Dear Pinar :

so , youshould change cohesion base on plastic strain and it needs that you do some triaxil tests to know this behaviour or as you suggest follow what literature says as a firste estimate , sorry i never tried it before

Regards
Manooch

pinar_wanstreet <pinar_wanstreet <at> yahoo.com> wrote:
Manooch,
Thank you for your respond.
I am not keeping the cohesion constant. The value of the cohesion
will be changing with respect to the angle of friction. I am trying
to apply the shear strength reduction technique in order to obtain
the safety factor of a slope.

Regards,
Pinar

--- In ABAQUS <at> yahoogroups.com, manoochehr salehabadi
<m_salehabadi <at> ...> wrote:
>
> Dear Pinar:
>
> are you going to keep cohesion constant during your modelling ?
i am using Mohr-Coulomb for my modelling for rock and i have just
constant cohesion base on literature what i am going to do is keep
cohesion constant and change plastic strain as following:
>
> C Strain
> 1.7 MP 0.0001
> 1.7 MP 0.001
> 1.7 MP 0.01
> 1.7 MP 1
>
>
> i hope that it works
>
> Regards
> Manooch
> pinar_wanstreet <pinar_wanstreet <at> ...> wrote:
> Hello,
>
> I am fairly new to Abaqus. I am searching for any kind of advise
how
> to adjust these two (cohesion yield stress and absolute plastic
strain)
> for slope stability problems.
>
> Some sources suggest to add 3% of the young's modulus value to
> cohesion value. Is that reasonable?
>
> For instance: cohesion=c, and Young's Modulus= E
>
> Cohesion Yield Stress Abs Plastic Strain
> c 0
> c+(3%E) 1
>
> Thank you for any respond in advance,
> Pinar
>
>
>
>
>
>
> ---------------------------------
> No need to miss a message. Get email on-the-go
> with Yahoo! Mail for Mobile. Get started.
>
> [Non-text portions of this message have been removed]
>

---------------------------------
Don't be flakey. Get Yahoo! Mail for Mobile and
always stay connected to friends.

[Non-text portions of this message have been removed]

•  8
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

Hi,

It all depends on what problem you are trying to model. Non-linearity
takes care of time integration accuracy to a certain degree beyond
which UTOL has to be used.

Alternatively, you can use fixed time incrementation by removing UTOL
altogether and removing min and max time incrementation values. But
you need to be careful when doing that.

The bottom line is trial and error is the best method.

Regards,

Sai

--- In ABAQUS <at> yahoogroups.com, "bassamyam" <BASSAM.SAAD <at> ...> wrote:
>
> I am doing Consolidation nonlinear analysis (SOIL STEP. My analysis
> doesnot run unless I set up Max. Pore pressure change per increment to
> a high value (almost in the order of the applied load). I do not know
> if that affects the accurecy.. and how I can check that.?
>
> ABAQUS Doc says in the example :Wicking in a partially saturated
> porous medium, quote "UTOL, the pore pressure tolerance that controls
> the automatic incrementation, is set to a large value since we expect
> the nonlinearity of the material to restrict the size of the time
> increments during the transient stages of the analysis and we do not
> wish to impose any further control on the accuracy of the time
> integration"
> Those of you who can comment about this issue, are greatly thankful
> BASSAMYAM
>

•  8
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

hello,

I'd like to know if anybody could tell me if it is possible through
USDFLD CALL GETVRM('S', ...) to know the stress at the extreme fibers
of a beam section.
I read that Array(1)=S11 at integration points (default), by
three-point Simpson integration, and the default stress output points
if a beam section integrated during the analysis is used are the the
vertices of the section.
My question is:
At what integration point corresponds S11 (or is the max value?),
Can I get also the min value? If not, any ideas?
I need this to know the M and N forces at a section.

Thanks

•  8
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

Hello all!

I have a need to use a UEL for defining my Cohesive elements. Apart from
using the UEL i need to use the Continuum elements and specify element
deletion. I know that ABAQUS/Standard does not let me delete elements like
Explicit and in Explicit i cannot use UEL's.

So i wanted to know if i can define my material properties by UMAT for
Standard is it possible to delete elements in ABAQUS/Standard.

I would really appreciate if anyone could be of help on this matter.

Thanks!
--
View this message in context: http://www.nabble.com/Element-Deletion-in-ABAQUS-Standard-tf3376981.html#a9398725
Sent from the Abaqus Users mailing list archive at Nabble.com.

•  10
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

Hi Badger

stitch operation might be another option.

Hope it helps you.

Noman

musketeer89 <musketeer89 <at> yahoo.co.in> wrote:
Hi all,

Some basic issues I ran into recently. Is there a way that you can
'merge' faces which would be like the opposite of partitioning them ? I
did try 'create face' by selecting a closed loop of the edges of two
separate faces but it didn't work for some reason. I'm having some
issues with an imported IGES file and such an operation might be useful.

