What Is Fatigue Limit and Why Does It Matter in Material Selection

The fatigue limit is a small line in a material data sheet, but it can decide the life of a spring, shaft, bracket, welded frame, or rotating part. If you are comparing metals or new material grades for cyclic load service, this guide on material properties gives you a practical way to read fatigue data, question supplier claims, and choose tests that match real service.

What Is Fatigue Limit and Why Should You Care?

A tensile test shows how a sample behaves when it is pulled once. Fatigue is different. It is about repeated loading, often below yield strength, where a part looks safe at first but grows a crack after thousands or millions of cycles.

Stress Amplitude at Long Life

Fatigue limit is the stress amplitude a material can survive for a very high number of cycles in a defined test without fracture. Engineers usually plot stress amplitude against cycles to failure, so they can see how the material behaves under repeated load. For a motor shaft, gear tooth, hinge pin, or valve spring, that repeated stress often matters more than one-time static strength.

Fatigue Limit vs Fatigue Strength

Use the terms with care. Fatigue limit, also called endurance limit in many handbooks, points to a plateau on the S-N curve. Fatigue strength means the stress at a stated life, such as 10^7 cycles. This difference matters because many aluminum alloys do not show a true flat plateau.

A Property Tied to Test Conditions

Do not treat fatigue limit like density. The number depends on the way the test was run. ASTM E466-21 says axial fatigue testing is used to study how material, geometry, surface condition, stress, and similar factors affect fatigue resistance over many cycles. Source: ASTM International, 2021. A polished laboratory bar and a laser-cut bracket are related, but they are not the same thing.

How Is Fatigue Limit Measured in Real Testing?

Useful fatigue data comes from controlled tests, not from one catalog line. If you buy material for a high cycle part, the test method behind the number is almost as important as the number itself.

S-N Curves from Constant Amplitude Loading

Specimens are cycled at several stress levels. A high stress may break a sample after 30,000 cycles, while a lower stress may last two million cycles. The results are plotted as an S-N curve, with stress on one axis and cycles to failure on the other. The curve shape tells you whether the material has a clear knee, or whether life keeps dropping as cycles increase.

Runouts at a Chosen Cycle Count

A runout means the specimen survived the planned cycle count. It does not prove infinite life by itself. For steel, repeated runouts near the knee may support a fatigue limit. For aluminum, you often see fatigue strength reported at a set life instead. Public fatigue references commonly use examples such as 10^7 cycles, and some ASM aluminum tables report values at 500,000,000 cycles for stated test methods.

Standards That Keep Results Comparable

ISO 1099:2017 specifies axial, constant amplitude, force controlled fatigue tests at ambient temperature on metallic specimens without deliberate stress concentrations. ASTM E466-21 covers force controlled constant amplitude axial tests for metallic materials. These standards do not give one universal value for every buyer to use. They make supplier data easier to compare, which is what matters in sourcing work.

Which Materials Have a Clear Fatigue Limit?

Material family matters, but it is only the starting point. Heat treatment, cleanliness, grain structure, casting quality, surface finish, and manufacturing route can shift fatigue performance by a lot. One small inclusion in the wrong spot can spoil a good data sheet.

Steels and Titanium Often Show a Knee

Many steels show a visible knee on the S-N curve. A common handbook-level estimate starts near one half of ultimate tensile strength for smooth laboratory steel specimens, but this is only a first screening rule. Source: ASM Handbook Volume 19, Fatigue and Fracture, and SAE fatigue design references. Some titanium alloys also show good long life behavior, but you still need alloy-specific data before making a part decision.

Aluminum Uses Fatigue Strength at a Set Life

Aluminum alloys such as 6061, 7075, and cast A356 are usually compared by fatigue strength at a named cycle count and load ratio. In other words, the question is not only how strong the alloy is. You also need to ask how many cycles, at what stress ratio, in what environment, and with what surface finish. Without those details, two numbers on two supplier sheets may not be comparable.

Polymers and Composites Need Separate Data

For polymers, elastomers, and fiber composites, fatigue life depends strongly on temperature, moisture, fiber direction, strain rate, and creep. There is no single reliable public fatigue limit for broad groups such as carbon fiber or nylon. You need grade data, process data, and often part-level testing, especially when the part will see heat, vibration, or outdoor service.

What Factors Can Lower Fatigue Limit in Service?

Most fatigue cracks start at weak local points. The base alloy may be fine, but the part fails because the design or process gives the crack an easy starting place. This is where shop details matter.

Surface Roughness and Machining Marks

Cracks often start at the surface because bending stress is highest there and air or moisture can reach it. A turned shaft with spiral tool marks can lose fatigue strength compared with a polished specimen. Fine grinding, polishing, and controlled shot peening can help. The process still has to be repeatable, or the result will vary from batch to batch.

