex, multi-directional reality.

A Tale of Two Capacities: Static vs. Dynamic Loading

A bearing has two different measures of strength, corresponding to two different ways it can fail. It’s like a person having a limit to how much weight they can lift once (a static effort) versus how much they can carry for a ten-kilometer walk (a dynamic effort).

• Static Loading: This refers to the load on a bearing when it is not rotating or is rotating very slowly (under 10 RPM). Under a heavy static load, the danger is not fatigue but brinelling. This is the permanent indentation of the raceways by the rolling elements. Imagine pressing a marble very hard into a piece of wood; it leaves a permanent dent. A similar dent in a bearing raceway will cause noise, vibration, and premature failure once rotation begins. The bearing's ability to resist this permanent deformation is called its Basic Static Load Rating (C₀).
• Dynamic Loading: This refers to the load on a bearing when it is rotating. Here, the enemy is metal fatigue. As the bearing rotates, each point on the raceway and each rolling element is subjected to a continuous cycle of loading and unloading. Think of bending a paperclip back and forth; eventually, it breaks. Similarly, these millions of stress cycles eventually cause microscopic cracks to form in the steel, which grow until a small piece of the surface flakes off, a process called spalling. The bearing's ability to resist this fatigue failure is its Basic Dynamic Load Rating (C).

The distinction is paramount. A crane hook sheave might only rotate slowly and intermittently under a very heavy load; its sizing will be governed by static capacity. Conversely, the wheel bearing on a car rotates millions of times under a moderate load; its sizing is entirely a matter of dynamic capacity and fatigue life.

Table 1: Bearing Type Suitability for Different Load Conditions

Different bearing geometries are inherently better suited for different types of loads. This table provides a general guide.

Bearing Type	Primary Load Direction	Radial Load Capacity	Axial Load Capacity	Speed Capability	Misalignment Tolerance
Deep Groove Ball	Radial & Axial	Good	Moderate	Excellent	Low
Angular Contact Ball	Radial & Axial	Good	Excellent (one direction)	Excellent	Very Low
Cylindrical Roller	Radiale	Excellent	Low (some types)	Very Good	Very Low
Spherical Roller	Radial & Axial	Excellent	Good	Good	High
Tapered Roller	Radial & Axial	Excellent	Excellent (one direction)	Good	Low
Thrust Ball	Axial	None	Good	Moderate	Very Low
Cuscinetto a strisciamento	Radial & Axial	Very High	Very High	Low to Moderate	Moderate

The Static Load Rating (C₀): A Measure of Brute Strength

The Basic Static Load Rating, C₀, is a value found in every bearing catalog. It represents the load that will cause a specific, very small amount of permanent deformation at the contact point between the rolling element and the raceway. Specifically, it is the load that results in a total permanent deformation of 0.0001 times the rolling element diameter. This may seem like an incredibly small amount, but even this tiny indentation can be the start of problems in a high-precision application.

To properly size a bearing for static conditions, we must first calculate the Equivalent Static Load (P₀). This is a theoretical load that, if applied, would cause the same amount of contact stress as the actual combination of radial and axial loads acting on the bearing. The formula is:

P₀ = X₀ * Fr + Y₀ * Fa

Where:

• P₀ = Equivalent Static Load
• Fr = The applied radial load
• Fa = The applied axial load
• X₀ = A radial load factor (provided by the manufacturer)
• Y₀ = An axial load factor (provided by the manufacturer)

Once P₀ is calculated, we determine a Static Safety Factor (S₀):

S₀ = C₀ / P₀

The required safety factor depends on the application. For smooth, vibration-free operation, an S₀ of 2 or more is often recommended. For applications with heavy shock loads, an S₀ of 4 or higher might be necessary to prevent brinelling.

The Dynamic Load Rating (C): A Bearing's Endurance Story

The Basic Dynamic Load Rating, C, is a more abstract concept. It is not a load limit that cannot be exceeded. Instead, it is a reference value used to calculate fatigue life. By international standard (ISO 281), the C rating is the constant load that 90% of a population of identical bearings can endure for one million revolutions before the first signs of fatigue appear. This is the foundation of the L10 life calculation, which we will explore in detail later. It represents the bearing's endurance. A higher C value means the bearing can handle heavier loads for a longer period, or a given load for a much longer period.

