Seismosoft | Earthquake Engineering Solutions

I am using Seismostruct to model 3D multi-storey framed structure. The beams of my structure carry distributed gravity loads. Further, I want to shift the mass centers of the decks from their nominal position to take into account the accidental eccentricity. I can specify an additional mass to the beam sections and use the “Automatically transform masses to gravity loads” option to obtain the beam gravity loads. Unfortunately, in this case, the obtained center of masses will be automatically located into their nominal position. Conversely, I can locate all the mass of the deck into the point that I want by using lumped mass, but in this case I cannot apply a distributed gravity load on the beams. In conclusion, how can I apply distributed gravity loads on the beams of my structure and obtain a center of mass shifted from its nominal position?

emarino,

I think the answer to your question depends upon how you wish to model the eccentricity.

On a local scale, you could apply distributed torsions to beam elements to represent the eccentricity. The local torsion would simply be the product of the distributed gravity load and the local eccentricity.

If, on the other hand, you are referring to a global eccentricity, such as a multi-story building with rigid floor diaphragms, then you could create a node at the center of mass of each floor (which won't coincide with the geometric centroid) and use rigid links to tie the node to the corners (or to any other points you deem appropriate).

So, the answer to your question depends on what you wish to model and to accomplish. I hope these two general examples help you make a wise decision emarino.

Thank you huffte for your help.
I will clarify my question. I refer to a global eccentricity. I have modeled a multi-storey buildings with rigid floor diaphragms and I have created one node at the center of mass of each floor. All the mass of the building is concentrated into these nodes. The problem is, how can I apply (in the same model) a distributed gravity loads on the beams?

There is a good discussion of the limited distributed load capabilities in the Help System. Try starting the help system from SeismoStruct, Select "Project Settings", then select "Gravity and Mass". Best of luck and kind regards emarino.

Dear Edo,

You will probably need to model the distributed loading through the employment of point forces and bem end-moments.

Seismosoft Support

Dear seismosoft ,

do you have any suggested paper/ article or book helping us transform the distributed mass to point forces and beam end-moments?.

best,

zthaer,

A textbook on basics of structural mechanics or of structural analysis should include the type of background info you are looking for. There are a few of these included in the Bibliography page of the Help/Manual of SeismoStruct, but I have actually never checked if they include the description you seek or not.

Best,

Rui

Hi Dr.rui;

what i'm concerning about is that i checked the verification examples Ch4-infill-03.spf, the study of Negro, P. et al [1995] “Test on the Four-Story Full-Scale RC frame with masonry Infills”, [ find it online here http://www.iitk.ac.in/nicee/wcee/article/11_2045.PDF ] to compare the calculated lumped mass at the nodes with loading suggested by above paper ( super IMPOSED 2 KN/m2 , live load 2 kn/m2 , slab of 0.15 m thickness (24 kn/m3 density) and the remaining part of beams ) i found it identical but i could not see any masses in the rotational dimension. those masses only in x,y,z and the results comparing with experiment is more or less identical.

my question is , do i need just to calculate the masses and applying it in those 3 dimensional vectors without considering the end moment ?

best , and many thanks to find a time answering new learner questions

zthaer,

When using lumped masses distributed/spread throughout the plan or floors of a given building, it tends indeed to be appropriate to define such concentrated masses only in the translational degrees-of-freedom (typically only the horizontal ones, since seismic vertical vibration can be ignored in the majority of cases). Rotational inertia of lumped masses are normally defined only when e.g. one uses a single concentrated mass placed at the centre of gravity of the floor to represent the entire mass of that storey.

I note also that this issue (of lumped mass directions) is not necessarily related to the topic of beams gravity loading forces; indeed, transforming beam gravity loads into end point forces+moments is not done through the use of lumped mass elements, but rather through the application of forces and moments.

Best,

Rui

Dr.Rui ;
that is much clearer,
best;