Notches Welds and Sharp Corners

Holes, keyways, threads, weld toes, and sharp inside corners raise local stress. You may calculate nominal stress as 60 MPa, while a small radius sees much higher local stress. Welded joints need extra care because weld profile, toe defects, residual stress, and inspection quality often control fatigue life more than base metal strength. In real production, this is why a neat drawing is not enough.

Corrosion Temperature and Mean Stress

Corrosion pits are crack starters. High temperature can soften alloys or change residual stress. Mean tensile stress also hurts because the crack stays open for more of each cycle. A well-known case is Aloha Airlines Flight 243 on April 28, 1988, where the NTSB cited significant disbonding and fatigue damage at a fuselage lap joint. Source: NTSB/AAR-89/03, 1989. The lesson is not only for aircraft; coastal machinery and outdoor equipment face the same bad mix of moisture, stress, and time. See also: Application.

How Should You Use Fatigue Limit in Material Selection?

When you compare suppliers, ask for the test background before you compare prices. A fatigue value without test conditions can lead you in the wrong direction. A better material choice starts with the real load story of your part.

Match Data to the Actual Load Ratio

Ask for the load ratio R, waveform, frequency, temperature, environment, specimen type, and surface finish. A fully reversed test at R = -1 is not the same as pulsing tension at R = 0.1. If your pump rod never sees compression, do not buy material based only on a reverse bending number. The test should look close enough to the duty cycle you expect in service.

• Check cycle target and runout definition.
• Check whether the specimen was smooth, notched, welded, coated, or polished.
• Check whether the data came from room temperature air or the real service environment.

Compare Coupon Data with Part Geometry

A coupon is clean and simple. Your part has edges, holes, coatings, assembly stress, and maybe a small burr near a slot. Use coupon data to rank materials, then apply stress concentration factors, size factors, surface factors, and mean stress corrections such as Goodman or Gerber where suitable. This step keeps the data sheet from being used outside its real limits.

Add Testing for Critical and New Materials

For safety-critical parts, additive manufacturing, welded assemblies, or high cycle rotating parts, run a test that copies service. That may cost more at the start, but it is cheaper than a field failure. FAA rule 14 CFR §25.571 requires aircraft structure evaluation so catastrophic failure from fatigue, corrosion, manufacturing defects, or accidental damage is avoided through operational life. That shows why fatigue is a life-cycle issue, not a one-page property table.

Can Better Design Raise Fatigue Performance without Changing Material?

Yes, and this is often the lower-cost path. A better alloy can help, but geometry and process control may give the bigger gain. Many real failures are not material failures alone; they are detail failures.

Smoother Transitions and Larger Radii

A larger shoulder radius, a rolled thread instead of a cut thread, or a better hole finish can do more than a more expensive alloy. In a small bracket, moving a hole a few millimeters away from a bend line can reduce local peak stress. It sounds simple, but many workshops learn this through broken parts. Good fatigue design is often built from these small choices.

Controlled Surface Treatments

Shot peening, roller burnishing, nitriding, carburizing, and anodizing can help or hurt depending on the process. Shot peening adds compressive residual stress, which can slow crack initiation. A brittle coating may crack early, especially on a part that flexes. Ask how the treatment was qualified, not only what treatment name appears on the drawing. The same treatment name can mean different results from different suppliers.

Inspection Plans for Long Life Parts

Fatigue design also means inspection. Dye penetrant, magnetic particle, eddy current, ultrasonic testing, and visual checks each find different defects. If the component must run for ten years, set inspection intervals from crack growth data and real load history. Calendar habit alone is not enough for a part under repeated load.

FAQ

Q1: What Is the Simple Meaning of Fatigue Limit? A: It is the stress level a material can survive for a very high number of repeated cycles in a defined test without breaking.

Q2: Is Fatigue Limit the Same as Yield Strength? A: No. Yield strength is measured under one-time loading. Fatigue limit deals with repeated loading and is usually much lower.

Q3: Why Does Aluminum Often Need a Cycle Count? A: Many aluminum alloys do not show a true endurance plateau, so engineers report fatigue strength at a stated life, such as 10^7 cycles.

Q4: Can Surface Finish Change Fatigue Limit? A: Yes. Rough machining marks, pits, and scratches can start cracks early, while controlled polishing or shot peening may improve fatigue behavior.

Q5: Should You Trust a Supplier Fatigue Number Alone? A: Not without the test method, load ratio, specimen type, surface condition, environment, and cycle count. Those details decide whether the number fits your part.