Secondly, is there a way to impose nodal loads in ABAQUS CAE ? If the
concentrated load were to be within the plane of the geometry where
there is no CAE defined point in the geometry, is there a way other than
partitioning, to get a geometrical point at that location to be able to
apply concentrated loads ? Alternatively, why not apply a nodal load on
the mesh itself ? But I was not able to find this operation.

Thanks - Badger
Georgia Tech School of Aerospace Engineering

---------------------------------
Get your own web address.
Have a HUGE year through Yahoo! Small Business.

[Non-text portions of this message have been removed]

•  9
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

--- In ABAQUS <at> yahoogroups.com, "jocagalhao2000" <jocagalhao <at> ...> wrote:
>
> hello,
>
> I'd like to know if anybody could tell me if it is possible through
> USDFLD CALL GETVRM('S', ...) to know the stress at the extreme fibers
> of a beam section.
> I read that Array(1)=S11 at integration points (default), by
> three-point Simpson integration, and the default stress output points
> if a beam section integrated during the analysis is used are the the
> vertices of the section.
> My question is:
> At what integration point corresponds S11 (or is the max value?),
> Can I get also the min value? If not, any ideas?
> I need this to know the M and N forces at a section.
>
> Thanks
>
I'm using also URDFIL sub, but I get error code 144 or 148.

The sub is like this:
SUBROUTINE URDFIL(LSTOP,LOVRWRT,KSTEP,KINC,DTIME,TIME)
C
....
integer mom, axial, elem
common /var/ mom, axial
DO K1 = 1, 999999
CALL DBFILE(0,ARRAY,JRCD)
KEY = JRRAY(1,2)

if (KEY .EQ. 1900) then
if (array(1) .EQ. 4005555) then
....
endif
endif

if (KEY .EQ. 1901) then
if (array(1) .EQ. 920001) then
....
endif
endif

if (key .EQ. 13 .and. elem .EQ. 1) then
mom=array(5)
endif

if (key .EQ. 104 .and. no .EQ. 1) then
axial=array(4)
endif

end do
RETURN
END

SUBROUTINE USDFLD ...
...
integer mom, axial, EXC
common /var/ mom, axial

EXC=abs(mom/axial)
IF (EXC<=0.6*65) THEN
FIELD(1)=0
ELSEIF (EXC<=0.6*90) then
if (EXC3>0.6*65) THEN
FIELD(1)=1
endif
ELSEIF (EXC>0.6*90) THEN
FIELD(1)=2
ENDIF
C
STATEV(1) = FIELD(1)
C
RETURN
END

I think the error could be here:
DO K1 = 1, 999999 (I copied the line from an example of Abaqus Doc)
CALL DBFILE(0,ARRAY,JRCD)
KEY = JRRAY(1,2) (I copied the line from an example of
Abaqus Doc)

if (KEY .EQ. 1900) ask the elem number....

Is this ok? Or I must search another thing?

Thanks a lot for the help.

•  9
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

Hi Dharmendra,

> Thanxs a lot for your suggestions. now i am having one doubt. in the
> model geametry i can calculate the nodal displacement at the contact
> point (i.e., at the boundry nodes) using some formula, but not at
> the interior region of the mess. how to modify the interior nodes? i
> am getting problem here. is there any option in Abaqus, where we can
> mesh the geometry if we know only the boundry/surface nodes? or else
> i will have to do meshing in some other software, get the node
> definition and include that in the input file for submitting in
> Abaqus. i will wait for your reply.

I think you have 2 options to define the mesh programmatically:
a) script it, so that you create a model using the positions of the
boundary nodes, and then instruct /CAE to mesh it. You can do all this
with Python, and then use ABAQUS/Python to run the script and generate
the input, which you then run with the normal command line.

b) If the boundary nodes keep their numbers, an easier thing to do is
to create a "dummy" simulation where you assign all your elements a
linear elastic material, and apply the displacement you need to the
boundary elements. That way, all the interior mesh will deform
proportionally to their respective element size, and the mesh
properties should be kept (for small changes in the geometry: large
deformations in the geometry will destroy your mesh!). This is very
easy, the simulation should run very fast (because of the linear
elastic material), and then you use the new coordinates of the nodes
for the "real" simulation, without needing to update the element
connectivity.

Hope this helps,
Fernando

•  9
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

I think you're looking for the virtual topology functionality. Look in the
mesh module under tools.

Imposing nodal loads without a partition is difficult. Here's a roundabout
approximation: create a reference point where you want the force assigned.
Connect the reference point to to your geometry using a distributing
coupling.

