Learning Forum

[Klaus]

Hi Discovery Team,

We have tested discovery for a couple of nonlinear structure mechanic models (e.g. Snap fit or NL spring analyis). We found that the chances of success are quite low without additional debugging information. After some excessive search, we were able to find the solve.out from the run which prooved to be quite helpful in solving the problem. Question: Are there any plans to expose some key information from the solve.out to the Discovery user (like max. DOF Limit exceeded, too much contact penetration and the like...). We at CADDFEM do believe that it would raise the number of successful anaylsis without adding much complexity to the product. kind regards, Klaus (CADFEM ANSYS Support Staff)

[Charudatta Bandgar]

Hello Klaus

Thanks for testing discovery with non linear models and providing your observations. 

I will discuss your questions, observations and requirements with the developers and will get back to you with more information.

Regards.

[Klaus]

Hi Charudatta,

Thanks for your quick response. Our observation is that in our introductory seminars for discovery, we mostly had at least one participant that showed interest in something like a snap fit problem. Most of the time, those folks are designers, not engineers. Currently, they do not have a real chance with discovery to solve this, even with extensive training from us due to the almost complete lack of debugging capability . Please discuss with development wether we want to enable this type of user to solve nonlinear contact problems inside Discovery Analyze. That would require some solver feedback in the Analyze GUI. Alternatively, we would point them towards Mechanical in Workbench, which is clearly able to solve this kind of problem. We at CADFEM are in favour of enabling designers to solve at least some basic nonlinear problems without having to switch the GUI, since we train them to use Discovery. kind regards, Klaus

[Charudatta Bandgar]

That is understand Klaus This is a limitation in Discovery, and there isn't a timeline as to when these things will be exposed in future versions, but it is on high priority for the developers. I agree that it will be really beneficial for such non linear runs, but at this moment we might have to ask them to go for mechanical in workbench if it is too complicated for Discovery to handle. We thank you for sharing your feedback. We will share your idea/input with the product development team. Regards.

[Klaus]

Hi Charudatta,

OK. Thanks for the reply. I just posted this in the Discovery Idea section.

Regards, Klaus