Since being created to help car manufacturers in the 1950s, fatigue management software has developed from an in-house pet project to a viable commercial product. These days, many large mining companies use the technology to reduce costs and man-hours when designing rigs and machinery.
“Fatigue used to be the black art that produced financial black holes,” says Professor John Draper, chairman of fatigue analysis vendor Safe Technology and author of Modern Metal Fatigue Analysis.
“Potential errors were hidden behind the euphemism of ‘safety factors’ and manufacturers paid the price for overweight components that cracked prematurely, a seemingly endless series of prototype developments, unpredictable warranty claims and loss of customer confidence.
“Traditionally, fatigue failures have been fixed by overdesign. But increasingly, engineers are under pressure to ‘design down’ to save weight and material costs. Overdesign is no longer a viable option and the need for sophisticated fatigue analysis tools has become increasingly apparent.”
Fatigue analysis software can help reduce development times by making designs correct before you actually cut a component, according to Professor Phil Irving, head of the damage tolerance group at Cranfield University.
“It can make the difference between doing a dozen fatigue tests and one or two, which saves a lot of time and money during development,” he says.
“In the aerospace industry, for example, certain regulations require testing and analysis of fatigue liability in sections of the aircraft. Fatigue software allows you to demonstrate analysis in that instance and it also helps to calculate life extensions for big structures like oil rigs. If you have good information, you can recalculate fatigue life and extend the life of your assets. We’re seeing a lot of that in the North Sea [with offshore oil and gas installations] at the moment.”
Pushing the boundaries of fatigue
There are three major inputs on fatigue analysis software – component load measurements, stress analysis of the structure and material fatigue properties.
“The software interfaces directly to FEA [finite element analysis] software and is supplied complete with a database of material fatigue properties to which users can add their own data,” says Draper.
“It calculates where and when fatigue cracks will occur – the fatigue hotspots – plus the safety factors of working stresses for rapid optimisation and the probability of survival at different service lives, otherwise known as the ‘warranty claim’ curve. The results are presented as contour plots of fatigue lives, stress safety factors and probabilities of failure and plotted using standard FEA viewers and graphics software.”
BAE Systems, Rolls-Royce, Thales UK and Meggitt hope the latest developments in aerospace fatigue analysis will be found at Cranfield University, where the integrated vehicle health management (IVHM) centre was launched in November.While focused on aerospace, they are working on a product they say could be used in any industry where fatigue could be a concern.
Each of those companies has invested £1m in the joint university-Boeing project, which will lead research into vehicle monitoring and management in several areas, including fatigue.
“The IVHM idea came about because when companies move from selling products to services, as Bae has done, they need to know more about their assets,” says IVHM centre director Ian Jennions.
“In simple terms, IVHM involves a computer reading data sent via an acquisition unit from sensors around the machinery and suggesting appropriate action based on that.
“We’re researching various IVHM projects that will take between one and three years each. When it is ready for commercial use, IVHM will make a huge money savings for customers by reducing operating costs and helping to prevent component malfunction. The software will work across the aerospace, rail, marine, energy and car industries – anything in fact that utilises high value assets.”
For years, fatigue development software has been unable to cope with one of engineering’s most difficult areas – multiaxial states of stress. However, developers like nCode and Safe Technology have made huge advancements in this area since 2006.
“Fatigue analysis software research is in an exciting phase,” says Professor Draper.
“The last two to three years have seen a lot of the gaps in our knowledge filled in. We are now able to operate without making too many approximations and have software capable of taking into account things like multiaxial states of stress, so the results are much truer.”
When mining equipment giant Caterpillar was experiencing early field failures on a new piston, it turned to Safe Technology’s safe4fatigue to help eradicate the problem.
“Safe4fatigue has a wide range of fatigue algorithms for brittle and ductile materials including the principal stress method and Brown-Miller with mean stress correction, as well as surface finish factors,” explains Safe Technology’s Draper says.
“The software’s easy-to-use tools identify spikes and other signal anomalies and is integrated into our own fe-safe suite of software for the fatigue analysis of FEA models.”
The challenges for safe4fatigue on Caterpillar’s project were to accurately predict fatigue life from measured strain data, assess the impact of mean stress and surface finish on fatigue life and integrate and understand how the simulation and tests correlated.
“The minimum life was predicted by safe4fatigue at the notch area of the piston skirt, which correlated to our field observations,” says Caterpillar’s senior engineer Qingzhong Li.
“Considering the effect of surface finish also proved vital. Fatigue life decreased about 60% between polished and as-cast surface finishes. Overall, we found that safe4fatigue can accurately predict fatigue life from strain gauge data, and proved to be a useful complement to fe-safe and its other predictive and test methods for product design.”
The challenge with fatigue
Acquiring the information to feed into fatigue analysis software packages can be a costly business. Measuring load components and conducting stress analysis involves employing teams of people, keeping the subject operational and installing systems to conduct trials, which can run into thousands of pounds. Because of this, the data is not always as good as it should be, which can affect the accuracy of the fatigue life predictions.
“Fatigue analysis software is not always wonderfully accurate,” says Professor Irving.
“People get around this by inputting data for different materials or reducing the level of stress to get the optimum calculated fatigue life. This life can then be calibrated against an experimentally measured life. The effects of changes in material or stress can be accurately assessed; but absolute fatigue life can be difficult to calculate accurately.
“We did an experiment once where we asked 12 organisations to use their fatigue analysis software to calculate the fatigue life of a sample component on a helicopter. We then did an experimental measurement to find the correct answer, which was 400 flight hours. Whilst some of the companies were within a few hours, some suggested the fatigue life should be 250 hours and one said it was 12,000.”
However, Draper says he believes the advancement of research over the last couple of years means the accuracy of fatigue analysis software is on a par with anything else in engineering. “We can estimate what an allowable stress is to an accuracy of within 5% now,” he says. “That puts us on a par with anything else in the field.”
be a costly business.”
Fatigue in the future
Leading IT researcher Gartner says it expects software spending to increase 1.5% in 2010. However, the picture for the niche fatigue analysis software market looks far rosier, according to Professor Draper.
“It’s very difficult to put a figure on it but I’d say this market is growing by about 30% each year,” he says.
“I think this kind of growth will continue into the future as more companies and more industries are being forced into reducing costs. The high cost of raw materials means you can’t just throw metal at problems like fatigue anymore.”
Although acknowledging that cost of procuring information to feed into the software may prevent some companies from using it, Professor Irving also predicts a bright future for the market.
“Fatigue analysis software will become more widely adopted as it becomes further integrated with CAD packages,” he says.
“I think there needs to be more transparency and a set of standards introduced, so there is more uniformity on fatigue life predictions. But ultimately, as a service experience, I expect the market will grow and the software will become more accurate as time progresses.”