Author Topic: Viscoelasticity  (Read 5243 times)

Juan

  • Jr. Member
  • **
  • Posts: 20
Viscoelasticity
« on: March 18, 2013, 05:00:29 PM »
Hello All,

I am new working with FEAP modeling simple viscoelasticity on a 2D mesh of 4x4 square elements with a constant stress applied on one side and restrained on the opposite side.

I have written the Input script below but the values of displacement and stress I obtain on the output are the same for all the time steps.

I have checked the values of “qi” (Viscoelastic Strain) as they enter the subroutine “viscoe.f” (../feap83dist/ver83/elements/material/small/viscoe.f) and they are always zero for all the time steps. I have also tried with different time steps and loop sizes.

Any suggestions to obtain actual viscoelastic behavior?

Thank you so much,

Juan G.
 --------

FEAP * * Example_Viscoelasticity
  25 16 1 2 2 4

MATErial,1
  SOLId
    PLANe STRAin
    ELAStic ISOTropic 200.0e9 0.4
    VISCoelastic term1 0.7 10
                  ! Blank termination record

COORdinates ALL
  1 0 0 0
  2 0 0.25 0
  3 0 0.5   0
  4 0 0.75 0
  5 0 1 0
  6 0 0 0.25
  7 0 0.25 0.25
  8 0 0.5   0.25
  9 0 0.75 0.25
  10 0 1 0.25
  11 0 0 0.5
  12 0 0.25 0.5
  13 0 0.5 0.5
  14 0 0.75 0.5
  15 0 1 0.5
  16 0 0 0.75
  17 0 0.25 0.75
  18 0 0.5 0.75
  19 0 0.75 0.75
  20 0 1 0.75
  21 0 0 1
  22 0 0.25 1
  23 0 0.5 1
  24 0 0.75 1
  25 0 1 1
                  ! Blank termination record
ELEMents
  1 1 1 1 2 7 6
  2 1 1 2 2 3 8 7
  3 1 1 3 4 9 8
  4 1 1 4 5 10 9
  5 1 1 6 7 12 11
  6 1 1 7 8 13 12
  7 1 1 8 9 14 13
  8 1 1 9 10 15 14
  9 1 1 11 12 17 16
  10 1 1 12 13 18 17
  11 1 1 13 14 19 18
  12 1 1 14 15 20 19
  13 1 1 16 17 22 21
  14 1 1 17 18 23 22
  15 1 1 18 19 24 23
  16 1 1 19 20 25 24
                  ! Blank termination record

BOUNdary restraints
  1 0 1 1
  6 0 1 0
  11 0 1 0
  16 0 1 0
  21 0 1 0
                  ! Blank termination record
FORCes
  5 0 2.5e6 0.0
  10 0 5.0e6 0.0
  15 0 5.0e6 0.0
  20 0 5.0e6 0.0
  25 0 2.5e6 0.0 
                  ! Blank termination record
END

BATCh
  DT,,1
  TRANsient BACK
  LOOP,,2
    TIME
    LOOP,,10
        TANG,,-1
        OUTP TANG
        TANG,,1
    NEXT
    DISP ALL
    STREss ALL
  NEXT
END
STOP
------------

FEAP_Admin

  • Administrator
  • FEAP Guru
  • *****
  • Posts: 993
Re: Viscoelasticity
« Reply #1 on: March 20, 2013, 10:41:54 PM »
The viscoelasticity seems to be working fine.  I have modified your input file to generate a time history file of the displacement of node 5 (dof 1).  Run the file and it will generate a file Pvisc*.dis.  Plot the first column time versus the second column ux @ node 5 and you will see that it is ok.

Note two changes.  One I changed the time integrator to Newmark which makes better sense than backward Euler for a mechanics problem.  Two your mesh had one element (element 2) with five nodes so I fixed that.

Jacopo

  • Jr. Member
  • **
  • Posts: 12
Re: Viscoelasticity
« Reply #2 on: October 12, 2020, 11:49:02 PM »
Dear FEAP community,
I think I have a similar issue, since I am not able to reproduce viscoelastic effect in FEAP8.3. I ran the same input file from the last post both in FEAP8.3 and in FEAPpv4.1, with the results shown in Fig.1 for the horizontal displacements of node 5. As you can see no creep effects are present in FEAP8.3. How could I fix the problem?
Thanks in advance,
Jacopo

Prof. R.L. Taylor

  • Administrator
  • FEAP Guru
  • *****
  • Posts: 2649
Re: Viscoelasticity
« Reply #3 on: October 13, 2020, 10:51:14 AM »
Post your input file.

Note that version 8.3 is very old -- you should consider updating to the latest version.

Prof. S. Govindjee

  • Administrator
  • FEAP Guru
  • *****
  • Posts: 1164
Re: Viscoelasticity
« Reply #4 on: October 13, 2020, 03:58:01 PM »
Your version 8.3 is not reading the VISCOelastic material properties.  Look in your output file and you should see that they are not being printed.

I note that in version 8.6 the syntax changed to
Code: [Select]
VISCoelastic,DEVIatoic,#,# and
Code: [Select]
VISCoelastic,VOLUmetric,#,# but this should not be a problem with version 8.3.
You can debug what is going on by looking around line 1404 in elements/material/inmate.f (the routine that reads the material properties).

Jacopo

  • Jr. Member
  • **
  • Posts: 12
Viscoelasticity
« Reply #5 on: October 16, 2020, 04:27:44 AM »
Dear Professors,
thank you for your prompt response. Please find in the attachments both the input and the output files. Actually the VISCoelastic material set is red by the program and printed in the output file. I've tried debugging the line of code suggested by Prof. Govindjee, but I did not come to a solution.
Jacopo 

Prof. R.L. Taylor

  • Administrator
  • FEAP Guru
  • *****
  • Posts: 2649
Re: Viscoelasticity
« Reply #6 on: October 16, 2020, 09:59:11 AM »
remove the transient Newmark command -- your problem has no mass, for a quasi-static problem you can use the defaults