pattern plain

[kgkoktsi]

Dear all,

I have modeled a soil-structure-interaction system and the soil is modeled with quad elements. The applied load is a time series with the following command:

pattern Plain $patternTag $tsTag {
load $nodetag
}

where the $nodetag is a node at the bottom of the soil profile. The command "equal" has been used in order to connect the degrees of freedom of all the nodes that belong to that level.

The time series response (accelerations and displacements) in the top of the structure are expressed in relative or absolute values?
I suppose that they are relative. Otherwise, the top of the structure develops lower acceleration than the interface of the soil-structure (!).

Thanks a lot!

[vesna]

It is relative.

[kgkoktsi]

Vesna, thank you for your quick response.
Now, I'm dealing with another problem.
I want to perform a dynamic, time history analysis by considering the gravity load.
The gravity load is imposed as nodal load with the pattern Plain Linear.
Firstly, I run the gravity analysis. Then, I write on the script:
setTime 0.0
wipeAnalysis
and then I perform the dynamic analysis.
But, I think that the gravity load (which is nodal with the linear plain pattern) is still increasing for as much steps as the time series consist of.
Is there any way to elude that problem?
Thanks in advance.

[vesna]

There are two ways to solve that problem:

1. Change time series of gravity load to Constant
or
2. replace "setTime 0.0" with "loadConst -time 0.0"