Just as with static loads, we must convert our real-world combined loads into a single Equivalent Dynamic Load (P). This is the theoretical constant radial load that would have the same effect on bearing life as the actual combination of loads and rotations. The calculation is similar in form to the static one, but uses different factors:

P = X * Fr + Y * Fa

Where:

• P = Equivalent Dynamic Load
• Fr = The applied radial load
• Fa = The applied axial load
• X = A radial load factor (from the catalog)
• Y = An axial load factor (from the catalog)

The values of X and Y are not constant; they depend on the ratio of axial to radial load (Fa/Fr) and the specific geometry of the bearing. Bearing catalogs provide tables or formulas to determine the correct X and Y values. This calculation is a central pillar of learning how to size a bearing for rotating applications. It translates the complex reality of combined loads into a single, usable number for life calculations.

Factor 2: The Demands of Speed and Rotation

After grappling with the forces a bearing must resist, our attention must turn to the second great parameter of its existence: speed. Motion is the bearing's purpose, but motion itself brings its own set of challenges. Speed generates heat, creates centrifugal forces, and places immense demands on the lubricant. A bearing that is perfectly adequate in terms of load capacity can fail catastrophically if its speed limitations are not respected.

More Than Just RPM: Defining the Speed Profile

When we talk about speed, the first number that comes to mind is usually revolutions per minute (RPM). While this is a foundational metric, a thorough analysis requires a more nuanced understanding of the machine's speed profile. Is the speed constant, like in an industrial fan that runs continuously for months? Or is it highly variable, like in the spindle of a CNC machine that rapidly accelerates, decelerates, and reverses?

Consider the centrifugal forces at play. As the bearing rotates, the rolling elements are thrown outwards against the outer ring. At very high speeds, this force can become significant, increasing the contact stress and generating extra heat. The bearing's cage, which serves to separate the rolling elements, is also subjected to these forces and must be strong enough to withstand them without breaking. For applications with rapid acceleration, the inertia of the rolling elements becomes a factor. If the load is too light and the acceleration too high, the rolling elements may not be driven by the inner ring's rotation and can instead slide or skid along the raceway. This skidding, or smearing, is a destructive phenomenon that scrapes material from the surfaces, leading to rapid failure. Therefore, understanding the full dynamic character of the rotation—constant speed, variable speed, acceleration, and deceleration—is paramount.

The Speed Limit: Understanding Limiting and Reference Speeds

How do we know how fast a bearing can go? Bearing catalogs provide two key values to guide us: the Reference Speed and the Limiting Speed. It is easy to confuse them, but they represent different concepts.

• Reference Speed (ISO 15312): This is a thermal reference value. It is based on a standard set of operating conditions (a specific load, a specific lubricant viscosity, a specific heat dissipation from the housing) and represents the speed at which the bearing will reach a thermal equilibrium at a modest temperature rise (e.g., 70°C). It is not a hard limit. You can operate above the reference speed, but it serves as a warning. It tells you that if you exceed this speed, you must perform a more detailed thermal analysis. You might need to use a lower-viscosity lubricant, provide external cooling, or accept a higher operating temperature. It's like a "check engine" light for speed.

• Limiting Speed: This is a mechanical limit. It is the maximum speed at which the bearing can be operated without risking mechanical failure of its components, such as the cage or the seals. It is determined by the strength and design of the cage, the quality of the raw materials, the lubrication method, and the precision of the bearing's manufacture. Exceeding the limiting speed is courting immediate disaster. The cage could disintegrate, causing the bearing to seize instantly. This is a hard "redline" that should not be crossed.

For many standard applications running at moderate speeds, the operating speed will be well below both the reference and limiting speeds, and no special consideration is needed. But for high-speed machinery like machine tool spindles, gas turbines, or turbochargers, the speed analysis becomes the dominant factor in bearing selection, often outweighing the load considerations.

The Dangers of Skidding and Smearing at High Speeds and Low Loads

There is a paradoxical failure mode that can occur in high-speed applications: failure due to insufficient load. It seems counterintuitive. How can not having enough load be a problem? The phenomenon is called skidding.

