flexible cantilever does not act as expected

[HannsM]

Hello,

I'm trying to verify the behavior of a 1-dimensional cantilever (crossection: solid circular, diameter: 20 mm, length: 270 mm) in MSC.Adams. I already tried a lot of different options so right now I don't have any more ideas what to try next.

I build the part in Nastran/Patran and export the modal neutral file (.mnf) to Adams.
To see some screenshot look at my post at a german forum (unfortunaly nobody is answering there...).
You can also see there, that the same simulation in Nastran/Patran (with the same model that ist later exported) shows the correct results.

By looking at the displacements in Adams it seems as if the x-coordinate of the free end is in some way fixed. But I did not create any joints or constraints at this end.
I also was not able to stretch the part by pure tension in the x-direction. I can't not say why.
Similiar to a tutorial I found for MSC.Adams I also tried to use gravity and a small sphere at the end of the lever to bend it. While it worked on the tutorial files, it doesn't work for me.
I don't think that I'm using to few nodes or mode shapes, at least I tried up to 10000 nodes an 32 mode shapes.

Thanks a lot for your help
Hanns

 

[GregLocock]

What's that green hatching near the tip of the cantilever?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[HannsM]

Hello Greg,

the green hatching at tip comes from the single force being "Fixed in Space" meaning it does not rotate while the tip is moving. Instead it always stays in y-direction. I can get the green hatching "away" be setting the force on "Moving with body". Unfortunately this does not change the general problem.

What I found out in the closed MSC-Forums is that I have to create my .mnf file with interface nodes. I did this by following this guide:

http://esteq.co.za/2013/05/15/create-flexible-bodies-for-adams-using-patran/

With the new .mnf file (including interface nodes) I know get the correct numbers for the displacement. Unfortunaly the displacement shape still looks the same.

For a few moments I had the correct bending shape after importing yet again a new .mnf file with translation of the CBARs to CBEAMs (in accordance to the above mentioned guide). But than the numbers for the displacement were way off. Unfortunatly I didn't save nor took a screenshot and now I can't even recreate this situation.

I hope to get it working and will update if I know something more. I just saw that I can also attach files here so I upload my last version if someone is interested.

Greetings
Hanns