In 2006 I began part-time study at the University of Bolton for their MSc in Electronic Product Development. The first module that I took was titled “Design For Thermal Issues” and, as the name suggests, was all about the thermal aspects of electronic design.
As part of the coursework, students were given the task of thermally optimising a 15U-high telecoms cabinet – the sort of thing you might see in the server room of a small company or institution. We were asked to look at several aspects of the design and determine how varying those aspects would affect the overall thermal behaviour of the system. Of course, it was not feasible to do this with an actual, real-life telecoms cabinet and so the main part of the investigation was carried out using the leading thermal simulation software package, FloTHERM.
By 2006 I was already a fairly experienced FloTHERM user, having been responsible for the introduction of the software at my employer’s site in Manchester. I had built up a good working relationship with the folks at Flomerics (who owned FloTHERM, later taken over by Mentor Graphics), and I had the good fortune to be loaned a fully working version of FloTHERM for use at home for the duration of the project. The alternative to this would have been to use the software remotely on the University servers via Citrix – not ideal for someone who lives in an area where broadband coverage was (and still is) rather flaky. I was, and remain, very grateful to my colleagues at Flomerics/Mentor for this extremely generous gesture.
Having the luxury of the software running at home, I was able to dive very deeply into the optimisation of the cabinet design, often leaving the more complex simulations to run overnight. The result was a report that runs to just over 14,000 words. Due to my natural enthusiasm for the subject, this never felt like hard work, and I’m sure I could have found more to write about given sufficient time! The length of the report, plus the necessity of including numerous tables, graphics, graphs and so on makes it near impossible to reproduce directly on this page. I have therefore converted the original Word document to PDF format, which may be downloaded via the link below.
If you are researching a similar subject yourself, and have come to this page via a Google search, then I hope you find my work useful in your own endeavours. As is so often the case with thermal analysis, some of the results were surprising, counter-intuitive but ultimately educational. My particular favourite being the conclusion that making bigger holes in the sides of the rack unit does not necessarily result in better ventilation! That seems wrong at first glance, but does actually make sense if you think about it hard enough.
I welcome any contributions you may like to make in the Comments section, and now here is the link. Enjoy!