For a bearing to roll correctly, there must be enough force pressing the rolling elements against the raceway to generate the traction needed to make them turn. When a bearing is operating at very high speed but under a very light load, the rolling elements can lose this traction. Instead of rolling, they begin to slide or "skid" against the raceway. This sliding contact, under high speed, generates intense localized heat. The heat can be so high that it causes microscopic welding and tearing of the surfaces, a destructive process known as smearing. The raceway surface becomes damaged, and the bearing quickly fails.

This is a common problem in applications like vertical pumps, where the upper bearing might see very little radial load, or in machine tool spindles during idling. To combat this, engineers must ensure a minimum load is always present on the bearing. If the application itself does not provide this minimum load, it may be necessary to preload the bearing. Preloading involves installing the bearing in such a way that a permanent internal load is created, often by using a spring or by precisely adjusting the position of two opposing bearings. This ensures that even when the external load is zero, there is enough internal force to prevent skidding. The process of how to size a bearing for high-speed applications must therefore include a check for minimum load, not just maximum load.

Factor 3: Calculating Bearing Life and Ensuring Reliability

With an understanding of the loads and speeds our bearing will experience, we can now address a question that lies at the heart of the selection process: How long will it last? This is not a question with a single, deterministic answer. The failure of a bearing due to fatigue is a statistical phenomenon. Even among a batch of identical bearings produced under the most stringent quality controls, there will be a distribution in their individual lives. Our goal is to use statistical methods to predict this life with a specified degree of confidence.

The L10 Life Equation: Predicting Longevity with Confidence

The most widely used measure of bearing longevity is the Basic Rating Life, or L10 life. The "10" signifies that it is the life that 90% of a sufficiently large group of identical bearings are expected to reach or exceed before the first evidence of material fatigue develops. Phrased differently, it is the life at which there is a 10% probability of failure due to fatigue.

The L10 life in millions of revolutions is calculated using the following fundamental equation:

L10 = (C / P)^p

Where:

• L10 = Basic rating life in millions of revolutions
• C = Basic Dynamic Load Rating (from the catalog)
• P = Equivalent Dynamic Load (which we calculated earlier)
• p = The life exponent, which is a constant based on bearing type:
  • p = 3 for ball bearings
  • p = 10/3 (approximately 3.33) for roller bearings

Notice the powerful influence of the load (P) and the load rating (C). Because of the exponent 'p', a small change in load has a large effect on life. For a ball bearing, doubling the load reduces the life by a factor of eight (2³). Conversely, halving the load increases the life by a factor of eight. This illustrates why an accurate calculation of the equivalent dynamic load P is so profoundly important.

Often, expressing life in millions of revolutions is not as useful as expressing it in operating hours. We can convert it easily:

L10h = (1,000,000 / (n * 60)) * L10

Where:

• L10h = Basic rating life in operating hours
• n = Rotational speed in RPM

What Does "L10" Truly Mean? A Statistical Perspective

It is vital to grasp the statistical nature of L10. It is not a guarantee. If you install a bearing with a calculated L10 life of 20,000 hours, it does not mean the bearing will fail at precisely 20,001 hours. It means there is a 10% chance it will have failed by 20,000 hours. The median life (L50), the point at which 50% of bearings will have failed, is typically about five times the L10 life. The existence of a few bearings that fail much earlier than the L10 prediction is a statistical certainty.

For applications where failure is exceptionally costly or dangerous (think of an aircraft engine or a medical device), designers may need to calculate life for a higher reliability than 90%. For example, they might calculate the L5, L2, or L1 life (the life that 95%, 98%, or 99% of bearings will survive, respectively). These lives can be estimated from the L10 life using factors derived from Weibull statistics, the branch of statistics that describes the failure rates of materials. For instance, the L1 life is approximately 0.21 times the L10 life. This means that to achieve 99% reliability, you need a bearing with a calculated L10 life that is almost five times longer than your target service life.

Table 2: Typical L10 Life Requirements for Various Machines

The "correct" amount of life depends entirely on the application. A hand tool used intermittently has very different requirements from a power station generator that must run for years without interruption.

