Hypermesh Optistruct Tutorial [Global-Local Analysis]

[@emmanuelvassilakis5630]

Hi and happy new year. Nicely presented. I am glad I've helped.

[@Aeroswap]

Happy new year to you too! Your help in making this was invaluable. Thanks a lot :)

[@studyrelated9622]

Awesome video! Keep posting more such videos! Grateful!

[@Aeroswap]

Thanks Tanya! I'm happy it was helpful :)

[@ravindermuppdi7839]

Thank you for such great information...it's reallly help ful for me

[@Aeroswap]

Good to know it was helpful :)

[@onkarjalkotkar5647]

It's tutorial so good. I am waiting for next Tutorial

[@Aeroswap]

Thanks Onkar! The next video will be out soon :)

[@mehmetfurkankesal5794]

In one word you're fantastic. Please make a video about composite optimization :)

[@Aeroswap]

Hello Mehmet, I'm glad the video was helpful. Thanks for the suggestion. I will soon start working on a composite optimization tutorial :)

[@sparcksp]

Amazing technique! Maybe you could create a video from global to local with different element types (from 1d beam to 2d shell)? Cheers.

[@Aeroswap]

Yeah, I will experiment more with this and see how it goes! Thanks for the suggestion :)

[@kenkim2687]

Happy New Year Swap! Thank you for showing the breakout modeling method. It is very interesting to see how the hypermesh freebody tool can be useful to apply the previously-collected gridpoint force results from the global model to the local model as a new load collector. I like how it automatically creates LOADADD card that includes all the gridpoint forces and moments. Also thank you for proving this method showing identical results between the global and local breakouts. My question regarding this gridpoint force approach when you further investigate the local refined model, I see you did fine-mesh and assigned anchor nodes to map out those gridpoint forces/moments at the same exact nodal locations. I am curious, if any interpolation is needed for the other refined nodes (unmapped) and by doing this would stress results be changed? Also curious where I can find the interpolation feature in the hypermesh. My question regarding BCs when you do the global to local, For the mount components they already have the original BCs so it's easy to breakout with the gridpoint forces and run the analysis. How about if you breakout the bracket component? I would assume applying the gridpoint forces/moments on the welded edges, but how would you define the BCs? Would there be any arbitrary BCs to run the analysis?

[@Aeroswap]

Happy new year Ken! Yes, I am also interested in interpolating the grid point forces to other nodes. I tried some experiments but no luck till now. That is the reason I had to reduce the node density along the weld edge. I will surely let you know if I figure out a workaround for this! As for the other components, I think the case will be treated as an inertia relief analysis. Since there are no fixed constraints on the other components, maybe 'Support' type of BC should be assigned at the weld edges along with the existing grid point forces. I will do some tests with this method and see how it goes. Will get in touch with you via mail soon :)

[@emmanuelvassilakis5630]

@@Aeroswap I am not sure if hypermesh can do it automatically, however one might try to distribute the forces on the additional nodes using hand calculations. As to the other question posed, the answer is yes it will be an inertial relief case: just add the following parameters in the .fem file: PARAM,AUTOSPC,NO and PARAM,INREL,-2 PS: Good questions by Ken Kim.

[@Aeroswap]

Thanks for the clarification Emmanuel. But going back to hand calculations seems very time consuming if we want quick results. They should include some interpolation tools in future versions of Hypermesh, it will be very helpful ! Good to know my thinking was on the right track! We are automatically converting a linear static analysis to inertia relief. This is great. @Ken Kim, I hope this answers your questions!

[@emmanuelvassilakis5630]

@@Aeroswap No problem Swapnil. In the examples I've sent the inertia relief method is implemented, therefore you can check yourself how it works. As to the interpolation mentioned, allow me to have a look and should I find something useful I'll let you and your viewers know.

[@emmanuelvassilakis5630]

Here is a first attempt of an educated guess regarding remeshing the edges connected to the tubes: As explained in the video, extract the loads from the global model to the local one. Create a summation node on each side of the bracket. Calculate the summation of loads on each of the created nodes (post-->free body). Create a load collector to store each summed load. Remesh your model. Distribute the summed loads to the new edge mesh using rigid elements (preferably interpolation elements RBE3s). Continue your setup as shown and solve your model. I haven't tested if the above method works or not.

[@allonmessenberg3129]

I just want to thank you very much for your tutorials. They are great. Would it be possible for you to add some additional details to the selections you make? For example (in this video): Why are we outputting 3 different formats? Why are we setting the "Post" option to -1? Why are we setting "autospc" to "no" in this case but not in others? What happens if we allow "autospc"? Thanks again.

