http://phys.org/news/2014-04-loop-computer-aided.html

Source - Phys.org, University of California - San Diego
Author - Tiffany Fox
Date - 10 April 2014

 

Pretty good article. One of the problems we've had with new engineers coming out of some schools is that they're completely in the digital world and don't realize how parts, machines, designs, etc are actually not identical to what's on their computer simulation. They'll point at their model and say, "this is my product" when in fact it's just an idealized model of the product in an ideal digital environment. It's nice seeing schools giving the students some more hands-on experience, so they can see the feedback loop between design, build, test, iterate, etc. Thanks for sharing.

 

Yeah, I agree with you about the idealized models. I don't remember being tought very much about error propagation or tolerance stacking. I found out about it pretty fast, though, when it was time to manufacture a mechanism I designed for one of my jobs.

 

I had only a superficial reading of the article but I don't see what's wrong with CAE here. The 6.8 earthquake have simply to be part of the design requirement. Nothing to blame on the CAE design.

By the way, tolerances management is part of any good 3D CAD software. You can integrate them in your model, play with them in your simulation and get a model of the response curves of your design. It can be pretty quick to solve the response model of a complex design. Much quicker than doing oversimplification, then bench test.

As I have said before, the problem is not with the tools but how and for what you use it. You have to assess the time spent and the overall cost. I'd agree however, in a naive discussion that young engineers might tends to be more addicted to their CAD station than paper and pen.

But hey... Some of the older are e-mailing daylong, conversing loudly on the phone and still think at the end of the day that they had been performing well

 

Really great article. One of the issues we've had with new specialists leaving a few schools is that they're totally in the advanced world and don't understand how parts, machines, outlines, and so forth are really not indistinguishable to what's on their machine simulation.yeah, I concur with you about the admired models. I don't recollect being tought truly about lapse proliferation or tolerance stacking. I got some answers concerning it really quick, however, when it was time to fabricate an instrument I intended for one of my employments.

 

Trial and error are still the alpha and omega of engineering .

That's the very basis of simulation


Example:
[video]http:\\srdfmc.free.fr/images/SRDFMC - BallisticTest1.swf[/video]


...And a view of an intermediate test:

[video]http:\\srdfmc.free.fr/images/SRDFMC - BallisticTest 1-4.swf[/video]


Those were the first ones of a long series