Taking up the strain

Written by: Tom Shelley | Published:

A better knowledge of strain measurement techniques could help designers to improve their products. Lou Reade reports on Eureka’s round-table event

Every time a plane falls out of the sky, or any kind of building collapses, you can guarantee that an engineer somewhere will get it in the neck.
Because it’s likely that the ultimate cause is some kind of structural failure – a result of the hidden stresses and strains contained within the material.
Richard Burguete, experimental mechanics specialist at Airbus UK, believes that a better knowledge of strain measurement could help engineers to improve their products.
“I think about half of all products out there have not been through any kind of strain measurement,” he says. “But every industry could benefit from using it.”
Eureka recently hosted a round-table discussion on the subject of strain measurement to debate some of its finer points: what are the key technologies; where does computer simulation fit in; and how might the techniques be integrated into existing design processes?
The event was organised in conjunction with the British Society of Strain Measurement (BSSM), which represents users and suppliers of the technologies – as well as advancing the understanding of strain measurement, stress analysis and related areas – such as through training, and by setting standards.
There are three main ways to determine strain data within materials – which can then be used to optimise design: using strain gauges; with ‘optical’ strain techniques; or using computer simulation. Each has its own advantages.
Strain gauges are the workhorses of the field: they produce reliable results – as long as they are correctly applied – and have the added benefit of being inexpensive. They have also been around for 70 years, so are sometimes seen as less trustworthy than newer optical or simulation techniques.
“Many people don’t believe that strain gauges will give you a reliable result – I’ve heard that from several sources,” says Anton Chittey, senior technical support manager at strain gauge supplier Vishay. “There’s a perception that they’re not reliable, but there’s no reason why a strain gauge should not last for 20 or 30 years.”
Ian Ramage, managing director of Techni Measure – which supplies similar devices – points out that because they need some skill to fit, they can produce false results.
“Lots of people dabble with strain gauges,” he says. “Lot of catalogues sell them – and people just buy them, stick them on – and don’t get the right result. They need to be taught how to use them properly.”
For Janice Barton – BSSM chairman, and professor of experimental mechanical at Southampton University – a strain gauge is a solid, dependable instrument.
“If you’re not sure of something, put a strain gauge on it,” she says. “We know that readings from them are 99% reliable – as long as you’ve applied them properly. There’ll always be a place for them.”

Optical measurement
While strain gauges produce accurate and detailed point data, the newer optical techniques produce a strain ‘map’ – showing the strain variation within a component.
Rob Wood, applications specialist at Dantec Dynamics, explained: “There are three main differences between strain gauges and optical techniques like ours: optical techniques produce full field measurement, instead of spot or point measurement; they’re non-contact, which can be a benefit in certain applications; and they’re 3D – strain gauges generally only perform in 2D.”
David Hollis, applications consultant at Lavision, added: “Our main technology is digital image correlation (DIC). You apply a random speckle pattern to a surface and image it with a CCD camera, then split the image into smaller subsets and track the motion of the pattern within those subsets. By looking at the difference in displacement between the subsets you can get the strain out.”
There is a long list of optical techniques – each of them slightly different – that can be used to measure strain. These include: photoelasticity; digital image correlation; speckle pattern interferometry; and thermoelastic stress analysis (TSA).
And then there is computer simulation – be it finite element analysis (FEA), computational fluid dynamics (CFD) or multi-physics methods.
When techniques like FEA became generally available, one of the first things it did was destroy sales of strain gauges. But where once there was competition, now there is harmony – as suppliers and users agree that simulation and physical testing should complement one another.
“Most stuff can be solved in FEA with a degree of reliability,” says Tim Morris of Nafems – which represents simulation companies. “But for more complex algorithms you need a lot of validation. Even where it’s proved there’s still lots of uncertainty. There should always be some kind of validation through testing.”
Kevin Potter of Bristol University – and a director of its optical strain spin-out company, Imetrum – says: “FEA is an incredible tool but it needs to be treated with skill and understanding. You must model the material properties – and get them right – before you can model the structure.”
And that means using the two types of technique side by side.
“Integrating strain measurement with simulation techniques could deliver greater benefits than people are aware of,” said Burguete. “We need to help people to access these technologies much more easily.”
Historically, there has been a split between the two camps – with engineers tending to specialise in one or the other. But this, says Tim Morris, needs to change.
“When FEA came out there was talk about it replacing testing, and a lot of people felt threatened,” he said. “The right way to view it is in conjunction with testing. Validation is the key. We want simulation and test to be used together in an effective way, but very few people have enough skills on both the test and FEA sides. Ideally, you need a combination of those skills.”
Kevin Potter agreed, saying: “I think all these techniques are complementary.”
Barton pointed out that strain measurement needs to be well understood in order to produce good results.
“You need people who understand stress analysis,” she said. “If you don’t understand the concept of stresses and strains – in a deep way, not just a ‘force over area’ kind of way – then you will get results out that look good but aren’t. It’s what I call feasible rubbish.”
At the end of the day, many designers are producing parts without recourse to these techniques – some of them expensive, all of them needing some degree of specialist knowledge. So why should they consider taking them up?
“Keeping the lawyers off your back is the first thing. If you haven’t got a proper process that’s certified and nailed down all way along you’re going to get sued,” said Potter. “There’s also an increasing understanding that materials are complex. The idea that you can use two numbers to characterise steel is long gone – you need more.”
For Barton, it all comes down to confidence.
“It depends how confident you are with your simulation,” she said. “If you’re 100% confident that your simulation will give you the right answers, you don’t need testing. But that’s very unwise.”

* Eureka intends to run a series of round-table discussions. If you have an idea for a subject, contact Lou Reade on lreade@findlay.co.uk

Round-table participants
Clive Bancroft, EPT
Janice Barton, BSSM/Southampton University
Richard Burguete, Airbus UK
Anton Chittey, Vishay
David Hollis, Lavision
Tim Morris, Nafems
Kevin Potter, Imetrum
Ian Ramage, Techni Measure
Rob Wood, Dantec Dynamics
and Lou Reade, Eureka

The integration issue
If you do decide that you want to use these techniques and embed them into your business, it’s not always that simple. Because some of the techniques are costly – and their payback is hard to quantify – it often takes a leap of faith according to Clive Bancroft of EPT, which specialises in integrating systems into companies.
“Integration is not technology as far as we’re concerned,” he says. “The biggest issue is people. I’ve never come across anybody who’s responsible for integration. The problem is that you can’t put a business case together for integration because you’re dealing with very complex processes – such as aerospace components.”
He says it is a delicate balance between the engineers – who are technically aware, and can see the benefit of such techniques – and senior managers, who sign the cheques and need to see a business case for such an investment.
“You need to get into the right level of the business,” he says. “There’s a disconnect between the techies and business people. I’m trying to join up the language and find that common ground.”

This material is protected by MA Business copyright see Terms and Conditions.
One-off usage is permitted but bulk copying is not. For multiple copies contact the sales team.


Please view our Terms and Conditions before leaving a comment.

Change the CAPTCHA codeSpeak the CAPTCHA code