Machine Type	Required L10h (Operating Hours)
Instruments, household appliances, agricultural machines (intermittent use)	500 – 2,000
Machines used for short periods or intermittently: electric hand tools, elevators	4,000 – 8,000
Machines running 8 hours per day, not always at full load: gear drives, cranes	10,000 – 25,000
Machines running 8 hours per day, at full load: general industrial machinery, blowers	20,000 – 30,000
Machines for continuous 24-hour operation: pumps, compressors, conveyor systems	40,000 – 60,000
Machines requiring very high reliability: large power station equipment, paper machines	100,000 – 200,000

Moving Beyond Basic Life: The Adjusted Rating Life (a_ISO)

The basic L10 life calculation makes a significant assumption: that the operating conditions are ideal. It assumes a perfect lubricant film separates the surfaces, the bearing is perfectly clean, and there are no unusual stresses. In the real world, these ideal conditions are rarely met. To provide a more realistic life prediction, modern bearing theory uses the Adjusted Rating Life, often denoted as Lnm.

The concept is to take the basic L10 life and modify it with a factor that accounts for these non-ideal conditions. The most common method is defined by ISO 281:

Lnm = a1 * a_ISO * L10

Where:

• Lnm = The adjusted rating life
• a1 = A life adjustment factor for reliability (a1 = 1 for 90% reliability)
• a_ISO = A life modification factor that considers lubrication, contamination, and the material's fatigue load limit.

The calculation of a_ISO is complex, involving several sub-factors. It acknowledges that if lubrication is poor or contamination is present, the life will be shorter than the basic L10 prediction. Conversely, if the lubrication is exceptionally good and the bearing is very clean, the life can be significantly longer.

The Contamination Factor (ηc): The Enemy Within

One of the key inputs to the a_ISO factor is the contamination factor, ηc. Contaminants—dirt, dust, water, metal particles from wear—are the arch-nemesis of a rolling bearing. When a tiny, hard particle is trapped between a rolling element and a raceway, it creates an immense pressure point. This pressure point, a stress concentration, acts as a starting point for a fatigue crack. A bearing that might have lasted for years in a clean environment can fail in weeks if contaminated.

The ηc factor ranges from 1 (for perfectly clean conditions, like those achieved with ultra-fine filtration) down to 0 (for heavily contaminated conditions where early failure is almost certain). Selecting the correct ηc value requires an honest assessment of the operating environment and the effectiveness of the sealing arrangement. For example, a bearing in a food processing plant that is frequently washed down with high-pressure water requires a very different sealing solution and will have a different ηc value than a bearing sealed for life inside a clean electric motor. High-quality Cuscinetti ad anello rotante used in heavy construction equipment are designed with robust sealing systems specifically to achieve a better ηc value in harsh environments.

The Lubrication Factor: The Lifeblood of the Bearing

The second key input to a_ISO is the quality of the lubrication regime. The primary purpose of the lubricant is to create a thin film of oil that completely separates the rolling elements from the raceways. This condition is called elastohydrodynamic lubrication (EHL). When a full EHL film is present, there is no metal-to-metal contact, and the theoretical fatigue life can be very long.

The effectiveness of the lubricant is captured by the viscosity ratio, kappa (κ). Kappa is the ratio of the actual lubricant viscosity at the operating temperature to the minimum required viscosity for adequate separation of the surfaces.

• If κ > 4: Excellent lubrication. The surfaces are completely separated. The a_ISO factor can be significantly greater than 1, leading to a life longer than the basic L10.
• If 1 < κ < 4: Good lubrication. A full EHL film is established. This is a typical design target.
• If κ ≈ 1: Marginal lubrication. There may be some intermittent contact between surface asperities (microscopic peaks).
• If κ < 1: Poor lubrication. Significant metal-to-metal contact occurs. Surface distress and wear are likely, and the bearing life will be much shorter than the L10 prediction.

Calculating kappa requires knowing the bearing's operating temperature, as lubricant viscosity is highly sensitive to temperature. This shows the deep interconnection between all the factors in how to size a bearing. A decision about speed affects temperature, which affects lubricant viscosity, which in turn affects the adjusted life calculation.

Factor 4: Navigating Environmental and Operational Realities

A bearing does not exist in a vacuum. It lives and works within a specific environment, and that environment can be either a gentle cradle or a hostile crucible. Temperature extremes, corrosive chemicals, abrasive dust, and powerful vibrations all exert their influence on the bearing's performance and longevity. A designer who has meticulously calculated loads and life based on ideal conditions will see their work undone if they fail to account for the realities of the machine's habitat.

The Influence of Temperature on Fit, Clearance, and Lubrication

Temperature is a pervasive influence. Nearly every aspect of a bearing's function is sensitive to it.

