Part One

Several years ago our core product started to have problems with the bearings. Fine dust was penetrating the seal to the prepacked bearings and within a short time the bearings failed. As an issue it was serious as the failures were quite random and if the failures occurred during high winds and went unnoticed, the bearings would seize and the top of the unit would shear off and fall to the ground. So far nobody has been decapitated but there have been some pretty scared cats.

Obviously this situation could not last and the Australian manufacturer was not showing a lot of interest in solving the problem despite threats of law suits or even worse. I decide to try and find a solution to the problem. The worst failures were a bottom bearing that was exposed to fine dust, although the top bearing was enclosed, the dust in this region is so fine also it was making its way into the bearing. At least if the bottom bearing could be sorted out, the top bearing would not be so much of a problem.

With this in mind I started fabricating different models without a bottom bearing. The results were not very encouraging but I persisted. Previously before the bearing problem I had met up with a scientist from Melbourne who did some work for us on imaging thermal flow through a turbine hall. His expertise lay in CFD Modeling and he had used his skill at this to obtain a Doctorate from the University of Adelaide. Each time I completed a prototype I sent it to him for analyses but the results were not anything to get fires up over.

Then, out of the blue he emailed me and asked if I would be interested in a model that would do the same job but without bearings. On the plane immediately, I went to Melbourne to discuss the idea in detail. In its preliminary stage it looked good, that was 18 months ago. We agreed that the scientist would become my partner in a research and development company to develop the intellectual property and exploit it. I would put up the cash and he would put up the IP.

Like all good ideas, the devil is in the detail and what looked good on paper was going to take some time to work out and test. First step was to form the company. After some checking around we found that Adelaide would be more conducive to this small enterprise so we shifted our operations to there. Besides the University of Adelaide was my partner’s alma mater and they were willing to be more helpful than the establishments we contacted in Melbourne.

Creating a company was a far more time consuming process than I envisaged and it required yet another trip to Adelaide. These excursions to Adelaide were to become a regular habit. I became weary of the plane flights although Adelaide turned out to be a far more agreeable than I anticipated, particularly during March when they have their Fringe Festival that dominates the city’s life for a month.

After having legally established our identity the next step was to file a claim for intellectual property protection on the design of our invention. This too was a slightly longer but necessary process than predicted. We needed to get physical prototypes made and up until the IP was protected we had to obtain a non-disclosure agreement with each party before we could receive quotations, this was a burdensome task.

Now we were in a position to commence CFD Simulation and wind tunnel testing. The University of Adelaide had just installed a new high tech wind tunnel and we were able to negotiate a favorable fee for the extensive testing that was needed. In addition, we were able to secure a grant from the South Australian Department of Manufacturing, Innovation Trade Resources and Energy to assist in developing our invention to the manufacturing stage. Again this took longer than expected.

The test procedure was complicated. As the wind tunnel, despite its size could only accept test pieces of a certain dimension to avoid inaccuracies due to the “blockage factor”, we had to scale down our device. Moreover we needed a bench mark so we had to use a traditional model of a known capacity to measure the improvement in our design. After the wind tunnel was calibrated, the traditional appliance was tooled up with instruments, its performance measured over a range of parameters then the results were compared with a CFD model that had been created. By comparing the values, the CFD model could be calibrated to the physical results.

We could now compare the prototype tested in the wind tunnel to the CFD model and use these results as the means to determine the improved performance of our apparatus. The whole procedure, as expected took much longer than expected and was not completed until some three months later and also after some anguish. However, in the end we had a unit that was a 25% improvement on the old product, at least on paper. And just as important it was static and we had eliminated the bearings. Time to open the champagne.

But a bit too early!

Read Part Two