elastic / inelastic frame

[noxino]

Does anybody know why using a 1 bay 1 story INELASTIC frame subject to a distributed load (distributed mass) I find the basic moments to be greater than those of a 1 bay 1 story ELASTIC frame?

[ricardomonteiro]

Hi noxino.

Have you checked if your structure is responding in the linear elastic range? That might be a possible cause for the differences you report.

Moreover, according to SeismoStruct help system, as opposed to its inelastic counterparts, elastic frame elements do not call for the definition of any material or section types, the reason for which the user needs to define their self-weight here, through the assignement of an adequate value of specific weight. Hence, could you be considering an additional self weight in the inelastic elements, which is not being accounted for in the elastic ones?

Ricardo

[noxino]

I'm in the elastic range. I did not apply additional mass to the section, only a mass distributed. varying the mass, the curvature and the reactions to the base increase linearly. so I think it's a software problem!
This is the link of the file
http://www.mediafire.com/?l2spv44pla9p4v8
a

[noxino]

does anybody know why using, doubly clamped beam inelastic frame subject to a concentrated load in the middle span, i have an action of compression along the beam? when there should be?

[seismosoft]

De-activate the 'geometric nonlinearity' option (in the Settings menu) if you are interested in linear response only.

Seismosoft Support

[noxino]

I didn't use the 'geometric nonlinearity' option.
I’ll try to explain better my problems.
First, using inelastic frames with inelastic material (in the core and in the cover) on a simply supported beam with uniform load (mass distribuited), I‘ve an action of compression along the element (on the restrain). Instead, using inelastic frames with elastic material, I don’t have a compression along the element.
Second, using inelastic frames with inelastic material (in the core and in the cover) on a doubly cramped beam with uniform load (mass distribuited), I‘ve an action of compression along the element (on the restrain). Instead, using inelastic frames with elastic material, this action of compression along the element is less than in the case of inelastic material.
Is this a problem due to the software or to the theory of inelastic frames?

[seismosoft]

Dear noxino,

We are indeed surprised with what you describe: horizontal reactions being obtained in a beam subjected to vertical loads only, with geometric nonlinearity disabled.

Hence please do send us the model files, so that we may take a look at it.

Seismosoft Support

[noxino]

Thank you!
these are the files:
http://www.mediafire.com/?dy9menn5vynv9ia
http://www.mediafire.com/?d8fbi4ycc9mx1a5

[seismosoft]

Noxino, quick note simply to say that we are looking into the files you sent us.

Seismosoft Support

[seismosoft]

Dear Noxino,

We have now looked into the files you sent us and can confirm that the software is reproducing correctly the geometrically linear response of the simply supported and double-clamped beams that you have modelled with infrmFB elements. Indeed:

- if an elastic section is considered, no axial load should be developed, as classic structural analysis theory attests. This was the behaviour we observed with both models you sent us.

- if an inelastic reinforced concrete section is considered instead, an axial load must indeed develop, due to the fact that the neutral axis is not located at the section's baricentre and hence the element elongates. Again, this was the behaviour we observed with the software.

On a separate note, in carrying out verification exercises like this, you might want to consider modelling the distributed load through the employment of section additional mass, rather than through the use of 'dmass' elements, since the former will allow you to make use of the 'stress recovery' facility to obtain correct member forces (check the Help System for further details). Using both displ- and force-based convergence criteria may also be advisable.

Seismosoft Support