First, consider the materials themselves. The shaft, the housing, and the bearing rings are typically made of different materials (steel and cast iron, for example) which expand and contract at different rates when heated. This differential thermal expansion can dramatically alter the bearing's fit and internal clearance. A bearing that has the correct clearance at room temperature might become too tight when the machine heats up to its operating temperature, leading to excessive preload, high friction, and rapid failure. Conversely, in a cryogenic application, a fit that is correct at room temperature might become loose at low temperatures. A thorough thermal analysis must predict the operating temperatures of the shaft and housing to ensure that the correct fits and clearances are specified for the operating condition, not just the assembly condition.

Second, as we have seen, temperature has a profound effect on the lubricant. As temperature increases, the viscosity of oil decreases. An oil that provides a perfect lubricating film at 40°C might be as thin as water at 120°C, offering almost no protection. Grease is also affected; its base oil will thin out, and at very high temperatures, the thickener can break down, causing the oil to bleed out and leave the bearing dry. The selection of a lubricant and the prediction of bearing life are meaningless without a realistic estimate of the operating temperature.

Guarding the Gates: Sealing Solutions and Contamination Control

The battle against contamination is won or lost at the bearing's boundaries. The seal is the gatekeeper, tasked with two jobs: keeping the lubricant in and keeping the contaminants out. The choice of sealing solution is a critical part of the bearing sizing process and depends entirely on the severity of the environment.

• Shields (Z, ZZ): A shield is a non-contacting metal disc that fits into a groove in the outer ring. It provides a small gap with the inner ring. Shields are effective at retaining grease and keeping out large solid contaminants. Because they are non-contact, they generate no friction and have no speed limit. They offer no protection against liquids or fine dust.

• Contact Seals (RS, 2RS): A contact seal is typically made of a synthetic rubber material that is bonded to a steel insert. It has a flexible "lip" that makes light contact with a ground surface on the inner ring. These seals are highly effective at excluding moisture and fine dust. However, because they make contact, they generate some friction and heat, and they have a limiting speed beyond which the lip can be damaged.

• Labyrinth Seals: For high-speed or harsh environments, more complex sealing systems are needed. A labyrinth seal consists of a series of intricate, non-contacting pathways that are difficult for contaminants to navigate. They are often used in combination with other sealing methods.

• Taconite Seals: In extremely contaminated environments like mining or cement production, heavy-duty taconite seals are used. These are robust multi-stage labyrinth seals, often with an external grease barrier that is periodically purged to flush away contaminants.

The decision of which seal to use is a trade-off. A more effective seal may create more friction or have a lower speed limit. The designer must weigh the need for protection against the performance requirements of the application.

The Unseen Forces: Vibration and Shock Loads

Our initial load analysis focused on the steady forces acting on the bearing. Many machines, however, also generate vibration or are subjected to sudden shock loads. A rock crusher, a jackhammer, or the suspension of an off-road vehicle are extreme examples. These events impose forces on the bearing that can be many times higher than the normal operating load.

If these shocks occur while the bearing is stationary or rotating slowly, they can cause false brinelling. This is a type of surface damage that looks similar to true brinelling but is caused by a combination of load and micro-motion (vibration). The rolling elements fret against the raceway, creating polished depressions that can lead to noise and premature failure.

When shocks occur during rotation, they are treated as peak loads that must be considered in the fatigue life calculation. It is not sufficient to use an average load; the damaging effect of these high-magnitude, short-duration loads must be accounted for. Often, this is done by applying a "service factor" or "application factor" to the calculated load, effectively increasing it to build in a margin of safety against these unpredictable events. For applications with severe vibration, a bearing with higher internal clearance (like C3 or C4) may be required to accommodate the shaft deflections, and a more robust cage material (like brass or steel instead of polyamide) might be chosen for its superior strength.

Factor 5: Accommodating Physical Constraints and Assembly

Our journey so far has been in the world of forces, speeds, and temperatures. We have determined the performance characteristics our bearing must have. Now, we must return to the physical world of solid objects. The bearing must fit into the machine. It must be mountable, dismountable, and have the correct internal geometry to function properly. This final set of considerations bridges the gap between theoretical calculation and practical reality.

