With an aging stock of manufacturing equipment, how can we improve reliability?

James Westley, Senior Engineer within our Mechanical and Analytical group, discusses how low cost sensors and clever algorithms could provide an answer.



There are a lot of old machines in the world today, in fact we have an aging population of machines and infrastructure, the average age of manufacturing facilities and equipment in America is 20 years, Japan is 15 years. This partly due to the economic boom in the 80’s and 90’s followed by the bust and stagnation of recent times. There is nothing wrong with these machines but as growth picks up and new competitors emerge with more reliable and productive equipment, how can these factories keep up?

We need to teach these old dogs some new tricks!  Thanks to major advances in sensor, algorithm and wireless technology over the last 2 decades, there is now an opportunity to cost-effectively retrofit older machines with condition and performance monitoring systems.

So what is condition and performance monitoring and how can it improve reliability? Have you ever been driving your car and noticed a funny smell, noise or vibration and thought, “That’s not normal, there is something wrong, I should probably take this into the garage!”  Condition and performance monitoring does exactly that, but using a combination of sensors, algorithms and data analytics that are far more sensitive than our senses to pick up the onset of degradation well in advance of failure or performance drop. This enables predictive rather than reactive maintenance; replacement parts can be ordered whilst the machine is still running and re-fit can be planned to minimise disruption, saving hours or days of downtime.

The benefits of this technology have already been demonstrated across different sectors. The fitting of an in-line oil condition monitoring system at the Edison Big Creek hydro-generating facility in California has eliminated the need for bearing re-works and has saved hundreds of man-hours each year. Vibration monitoring on six wind turbines at Slitevind AB in Sweden has resulted in a saving of 34,000EUR/annum in lost production time and repair costs. Cambridge Consult’s DropTag can be stuck to the side of an object to measure pressure, temperature, vibration and shock throughout the duration of a manufacturing or logistics process.

So why haven’t we always done this? The retro-fitting of old machines is now a cost-effective option due to the drop in price of sensors and wireless connectivity platforms, which has been dramatic over the last 20 years, partly thanks to the popularity of smart phones, the initial investment is now a fraction of the capital cost of the host machine. It should be understood that it is not a one-size-fits-all affair; not every machine will benefit from this approach and we can’t just slap a vibration sensor onto the side of anything and expect meaningful data. There is some time and cost associated with carefully designing, installing and calibrating these systems on a case-by-case basis. Even so, the investment is offset by the improved reliability and performance for critical machinery.

This is a really exciting area, the uptake of condition monitoring is accelerating and there is a lot of potential to improve productivity in many sectors without the need to investment in expensive new equipment.

James Westley

James is a Senior Engineer within our Mechanical and Analytical group with 5 years of experience in the energy sector. He is a composite specialist interested in projects which promote sustainability.