On a recent trip to the local play park with my children, I came across a serious and catastrophic fatigue failure on one of the play structures.
A closer look at the installation revealed that the design and installation had not been integrated such that the installer had clamped the spring exactly where the highest bending moments would occur.
This demonstrates a lack of understand on the part of the installer and maybe an oversight on the manufacturer's behalf in providing accurate and detailed installation instructions or worse still, a lack of fatigue engineering understanding by the design team. Furthermore in this case, whilst the design is meant for children and the mass of these users would be less than say 20 kg, it is also possible for an element of misuse to occur by heavier older users. Therefore an element of over design is required. However, by detailed failure investigations, it would be possible to prove or otherwise not that the design was repetitively misused.
The material condition may be a contributing cause of the failure as well but this would require a detailed study of the crack surfaces (fractography) and a check on the material composition (metallography). Also there may have been a crack initiator on the surface which again requires closer study under a microscope (optical or electron microscopy).
This short overview shows that expert fatigue investigations can definitely add value and could lead to a clear understanding of whether the design, the materials, the manufacturing process, the installation or the use of the product was the prime reason for failure and what other contributing factors played a role in it.
The Fatigue Expert team can help with your fatigue design and failure investigations as well as fatigue test programmes and material performance qualifications.
A quick review of the job sites shows that many structural and stress analyst jobs also ask for fatigue experience. Clearly there is a link between these two engineering disciplines since all fatigue analyses require stress data, but that is where this link stops. The skill set for a stress analyst is very different from that required for a fatigue engineer as discussed below...
... employers would do well to heed the following!
The Stress Analyst needs to understand how structures carry loads, how to take into account boundary conditions in stress analysis that reflect engineering reality as close as possible, and how to use FEA effectively to obtain reasonable stresses that are not influenced by the way the FE model has been constructed or constrained. Beyond these basic stress analysts skills there are further skills such as calculating the effect of a wide range of loads, contacts, non-linear effects etc.
The Fatigue Engineer requires deep underlying knowledge of loads, the influence of geometry on fatigue, and materials, with a special focus on design methods to prevent or minimise fatigue. On top of this, the fatigue specialist has expertise to apply fatigue, crack growth and damage tolerance modelling methods to determine the fatigue life expectancy of a structure. His/Her materials knowledge includes a good understanding of how fatigue cracks develop in different materials and how to determine suitable fatigue properties for use in fatigue modelling. Fatigue engineers also take on a detective role when dealing with fatigue failures, using all their skills to determine the cause of fatigue failures from a wide range of potential reasons and provide guidance on how to avoid them in future.
Why use a Fatigue Consultant? As can be seen, these two roles are quite different. It is no mistake that the sophisticated product development world used by the aerospace industry separates these job functions and employs highly trained specialists for each separate role (see below for one reason why!). In smaller product development teams, the budget may not support two jobs: stress analyst and fatigue specialist. This is where the use of a fatigue consultant is the most effective method of getting access to the appropriate skills without adding the overhead of a permanent employee.
Using unskilled staff to do fatigue work can cost you much more than a delayed project! No, I am not scare mongering but fatigue failures can destroy a business - the Comet story of a UK business leading the world in jet powered aircraft became the story of a series of planes dropping from the sky, due to catastrophic fatigue failures!
DeHavilland, and hence the UK, lost their hard won lead in modern passenger jet planes to Boeing (and more recently Airbus Industries).
Fatigue engineering kicked off over 150 years ago in the axle world and here we are in the 21st Century and still fatigue of railway axles is yet again under discussion ... This time the activity is in the East, in China, where a modern industrial revolution is going on every day.
As one wise person said "What has been will be again, what has been done will be done again; there is nothing new under the sun."
(Ecclesiastes 1v9, NIV Bible)
The lower Image shows a typical axle fatigue surface
It seems that fatigue is still not understood by the general engineering and management world. Even a local expert, Dr Zhao Jiaxi, is trying to influence the thinking, suggesting "that the debate could be settled once and for all by cutting the suspect axles apart and examining the structure inside". (Ref 1.)
