This content has been marked as final. Show 2 replies
Hi Brian, can you post the model (no results) or send it to me directly? It will be easier to help if I can see it.
To put some closure to this post, after looking at Brian's model, it appeared that the lower natural freqs calculated by COSMOSWorks were correct based on the model inputs. My suggestions were to double-check material properties and the basis behind the expectation of higher modes.
A good example of this is a motorcycle bracket I was called about a few years ago. This designer knew what the natural freq range should be based on similar historical parts. His analyses suggested a much lower natural freq and he wanted to know why. (Of course, he assumed the solver was wrong!) It turned out that, thanks to our extremely flexible units capabilities, he'd entered in the wrong units for density and essentially raised his mass by an order of magnitude. This, subsequently dropped his natural frequencies.
Another suggestion, actually a warning, is to refrain from adding restraints or adding counter-intuitive loads to try and force a response which matches expectations. Explore adjustments to your loads and restraints to make them more realistic, as you would in any study, but be careful not to force an end-result artifically. I see this in both static and frequency models over the years.