How Do Fatigue Properties Decide Whether New Materials Last or Fail?

Why Do Fatigue Properties Matter More Than Static Strength?

Fatigue properties often decide whether a part stays in use or comes back with cracks, especially when the load repeats every second, every shift, or every trip. If you buy, specify, or compare new materials, do not stop at tensile strength and hardness. The real check is simple: what happens after thousands, millions, or even hundreds of millions of cycles? For a wider view of related material behavior, you can also review the materials properties section.

Repeated Loads Create Hidden Damage

A single static test can make a material look safe on paper. Fatigue loading is different because the damage builds up little by little. A shaft, spring clip, battery tab, bracket, implant, or aircraft fastener may work below yield strength, yet small cracks can still start and move through the part. Cambridge University Press, in Mechanical Behavior of Materials, Chapter 17, reports an estimate that about 90% of service failures in metal parts are caused by fatigue. IEEE Technology Navigator gives a broader public estimate of 50% to 90% of metallic structural failures in service. The exact share is not the same in every industry, but the point is clear enough: cyclic loading needs real attention before purchase or approval.

Static Strength Does Not Tell the Whole Story

Tensile strength tells you how a sample behaves in one pull. Fatigue strength tells you how it handles repeated stress over time, and that is closer to many real parts. A high-strength aluminum alloy may save weight, but it may not have a clear endurance limit like some steels. A hard coating may improve wear resistance, while a rough or cracked coating can reduce fatigue life. A new composite may look good on a datasheet, then become sensitive to holes, edges, moisture, or impact marks in service. This is why buyers should treat static data as the first check, not the final answer.

Service Conditions Shape Real Life

Fatigue life depends on load level, load ratio, temperature, surface finish, corrosion, residual stress, geometry, and manufacturing defects. One small notch near a bolt hole can matter more than a clean tensile number. In a plant, motor vibration can add millions of cycles before the maintenance team sees a problem. In a vehicle, every road bump adds another small stress event. In electronics, thermal cycling can bend solder joints and metal foils again and again. Fatigue is usually not sudden at the start; it is repeated damage that finally becomes visible.

Which Fatigue Properties Should You Check First?

You do not need every fatigue number for every order. You need the numbers that fit how the part will be used. A cosmetic panel, a load-bearing spring, and a medical screw should not be checked with the same plan. Start with the properties below, then check whether the test conditions match your service conditions.

S-N Curve and Fatigue Strength

The S-N curve, also called the stress-life curve, shows stress against cycles to failure. It helps you see the stress a material can take for a target life, such as 100,000, 1 million, or 10 million cycles. ASTM E466-21 describes force-controlled constant-amplitude axial fatigue testing for metallic materials and states that this type of test is used to study the effects of material, geometry, surface condition, stress, and similar factors on fatigue resistance for large numbers of cycles. For sourcing work, this detail matters. If a supplier gives S-N data, ask whether it follows a recognized method and whether the specimens match your part form.

Endurance Limit or Fatigue Limit

Some steels show a practical endurance limit, which means a stress level below which fatigue failure may not occur in a defined test setting. Many aluminum alloys, magnesium alloys, and copper alloys do not show such a clear flat area, so engineers often use fatigue strength at a chosen number of cycles. This often causes confusion in trade talks because different suppliers may use different cycle counts. If a datasheet says “fatigue limit,” check the cycle count, stress ratio, sample type, and surface condition. A value at 10 million cycles is not the same as a value at 500 million cycles.

Crack Growth and Fracture Toughness

For critical structures, fatigue is not only about when a crack starts. It is also about how fast that crack grows after it starts. FAA Advisory Circular AC 25.571-1D, issued for transport category aircraft structure, gives guidance for damage-tolerance and fatigue evaluation, including widespread fatigue damage and the limit of validity of engineering data. Aircraft rules are strict, but the same basic lesson applies to many industries. A safer design often needs a clear view of how damage grows before final fracture.

How Do Standards Make Fatigue Data More Trustworthy?

Fatigue data can have wide scatter. Two samples from the same alloy may fail at different cycle counts, even under the same nominal stress. This does not make the test useless. It means the method, sample count, and statistics must be clear. A polished lab bar and an as-cast part with pores can give very different results.

ASTM E466 for Axial Metallic Testing

ASTM E466-21 is widely used for force-controlled, constant-amplitude axial fatigue tests of metallic materials. In this test, the specimen is loaded again and again in a controlled pattern until it reaches a defined failure point or runout. If you compare suppliers, ask for the standard version, specimen geometry, stress ratio, test frequency, number of samples, and runout rule. A clean curve without these details is closer to a sales picture than engineering data.

ISO 12107 for Statistical Planning

ISO 12107:2012 covers statistical planning and analysis of fatigue test data. The standard points to planned test levels and data treatment, not just one or two good-looking results. This is useful when you buy new materials with a short market history. A single best result may look strong, but the lower side of the scatter band is often the safer number to use in design.

Traceable Test Reports and Conditions

A good fatigue report should list material grade, heat lot, processing route, heat treatment, surface preparation, specimen drawing, loading mode, stress ratio, environment, temperature, failure definition, and sample count. These items help you check whether the data can be used for your part or only for the lab sample. For molded polymers, add moisture condition and molding direction. For composites, add layup, fiber direction, void content, and open-hole data if bolts are used. It may feel like extra paperwork, but this paperwork can prevent costly re-testing later.

What Material and Process Factors Change Fatigue Life?

