This week’s CFD news may take you some time to get through because it includes several worthwhile, long reads. One is about simulation driven design and democratization, three are about computing including HPC, GPU, and AWS, and another fluid mystery has been solved. Shown here is a teaser of Image of the Week, a CFD solution of an air taxi done by NASA.
- Join us next Tuesday, June 23rd, at 10:00 a.m. (central) for a webinar with Dr. Amine Ben Haj Ali from Bombardier during which we discuss how he automated the generation of over a quarter million high-fidelity meshes since 2017. (Registration required.)
- Maybe you’re interested in the performance of marine propellers.
- Or optimization of automotive components.
- UAV icing?
- Jumpstarting Simulation-Driven Design, an article in Digital Engineering, is very much worth your time. It focuses on technologies and techniques for moving simulation to the early stages of the design process where it can be most impactful. Basically, it’s about various approaches to democratizing simulation. [Yes, the “D” word.]
- I won’t claim to understand all of the tech cited in A New Era in High Performance Computing, but the idea of HPC systems that need to be flexible and evolvable (versus the old way of buying a system and simply using and maintaining it until it was obsolete in five years) makes sense. You also see that concept echoed elsewhere. For example, the Air Force’s NGAD program (next generation air dominance – think F-22 replacement) is discussed as a platform onto which upgrades and changes will be incorporated as they become available during production, almost a continuous upgrading.
- How to achieve 37% better price/performance when running OpenFOAM on Amazon AWS.
- It is well worth your time to spend 30 minutes with the video describing how Altair used GPUs to accelerate AcuSolve’s performance.
- Another fluid dynamics mystery bites the dust: time dependent dispersion coefficients for Taylor dispersion theory.
- New [to me] is the CFD blog Fluid Dynamics Using the Computer.
- Truly new is AirShaper’s Passion for Aerodynamics blog.
- On the topic of after-market, CFD-designed, 3D printed parts for your McLaren 720S, if it’s called a supercar originally how much more super can you make it? Isn’t this like putting ketchup on a Wagyu ribeye?
- ANSYS announced their Art of Simulation Image Contest 2020. Entries are due 02 October. [I’m guessing this is a rebranding of their Hall of Fame.]
- Esgee Technologies announced the release of OverViz v2.4.
- ITI released CADfix PPS Release 3 for geometry model simplification.
- ESI’s 8th OpenFOAM Conference 2020 (13-15 October) is going virtual.
- CFD for “aero extensions” for a racing bike which I think are where the rider rests their forearms.
- CFD for brake cooling.
- Here’s Visualizing Data’s best of the visualization web for March 2020.
- The latest topic in OnScale’s blog series on FEA meshing is mesh convergence.
When considered within the context of his entire body of work, Texas painter Matt Clark’s Ahau (shown below) is an outlier. Not thematically, but expressively. He has written that his works operate as a mirror to himself and a lens to the world. And when one considers the full scope of his work, one finds canvases with a much more organic application of paint versus the hard-edged Ahau (for examples, see the artist’s website). Yet this faceted, Voronoi-like work exhibits the effects of layering and color changes that are the hallmark of his other works.
It makes one wonder, why is Ahau so unique? Is its uniqueness why the ArtSpace111 gallery features it?
Update: Matt Clark replied to an email I sent him about this work. While he agrees it was a bit of a departure from his other work, many of the techniques in it are being utilized in his new work. Originally, Ahau began by trying to imprint marks from another canvas onto this one. As that process continued, the structure of this image began to appear as well as the direction to its completion.
Bonus: The June 2020 edition of ACM’s Journal of Programming Languages has a special focus on the history of programming languages that is available, at least for the time being, at no cost. For example, the second language I was taught (in 1980) was APL (don’t laugh) and the journal includes APL since 1978. More germane to a lot of us is Thriving in a Crowded and Changing World: C++ 2006–2020. The rest covers Clojure through Verilog and everything (alphabetically) in between.
Something missed from last weeks update: FLOW-3D Cast v5.1 was released. Transcript below.
Flow Science is pleased to announce the release of FLOW-3D CAST v5.1. This release provides casters with a complete simulation tool for modeling all the casting processes commonly found in modern foundries. Highlights of the FLOW-3D CAST v5.1 release include new process workspaces for investment casting, sand core making, centrifugal casting, and continuous casting, an expansive database of exothermic risers, and a brand new, chemistry-based alloy solidification model.
A major development in FLOW-3D CAST v5.1 is our new, award-winning alloy solidification model. This ground-breaking achievement provides casters with the ability to predict casting strength characteristics and reduce material usage while still meeting product safety and performance requirements.
The new process workspaces greatly reduce simulation setup and results analysis time. The Investment Casting Workspace features a fast and accurate shell mold generation tool and a full radiation model. The Sand Core Making Workspace models sand core shooting as well as all the hardening processes. The Centrifugal Casting Workspace provides the most accurate simulation tool available for all centrifugal processes. Finally, the Continuous Casting Workspace simulates all continuous casting processes including direct chill.
Additionally, databases for heat transfer coefficients, air vents, HPDC machines, and risers provide information at the users’ fingertips. The new exothermic riser database along with the solidification hotspot identification tool lets users quickly and easily place an appropriately sized exothermic riser at locations where shrinkage is predicted to occur.
Only one thing was missed? That would be amazing if I only missed one thing 😉
Thank you for the update. Actually, the announcement of FLOW-3D CAST v5.1 is in my bookmarks. Due to time and other constraints, sometimes I don’t get through all my bookmarks before pressing Publish.
So don’t take the “This Week” in “This Week in CFD” too literally.
Thanks for mentioning the AirShaper blog John!
We look forward to a contribution from you there, should you find it relevant / find the time 🙂
Hopefully, we can deliver the same quality as the Pointwise blog.
You’re welcome. I’m confident your blog will be interesting reading.
As for time for writing…