On 12 Mar 2007 03:58:30 -0700, K.N.Ahsan Noman <noman_buet <at> yahoo.com> wrote:
>
> Hi Badger
>
> stitch operation might be another option.
>
> Hope it helps you.
>
> Noman
>
> musketeer89 <musketeer89 <at> yahoo.co.in <musketeer89%40yahoo.co.in>> wrote:
> Hi all,
>
> Some basic issues I ran into recently. Is there a way that you can
> 'merge' faces which would be like the opposite of partitioning them ? I
> did try 'create face' by selecting a closed loop of the edges of two
> separate faces but it didn't work for some reason. I'm having some
> issues with an imported IGES file and such an operation might be useful.
>
> Secondly, is there a way to impose nodal loads in ABAQUS CAE ? If the
> concentrated load were to be within the plane of the geometry where
> there is no CAE defined point in the geometry, is there a way other than
> partitioning, to get a geometrical point at that location to be able to
> apply concentrated loads ? Alternatively, why not apply a nodal load on
> the mesh itself ? But I was not able to find this operation.
>
> Thanks - Badger
> Georgia Tech School of Aerospace Engineering
>
> ---------------------------------
> Get your own web address.
> Have a HUGE year through Yahoo! Small Business.
>
> [Non-text portions of this message have been removed]
>
>
>

[Non-text portions of this message have been removed]

•  9
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

Hi Deepak,

> > I would advise you to do the meshing, boundary conditions and
> > material assignment in CAE directly, it will be much easier for a
> > novice as there are examples and tutorials you can follow, and the
> > diagnostics tools will tell you when something is wrong with your
> > model.
>
>
> The reason i have developed a seperate C++ application is because,
> the shape of an implant can vary and be much complex than a simple
> hemisphere. I have formulated a geometric model that can represent
> all these implant shape and i guess this is not-possible to create
> interactively in ABAQUS CAE.

If your geometry responds to some mathematical formula, don't
underestimate the CAD capabilities of CAE (it is just a matter of
practice, just as everything else). But like I said, probably the best
approach (highest output for smallest input ) is to export the
geometry in STEP or ACIS from your program, and then let CAE do the
meshing, boundary conditions, material properties and input file
export (that way you also get a known-good file).

> Right now, i am using an open source mesh generator called TETGEN (
> http://tetgen.berlios.de/)to convert the surface mesh that my C++
> application generates to a solid tetrahedral mesh ( 4-node
> tetrahedral elements ). I then write this solid mesh in the ABAQUS
> INPUT FILE FORMAT (*.inp), so i can import it into abaqus.

I know TETGEN, and its tetrahedral meshes are generally good. That
said, I strongly discourage you to use linear tets in your model, go
for quadratic tets or for hexas (there was a very nice discussion in
this list a few months ago about this topic, I recommend you to read
the archives).

> i have one question -----> can abaqus generate a solid mesh out of
> the syrface mesh my C++ application generates. Then i can get rid of
> the open source mesh generator ?????

ABAQUS needs an analytical definition of the surface in order to
operate well, anything else is just not robust enough.

> I understand why you would model the tissue as a hyperelastic
> material. But i dont understand why you would do so for the implant.
>
> To my understanding the implant is made up of a rubber-like elastic
> membrane filled completely and homogenously with saline solution.
>
> Dont you think the saline solution should be modeled with fluid
> elements.

Depends on what you want to achieve exactly, you didn't tell us this
part. Externally, the implant is a "soft solid", and displays
incompressibility (due to the fluid inside and the intrinsic
incompressibility of rubber), so using a hyperelastic model to model
this "apparent" solid is a good approach. Additionally, it is just
easier to have only one material for the implant, instead of the outer
shell (you need to know exactly its dimensions etc), plus the fluid
elements.

> How about modeling it this way:
>
> I would use a tetrahedral mesh with acoustic elements (AC3D4).
>
> I would approximate the saline solution to be water. Then i would
> assign the following material properties:
>
> ---- Density, the same as water
>
> ---- Acoustic Medim ---- Bulk modulus same as water.

I would try the simpler way of hyperelastic material first--messing
with acoustics is not something you want to do as a beginner...

> And one more question ---- how should i model the gravitational load
> on these tetrahedral elements. AS a body force ( like gravity load
> in ABAQUS CAE ) or as a hydrostatic pressure.

As a body force (remember, you only need the gravity *acceleration*!!!
The load is defined by the acceleration and the density, which you
define with the material).

> And should i model the rubber-like elastic membrane as well ???

If you insist on the fluid element approach, then yes, otherwise I
suspect ABAQUS will complain...

Hope this helps,
Fernando

•  9
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online

--- In ABAQUS <at> yahoogroups.com, "eric_powercn" <eric_powercn <at> ...> wrote:
>
> question1:
> How to determine the element status(break or good)?when the
> integration points are all failed in an element,we defined that the
> element is broken. but we can only determine the integration point
> status, when an element include one more integration points,what
> should i do? how to determine the element status?

Unless you are using the latest "extended functionality" version of
ABAQUS, /Explicit elements have 1 integration point, as far as I know
(please somebody correct me). Additionally, in /Explicit you get a
*vector* of integration points and their corresponding variables, so
that you return the corresponding stress and stiffness values --
ABAQUS determines everything else.

Hope this helps,
Fernando

•  9
  New Members

• Computer science program
• Science program
• Abaqus
• Computer science program on line
• Science program online