Life and Reliability

Our life and reliability services provide you with the data you need to maximise the probability of issue-free performance for your product or process. The field, here, is vast, but most problems come down to statistical modelling and/or engineering analysis of one kind or another. We will work with you to understand and document your key needs, and deliver the engineering that satisifies them.

Reliability and Demonstration Testing

How long will your product last, and how long will its individual elements last before needing maintenance or replacement? Our analytical Life/Reliability services can help you design in target performance levels but sometimes you’ll want to consider such questions in a different way. You may wish to build and test something and measure what happens, as either an alternative or complementary approach to analytical considerations.

The language of Reliability and Demonstration Testing is statistical rather than having its basis in continuum mechanics and the behaviours of materials. For example, you might want to design your products such that 90% of them last for a year in their most arduous use case and, to complete the specification for such a statement, you’ll want to assign and understand a confidence interval that relates to this.

Reliability statistics with various Weibull plots

If you are lucky, and neither your product nor its testing are especially costly, maybe you can test 50 of them in exactly this scenario, and derive some level of confidence that your target can be met. In most instances, though, this will not be practicable.

As an example scenario, what if you can afford to build and test two, and you can test them continuously for two years? If both items reach the end of this testing, what do you know and how confident are you in this result? What if one of them fails after nine months and the other lasts just over a year?

More generally, you might be faced with a position where you know your costs for creating test units and you know the costs involved in testing them, but what you would like to know is how to best assign this overall budget to get you the most confidence you can in a particular position within a target timescale.

Any of these scenarios, and anything in between, can be evaluated by us for and with you, as needed.

Structural Finite Element Analysis FEA

A core element of our business is helping customers understand what happens to something in use in its physical environment. What happens when this component sees a load of a certain magnitude in a particular position? How much do things move and how much of the available clearance is taken up? What levels of stress are present in the component in this scenario, and what are the implications of this?

Structural FEA results screen showing stress levels in a loaded buckle

For relatively simple situations, we can either provide you with the output from manual calculations or provide you with template tools to enable you make these same assessments yourself should you wish to explore the effects of design changes. If you are worried about the effects of overload we can combine knowledge of materials with strength and failure theories to translate these numbers into appropriate levels of concern and advise on corrective actions to take with your designs.

A fully-meshed FEA moel of a crankshaft

Once geometries start to get complicated, interactions with other components become important, or other physical factors like temperature and friction come into play, it is often the case that manual methods no longer suffice and a different approach is needed, and this is where our Structural FEA services come in. We will work with you to complete the specification for your desired outcome if needed, and take the whole project from start (generating the CAD information) to finish (most often, a detailed report giving all results, including that all too important implication assessment) or just help with steps along the way.

Fatigue Analysis

It is conventionally stated that around 90% of mechanical failures can be linked to fatigue, or the repeated application of loading cycles at levels well below those that would be considered to be overloading. There are many ways to obtain the data needed to make assessments in this field, and there is a surprisingly large variety of techniques that can be brought to bear upon different problems.

Analytically, the most straightforward use case is to review a fully-reversed loading event series versus a known fatigue strength for an elastically loaded item. What this means is that, for the region in which you are looking, one loading cycle will generate both a maximum (tensile) stress and a minimum (compressive) stress of equal magnitude, this loading cycle is repeated, and the events themselves can be represented by Hooke’s law, such that stress and strain are proportional to each other. For this scenario, a stress is compared with a strength, and an assessment of a safety factor can be made, or a prediction of a lifetime. One important thing to remember, though, is that fatigue is a stochastic process – it cannot be predicted precisely, and you must always factor in a degree of scatter when real world events are compared to analytical predictions.

Beyond this, things get more complicated, as illustrated by our example situations below:-

  • The loading situation is not simply fully-reversed, and you need to make corrections based around the mean stress that is present as well as the stress amplitude.
  • You know that the maximum stress levels you have ascertained from some form of analysis are way higher than the static strength of the material to be used, and you want to estimate the effects of real-world plasticity from your elastic model.
  • Your loading regime is not simply repeated, cycle after after cycle and is, instead, variable.

Whatever the problem is that you are faced with, PDE can support with you with its resolution with both Stress Life and Strain Life approaches to generating fatigue and life analytical data.

Fracture Mechanics

A typical journey through a product develoment team for a critical, loaded component might be investigation of strains and stresses in a static load case, followed by a fatigue investigation and some targeted material testing if there is a requirement to ensure a minimum life to, for example, align engineering considerations with warranty planning for a business.

Whilst the above is often a sensible route that demonstrates due diligence and allows design decisions to be taken based on good data sets, it is not always the complete story. For many components, significantly exceeding an accurately derived fatigue strength or endurance limit will clearly lead to a fairly predictable failure but, for others, they may not fail at all. Fatigue engineering is, in reality, looking to determine the time to formation of a crack, usually from an exposed tensile surface. Should this crack form in some stress-concentrating feature, say, and then propogate into a volume of material where the levels of applicable stress drop off significantly, will there be an eventual catastrophic failure or will the item reach some sort of equilibrium?

LFEM, linear elastic fracture mechanics looking at life after fatigue failure initiation

Often used in engineering areas where field management of inspected items is a critical cost driver, fracture mechanics analyses are often paired with NDT (Non-Destructive Testing) techniques to determine how long something can be left in service before being replaced, or before another inspection is scheduled, but they can also be used in more general terms to understand what happens should crack formation occur on any loaded item.

Please contact us if you have a project that you think could benefit from such an analysis.