NVIDIA RTX is spinning new cycles for designs. Trek Bicycle is utilizing GPUs to deliver design ideas to life.
The Wisconsin-based firm, one of many largest bicycle producers on the earth, goals to create bikes with the highest-quality craftsmanship. With its new accomplice Lidl, a world retailer chain, Trek Bicycle additionally owns a biking group, now known as Lidl-Trek. The group is competing within the annual Tour de France stage race on Trek Bicycle’s flagship lineup, which incorporates the Emonda, Madone and Velocity Idea. Most of the group’s equipment and gear, such because the wheels and street race helmets, have been additionally designed at Trek.
Bicycle design entails complicated physics — and a key problem is balancing aerodynamic effectivity with consolation and journey high quality. To deal with this, the group at Trek is utilizing NVIDIA A100 Tensor Core GPUs to run high-fidelity computational fluid dynamics (CFD) simulations, setting new benchmarks for aerodynamics in a bicycle that’s additionally snug to journey and handles easily.
The designers and engineers are additional enhancing their workflows utilizing NVIDIA RTX expertise in Dell Precision workstations, together with the NVIDIA RTX A5500 GPU, in addition to a Dell Precision 7920 operating twin RTX A6000 GPUs.
Visualizing Bicycle Designs in Actual Time
To kick off the product design course of, the group begins with consumer analysis to generate early design ideas and develop a variety of concepts. Then, they construct prototypes and iterate the design as wanted.
To enhance efficiency, the bikes must really feel a sure manner, whether or not riders are taking it on the street or the path. So Trek spends numerous time with athletes to determine the place to make vital adjustments, together with tweaks to geometry and the flexibleness of the body and taking the sting off of bumps.
The designers use graphics-intensive functions instruments for his or her computer-aided design workflows, together with Adobe Substance 3D, Cinema 4D, KeyShot, Redshift and SOLIDWORKS. For CFD simulations, the Trek Efficiency Analysis group makes use of Simcenter STAR-CCM+ from Siemens Digital Industries Software program to benefit from the GPU processing capabilities.
NVIDIA RTX GPUs supplied Trek with a large leap ahead for design and engineering. The visualization group can simply faucet into RTX expertise to iterate faster and present extra choices in designs. They will additionally use Cinema 4D and Redshift with RTX to supply high-quality renderings and even to visualise completely different designs in close to actual time.
Michael Hammond, the lead for digital visible communications at Trek Bicycle, explains the significance of getting time for iterations. “The quicker we are able to render a picture or animation, the quicker we are able to enhance it,” he stated. “However on the identical time, we don’t need to lose particulars or spend time recreating fashions.”
With the assistance of the RTX A5500, Trek’s digital visible group can push previous inventive limits and attain the ultimate design a lot quicker. “On common, the RTX GPU performs 12x quicker than our community rendering, which is on CPU cores,” stated Hammond. “For a render that takes about two hours to finish on our community, it solely takes round 10-12 minutes on the RTX A5500 — which means I can do 12x the iterations, which results in higher high quality rendering and animation in much less time.”
Accelerating CFD Simulations
Over the previous decade, adoption of CFD has grown as a vital instrument for engineers and gear designers as a result of it permits them to achieve higher insights into the habits of their designs. However CFD is greater than an evaluation instrument — it’s used to make enhancements with out having to resort to time-consuming and costly bodily testing for each design. For this reason Trek has built-in CFD into its product improvement workflows.
The aerodynamics group at Trek depends on Simcenter STAR-CCM+ to optimize the efficiency of every bike. To offer a cushty journey and clean dealing with whereas reaching the perfect aerodynamic efficiency, the Trek engineers designed the newest technology Madone to make use of IsoFlow, a novel function designed to extend rider consolation whereas lowering drag.
The Simcenter STAR-CCM+ simulations profit from the pace of accelerated GPU computing, and it enabled the engineers to chop down simulation runtimes by 85 days, as they may run CFD simulations 4-5x quicker on NVIDIA A100 GPUs in comparison with their 128-core CPU-based HPC server.
The group also can analyze extra complicated physics in CFD to higher perceive how the air is transferring in real-world unsteady circumstances.
“Now that we are able to run increased constancy and extra correct simulations and nonetheless meet deadlines, we’re capable of scale back wind tunnel testing time for vital price financial savings,” stated John Davis, the aerodynamics lead at Trek Bicycle. “Inside the first two months of operating CFD on our GPUs, we have been capable of cancel a deliberate wind tunnel check as a result of elevated confidence we had in simulation outcomes.”
Be taught extra about Trek Bicycle and GPU-accelerated Simcenter STAR-CCM+.
And be part of us at SIGGRAPH, which runs from Aug. 6-10, to see the newest applied sciences shaping the way forward for design and simulation.
