Fatigue Explained

Ever find yourself in a situation where you needed to break something in two (say a wire), but you didn’t have any tools to do it? Did you pull at both ends until it broke? Maybe at first, but then you likely bent the wire back and forth until it weakened enough at the bend point to break in half. Without really thinking about it, you have utilized the principle of fatigue.

Let’s go back and look what is actually happening with the wire. While the wire is sitting on your desk, it is considered to be “at equilibrium.” When you start bending the wire, you are introducing stress. Bending both ends downward introduces tension in the top part and compression in the bottom part of the bend. Bending upwards reversed the positions of stress, introducing compression in the top and tension in the bottom area of the bend.

Now here comes a key part: cycles. You are not just bending the wire once or twice; you are bending the wire back and forth repeatedly. Every time you bend it once, you are introducing this sequence: positive stress then zero stress then negative stress. That sequence is one cycle. Bending it repeatedly causes this stress to repeat over and over again, producing multiples of these stress cycles. Multiple cycles of stress produces “fatigue” in the bend of the wire, making it progressively weaker at that point. This fluctuation of stress over time is the root cause of failure due to fatigue.

In this example we are just using two hands and a small wire. Imagine a machine that turns a shaft 1000 rotations per minute. If it is not designed/built well, it may have an ever-so-slight imperfection. That imperfection can allow fluctuations that are insignificant and unnoticeable at first. However, at 1000 rpm, that’s 1,440,000 rotations a day or close to 263 million rotations over 6 month period! Given time, that imperfection can contribute to a micro-crack that begins to grow with every fluctuating cycle. Given enough time or enough of a fluctuation (big fluctuations need less time/small fluctuations need more time), an abrupt failure can occur.

This is different from failures involving a static (non-cycling/repeating) load. If you were piling numerous bricks onto a scaffold, you may hear some creaking & snapping and see some swaying & splintering that would give you advanced warning that the scaffolding was about to come crashing down. Not so with fatigue failures. With fatigue failures, everything will seem normal and then suddenly it is not. The failure is unpredictable. Damage from fatigue is cumulative; so resting the machine or material does not help it to recover.

This type of failure should not be confused with normal “wear & tear”, which refers to normal damage that is inevitably experienced the life cycle of a material.

Remember a couple sentences earlier in this article when it was said that fatigue failure is unpredictable. Well, that’s incorrect. It is predictable and much effort is put into designing around that inevitable break. How can it be both predictable and unpredictable? Let’s explore. Let’s start with the unpredictable. Let’s start with a look back to the year 1842 in France. Didn’t see that coming did you? 1842 France is relevant because of a train traveling from Versailles to Paris derailed unexpectedly that year. The front locomotive broke an axle and the following carriages crashed into the wrecked engines and caught fire. 55 people lost their lives. The crash was witnessed by a locomotive engineer and was one of the first events that got people thinking about how something such as a train axle could unpredictably fail.

Engineers (educated scientists, not the locomotive-operator kind with the funny hats) studied and tested on this concern until finally in 1870 one particularly bright engineer not only acknowledged the modern idea of fatigue, but also concluded that cyclic stress range is more important than peak stress and introduced the concept of an endurance limit. What does this mean? We now know that if you cycle a material from one stress to another over and over again, it can eventually fail (a railroad wheel comes under load when in contact with track and then that stress is relieved when the wheel rotates up – a classic fatigue environment). After engineers became aware of this phenomenon, they were highly motivated due to the rapidly expanding railway system to figure out how to design railroad axles so that this would not continue to happen.

The basic science looks like this: A material has an ultimate stress point. If you bend it to that maximum point it will break even before it cycles once. If you decrease the stress a bit from the ultimate stress (breaking) point, you will be able to cycle it (bend it back & forth) a few times before it breaks. If you decrease the stress more, you can cycle even more times. This relationship between the fluctuation or amplitude of stress and the number of cycles trends down (the less stress = the more cycles needed until failure), but not forever. Once you reach a low enough stress amplitude, the number of cycles needed to get the material to fail increases dramatically. So, if the stress amplitude is below this established limit, the number of cycles needed to make it fail jumps to infinity! Another way of saying this is that the material never fails after this limit. This limit is known as the endurance limit.

If you are building something where cyclic loading is the nature of the design, then you want to build it so that it operates below its endurance limit. So, prediction of failure is utilized in the design/build portion of a product to establish its endurance limit. But once it is built, fatigue failure can still occur unpredictably because of a manufacturing defect or placing loads upon it that take it beyond its intended design.

So as you go about your day, look around you. Play a game. The things you see – are they under static (non-moving load), such as an advertising billboard sign or are they under cyclical loading such as advertising billboard sign swaying in the wind? Are they “at equilibrium?” Are they experiencing normal wear & tear? Hopefully, they are all operating under their designed endurance limit!

CED has engineers working tirelessly within its ranks who can evaluate material to determine if a failure has been the result of fatigue. Submit a claim on our site to see


Featured Engineers:

Douglas E. Bishop, EIT Mechanical Engineer

Joel J. Schubbe, Ph.D. Materials Science Engineer

Connect with a CED Engineer in your region.

Submit a case or claim online.