Fatigue performance is not decided by chemistry alone. Processing can raise or lower fatigue life because it changes microstructure, defects, surface state, and residual stress. This is where new materials can be useful, but it is also where loose comparisons create trouble.

Surface Finish and Notch Sensitivity

Fatigue cracks often start at the surface. A rough machined groove, sharp corner, weld toe, stamping burr, or corrosion pit can raise local stress. Polishing, shot peening, burnishing, and better radius design may improve fatigue behavior when they are applied in the right way. For example, a stainless steel part with a smooth radius near a hole can last much longer than the same steel with a sharp punched edge. The alloy is the same, but the fatigue result can be very different. See also: Application.

Microstructure and Heat Treatment

Grain size, inclusions, phase distribution, and heat treatment all affect fatigue. In steels, clean melting and controlled heat treatment can reduce the risk of cracks starting at inclusions. In aluminum alloys, aging condition and product form can shift fatigue strength. ASM International handbooks on steels and aluminum alloys treat fatigue behavior as linked to composition, mechanical properties, microstructure, product form, and processing. The practical takeaway is simple: ask for process history, not only the grade name.

Defects in Additive Manufacturing

Additive manufacturing can make shapes that casting or machining cannot make easily. Fatigue is often the harder part, especially for metal parts with internal defects or rough surfaces. NIST has stated in its additive manufacturing fatigue and fracture program that metal AM has not been broadly used in fatigue and fracture critical applications despite industrial need. NIST also shows lack-of-fusion defects in AM Ti-6Al-4V as crack initiation sites in high-cycle fatigue fracture surfaces. For buyers, that means powder quality, build orientation, scanning parameters, heat treatment, surface finishing, and inspection method should all be discussed before the order is confirmed.

How Can You Compare Suppliers Without Getting Misled?

Supplier datasheets are useful, but they rarely show the full fatigue picture. Compare data under similar conditions first, then check the gap between lab specimens and your real part. A bit of caution here is normal purchasing work, not a sign of distrust.

Match the Loading Mode

Axial fatigue, rotating bending, flexural fatigue, torsion fatigue, thermal fatigue, and rolling contact fatigue do not give the same results. If your part bends in service, an axial S-N curve may still help, but it may not be enough. If your product sees start-stop torque, ask for torsional or multiaxial data. If temperature change drives the stress, thermal cycling data may matter more than room-temperature mechanical fatigue data.

Check the Specimen and the Real Part

Small polished specimens usually perform better than real parts with holes, welds, threads, edges, porosity, and assembly marks. That does not make specimen testing bad; it makes it a baseline. For high-value parts, you may need coupon testing first, then subcomponent testing, then full part validation. A bicycle crank, a robot arm joint, and a compressor blade all need different proof levels because their risks and loads are different.

Ask for Scatter, Not Only Averages

Fatigue is statistical. A supplier may show an average life of 1 million cycles, while the lowest tested sample failed at 380,000 cycles. That lower result can matter more than the average if the failure cost is high. Ask for the number of specimens, runouts, failed samples, and confidence basis. If the supplier cannot share raw data, request at least the stress levels, life range, and failure locations. No reliable public data should be invented to fill gaps; when data is missing, it should be marked as missing.

How Should You Use Fatigue Properties in Material Selection?

Good material selection is not about chasing the highest number. It is about choosing a material and process that match the load history, production route, cost target, and inspection plan. Fatigue data becomes useful only when you connect it to the work the part must do.

Define the Duty Cycle Early

Before choosing a material, write down the expected load cycle. Include peak stress, minimum stress, mean stress, temperature, environment, vibration, impact events, and desired life. If the part runs 24 hours a day at 30 Hz, it reaches about 2.6 million cycles per day. That quick calculation can change the whole material discussion. What looks like a long lab test may represent only a few days of service.

Use Safety Factors With Care

A simple safety factor on tensile strength may not protect against fatigue. You may need a fatigue design factor, a knockdown factor for surface finish, or a damage-tolerance plan. Corrosion, weld quality, and machining marks can reduce life fast, even when the base material is strong. For safety-critical parts, inspection intervals and nondestructive testing may need to be part of the design from the start.

Balance Performance, Cost, and Inspection

A premium alloy may reduce weight, but it may also need tight surface control. A lower-cost steel may be heavier, but it can be easier to inspect and repair. A composite may resist corrosion, while it may need special checks for impact damage. For new materials, the better choice is often the one with stable, repeatable fatigue behavior and clear quality control. High numbers are useful, but repeatable production is usually more important for buying decisions.

FAQ

• Q1: What Are Fatigue Properties? A: Fatigue properties describe how a material behaves under repeated loading, including fatigue strength, S-N curve behavior, endurance limit, crack initiation, and crack growth.
• Q2: Why Can a Part Fail Below Its Yield Strength? A: Repeated stress can create tiny cracks at surfaces, notches, inclusions, pores, or corrosion pits. These cracks grow cycle by cycle until final fracture occurs.
• Q3: Is a Higher Tensile Strength Always Better for Fatigue? A: No. Higher tensile strength can help in some cases, but surface finish, defects, heat treatment, residual stress, and environment may control fatigue life.
• Q4: Which Standard Is Common for Metallic Fatigue Testing? A: ASTM E466-21 is commonly used for force-controlled constant-amplitude axial fatigue tests of metallic materials. ISO 12107:2012 is useful for statistical planning and analysis.
• Q5: What Should You Ask a Supplier Before Buying a Fatigue-Critical Material? A: Ask for test standard, specimen geometry, stress ratio, sample count, S-N data, surface condition, processing history, failure locations, and any limits on using the data for real parts.