To Fail or not to Fail… That is the Question!

To Fail or not to Fail… That is the Question!

November 23rd, 2016

When a product breaks, typically after a warranty expires, then a common utterance is “well I bet it was designed that way” and in all truth it most probably was.

We’re all too aware that the products we buy won’t last forever and will eventually fade into a sad obsolescence, but what we may not always consider is that these products deliver useful lifetimes precisely because they have been engineered with the end user’s requirements in mind. The design challenge is not to always to create the best or the cheapest product, but to demonstrate enough value to meet customer expectations.

firth-road-bridge
The Forth Road Bridge opened in 1964. It had an expected usage of about 30,000 cars per day in each direction over a period of 120 years. The number of load cycles per annum was exceeded, with other unforeseen occurrences reducing the effective strength of the bridge and it eventually suffered closures in early 2016. Free stock image courtesy of photoeverywhere.co.uk.

The illusion of choice.

If we are truly honest with ourselves then we might acknowledge that we don’t really know very much about the products we buy, so our expectations as consumers can be pretty fickle and rely on a multitude of sources. I freely admit that I can’t really tell the difference between washing machine performance grades, as much as I believed I could when I bought one. So I now just revel in the little song it plays whenever it finishes a spin cycle, coddling my belief that I definitely made the right purchase – and it hasn’t failed yet so it must have been right and anyone who didn’t buy one like mine is a fool – right?

For the typical consumer it becomes like the game of buying the second cheapest bottle of wine at a posh restaurant. You don’t want to spend a fortune because you aren’t a spendthrift, but you don’t want the cheapest bottle either as it makes you look a penny pincher. So the house wine is popular because of where the product sits on the menu rather than due to anyone’s knowledge of its contents.

The restaurant probably also got a good deal on it from their supplier.

But if the sommelier were to tell you that it’s cheaper because 1 in every 20 bottles of the house-red tasted sour, enough to disturb your meal but not quite ruin it, would you pay more for a ‘better’ bottle or would you take the risk for a cheaper meal 95% of the time? Being able to present the same information about our own products can be a very powerful message.

Reputation is easier kept than recovered.

Many designers and design engineers aim to predict and to mitigate failures, be they design, logistical, or process related. But given a long enough timeline, or a large enough sample size, then dominant failure modes will always prevail somewhere at some time. Understanding what these failure modes are and how they affect performance can help us focus our efforts on the areas that have the greatest impact.. By limiting our attention to a manageable number of failure modes, we can minimise their occurrence to a level that aligns with the customers’ expectations; improving our customer’s perception of our products and maintaining our reputation.

So it is very much in a manufacturers best interests to know all they can about their products to achieve balance between design reliability and design conservatism. If items fail earlier than expected or a major design flaw is found after a general release, then the cost of putting it right is significant. You might have turned off your Galaxy Note 7 by now if you’ve bought one.

A bad workman blames his tools.

To help engineers predict these events and design against them, there are many qualitative and quantitative methods that have been developed to understand and replicate real world responses to particular conditions. The skill is in correctly understanding and interpreting the real-world conditions to ensure that these are correctly replicated in the analysis.

Bob Petch, Senior Engineer and Christopher Smith, Senior Engineer

To discuss any of the points raised in this blog or for more information on the industry leading software tools we use at Wilde Analysis for fatigue analysis, please contact us or leave a comment below.

Leave a Reply

Your email address will not be published. Required fields are marked *