Democratization? Cloud?
- The Metro College of Technology (Toronto) is adding CFD and FEA to the curriculum for its Mechanical Engineering Design program. [I’m guessing this is an example of what some might call “democratization.”]
- Along those same lines, Desktop Engineering’s Kenneth Wong engaged ANSYS with a simple request: get a structural engineer started with CFD in just a few minutes. Listen to and watch the results.
- Cloud computing advisory: Monica Schnitger provides valuable insight for CAE vendors thinking of offering cloud-based services. Adobe’s recent announcement that they’re changing their Creative Suite from a traditional deployment model to the cloud generated a great deal of customer angst. You have to wonder whether
- the angst was just FUD-driven and could be ameliorated with appropriate marketing or
- the angst reflects real hurdles that have to be overcome for adoption of the cloud.
Software
- Take 10 minutes to see and hear what’s new in COMSOL 4.3b including a new CFD solver and improved sweep meshing.
- I have only now become aware of LEMMA and their CFD solver, ANANAS.
- FDS 6 (Fire Dynamics Simulator) is targeted for release by the end of May.
- tetra4D released 3D PDF Converter 4.6 for the conversion of 3D CAD data into PDF documents. [If you know of good applications of 3D PDF in the CFD business please let me know.]
- The March 2013 FieldView newsletter is out and highlights a 25:1 productivity improvement for BYU’s simulation of fans.
Time accurate simulation of a multistage fan. Simulation by BYU. Image from Intelligent Light.
Events
- Chalmers University will offer the course Applied LES, DES, and URANS for Industry on 28-30 August on Gothenburg.
- TFAWS 2013, the Thermal & Fluids Analysis Workshop, will be held (tentatively) on 29 Jul – 02 Aug in central Florida. The submission deadline for abstracts has been extended to 17 May.
- NAFEMS will launch Professional Simulation Engineer (PSE) Standards at the NAFEMS World Congress (June, Salzburg). The PSE will define and track “the competencies and skills that simulation engineers should possess.” [This will definitely be something worth being aware of.]
News in Brief
- Energy services company Senergy has acquired CFD talent from Formula 1.
- CD-adapco is hiring for a number of CFD related positions including Product Manager.
- The National Institute of Health has awarded $2.4 million to Kitware for the extension of its visual toolkit VTK to medical applications.
- Mentor Graphic’s Engineering Edge newsletter (Vol. 2 Issue 1) is now available including interesting articles on FloTHERM XT and Improving 1D Data with 3D CFD. Registration required. [Caution: The online version of the newsletter is one of those faux magazine designs that drives me nuts.]
Temperature in an engine block, simulation and image by Mentor Graphics.
Flowing Visual Effects
Stoke MX is a visual effects plugin from for Autodesk 3ds Max from Thinkbox Software that is used in animation and movies for creating moving clouds of particles. Their latest version can do some pretty amazing stuff as this video shows. (Read more about Stoke MX on GraphicSpeak.)
Screen capture from a video by Tobbe Olsson of a particle cloud simulation done with Stoke MX. (Click to play video.)
RSS - Posts
Pingback: This Week in CFD | Another Fine Mesh | CAE FEA ...
In regards to ANSYS’s flow through a value, I’m scratching my head a little. This type of flow is not my area of expertise, but I would think it is very difficult. There was a lot of hand waving about picking the right turbulence model and best practices. Of course there was hand waving about input conditions and thermal stuff. However, the comments to that discussion seemed to like it. I disagree. Just my opinion. And yes, I believe there is a type of “democratization” occurring. (I’m probably using the word differently than others) People are being told that CFD is a snap. Individuals will then “vote” on whether all this marketing is true or not, or at least correct enough for people’s needs. I did look at some papers which compare CFD to valves, and, not surprisingly, some results were good and some were very poor (error 30-50% on pressure loss). Unfortunately, it doesn’t always seem easy to guess when it will be bad. Well, some of the bad cases were obvious. But I was a little surprised at when some went bad. I was also surprised at some of the behavior of the bad cases. It’s one thing to be “off” and get the trend. But in some cases the trend wasn’t even correctly established.
BTW, a disclosure. I was looking at large valves (diameters of a few feet) rather than small ones. Hopefully small valves compare better since the vortices they kick up are larger relative to the valve itself. A very quick search (diameters less than a couple of inches) seems to indicate that may be true.
Hi Martin. Sorry for the delay in responding. Having started my CFD career doing internal CFD (aircraft inlets and nozzles) I’ll absolutely agree that initial conditions are extremely important to getting the right answer. All assumptions from the core properties of the flow to profiles down through the boundary layer will effect the results. However, the valve is a completely different type of problem for which I don’t have much experience (i.e. low speed versus transonic, perhaps a denser fluid than air, etc.). But I’d like to think that the basic issues are similar. It’s a case of knowing what you don’t know – in other words, I don’t really know what the inflow looks like therefore it might behoove me to do a brief study to determine sensitivities.
Re CFD in the Cloud: check out and sign up for the UberCloud HPC Experiment at http://www.hpcexperiment.com
The UberCloud HPC Experiment is a very successful and rapidly-growing (over 450 participating companies/individuals) crowd-sourced volunteer research project to investigate and document the practical steps and hurdles encountered when moving simulation engineering work in “missing middle” (SME) companies onto the Cloud. hpcexperiment.com