Fatigue testing allows analysts to assess the failure of materials and parts under repeated fluctuating loads that are lower than the material’s ultimate tensile strength or yield strength. This helps manufacturers to understand the progressive damage caused by multiple loads over time, none of which are sufficient on their own to fracture the material. This method is used in many applications, ranging from testing of automobile and aircraft parts, to construction materials, to medical devices and more.
While a component in the real world might take many years or decades to ultimately fail, laboratory fatigue testing using specialized machinery shortens this time by applying the loads in quick succession, often at a frequency of several cycles per second depending on the application. Even so, fatigue testing can still be time consuming, with some tests taking days or weeks to complete the necessary number of cycles.
As these tests are necessary to ensure not only product quality but also the safety of critical structural components, vehicles, aircraft and medical products, acceleration of fatigue testing can help save costs and shorten time to market, but must be done with care to maintain the accuracy and reliability of the results data. Here are a few strategies and tips to help speed up your fatigue testing and ensure results that meet the standards of consumers, clients and regulators.
1. A Small Load Increase Can Make a Big Difference
One of the main factors that will impact the duration of your experiment time (i.e. how long it takes the sample to fail) is the magnitude of the load being applied during each cycle of the test. The closer this load is to the material’s ultimate tensile strength, the fewer cycles it will take to induce failure, and the shorter the test duration will be.1 Choose a stress load that is far lower than the ultimate tensile strength, and you could end up waiting around forever for a failure to occur; choose a load that is too high and the test could be over as soon as it begins, which doesn’t provide very useful fatigue life insights. This is why it is important to optimize your loading conditions to produce valuable data points without enduring excessive test durations.
If you are looking to speed up your current fatigue testing regimens, increasing the load is one option to reduce the total test cycles and shorten the test. However, what’s crucial to remember is that increasing the load reduces the number of cycles exponentially, not linearly2; in other words, it is not as simple as doubling the load to halve the runtime. For example, for many types of steel, a small load increase of just 15% decreases the test time by 50%—for most aluminum varieties, this small increase could reduce the time by 60%.3
Additionally, a change in load could alter the type of failure that occurs as a result of the load being applied. If you are running a high cycle fatigue test assessing repeated elastic deformation, you do not want to push that test into the low cycle, plastic deformation region where the results would no longer be representative of ordinary use.3 In this case you will want to take care that the new loading stays below the material’s yield strength and does not lead to the formation of a plastic hinge, which would alter the load path and distort the results.2 Therefore, the more resistant the sample material is to the plastic deformation, the more flexibility you may have to accelerate testing by increasing load.
2. Temperature and Natural Frequency Matter When Increasing Loading Frequency
Increasing the frequency of the loading cycles in your test may seem like a straightforward method for reducing test time, but your instrument’s frequency capabilities are not the only factors you should consider when taking this approach. Ramping up the frequency of loading cycles can cause heat to build up in the sample without enough time to dissipate, and this heat can contribute to deformation, distorting the results of the test. If you are increasing the frequency of your test, you should monitor the rise in temperature in the sample; one rule of thumb is to ensure that the temperature during the new test is no more than 6°C (10°F) higher than seen during the original test.3
Another important consideration when increasing the frequency is the natural frequency of the sample – this is the frequency at which the part tends to vibrate on its own when disturbed. If a force applied to an object is close to or equal to the natural frequency of the object, a resonance is excited and the object oscillates at a higher amplitude than it would when the force is applied at a non-resonant frequency. This resonant condition can increase the deformation in the sample, so the frequency of the test should always be kept well below the sample’s natural frequency to avoid this problem. If you want to accelerate your fatigue testing by increasing the frequency of your loading, you should perform a modal analysis on the sample to determine its natural frequencies and keep the maximum frequency for your test below ⅓ the sample’s lowest natural frequency.2
3. Omitting Low-Damage Cycles Speeds Up Variable Amplitude Loading Analysis
Variable amplitude loading (VAL) fatigue tests can allow for more realistic and comprehensive analyses of how the different levels of stresses on a component over its lifetime will contribute to damage progression. Real-life load histories can be obtained by monitoring a component during normal use, and these histories can be used to design simulated VAL fatigue tests in the lab. However, variable load histories from real-life scenarios often include many load cycles of low amplitude that cause little-to-no damage, and playing out all of these low damage cycles on a testing machine can add a lot of time with little impact on results.
