## You are here

# Comments

### Re: Determination of stress intensity factor of FGM materials

*In reply to Determination of stress intensity factor of FGM materials*

Dear Adjal

I think that the problem that you describe is very well tackled in this paper:

http://link.springer.com/article/10.1007/s10999-014-9265-y?sa_campaign=e...

Especially in section 2.3.3

Best regards,

Emilio Martínez Pañeda

SIMUMECAMAT Research Group

### Do u find the reason eventually?

*In reply to ABAQUS-cohesive elements in compression*

Dear Saurabh,

I happened to meet the same problem again when I simulated compression process of MMCs with cohesive element. Reaaaally puzzled by the overlap between the matrix and particle. Could you please share some advice about it?

Thanks very much.

Charlie

### plz see to this problem &

### water-retaining hydrogels

*In reply to Stretchable and Transparent Hydrogels as Soft Conductors for Dielectric Elastomer Actuators*

Most hydrogels dry out in the open air. However, hydrogels can be made to retain water by using descicant. Such hydrogels are highly stretchable, transparent, water-retaining, ionic conductors.

At room temperature, a saturated aqueous solution of NaCl has a relative humidity of 75%, but a saturated aqueous solution of LiCl has a relative humidity of 11%. Thus, a hydrogel containing LiCl is an excellent stretchable, transparent, ionic conductor.

### Biographies of Newton, Young, Maxwell and Kelvin

*In reply to Three interesting, recent books*

Dear Arash: Thank you for the suggestions for the summer readings. I have read the last one, and appreciate your comments. I'll follow up on your suggestion of Poincare. Here are a few more suggestions:

Isaac Newton, by James Cleick. There must be many biographies of Newton. This one is short and well wriiten. The author is an excellent professional writer, not a scientist.

The last man who knew everything: Thomas Young. This is a very easy reading. I read it several years ago. We all know about Young's modulus and Young's diffraction pattern. But Young was a medical doctor, deciphered the Rosetta Stone, and discovered the mechanism of the accommodation of the eye.

The man who changed everything. The life of James Clerk Maxwell. Also a very easy reading. I was hoping to learn how he discovered his equations, but ended learning about many of his other activities.

Degrees Kelvin. Reading biographies of old masters always helps to bring life to old subjects. In this case, Kelvin's involvements in many subjects make an interesting reading of the time and people.

### Mooney Rivlin Constants for material model

*In reply to Mooney Rivlin Constants *

how to calculate Mooney Rivlin Constants for Polyurethane material model used in ABAQUS FEM package??

### hi akshan

*In reply to One time calculation in UEL*

akshan,

i would do the following.

1) at the begining of increment 1 or at time t=0, you compute the eigen values in your uel. use an if loop to achieve this.

2) at the end of increment 1, use uexternaldb to compute the eigen values and store the values in a variable.

3) from increment 2, use the stored eigen values from step 2.

4) the datatype of the variable would be common. this kind of variable can be accessed across multiple subroutines and further the stored data is not deleted during increments.

5) further in uexternaldb, you can use an if loop to control when to calculate the eigen values.

ex: if (kinc .eq. 1) then

calculate eigen values

end if

using the above loop, the eigen values are calculated only once at the end of increment 1

i hope this helps

brunda

### factor in your code

*In reply to Isotropic Hardening VUMAT Problem*

Get the file

http://imechanica.org/files/Writing User Subroutines with ABAQUS.pdf

it may show up as

http://imechanica.org/files/Writing%20User%20Subroutines%20with%20ABAQUS...

A code derivaion is given in:

COMPUTATIONAL METHODS FOR PLASTICITY THEORY AND APPLICATIONS

EA de Souza Neto

D Peric

DRJ Owen

publisher: Wiley

Good luck

Frank

### Professor of Mechanics

### hello

### Hi Brunda,

*In reply to One time calculation in UEL*

Hi Brunda,

Thanks for your response and I'm sorry for my late reply on your comment.

I checked uexternaldb. I couldn't realize that if there is any way to make this subroutine to do some calculations just in the beginning of analysis or not. If there is how can I do that?