The Marriage of Shaft and Housing: Selecting Proper Fits

A bearing cannot do its job if its rings are spinning relative to the shaft or housing. To prevent this, the bearing rings are mounted with an interference fit (also known as a press fit). This means the shaft diameter is made slightly larger than the bearing's bore, or the housing bore is made slightly smaller than the bearing's outer diameter. When assembled, the elastic compression creates a strong frictional grip.

The amount of interference is not arbitrary; it must be carefully calculated. Too little interference, and the ring may slip under load, causing wear and fretting corrosion on the shaft or in the housing. Too much interference, and two bad things can happen. First, the ring can be stretched or compressed so much that it is permanently deformed or even cracked. Second, the heavy press fit will "squeeze" the bearing, reducing or eliminating its internal clearance. This can lead to excessive preload, high operating temperatures, and a very short life.

The correct fit depends on the type and magnitude of the load.

• If the load is rotating relative to the ring, a tight interference fit is required. For example, in a typical wheel hub, the shaft is stationary and the housing (the wheel) rotates. The load is therefore rotating relative to the outer ring, so the outer ring needs a tight fit in the housing. The inner ring sees a stationary load, so it can have a looser, often "push" fit on the shaft.
• If the load is stationary relative to the ring, a looser fit (often called a clearance or slip fit) is permissible.

Bearing manufacturers and international standards (like ISO 286) provide detailed tables of recommended shaft and housing tolerances (e.g., k5, m6 for shafts; H7, K7 for housings) for different bearing types, sizes, and application conditions. Selecting the correct tolerance class is a foundational step in machine design.

Internal Clearance: Giving the Bearing Room to Breathe

Every bearing is manufactured with a small amount of internal "play" or looseness. This is the internal clearance. It is the total distance that one ring can be moved relative to the other, either radially (Radial Internal Clearance) or axially (Axial Internal Clearance). This clearance is not a defect; it is a carefully controlled and absolutely necessary feature.

Why is it necessary? Because, as we have seen, the bearing's operating conditions will reduce this initial clearance.

• The interference fit on the shaft and in the housing squeezes the bearing, reducing its clearance.
• Thermal expansion, where the inner ring typically runs hotter than the outer ring, causes the inner ring to expand more, further reducing the clearance.

The goal is to select a bearing with enough initial "unmounted" clearance so that after it is mounted and has reached its operating temperature, it still has a small amount of positive "operating" clearance. A bearing with zero or negative operating clearance (a state of preload) will generate excessive heat and fail quickly unless it is specifically designed for such a condition (like angular contact ball bearings mounted in pairs).

Bearings are available in different clearance classes.

• CN (Normal) Clearance: The standard class, suitable for most applications with normal fits and operating temperatures.
• C3 Clearance: Greater than Normal clearance. This is very common and is used when there are tight interference fits, a significant temperature difference between the inner and outer rings, or a need to accommodate some shaft deflection.
• C4 Clearance: Greater than C3. Used in high-speed applications or those with very large temperature gradients, like vibrating screens or paper machine rolls.
• C2 Clearance: Less than Normal. Used in special applications where high running accuracy is needed and there is little to no heat or press fit to reduce the clearance.

Choosing the correct clearance class is a vital part of how to size a bearing. It requires an analysis of the fits and temperatures to predict the final operating clearance.

Mounting and Dismounting: A Practical Consideration in Sizing

A bearing is a consumable item. Eventually, it will need to be replaced. A design that makes this replacement difficult or impossible is a poor design. The sizing and selection process should also consider how the bearing will be installed and removed.

For small bearings, mounting can often be done with a press or by heating the bearing to expand it before sliding it onto the shaft. For large bearings, these methods become impractical. Large roller bearings, for example, are often made with a tapered bore. The inner ring has a tapered hole (typically a 1:12 or 1:30 taper). This bearing is mounted on a matching tapered shaft seat or, more commonly, on a tapered adapter sleeve or withdrawal sleeve.

The sleeve system is ingenious. The bearing can be slid easily onto the loose sleeve. Then, a hydraulic nut or a lock nut is used to drive the bearing up the taper. As it moves up the taper, the inner ring expands slightly, and the internal clearance is reduced. By monitoring the reduction in clearance with a feeler gauge, the mechanic can set the interference fit with extreme precision. Dismounting is also made much easier; hydraulic oil can be injected between the sleeve and the bearing, "floating" the bearing off the taper. When selecting a large bearing, considering a tapered bore and sleeve arrangement can save enormous amounts of time and effort during maintenance.

