Position of Gauss Integration Points in Quad & QuadUP Elm.

A forum dedicated to users with questions regarding soil materials and elements.

forum currently locked

Moderator: Moderators

Locked
hsafti
Posts: 68
Joined: Thu Apr 22, 2010 5:38 am
Location: Braunschweig

Position of Gauss Integration Points in Quad & QuadUP Elm.

Post by hsafti » Fri Mar 25, 2011 3:12 pm

Dear all,

This question I found no answer for it in the manual. Do the Gauss points for quad and quadUP elements coincide with the element nodes? Also, how can one know (inside the Tcl script) which node is which Gauss point (if they are the same)?

To elaborate: In the recorder element, one specifies a Gauss point (from 1 to 4). Which node (of the element) is the first Gauss point and so forth? Or else how can one obtain the coordinates of the Gauss points of an element (for post processing)?

Thanks a lot and best regards,
Hisham ElSafti
Hisham El Safti

hsafti
Posts: 68
Joined: Thu Apr 22, 2010 5:38 am
Location: Braunschweig

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by hsafti » Mon Mar 28, 2011 6:34 am

I found that changing the number of the Gauss point does not change the response .... Therefore I am assuming that an element only has one Gauss point .... My question would be where is it?

Thanks again
Hisham El Safti

fmk
Site Admin
Posts: 5883
Joined: Fri Jun 11, 2004 2:33 pm
Location: UC Berkeley
Contact:

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by fmk » Tue Mar 29, 2011 4:10 pm

there are 4 gauss points and you cannot change the number of gaus points in the quad or quadUP elements, it is not an option! .. the locations are the standard 2 point gauss-legendre .. the locations, expressed in [-1,1] are:

pts[0][0] = -0.5773502691896258;
pts[0][1] = -0.5773502691896258;
pts[1][0] = 0.5773502691896258;
pts[1][1] = -0.5773502691896258;
pts[2][0] = 0.5773502691896258;
pts[2][1] = 0.5773502691896258;
pts[3][0] = -0.5773502691896258;
pts[3][1] = 0.5773502691896258;

hsafti
Posts: 68
Joined: Thu Apr 22, 2010 5:38 am
Location: Braunschweig

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by hsafti » Tue Mar 29, 2011 11:31 pm

I used the following recorders and they all gave the same results (Ver. 2.2.0)

eval "recorder Element -time -eleRange 1 $NumOfElements -file ElementsStress.out -dT $deltaT material 1 stress"
eval "recorder Element -time -eleRange 1 $NumOfElements -file ElementsStress.out -dT $deltaT material 2 stress"
eval "recorder Element -time -eleRange 1 $NumOfElements -file ElementsStress.out -dT $deltaT material 3 stress"
eval "recorder Element -time -eleRange 1 $NumOfElements -file ElementsStress.out -dT $deltaT material 4 stress"

OpenSees generates no error for any of these recorders. Nevertheless, the values retrieved for the four cases are equal. I aim to make a color map of the stress output. Therefore, I need to know:
1. Should there be 4 different results for the recorders stated above?
2. If so, where are there locations (points 1,2,3 & 4)?
3. If there is only one stress/strain output value per element, what is the definition of the point (inside the element) at which this value can be assigned to produce the most proper color map (for stress or strain).
Hisham El Safti

fmk
Site Admin
Posts: 5883
Joined: Fri Jun 11, 2004 2:33 pm
Location: UC Berkeley
Contact:

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by fmk » Wed Mar 30, 2011 9:44 am

either the version you have has a bug or that is the result for your model .. the current version is 2.2.2.f .. if i use the current and place the following recorder commmand
into Example6.1.tcl (http://opensees.berkeley.edu/WebSVN/lis ... pleScripts) before the start of the dynamic analysis:

recorder Element -file ele.out -ele 10 stresses
recorder Element -file ele1.out -ele 10 material 1 stress
recorder Element -file ele2.out -ele 10 material 2 stress
recorder Element -file ele3.out -ele 10 material 3 stress
recorder Element -file ele4.out -ele 10 material 4 stress

i get the following first line in the e.out file (the other files agree)
2.41326 -0.322342 1.87231 0.877006 -4.93111 1.19895 -0.385546 -5.35196 0.238792 1.15071 -0.743193 0.912154

NOTE: if you want you can have all rsults put in a single file.

hsafti
Posts: 68
Joined: Thu Apr 22, 2010 5:38 am
Location: Braunschweig

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by hsafti » Wed Mar 30, 2011 10:53 am

Thanks a lot for your posts. They help a lot.

I used the nDMaterial PressureDependMultiYield model with the QuadUP element. Is this why I get an equal value?

Please refer me to a reference where I can learn more about the position of the Gauss points. I really missed your definition :(

Thanks again.
Hisham El Safti

fmk
Site Admin
Posts: 5883
Joined: Fri Jun 11, 2004 2:33 pm
Location: UC Berkeley
Contact:

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by fmk » Thu Mar 31, 2011 9:23 am

any basic textbook on the finite element method .. or you can google gaus-legenndre integrration .. it is just standard numerical integration.

hsafti
Posts: 68
Joined: Thu Apr 22, 2010 5:38 am
Location: Braunschweig

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by hsafti » Thu Mar 31, 2011 11:55 am

I got it ... thanks
Hisham El Safti

noxon
Posts: 1
Joined: Fri May 06, 2011 2:26 pm

Re: Position of Gauss Integration Points in Quad & QuadUP El

Post by noxon » Sat May 07, 2011 1:17 pm

thanks for sharing your knowlege, i had the same doubt and also couldnt get any reply into the manual :( , once again thanks...
Feeling close to the iguanas care and iguanas habitats world.

Locked