Robot Structural Analysis Forum
Welcome to Autodesk’s Robot Structural Analysis Forums. Share your knowledge, ask questions, and explore popular Robot Structural Analysis topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Reply
Message 1 of 27
hussamshihabi
7821 Views, 26 Replies

seismic loading

hello to all

i want to ask if there is document on how does the robot analyse seismic loads is there a clear vision on how it works i mean i have been using ETABS to resolve seismic cases...so i need the philosophy in which ROBOT works to resolve the cases ... any way another question .... is there a video tutorial on modeling and analysing and designing a building with all wind loads and seismic loads application with the design of all the structural elements (shear walls , foundations , frames , bracing systems .... etc. ) thanx for all the help

26 REPLIES 26
Message 21 of 27
t.sautierr
in reply to: Pawel.Pulak

Pawel, I think your message raises some more questions 🙂

 

1) For FE meshed pannels, finally, what is the difference between a modal analysis with consistent matrix and lumped (with or without rotation)? Actually meshing the pannel creates the lumps, isn't it? Lumped matrix creates a true difference for non-divided bars by adding calculation nodes? 

2) what is the purpose of adding excentricies for the modal analysis ? (I must confess I use modal analysis mainly for seismic design and then use the excentricities only for in the seismic analysis, other times I use modal analysis there is no matter to consider excentricities eg: vertical eigen frequency for a stair).

3) Related to 2) : do we experience the same effect if the excentricies is defined on ly in the seismic analysis after having performed the modal analysis ? in other words is this parameter from seismic analysis re-used in the modal analysis and overwrites what is defined in the modal analysis. I wondering if it is the case beause Robot doesn't remember from which story the mass came.

 

4) For bar model, I think that there is a quite difference between lumped ou not lumped, if it is consistent then the program is dividing the bar into calculation nodes and lumps the matrix but without any posibility for the user to see what is done, and with lumped analysis then is consider only the nodes set by the user. Am I right?

 

thanks a lot for your answers, I'll let you net if I have others 🙂

Message 22 of 27
Tuctas
in reply to: Pawel.Pulak

Hello Pawel,

 

Any news about the question?: "If we enforce some elements and nodes to belong to a story in a way out of the "normal" procedure of the program, we should wait the results to be "as expected" (I.e the shifting of mass due to eccentricity actually to follow the assignment of stories, even if they are manually assigned) shouldn't we?" 

 

Regards

Message 23 of 27
Pawel.Pulak
in reply to: Tuctas

Hi Tuctas,

thank you for the PM reminder to answer this post:)

Yes it works this way. The mass eccentricity per story is calculated basing on the current assignment of stories (the default automatic one or the manual assignment, defined by the user)

 

I have double-checked it using your model related to this post and sent to Rafal in January 2013 in stories.rar 🙂

 

---------------------------------------------
If this post answers your question please click the "Accept as Solution" button. It will help everyone to find answer more quickly!

 

Regards


Pawel Pulak
Technical Account Specialist
Message 24 of 27
Tuctas
in reply to: Pawel.Pulak

Thanks Pawel for your answer.

 

So, the only case in which a "manual modification" (i.e manual assignment) cannot lead to results "as expected", about the stories and their eccentricities, is the case (or analogous cases for frame structures with line members and not necessary surface F.E) that you described in this discussion (07-16-2014). I.e the mass corresponding to the "upper" finite elements (surface or linear) comming from the bottom of the 3rd storey that are connected to the top of the 2nd storey will always belong to (and thus, follow the eccentricities of) the lower storey, isn't it? But if there is no any distributed loading along these "upper" elements (that are connected with the top of the lower storey) that will be converted to masses and which the half of them will belong anyway to the lower storey, then the results will be almost "as expected" (if we assume that the self weight of these elements, that will also be distributed as above, is low especially for linear elements).   

