Did you know that the weight of a bicycle becomes relevant only when the grade of the road you’re riding on is 8% or higher? The bike’s aerodynamics come into play on flats, uphills and downhills and, it turns out, have a significantly greater impact on a cyclist’s overall performance. According to Phil White, co-founder of Cervélo, makers of the bikes involved in 10 medals at the last summer Olympics (10!), drag accounts for around 80% of the resistance experienced by a rider on a flat road. Using CFD, said Mr. White, improved the aerodynamics of Cervélo’s bike, giving the rider a 6 to 30 watt advantage on the road. While that doesn’t sound huge, given that Lance Armstrong generates 400 to 500 watts when powering uphill, it can mean the difference between medaling and not.
Mr. White delivered the keynote at CD-adapco’s Global User Conference in Noordwijk an Zee, the Netherlands, to an audience of five hundred CFD experts from 35 countries. His was one of 100 presentations and workshops held over 3 days at the Grand Hotel Huis ter Duin. The sun was shining, the sea looked infinite, but we were committed: CFD, indoors. [I did duck out for a half hour walk to blow away the jetlag cobwebs and I’m definitely going back there on vacation some day. Amazing place.]
Before the start of the formal conference, I had the opportunity to sit in on hands-on sessions where I got to play with STAR-CCM+ and try out the highlights of the V7.02 release. CD-adapco says that its aim with this release was to further shorten the product development timeline while delivering greater insight into the products its customers make. To that end, V7.02 includes the first release of overset meshes, indirect mapped interfaces and solution history. Very briefly, overset (aka overlapping) meshes are individual meshes around each moving object in a simulation which then can be moved at will over a background mesh. This is used to examine the interactions between multiple moving objects, such as cars or bikes overtaking one another. Indirect mapped interface enables use of non-conformal meshes between solid and fluid domains, for conjugate heat transfer simulation where temperatures are solved for both fluid and solid regions — such as gases heating in an engine manifold. Finally, solution history captures and displays the complete history of a simulation so that analysts can compare and visualize results from their analyses. (For more details, go to http://www.cd-adapco.com/news/2012/03-08_starccmv7.)
For many, including me, STAR-CCM+ is an intimidating product. From the very beginning, CD-adapco wanted to create an integrated environment for engineering simulation, which means that all of the components needed to create, pre-process, solve and post-process simulations are in a single user interface. CD-adapco’s software engineers have clearly listened to their users, grouping all of the elements in ways that are logical once you’ve learned how STAR-CCM+ is meant to work, if not intuitive at first sight. With ZERO experience (but excellent instruction) I was able to define overset meshes (although my overtaking Formula One cars never did look as pretty as the instructor’s), mesh an engine manifold and the gases within in my first-ever conjugate heat transfer and then use those results to craft a pretty spiffy AVI using solution history. I am by no means a CFD expert, but was able to accomplish quite a lot in a few hours. As always, though, the devil is in the details.
The discussion in the classes highlighted one of the fascinating aspects of CFD, indeed all advanced CAE, and something that permeated the conference: there is no one perfect approach. All CAE is the application of math, and our best understanding of the underlying physics, to a particular problem. When we were working on our overset meshes, for example, 35 people in class had maybe 15 different possible approaches to the problem – everything from whether the road moves under the cars or the cars move over the road to how fine a mesh is really needed. The “correct” answer depends on what, exactly, the analyst is trying to discover, what can and cannot be assumed or simplified, and how long she is willing to wait for the answer.
In between lessons and sessions throughout the conference, I had the chance to speak with many CD-adapco users. Not surprisingly, these folks are true fans, sold on CD-adapco’s innovation, level of service and support, and future direction. If anything, they want the company to move faster, add more types of physics and mathematical models to their solutions.
During his keynote, company co-founder and CEO Steve MacDonald told the audience that the company is reinvesting “significantly” to grow its R&D capability,saying that “We’re adding a whole lot of physics” – CFD, electromagnetic, casting, DEM, and “stress as we move forward”. He also spoke of the company’s global expansion, putting support teams close to customers in Japan, India and Brazil.
Senior VP of Product Management, Jean-Claude Ercolanelli elaborated on these themes during his presentation, giving a more detailed product roadmap and also introducing the company’s new iPad app, the first in what will be a series of mobile products that will enable users to check progress on a submitted job, Power-on-Demand usage and remaining credits, and access CD-adapco news. (As of today, the app isn’t in Apple’s store yet; download at www.cd-adapco.com/iPad from within the iPad internet browser) I played with the app a bit using M. Ercolanelli’s account (so that there would be usage to monitor) and I think this could be a handy way to step away from the office and still stay in touch. Will we ever be able to start a simulation from an iPad? With a vastly simplified UI, perhaps.
M. Ercolanelli concentrated his team’s presentation on releases planned for 2012 and 2013, saying that the 2012 releases will focus on user experience, geometry creation, meshes, physics, solvers, and performance. At the end of his session, M. Ercolanelli took questions from the audience and I was once again impressed by CD-adapco’s relationship with its customers. In a discussion that ranged from big-picture to very specific enhancement requests, the entire CD-adapco team was open in discussing its plans and amenable to considering directional and priority shifts as it listened to customer requests.
After the keynotes, it was off to parallel tracks. I sat in on many fascinating presentations and I will blog about some of them in a Part 2 about CD-adapco’s Global User Conference sometime over the next week or so. But in my first 24 hours in the Netherlands, I modeled a conjugate heat transfer problem in a car manifold, learned about reducing drag on a racing bike and saw again the very strong ties between CD-adapco and its customers. Not bad.
Note: CD-adapco graciously covered expenses and registration for the event but did not in any way influence the content of this post.