A Worked Example: Sizing a Bearing for a Conveyor Pulley

Let's bring these five factors together in a practical, step-by-step example. Imagine we need to select the two bearings to support the head pulley shaft of a bulk material conveyor system.

Step 1: Defining the Application and Gathering Data

First, we gather the knowns:

• Machine: Conveyor head pulley.
• Shaft Diameter: 100 mm (this is our starting point for size).
• Rotational Speed (n): 90 RPM.
• Operating Schedule: 16 hours/day, 5 days/week.
• Environment: Dusty, outdoor location (coal transport).
• Loads: The weight of the pulley, shaft, and the tension from the conveyor belt result in a calculated radial load (Fr) of 45,000 N on each bearing. There is no significant axial load (Fa = 0).

Step 2: Calculating Loads and Equivalent Load

• Load Analysis: The load is purely radial.
• Static Check: We need to ensure the bearing can support the load even when stationary. Let's make an initial selection of a Cuscinetto a rulli sferici, which is excellent for heavy radial loads and can tolerate the shaft deflection common in conveyors. A 22220 E series bearing has a 100 mm bore. From the catalog, its Basic Static Load Rating (C₀) is 765,000 N.
  • Since Fa = 0, the Equivalent Static Load P₀ = Fr = 45,000 N.
  • The static safety factor S₀ = C₀ / P₀ = 765,000 / 45,000 = 17. This is very high, well above the recommended minimum of 2-3 for such applications. The bearing is perfectly safe from a static perspective.
• Dynamic Load: We need to calculate the Equivalent Dynamic Load (P) for the life calculation.
  • Since Fa = 0, the calculation is simple: P = Fr = 45,000 N.

Step 3: Determining Required Bearing Life

• Operating Hours: The machine runs 16 hours/day. From our table, a machine for continuous daily operation should have a life of at least 40,000 hours to be considered reliable. Let's set our target L10h life to be 50,000 hours.

Step 4: Initial Bearing Selection and Life Verification

• Initial Choice: We already provisionally selected a 22220 E Spherical Roller Bearing.
• Catalog Data: From the catalog, its Basic Dynamic Load Rating (C) is 655,000 N.
• Life Calculation:
  • The life exponent for a roller bearing is p = 10/3.
  • L10 = (C / P)^p = (655,000 / 45,000)^(10/3) = (14.56)^(10/3) = 6,695 million revolutions.
  • Now, convert to hours: L10h = (1,000,000 / (n * 60)) * L10 = (1,000,000 / (90 * 60)) * 6,695 = 185.2 * 6,695 = 1,240,000 hours.

Our calculated life of over 1.2 million hours is vastly greater than our requirement of 50,000 hours. This tells us our initial selection is very conservative and likely oversized. This is not necessarily bad; it means the bearing will be extremely reliable. However, a smaller, less expensive bearing might be adequate.

Let's try a smaller spherical roller bearing, a 22218 E, which has a 90 mm bore. We would need an adapter sleeve to fit it to our 100 mm shaft, but this is a common practice.

• 22218 E Data: C = 500,000 N.
• Life Calculation:
  • L10 = (500,000 / 45,000)^(10/3) = (11.11)^(10/3) = 2,535 million revolutions.
  • L10h = 185.2 * 2,535 = 469,000 hours.

This life is still far in excess of our 50,000-hour requirement. This illustrates that for low-speed, high-load applications, bearing life is often very long, and the selection is often governed more by static capacity and physical size. Our initial 22220 E choice is a robust and safe one.

Step 5: Final Checks

• Speed: The operating speed is 90 RPM. The reference speed for a 22220 E bearing is around 2,200 RPM. We are well below this limit, so speed is not a concern.
• Environment & Sealing: The environment is dusty. The standard open 22220 E bearing will need to be housed in a robust plummer block housing with effective seals. A four-lip or labyrinth seal housing would be appropriate for this environment to keep coal dust out.
• Clearance & Fit: Given the heavy load and potential for temperature variations, a C3 internal clearance would be a prudent choice to ensure the bearing does not get preloaded during operation. The inner ring will be mounted on an adapter sleeve, allowing the fit to be set precisely during installation. The outer ring will have a clearance fit in the housing (e.g., H7) to allow for thermal expansion of the shaft.