The case is politically driven too where measurements made have been disputed. Clearly the issue is raising a lot of arguments. What is clear, as pointed out by Dr Zhao and well known amongst fatigue engineers, is that a materials investigation of the actual material using sectioning of the suspect parts and a metallurgical examination would settle the discussion.
There is a story about a steering wheel manufacturer who was receiving a number of warranty claims from the police force for steering wheels that had failed in fatigue. As they investigated the possible reason for these failures, nothing seemed to point to the cause; the material was within specification and even with very high stress concentration factors, the calculations still showed that fatigue failure was highly unlikely. The engineers were puzzled as the loading on a steering wheel in their view was obvious and well understood. Why were their steering wheels failing in the hands of the police?
Getting closer ....
One engineer suggested that they should accompany the police to see what they were doing to the steering wheels. After a few weeks of doing this, nothing came to light. Then one afternoon, the engineer was again in a police car which had parked up at the side of the road. As he was talking to the police officer they got onto the subject of what they did when things were quiet. “Oh he said, if it is really quiet, we wind the seat back down and have a snooze - the radio wakes us if there is some need for action”. “Show me how you do this” asked the engineer and the officer obliged. However, to pull himself upright again, he held onto the steering wheel and gave a good yank. There it was, the unexpected load case that had been overlooked.
Using this knowledge
You could argue that this loading on the steering wheel was outside the design specification or beyond normal use but the key point is that it was unexpected. Often fatigue failures occur due to unpredicted loading, manufacturing process effects, materials specification variation, or even poor maintenance practices. To understand how a fatigue failure occurrs, software tools can be used to model the fatigue scenario and explore likely causes. In so doing, some scenarios can be discounted as unreasonable but nothing should be discounted too quickly. For example, stress concentration factors of 15 might seen exorbitant but there could be some machining marks that left very sharp surface crack like features in a radius.
Getting to grips with Fatigue!
Don't delay - why not speak to fatigue-expert about working with your engineers to train them in the use of fatigue modeling as an investigative tool.
I never expected to see a crack in a pen casing - they are normally chewed! But that is exactly what happened to my pen. There are some small cracks extending from the grip area up the cylinder. The material is an unidentifiable plastic but probably polystyrene.
There seems to have been little attention paid to the possibility of fatigue cracks initiating but clearly there is a bending load on the pen cylinder and the joint above the grip section is critical! The surface and finish of the end of the middle section tube is not perfect so there are crack initiation points.
I have a now acquired a new pen and noticed the fit of the grip section into the middle tube is very tight, so there might also be some rather high stresses in the tube. Of course the assembly requires a push fit to stop the pen falling apart. However, balancing this with the bending stresses is a critical design requirement. I also think a steel or aluminium tube would work better but that is more expensive and these pens are sold for less than £1!
When steel (or any metal) plate is passed through a forming or pressing process, the microstructure is changed and this means the fatigue behaviour also changes. Therefore using fatigue curves developed from native un-formed plates or bars in fatigue life calculations is not suitable and there is absolutely no reason why these data should correlate with the fatigue performance of the formed component.
This is why manufacturers often just test components as they are more likely to represent the real materials conditons where cracks initiate. Then all they need to worry about is getting the loads applied to the component to correctly represent the real world loading. However, this is slow and expensive process if used to optimise the design of components.
The problem comes when trying to use calculations in life prediction tools that require the materials data to properly represent the conditions in the component. The real conditions should be replicated in the fatigue test materials to develop fatigue curves that represent the actual fatigue performance of the component in the specific location where cracks initiate.
Fatigue-Expert has an excellent track record in helping its clients to obtain the correct materials data by designing a process to obtain these, executing that process, and applying the new data to the design process seamlessly. If you are struggling to correlate fatigue life estimates with tests, please give us a call - we can help!
159 Beacon Hill Road
Phone: +44 (0) 7980 694808