cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Question about weight link and distance of the dependent side

Highlighted
Regular Member

Question about weight link and distance of the dependent side

Hello  all

 

I have a big structure to mesh and I´m interested in performing a shock/dynamic analyse using Simulate 4.0. I am in a learning process in this type of analysis.

 

To reduce mesh size, I replaced the moving part by its CG as an advanced mass.

 

To connect this mass to its anchor structure using rigid links is not correct because the connection itself has some bending freedom. So, I thought to make individual connections to each anchor structure using weight links to avoid this over stiffening of the structure. A kind of compromise.

 

However, it is not possible to create different weight links sharing the same dependent side. To overcome this limitation, I created additional dummy points and connected them with the advanced mass using rigid links – I thought it reproduces well the original situation as the moving mass is a lot stiff than the anchor structure.

 

But firstly, I placed these dummy points at an arbitrary position. Later, I figure out that the “length” of the weight links was giving a huge difference in my dynamic analysis.  I figured out when placing these dummy points very close to each anchor connection.

 

I´m including a simple model where it shows that the eigenfrequency of the model is already influenced by the distance of the “dummy” points - the "length" of the weight link.

 

I know that it may be dangerous to use rigid/weight links but sometimes it can be of great help.  On the other hand, I may have misunderstood the theory behind the weight link and thus the influence of the distance of the dependent and independent side.

 

I think I remember some old Simulate tutorials connecting the CG of a structure using weight links and the difference doing so with rigid links. But in this case the length was the full length between the CG and the anchor points. I think the result would have had completely different doing as described above.

 

Instinctively, from now on, I would model the weight link using as close as possible the position and the placement of the original connection. In opposite to the rigid link, this can change a lot the result.

 

I´m open to understand the real “problem” here.

 

Thanks

Announcements