You mentioned COMMON TYPE variables. I didn't get what do you mean. would you please let me know more about it?

Let's say I have UEL and UEXTERNALDB subroutines. UEXTENALDB is performing some computations at the beginning of analysis and the results (which is a matrix) is needed in all of the time increments (UEL need them). How can I call the UEXTERNALDB's output inside of the UEL in every time increment?

It would be very kind of you to help with this issue.

Best Regards,

### Iterative Solution for gamma

*In reply to Plasticity integration: satisfaction of the consistency requirement*

When f is a nonlinear function of gamma you have to use the Taylor linear expansion in an iterative fashion so that finally f = 0. Consider the following

f(gamma_i + delta_gamma_i) = 0 -> f(gamma_i) + df(gamma)/dgamma x delta_gamma_i = 0 -> delta_gamma_i = -f(gamma_i)/[df(gamma)/dgamma] (1)

where df(gamma)/dgamma should be evaluated at gamma_i. Computing delta_gamma_i one updates gamma using

gamma_(i+1) = gamma_i + delta_gamma_i (2)

You have to check whether this value of gamma satisfies f i.e. f(gamma_(i+1)) = 0 within a given tolerance. If so the value is the desired value of gamma otherwise equation (1) should be used again with gamma_(i+1) instead of gamma_i so that finally the correct value of gamma is obtained. The procedure looks like this:

gamma = 0;

for (;;) {

delta = - f(gamma) / df_dg(gamma);

gamma = gamma + delta;

if (fabsl(delta) / fabsl(gamma) < tol)

break;

}

The loop iterates until delta is negligible which means that the correct value of gamma has been obtained.

Mohsen

### Anybody Walking around this

### Jahanshahi,

*In reply to Kuhn-Tucker Condition*

Jahanshahi,

In return mapping algorithm, only f=0 is enforced by applying Taylor's linear approximation of yield function (f= f_n + (df/dsigma)*delta sigma + (df/dq)*delta q = 0).

Then why the gamma calculated in return mapping algorithm will satisfy consistency condition?

### Any Lecture or something about periodic boundary conditions

*In reply to Lecture notes of interest to mechanicians*

Any Lecture or something about periodic boundary conditions please ? I am in urgent need of it

### University of Michigan Continuum Physics and FEM online lectures

*In reply to Lecture notes of interest to mechanicians*

As part of **Open.Michigan**, **Continuum Physics** and **Finite Element Method** lectures offered by ** Prof. Krishna Garikipati **are now available online on youtube and open.umich.edu. Lectures on Continuum Physics:https://www.youtube.com/playlist?list=PLJhG_d-Sp_JHvh47eZ8fSuUCUdp86i__y http://open.umich.edu/education/engin/continuum-physics/2013 Introduction to Finite Element Methods:https://www.youtube.com/playlist?list=PLJhG_d-Sp_JHKVRhfTgDqbic_4MHpltXZ http://open.umich.edu/education/engin/intro-finite-element-methods/2013

### Kuhn-Tucker Condition

*In reply to Plasticity integration: satisfaction of the consistency requirement*

When the return mapping algorithm comes into play it means that d(gamma)/dt is not zero. To maintian consistency (also implied by Kuhn-Tacker conditions) it is requred to enforce df/dt=0. f is a function of corrected stress which in turn is a function of d(gamma)/dt. Enforcing this condition leads to determination of d(gamma)/dt. Therefore it is evident that determination of d(gamma)/dt is equivalent to the satisfaction of consistency condition.

Mohsen

### When the return mapping

*In reply to Plasticity integration: satisfaction of the consistency requirement*

When the return mapping algorithm converged, will the consistency requirement be automatically satisfied?

### One time calculation in UEL

*In reply to One time calculation in UEL*

hi ashkan,

check out the user subroutine uexternaldb. this subroutine can be used to perfom the computations (sprinc?) at the end of the increment. further you can use common type variable to store the values and recall them during the next increment.

brunda

## Recent comments