TotalSim have been involved at the highest level of aerodynamic development in cycling since we were founded, you could say it is in our DNA.
When we were first asked to apply our motorsport experience to cycling, finding gains was simple. Improvements to the bike and helmet are easy to come by and everyone was happy. After the first pass, gains become harder to find and perhaps equally important, harder to measure.
Our initial process involved carrying out development in CFD, with verification with an athlete in a wind tunnel. When the gains were large this is straightforward and you can easily verify forward steps. As the gains reduce, you need to start gluing 5-10 pieces together to be detected and verified in the tunnel.
One of the problems with wind tunnel verification is the rider. The rider makes up a large percentage of the total drag and being human they have a tendency to wobble and move. These relatively small movements can be more significant than the small steps you are trying to measure, which becomes an issue.
The next difficulty comes with the desire to ensure robust performance. Maximising performance straight ahead is relatively easy, but unfortunately that is not reality. It might be a reasonable approximation on the track, but as soon as you move outside, the effect of cross wind becomes significant. Unlike Formula 1, a 2 mph cross wind has a significant effect upon the relative wind direction. As a result any development work must be done through a yaw polar and statistically averaged, to ensure robust performance gains. This is not a significant problem, but it does add to amount of work involved to ensure robust performance.
The flow around a cyclist is chaotic. As much as we want to, we cannot make a cyclist ‘aerodynamic’. The rider is irregular and bluff creating a wake and the separation behind the rider is highly transient. To ensure that the unsteady effects are captured, our experimental and computational methods have to be adapted. The wind tunnel sessions have to have multiple repeats and long sampling times to spot the trends.
Computationally we had to develop a transient methodology. The movie shown above, of a GB athlete, was made back in 2012 and demonstrated our transient CFD methodology, utilising DES to capture the unsteady flow structures around an athlete. Again, simulation time had to be extended and averaged over long time periods.
The next headache relates to the Reynolds number and a fluid phenomena known as transition. As velocity increases around a given object the boundary layer transitions between laminar and turbulent regimes. This can have a significant effect upon the separation point around an object and hence the drag.
To maximise performance for a rider and frame, it is essential to understand this phenomena. Unfortunately this is not straightforward, computationally or experimentally.
Computationally, TotalSim has developed a methodology utilising a solver capable of modelling transition, which has become a powerful tool in aiding understanding. Understanding and controlling transition has allowed us to make significant performance gains.
Finally going back to the athlete, in addition to minor motions and wobbles mentioned earlier, they also have to pedal! This introduces a huge amount of noise and complexity. Understanding the effects of leg and body motion on frame and leg separation is very difficult in a wind tunnel, due to the timescales involves.
TotalSim’s latest developments will hopefully allow us to make significant steps forwards in this area. With a transient CFD solver, with a pedalling athlete and the ability to pause, slow down and rewind computationally, we are confident we have an unrivalled toolset to push the boundaries and give our clients the edge they desire.
We cant wait to get started on a package for 2020.
Are you interested in our toolset and expertise for 2020? If so we want to hear from you and please get in contact.
- RT @SilvTechCluster: Exciting times ahead as STC member @TotalSimLtd opens its Sports Engineering Hub for world class athletes, housing 2 w… about 6 hours ago from Twitter Web App
- RT @SilverstonePark: 'Olympic' standard!The opening of Silverstone Park's @SSEHub (sports engineering hub) was broadcast to c.800,000 peo… about 1 day ago from Twitter Web App
- We are currently looking for an enthusiastic developer to join our web-based CFD platform team. Interested? Please… https://t.co/VfJVXZHxoW about 1 day ago from Loomly
- Learn how @Learn_HyperMesh can accelerate your CFD pre-processing time by joining the @Altair_Inc webinar, featurin… https://t.co/kClglICp9e about 2 days ago from Loomly