Airflow simulation. The image shows the flow of air over an open-wheel race car. The streamlines are coded by velocity magnitude where blue is low and red is high. In the case of aerodynamic design of an open-wheel race car, the initial concept model would represent the wheels, body, under tray (diffuser), rear and front wings. With this simple geometry setup you can freely move or modify the airfoils and body independently to find an optimal configuration based on lift and drag values from the Computational Fluid Dynamics (CFD) simulation. (Source: National Center for Manufacturing Sciences)
Rob, the current tools used for simulation at a number of levels are greatly improved over what was available in the past. On the other hand, I worked on manufactuing plant designs using simultation on the early 1980s. In the 1970s I worked for Link Simulation Systems. We had industrial plant simulations that were very high fidellity. Customers used them for training, but found they were so good that they could use them for plant tuning. By just running the math model, they could tune the process for changes in feedstocks.
On the other hand, it was much more difficult and visualization was much more expensive. The current crop of tools are also more general purpose. Becuause computing is much less expensive, a lot more simulation can be done. I also like that the tools apply generalized solvers. This makes for a lot less custom software.
Good points Naperlou. I went to a Siemens conference on CAE PLM last year and found that processing speed has greatly changed simulation. Calculations that once had taken weeks or months now take a couple hours. And the cost of the computing is significantly lower.
Admittedly, I'm behind the curve on this, naperlou, but I'm wondering: Do you still need to be an expert to do CFD and FEA? Or have these tools allowed for others to run the simulations? When I graduated from college in the mid-1970s, I did a little FEA work, and most of the older engineers that I worked with had never heard of it. When I took my first FEA class in school, there were a total of four students in it, including me. In my second FEA class, I was the only student. The number of people who could run FEA in private industry back then was very limited.
Chuck, I am glad you asked. I wrote about that in a blog on this site (http://www.designnews.com/author.asp?section_id=1386&doc_id=261758). If you look at the last paragraph you will get a feeling for what I mean.
CAE tools allow one to do the analysis, but you still need to know the field. These tools provide the solvers, but actually analyzing a particular situation still requires engineering knowledge.
Interesting slideshow, Rob. While Lou is right that simulation as a tool has been used for a long time, I learned a bit here about the range of use for simulation tools. For example, I didn't know about the Mars Rover landing simulation, even though I know NASA uses simulation in a lot of areas of technology.
N.L. is certainly correct in that the simulation tools and solvers are very useful, but not a substitute for understanding. Bob Pease was a big voice in pointing out that the results would never be more accurate than the model, and that without adequate understanding the model would usually be inaccurate. Besides that, without adequate understanding and insight it is hard to decide if the results are correct, or if there must be an error someplace. That can happen just by missing a decimal in some of the input data, for example. It takes insight and experience to understand if a result is "reasonable". We called that "the Maselowski criteria" at one company, after a favorite engineering instructor at LIT.
My other point is that many of the programs serve as a substitute for adequate visulaization skills. Not all folks can visualize functionality correctly, or even at all.
Thanks for the reply, naperlou. Even though I did some FEA early in my career, I still feel like I've lost touch with today's analysis techniques. The last paragraph in your article from last year suggests maybe I'm not as far behind the curve as I thought. Thanks.
Excellent post Rob. In my lifetime there are two marvelous technologies contributory to design and model making: 1.) Simulation and 2.) "Additive" manufacturing. I am blown away with the capabilities now available with simulation software. COMSOL, Workspace LT, Moldflow, Autodesk Simulation, etc all lessen, if used properly, design time. I do agree with William K. in that, when you think you have an acceptable design, you probably need to model that component or design. This is where 3-D printing or other additive process contribute. Rob, you and I are in the same generation -- remember: design, prototype, redesign, re-prototype, etc etc? You know the drill. No longer is this necessary with these two technologies available to us. Times are changing and for the better. Again, great post and thank you for keeping us informed.
The Industrial Internet of Things may be going off the deep end in connecting everything on the plant floor. Some machines, bearings, or conveyors simply donít need to be monitored -- even if they can be.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies.
You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived.
So if you can't attend live, attend at your convenience.