Seismosoft | Earthquake Engineering Solutions

dear friends , how can I model the gap link element?

Tanks

Hi khajehdehia.

The modeling of the gap element is really no different, conceptually, than any other link.

Two originally coincident nodes are needed as discussed in the Help System content topic "Element Classes" -> "Link Element" -> "Gap/Hook Curve". Call these nodes A and B. Node C is specified to define the axial direction and node D is specified to define the 1-3 local plane.

Node A represents one side of the link and node B the other side of the link. So, for example, node A might be connected to a node at a beam-column intersections node and node B might be connected to a brace coming into the joint.

If this doesn't help, I think you will get a lot of hits by doing a forum search on "link modeling".

Best of luck khajehdehia.

tanks a lot

I model the energy dissipative device with steel plates section in the shape of oval ,I apply the displacement at the top nod of that and I fully restrained the bottom nod of that ,but when I apply the displacement the nods displace out supporting plates ,in the lab model the device is placed between two plates , to solve that I must Use the gap element to each nod so that the upper and lower part act like the upper and lower boundary that not allowed the nods cross those boundary , I know I must Use the two coincide link for gap element ,like link element ,I want to know what is the D parameter in gap element and if I must enter the height of nod to restrained their displacement out of boundary ? and if I would connect the gap link to bundray or not how can I solve this problem?

tanks

As mentioned in the Help System in contents topic "Element Classes" -> "Link Element" -> "Gap/Hook Curve", "d" is the distance the two originally coincident nodes must separate by before the hook engages.

And in the same section, you will note that the other 5 DOF need not be gap/hook elements but may be defined by very stiff linear symmetric curves (see the same section for definitions of "very stiff" and problems which may arise of too large a stiffness is specified".)

I have never tried to make one of the coincident nodes a boundary support node, so I am not certain if that will create any problems. In the past, I have used a rigid constraint between a support node and the link node, but that was primarily for displaying graphics more indicative of the structure I was modeling. So you can try it both ways.

Best of luck.