[@Aeroswap]

Hi Allon, I am glad you liked the video tutorials posted on this channel. Thank you for pointing out the difficulties in understanding the need for certain steps in the analysis setup process. I will make sure to provide more details regarding every setting henceforth. Additionally, I will also remake this tutorial such that it is easier to understand and follow. Thanks again!

[@allonmessenberg3129]

I tried to import loads on version 2021.2 and couldn't do it. The FBD interface is different.

[@Aeroswap]

Yes, I have realized this issue. I will be recreating this tutorial soon with the new version of Hyperworks. Thanks for pointing this out!

[@allonmessenberg3129]

@@Aeroswap I was able to import the forces in version 2021.2 and get the same stress results. However, if my local mode does not include SPCs, then the model goes flying to infinity. If my local model does not include the location global SPC location, then I have to place new constraints on my local model in locations where they don't really exist. This will alter the results. I also don't understand why you kept the SPC in your model and why it would be necessary at all. A FBD is supposed to have forces and moments in place of constraints. Thanks.

[@Aeroswap]

Yes, you are right. It is possible to run the local analysis without redefining the SPCs. This can be done by triggering interia relief method where supports are created (automatically or manually) instead of SPCs. This will eliminate the need to create unwanted (extra) SPCs in the model. As I mentioned in my previous comment, I am working on a follow up video in which I aim to demonstrate a better way to perform global-local analysis. I will make sure to address all the issues you have pointed out in that new tutorial. Thanks!

[@shivukuderu]

Please make an example on FRF and Random. Thanks

[@Aeroswap]

Hi Shiva, I have already posted a tutorial about FRF using Hypermesh and Optistruct. Please check the most recent videos on this channel to find that tutorial.

[@saimanoj4165]

Hello sir ,can you make a video on metal cutting using LS Dyna

[@Aeroswap]

Thanks for the suggestion! I will see what I can do 👍🏻

[@ravindermuppdi7839]

will works for non linear materials and contacts simulations? also what is the difference between globacl-local and submodelling?

[@Aeroswap]

Yes, this will work for nonlinear analyses. Global-local analysis and submodelling is exactly the same thing. Just two different names commonly used to describe the method shown in this video.

[@igirehi8498]

Hi! Is it possible to look at the stresses of the composite material in the direction of reinforcement?

[@Aeroswap]

Yes, that is possible. Search for 'Stress linearization' in the Help documentation.

[@igirehi8498]

@@Aeroswap Thanks

[@ravindermuppdi7839]

Hello sir, how to add SPC bc in model..like load add card as forces ..also I want to add SPC in local run under static analysis loadstep

[@Aeroswap]

Create a new loadcollector and change its card image to 'SPCADD'. Using this loadcollector, you can now add SPCs. It works exactly like LOADADD.

[@ravindermuppdi7839]

Hello sir, how it will be done for abaqus profile.in hypermesh....pls help me out with any documentation.

[@Aeroswap]

Hi Ravinder, Abaqus is a completely different solver and the setup process for this type of analysis in Abaqus does not match with that in Optistruct. I am an Hypermesh and Optistruct expert however, I do not have much experience in using Abaqus. Therefore, I may not be able to help you in this regards. Good luck!

[@udaypatel8341]

Can local analysis of the component is possible without constraints or spc, because in assembly only one component is fixed and other parts connected by rivets or bolts.

[@udaypatel8341]

Can we do local analysis of other parts of this assembly, because in that case what be the constraints

[@ekramdanish9405]

Can you help me sir how to improve meshing ,

[@Aeroswap]

Sure, watch my 2D and 3D meshing tutorials to learn how to create a better mesh!

[@ekramdanish9405]

Can i have your number please

[@Aeroswap]

To avoid spam, I cannot share my mobile number or email in the comments. My email id is provided in the 'About' section of this channel. Feel free to contact me about any doubts or suggestions.

[@ekramdanish9405]

Hello sir

[@tejaspoojary]

Hello Sir, I have been following all your videos and all of them are really great! Thank you for spreading immense knowledge which you have. I had a doubt, kzbin.info/www/bejne/fpqkkIeroZd3gZY. In the link shared before, can we model and simulate in Hypermesh using OptiStruct? Can you help me with it?

[@Aeroswap]

Hey Tejas, I'm glad that my videos are helping you to learn something new! Yes, this analysis can definitely be performed using Hypermesh and Optistruct. Geometry cleanup and editing will be a crucial part of the setup. Feel free to contact me if you need any assistance. My contact details are provided in the 'About' section of this channel.