This worked example shows how the factors are not isolated but are considered in an iterative process to converge on a suitable and reliable bearing solution.

Frequently Asked Questions (FAQ)

What is the difference between a static and a dynamic load rating?

The static load rating (C₀) measures a bearing's ability to withstand heavy loads while stationary without suffering permanent damage (dents). The dynamic load rating (C) is used to calculate the bearing's fatigue life when it is rotating under load.

How do I choose between a ball bearing and a roller bearing?

Generally, ball bearings are suitable for higher speeds and lighter loads. Roller bearings, with their larger contact area, can support much heavier loads but typically operate at lower speeds. Your choice depends on the specific combination of load and speed in your application.

Why is my new bearing running hot?

A new bearing might run hot for several reasons: too much grease was applied, the internal clearance is insufficient for the operating conditions (wrong fits or clearance class), there is excessive preload, or the seals are causing high friction. The temperature should stabilize after a short run-in period; if it remains high, the installation should be investigated.

What does the C3 in a bearing part number mean?

C3 indicates that the bearing has an internal clearance that is greater than the "Normal" (CN) class. This extra "room" is often needed to compensate for tight press fits or for when the inner ring runs significantly hotter than the outer ring, preventing the bearing from becoming too tight in operation.

How important is lubrication for bearing life?

Lubrication is profoundly important. A properly selected lubricant forms a thin film that separates the metal surfaces, preventing wear and drastically reducing friction. Poor or contaminated lubrication is one of the leading causes of premature bearing failure.

Can I use a shielded bearing in a wet environment?

No, it is not recommended. Metal shields (Z or ZZ) provide a non-contacting gap and are effective against large dust particles but offer almost no protection against liquids or fine dust. For wet environments, you should use a bearing with contact seals (RS or 2RS) or an external sealing system.

What happens if I put too much grease in a bearing?

Over-greasing is a common mistake. As the bearing rotates, it has to churn the excess grease, which generates significant heat. This can cause the lubricant's properties to degrade and can lead to premature failure. Most bearings only need to be 30-50% full of grease.

Is a noisier bearing about to fail?

An increase in bearing noise, such as a rumbling or squealing sound, is often a very strong indicator of a problem. It can be caused by surface damage from fatigue (spalling), contamination, or improper lubrication. Any significant change in the sound of a running bearing warrants an inspection.

Conclusione

Reflecting on these principles, it becomes clear that the act of sizing a bearing is an endeavor of synthesis. It requires the designer to be part a physicist, calculating forces and motions; part a statistician, interpreting probabilities of failure; part a materials scientist, understanding the limits of steel; and part a pragmatist, considering the realities of assembly and environment. It is a process that resists oversimplification. A choice made solely on shaft diameter from a catalog is a gamble. A choice informed by a deep analysis of loads, a realistic expectation of life, a respect for speed and temperature, and a practical view of the operating environment is an act of sound engineering. By embracing this comprehensive approach, we move from merely picking a part to designing for reliability. We ensure that this small, often unseen component can faithfully perform its duty, allowing the grander machine of which it is a part to operate safely, efficiently, and enduringly.

References

American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.).

American Psychological Association. (n.d.). In-text citations. APA Style. Retrieved April 10, 2025, from

Harris, T. A., & Kotzalas, M. N. (2006). Essential concepts of bearing technology (5th ed.). CRC Press.

International Organization for Standardization. (2014). Rolling bearings — Dynamic load ratings and rating life (ISO Standard No. 281:2007). Retrieved from

International Organization for Standardization. (2018). Rolling bearings — Static load ratings (ISO Standard No. 76:2006). Retrieved from

Sharma, U. N., Karki, T. M., & Banjade, G. (2025). Understanding in-text citations in academic writing: A review of APA 7th edition guidelines. TULSSAA Journal of Social Sciences and Humanities, 12(1), 1-15. https://doi.org/10.3126/tulssaa.v12i1.77260

SKF. (2018). Bearing failures and their causes. SKF Group. Retrieved from

Stolarski, T. A., & Tobe, S. (2011). Rolling bearings in service: analysis and practice. John Wiley & Sons.

Zaretsky, E. V. (Ed.). (1997). STLE life factors for rolling bearings (SP-34). Society of Tribologists and Lubrication Engineers. Retrieved from