Message 25 of 27
Pawel.Pulak
in reply to: Tuctas

Hi Tuctas,

Yes you are right. The only source of inaccuracies when modifying mass distribution in relation to mass eccentricities per story can be from surface finite elements or bar elements existing on the higher story but having some nodes on the lower story. The masses assigned to these nodes will be considered when calculating eccentricity for the lower story and not the higher one.

 

Regards,


Pawel Pulak
Technical Account Specialist
Message 26 of 27
Pawel.Pulak
in reply to: t.sautierr

When touching this post I have noticed that some important questions from t.sautier are not answered yet. So here it is:)

 


@t.sautierr wrote:

1) For FE meshed pannels, finally, what is the difference between a modal analysis with consistent matrix and lumped (with or without rotation)? Actually meshing the pannel creates the lumps, isn't it? Lumped matrix creates a true difference for non-divided bars by adding calculation nodes? 

 

4) For bar model, I think that there is a quite difference between lumped ou not lumped, if it is consistent then the program is dividing the bar into calculation nodes and lumps the matrix but without any posibility for the user to see what is done, and with lumped analysis then is consider only the nodes set by the user. Am I right?


Lumped mass matrix does not add any calculation nodes for non-divided bars. Nor for the consistent mass matrix.


Here is a general description and differences between different types of mass matrix used in Robot:

  • Consistent - the original format of mass matrix - it is a sparse mass matrix with non-zero components not only on the diagonal but also outside of it. It has a similar structure to stiffness matrix - it is a sparse symmetrical matrix
  • Lumped with rotations - obtained by diagonalisation of sparse mass matrix - non-zero components only on the diagonal
  • Lumped without rotations - as above but all componnets on the diagonal corresponding to rotation DOFs (degrres of freedom) also set to zero

Of course the consistent mass matrix is the most precise - but the differences are visible mainly for very small, test models composed of several elements or when analysing very local vibrations.

Attached the document showing element mass matrices in some cases, describing the method of diagonalisation used in Robot and giving link to some publication discussing this topic.

 


@t.sautierr wrote:

2) what is the purpose of adding excentricies for the modal analysis ? (I must confess I use modal analysis mainly for seismic design and then use the excentricities only for in the seismic analysis, other times I use modal analysis there is no matter to consider excentricities eg: vertical eigen frequency for a stair).

3) Related to 2) : do we experience the same effect if the excentricies is defined on ly in the seismic analysis after having performed the modal analysis ? in other words is this parameter from seismic analysis re-used in the modal analysis and overwrites what is defined in the modal analysis. I wondering if it is the case beause Robot doesn't remember from which story the mass came.


Eccentricity defined in modal analysis results in modifying the mass matrix. It results in different results of modal analysis than without eccentricities - different frequencies and mode shapes. For instance using eccentricity in symmetrical structure will result in asymmetrical mode shapes.

When seismic analysis is run after such modal analysis it automatically inherits eccentricity - seismic forces are calculated from modal results already considering eccentricity.

 

Eccentricity defined directly in seismic cases ("Simplified approach" when using "Modal with automatic definition of seismi cases") is a simplified method, where the mass eccentricity is not defined in modal analysis and the mass matrix is not modified. Only the seismic forces are modified to mimic the influence of mass eccentricity.

 

When the eccentricities are defined in modal analysis they will result in the change of base shears in seismic case following this modal case.


It is different in case of the eccentricities defined directly in seismic cases - this method provides the same base shear for all eccentricities as without them. Only the values of reactions in specific supports are changed but the total base shears remain the same when changing eccentricities. So for instance for symmetical building if the mass eccentricity in Y is making it unsymmetrical the earthquake in X direction will still result in 0 base shear along Y and vice-versa.

 

---------------------------------------------
If this post answers your question please click the "Accept as Solution" button. It will help everyone to find answer more quickly!

 

Best regards,


Pawel Pulak
Technical Account Specialist
Message 27 of 27

i search for this topic also ... 

 thank you 

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report