The University of Sheffield Advanced Manufacturing Research Centre (AMRC) has spent the last 20 years transforming the UK’s manufacturing industry. They work to de-risk innovation, accelerate the commercialisation of R&D, and drive economic growth across the country.
This is just one example of their enterprising innovations…
The AMRC wanted to build motor housing (of different sizes), with cooling ducts (channels) around it. Their innovative solution to a well-known problem within the manufacturing industry had one catch though – the whole thing had to be 3D printed in one go. All fine, until they reached their desired cooling channel radius; the point at which the 3D printed product would collapse.
The AMRC team was trying to convert a mathematically perfect circle into a different shape using CAD. Unfortunately, that didn’t translate well within the package.
In abstract: clever geometry can open up a world of opportunities.
The product had to reliably support its own weight during a print, but the AMRC's existing tools didn't allow for them to 'sweep' the geometry. This meant that every time they wanted to manufacture a new product, the engineers would have to make all changes by hand – a very time consuming and inefficient solution, as you can imagine.
In reality: the engineers at the AMRC were trying to solve multiple problems all at once – like generating optimal wall thickness to channels, drawing arbitrary paths for the channels to follow and optimising the area of channel to maximise cooling – using a mathematically perfect shape.
This is where our Get Stuff Done engineering ethos really came into play as we knew they just needed an approximation of a circle.
We started the project with a Discovery to discuss the problem and possible solutions with the AMRC. We were able to break the problem down into smaller steps, by tackling the main, high-value problem first – auto generating optimised and perfect cooling channels.
With our proposed solution in mind, and their budget and time restraints accounted for, we dove straight into a two-week Innovation Sprint, on a quest to answer one fundamental question; is this doable?
A high-speed dive into the art of possible.
(Big up Fraser, whose experience and love of working with non-perfect shapes came in very handy here! Not to mention the rest of the team, who leveraged their own specialist skills to ever-so-speedily implement the mathematical hypothesis into a robust technical solution for the AMRC. You guys totally rock.)
Using OpenCascade – a powerful industry-standard library for working with complex geometry – and C++, we created a command line utility in which the AMRC’s engineers could simply input a few parameters to generate their desired specs. This parametrically generated geometry can then be converted into a CAD reader-friendly file format, ready for integration in the CAD model.
The next step is to build out the solution into an MVP and productionise it, with a view to enabling ‘further optimisation’ and ‘extra features’ later down the line.
Feeling inspired by this solution? The AMRC are actively seeking enthusiastic engineers to collaborate on refining and implementing critical features. They want to understand your specific requirements to develop a versatile tool that meets industry needs. Share your thoughts by completing the form here.
It’s great to work with a company so aligned with our own mission – to use technology to promote huge business transformations. We’re really happy with the first stage of this project – we now know what is possible – and once we’ve secured the next bit of funding we’re excited to get this idea off the ground and into production.
The AMRC is creating something new, something transformational, something intrinsically valuable. Even so, the most innovative of products could often do with an extra dose of innovation.