For this reason, many VAL fatigue tests are accelerated by truncating the load spectrum, omitting non-damaging or low damage cycles from the test. This process can be assisted by the rainflow counting algorithm, which filters out noise and identifies all fatigue cycles in a VAL history, plotting them on a rainflow matrix. The X and Y axes of the rainflow matrix represent the “to” load level and “from” load level, respectively, of each cycle, with the amplitude of the cycle being the difference between these two values. The number of occurrences of cycles with a given “from” and “to” value is visualized by a specific color on the matrix. The low damage, low amplitude cycles are found along the diagonal of the matrix from the top left to the bottom right, where the “to” and “from” values are similar. Typically, these cycles have the highest occurrence, and omitting them from the test can reduce runtime significantly.3,4
Similar to increasing the load on the sample, increasing the number of repetitions of high or medium amplitude in the variable load history can also increase the damage and further compress the test time. In one case study, increasing the medium loads increased damage by a factor of 1.87 and reduced the test time by 45.93% for a load history obtained by monitoring a track control arm with strain gauges.5
4. Consider Multiple Sample Testing Systems
It is necessary for several identical samples to be tested in order to ensure results are reliable, and the time it takes to test one sample after the other will of course extend your testing campaign. The ability to test multiple samples simultaneously can significantly decrease testing time and increase throughput, with one option being the use of simultaneous testing on separate machines, although this requires that the machines also have identical conditions to ensure consistency between the tests. Another option is the use of systems or fixtures that enable multiple samples to be tested in the same run on the same machine, which provides the advantages of increased sample capacity and higher confidence that test conditions are uniform for all samples.1 Capabilities of independent sample monitoring and fracture detection can help streamline data acquisition for multiple samples and control the test procedure as needed to ensure the integrity of all samples.6 Simultaneous testing using specialized systems and fixtures may be a suitable option to accelerate testing of small samples and devices such as small parts, test coupons and medical devices.
References
- “An Introduction to Fatigue Testing,” Webinar Presented by Troy Nickel, TA Instruments, 2016. https://www.youtube.com/watch?v=gBMhhsOkGH4
- “Methods for Accelerating Dynamic Durability Tests,” White Paper by Dr. Andrew Halfpenny, nCode International, 9th International Conference on Recent Advances in Structural Dynamics, 2006. https://www.ncode.com/images/Resources/Downloads/Whitepaper_nCode_MethodsforAcceleratingDynamicDurabilityTests_v2-Halfpenny.pdf
- “Some Thoughts on Accelerated Durability Testing,” Article by Peter Schaldenbrand, Siemens, 2019. https://community.sw.siemens.com/s/article/some-thoughts-on-accelerated-durability-testing
- “Rainflow Counting,” Article by Peter Schaldenbrand, Siemens, 2019. https://community.sw.siemens.com/s/article/rainflow-counting
- M. Jimenez, "Accelerated Fatigue Test in Mechanical Components", in Contact and Fracture Mechanics. London, United Kingdom: IntechOpen, 2017 [Online]. Available: https://www.intechopen.com/chapters/58621 doi: 10.5772/intechopen.72640
- “Advanced Test Control Techniques for Improved Insight into Individual Device Fracture in a Multi-sample Fatigue Test System,” White Paper by Kelsey Banas, Thomas Byron, Andrew Simon, Scott Anderson and Lito Mejia, MTS Systems. https://www.mts.com/en/articles/biomedical/msff-white-paper