Dynamic time history analysis problem

02-Getting started with the modelling
bakhtiyar
Posts: 46
Joined: 03 Sep 2012, 04:16

Dynamic time history analysis problem

Post by bakhtiyar »

Hi,
I want apply two different records to the a RC building in same time.firs record applying X derectin and second record applying Z derectin .define a second node with the same coordinates, that is then rigid-linked to the first, applying record 1 to the first node and record 2 to second one .but Dynamic time history stage the error please help me
huffte
Posts: 995
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Dynamic time history analysis problem

Post by huffte »

Bakhtiyar, I have accomplished something like this with the following method.

For each node - “A” - connected to the ground, add 2 nodes in the plane of motion:

1. a node - “B” - offset a small distance (say 0.5 meters) in the x-direction from the ground node
2. a node - “C” - offset a small distance (say 0.5 meters) in the y-direction from the ground node

The first added node - “B” - will be defined to be restrained in the x and z directions.

Then second added node - “C” - will be restrained in the y and z directions.

The original node - “A” - will be restrained for all rotations.

Define a rigid link with “B” as the Master and “A” as the Slave, with all degrees-of-freedom included.

Define a second rigid link with “C” as the Master and “A” as the Slave, also with all degrees-of-freedom included.

Apply the x-direction ground motion to node “B” and apply the y-direction ground motion to node “C”.

I have done this on very simple models of isolators. You will have to examine the response and see if you need to do some adjustment of restraints and links to capture your desired behavior.

Best of luck with your work bakhtiyar.
Tim Huff
bakhtiyar
Posts: 46
Joined: 03 Sep 2012, 04:16

Re: Dynamic time history analysis problem

Post by bakhtiyar »

thank you very much dear huffte...
bakhtiyar
Posts: 46
Joined: 03 Sep 2012, 04:16

Re: Dynamic time history analysis problem

Post by bakhtiyar »

Thank you very much for your response.how do difine time histoty stages
I want tow recoeds in same time apply The structure.
huffte
Posts: 995
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Dynamic time history analysis problem

Post by huffte »

For example, to apply the EW component of a particular record in the x-direction and the NS component of the record in the y-direction, suppose htta the duration of each record is 30 seconds and you wish to use a time step of 0.01 seconds.

Then you would apply the EW component to "B" nodes and the NS component to "C" nodes simultaneously using an "End of Stage" equal to 30 seconds with 3000 steps.
Tim Huff
fbianchi
Posts: 54
Joined: 29 Aug 2008, 04:02

Re: Dynamic time history analysis problem

Post by fbianchi »

Dear all,
probably I'm not understanding well the problem, but if you simply need to apply two different records (let's say EW and NS components of the same earthquake) at the same time, you have to:
1) load each record in the "Time history curves" module
2) define the appropriate time-history stage
3) move to the "Applied Loads" module and assign to a particular node (e.g. n111) the th curve in x direction (selecting the correct record). Then you can copy and paste that load and you only need to change the direction (y instead of x) and the name of the curve.
Following this procedure you are applying two different th loads at the same node..
huffte
Posts: 995
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Dynamic time history analysis problem

Post by huffte »

I think you do understand the problem fbianchi. I tried simpy applying the two components to the same node and it worked. I'm sure I have had problems doing this in the past - receiving an error message that loads are already applied to the selected node when trying to add the second component. But it works fine in the current version.

So, bakhtiyar, take the advice of fbianchi and simply apply two loadings to the ground nodes, for example the EW-component in the x-direction and the NS component in the y-direction.
Tim Huff
bakhtiyar
Posts: 46
Joined: 03 Sep 2012, 04:16

Re: Dynamic time history analysis problem

Post by bakhtiyar »

hi huffte
I tryed your advice .wehn i restrained nodes B and C in all degree of freedom then i recieved correct response
another quastion
if i want to modlling rc bulding with infill panle so i should input infill panle's mass or it is inputed by program(sismostruc)
thanks a lot for your consideration
huffte
Posts: 995
Joined: 22 Jul 2011, 10:19
Location: Cookeville, Tennessee, USA
Contact:

Re: Dynamic time history analysis problem

Post by huffte »

bakhtiyar, I believe that SeismoStruct will use the specific weight you enter when you define the infill element and automatically convert that to lumped masses, presumably at the 4 nodes of the infill element.

However, I have not modeled infills personally, so I cannot ensure this. I would suggest creating a very simle model for which you can easily discern whether or not the specific weight, element thickness, and element area are automatically used to generate element masses.

Best of luck.
Tim Huff
bakhtiyar
Posts: 46
Joined: 03 Sep 2012, 04:16

Re: Dynamic time history analysis problem

Post by bakhtiyar »

Hi
fbianchi, your Solution is perfect but if tow difrent record no same time step or duration how do difine time history stage?
Post Reply

Return to “02-Getting started with the modelling”