How to simulate a rigid beam? 1

[KiriakosCE]

I want to run a nonlinear static analysis of a 7 floor building with frames and walls.
I found in this site that, in order to simulate the plastic hinge at the bottom of the wall, i have to assign the wall as a frame section. So the beam which is framing into the wall, has to be broken into two different elements:
1)The normal section of beam, which is deployed until the point where the wall theoretically begins, and
2) A "dummy" rigid beam (inside the wall) which connects the "wall" frame section (centerline of wall), to the end point of the normal beam.
How can i simulate this rigid beam? I tried to use a material with no weight and a modulus of elasticity with a very high price, E=1.0e+11, But i observerd that these dummy beams create some giagantic forces and moments at the wall!

 

[271828]

Perhaps you should use more than one dummy beam in the wall to spread the load into the wall a little more evenly. For example, you could have three dummy frame elements in the wall arranged like an upside-down T. The forces in the vicinity of these members should be very high, but the overall equilibrium of the wall shouldn't be changed.

 

[KiriakosCE]

Thanks for the fast reply! I examined my model many times, and i found out that i had made an error during the input process of the model. My monumental mistake was that i didn't check out the mode shapes of the building after i completed the model! I would have realized that the 4rth floor was somehow fixed and could not move!
I am making the model from zero, again, and i will place the wall between two dummy beams (in a row), at every floor. (The transverse beam to the wall will be a normal one)
I was wondering if the command:
Assign > Frame > End Length Offsets, would work instead of these rigid "dummy" beams?

 

[271828]

I doubt rigid length offsets will be helpful here. Creating a good old fashioned dummy beam with huge A, E, and I, but tiny mass and weight should do the trick.

 

[stressed]

I agree with 271828 that end length offsets probably aren't the best solution.. If it were me, I would try joint constraints (body type). Joint constraints would accomplish the same thing as dummy rigid beams, but without the risk of numerical instabilities which can occur, depending on the stiffness of the beam compared to the stiffness of what it's connecting to.

 

[KiriakosCE]

Thank you for your posts. The dummies finally worked well.

I will try joint constraints too.