Welcome to the forums. Please post in English or French.

You are not logged in.

#1 2019-08-19 11:06:17

wojciechS
Member
Registered: 2016-08-16
Posts: 5

Strain results in axially compressed cylinder, large deformation

Hello,

as an exercise I want to model simple cylinder (diameter 50mm) which is axially compressed by 20% (height 50mm -> 40mm). If I am not mistaken, true strain along axis should be around -0.22 while post-processing in Gmsh I see -0.2, stress and reaction force are also incorrect. I have simulated similar case in Ansys where I get correct results.

I attach .comm and .med file. I would be grateful for any advice.

Regards,
ws

Last edited by wojciechS (2019-08-19 12:46:56)


Attachments:
compressed_cylinder_postpro.comm, Size: 6.63 KiB, Downloads: 24

Offline

#2 2019-08-19 11:07:32

wojciechS
Member
Registered: 2016-08-16
Posts: 5

Re: Strain results in axially compressed cylinder, large deformation

and .med file


Attachments:
compressed_cylinder_postpro.med, Size: 50.48 KiB, Downloads: 13

Offline

#3 2019-08-27 10:12:14

El3an0rR1gby
Member
Registered: 2019-08-27
Posts: 1

Re: Strain results in axially compressed cylinder, large deformation

Hello,

It seems you have post-processed strains under an hypothesis of "small displacements". You should try EPSG_ELGA option. You can check u4.81.04 for more info.

Regarding the stress, SIEF_ELGA is computed in STAT_NON_LINE. SIEF_ELNO is a spatial extrapolation of this quantity, which may give questionnable results. It is always best to look at ELGA quantities and then compare them to ELNO/NOEU quantities, which are easier to understand but contain extrapolation and averaging.

No idea for the reaction force...

Best regards,

Offline

#4 2019-08-27 10:39:38

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,569

Re: Strain results in axially compressed cylinder, large deformation

hello

are you sure about this line
it looks incoherent to me

ECRO_LINE=_F(
	D_SIGM_EPSI=10.0,
        SY=10.E100,
),  

jean pierre aubry

Offline

#5 2019-09-02 14:34:03

wojciechS
Member
Registered: 2016-08-16
Posts: 5

Re: Strain results in axially compressed cylinder, large deformation

Thank you for both responses.

@El3an0rR1gby
Thanks for advice. I should have begun with EPSG_ from the start as it is appropriate for large displacements. For both EPSG_ELNO and _ELGA I get identical results of EPZZ strain equal to -0.18. It confuses me even further because EPSG strain is smaller than engineering strain (-0.2) .

If I am reading results properly, stress results SIZZ for both SIEF_ELGA and _ELNO are the same, so problem isn't here.

I guess until I obtain correct strain results speaking about reaction force is premature.


@jeanpierreaubry
I added this line only to be able to use DEFORMATION='SIMO_MIEHE' in STAT_NON_LINE, as I found it in some script in different topic on this forum. Why does it look incoherent?


Regards,
ws

Last edited by wojciechS (2019-09-02 15:52:38)

Offline

#6 2019-09-02 18:58:44

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,569

Re: Strain results in axially compressed cylinder, large deformation

I added this line only to be able to use DEFORMATION='SIMO_MIEHE' in STAT_NON_LINE, as I found it in some script in different topic on this forum. Why does it look incoherent?

because SY is the elastic limit and such a high value as 10.E100 makes you never reach it in the calculation and the behavior is elastic all the way along

10E100 is 2E94 times the elastic limit i let you calculate the allowed displacement before you reach the elastic limit!!!!!!!!!!!!!!

Offline

#7 2019-09-06 17:55:31

wojciechS
Member
Registered: 2016-08-16
Posts: 5

Re: Strain results in axially compressed cylinder, large deformation

That was my intention because I am interested only in elastic behaviour, but I cannot use DEFORMATION='SIMO_MIEHE' without defining any elastic limit in material properties. Therefore I have written some dummy value for elastic limit to make sure that it would not be reached during simulation.

Offline

#8 2019-09-07 07:56:18

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,569

Re: Strain results in axially compressed cylinder, large deformation

ok
then why not using an ELAS material to begin with?

Offline

#9 2019-09-09 15:34:34

wojciechS
Member
Registered: 2016-08-16
Posts: 5

Re: Strain results in axially compressed cylinder, large deformation

What I want to achieve is to obtain correct true strain results. Compressed cylinder changes  its geometry significantly during loading therefore I have to choose DEFORMATION which allows large deformations (like SIMO_MIEHE). It is not possible to utilize SIMO_MIEHE without defining ECRO_LINE in material specification, even if I am interested only in elastic behaviour.

Maybe I am missing or doing something wrong. What bothers me mostly is the fact that the axial (ZZ) strains are not equal to -0.22

Offline

#10 2019-09-09 16:11:10

jeanpierreaubry
Guru
From: nantes (france)
Registered: 2009-03-12
Posts: 3,569

Re: Strain results in axially compressed cylinder, large deformation

if i print a .resu file from your files i get

 
LA VALEUR MAXIMALE DE EPZZ             EST  -1.99999999999999E-01 EN    2 MAILLE(S) : M235 

the vertical strain is 0.2 in Fortan meaning

and if i do the same calculation with an ELAS material i get the same results

why should it be 0.22 ?
imposed dispacement /length=0.01/0.05=0.2!!

you are probably more knowledgeable than me in SIMO_MIEHE so you are going to tell me in which manner it is different from ELAS below the elastic limit

Last edited by jeanpierreaubry (2019-09-09 16:11:52)

Offline