iMechanica - Computational Mechanics Forum - Comments

file download Request

Roozbeh Ashoori — Fri, 21 Nov 2008 11:50:50 -0500

hello darda

I may send you the file via email ;-)

Hope that it would help you!

The limitation of the site for Iran cannot be removed by me...

Request

darda — Fri, 21 Nov 2008 10:30:08 -0500

Hi Dear Adam Nieslony

Thank you for your reply, and sorry in writing

Unfortunately I couldn’t download your developed program on Mathworks
file exchange pages, because of some limitation in my country; I don't have
permission to access that site

I want to request you to send me the program by email,

My email addresses are: dardaei@modares.ac.ir and sadegh_dardaei@yahoo.com

Looking forward to hearing from you soon

Thank a lot.

Dear Neelesh

M. Khezri — Wed, 19 Nov 2008 09:07:23 -0500

I have some experiences about the matter which you mentioned. I think you should consider the highest required rank of derivatives in your problem. Usually the third order spline satisfies the requirements. I have used different weight functions for my problem but the differences in result were negligible.

Regards,

Mani

comment and paper

Jose E. Andrade — Mon, 10 Nov 2008 20:40:38 -0500

Dear Vinh Puh,

Both procedures yield the same answer, so this is a quick way to check. As you noted, linearizing from the weak form may not be straight forward, but seems pretty general. We used this approach in a two-field problem. Here is the paper with the linearization procedure (sorry, don't know how to attach paper):

http://geomechanics.civil.northwestern.edu/Papers_files/2phase.pdf

Regards,

Jose

It worked, thank you so much

PJtree23 — Sat, 01 Nov 2008 19:18:58 -0400

It worked, thank you so much Biswajit!

For my simulation I was running into errors with the S4R element however, in the example you posted the S3R element was used. I am now using the S3R element for all of my models with good success. I was curious about why this worked.

Could it be that the SR4 element is numerically "less sophisticated" than the SR3 element for curved surface modeling? Just a thought.

Anyway, another question about interpreting the von Mises stress plots, for the position selection I am selecting "Unique Nodal" and the graph contains two sets of data :

S, Mises (Avg: 75%) SP:1 PR:SHELLCBAV_SM-1 N:16_1

S, Mises (Avg: 75%) SP:5 PR:SHELLCBAV_SM-1 N:16_1

What do the terms SP:1 and SP:5 correspond to? (in bold)

Are they the outer and inner surfaces of the shell element?

Thanks,

FSDT

ramdas chennamsetti — Sat, 01 Nov 2008 06:21:27 -0400

If normal are straight and not perpendicular to center plane means it is following First Order Shear Deformation Theory (FSDT). Prof. J.N. Reddy's book on Theory of ELastic Plate (other one Composite plate is also nice) is a good one. It describes FSDT and TSDT also.

With regards,

- Ramdas

Re: Ansys Shell99 .ne. Kirchhoff theory

Biswajit Banerjee — Thu, 30 Oct 2008 23:56:53 -0400

The manual says

"Normals to the centerplane are assumed to remain straight after deformation, but not necessarily normal to the centerplane."

That is not an assumption of Kirchhoff theory. The element is probably closer to Mindlin-Reissner theory. Need to look at the detailed formulation to be able to say for sure. Could someone set the matter straight and normal?

-- Biswajit

Thanks a lot!!!

pirexengaer — Thu, 30 Oct 2008 12:59:41 -0400

Thanks a lot!!!

This element is for

ramdas chennamsetti — Thu, 30 Oct 2008 08:26:18 -0400

This element is for modeling laminated composite materials. It follows Kirchoff's Classical plate theory (KCPT) - suitable for thin composite laminates.

Regards,

- Ramdas

As far as i know you can

hijazi — Thu, 30 Oct 2008 01:15:05 -0400

As far as i know you can take help from the book named "Theory of beams and shells". It will be of great help to you. You can get the book from any library of mechanical department

Re: Ansys Shell99

Biswajit Banerjee — Thu, 30 Oct 2008 01:10:42 -0400

I found some information on it reasonably quickly (in ANSYS v 11).

The asssumptions appear to be :

14.99.2. Assumptions and Restrictions

Normals to the centerplane are assumed to remain straight after deformation, but not necessarily normal to the centerplane.Each pair of integration points (in the r direction) is assumed to have the same material orientation.There is no significant stiffness associated with rotation about the element r axis. A nominal value of stiffness is present using the approach of Zienkiewicz(39), however, to prevent free rotation at the node.

The references cited are

1) Ahmad, S., Irons, B. M. and Zienkiewicz, O. C., "Analysis of Thick and Thin Shell Structures by Curved Finite Elements", International Journal for Numerical Methods in Engineering, Vol. 2, No. 3, pp. 419-451 (1970).

2) Cook, R. D., Concepts and Applications of Finite Element Analysis, Second Edition, John Wiley and Sons, New York (1981).

3) Yunus, S. M., Kohnke, P. C. and Saigal, S. "An Efficient Through-Thickness Integration Scheme in an Unlimited Layer Doubly Curved Isoparametric Composite Shell Element", International Journal for Numerical Methods in Engineering, Vol. 28, pp. 2777-2793 (1989).

-- Biswajit

Rainflow for Matlab

Nieslony — Wed, 29 Oct 2008 03:52:41 -0400

Hi Sadegh darda,

If you are familiarly with the Matlab software then look on Mathworks file exchange pages:

http://www.mathworks.com/matlabcentral/fileexchange/3026

for rainflow-counting written by me. Have fun and best regards,

Adam Nieslony

contact in fsw

hamedjam — Sat, 25 Oct 2008 16:00:30 -0400

i have a problem in definition of contact of tool and workpace in fsw. when define contact i dont have any heat generation result in friction between tool and workpace!!!!

please help me.

Re: impossible for ABAQUS to report the stress or strains ...

Biswajit Banerjee — Thu, 23 Oct 2008 01:06:17 -0400

".. guess it is impossible for ABAQUS to report the stress or strains of a shell element".

I took that as a challenge and decided to pursue the matter. I don't use ABAQUS at present nor have I used it extensively in the past.

If searched the web an found an Abaqus manual at http://www.hlrs.de/v6.6/books/exa/default.htm. After reading TFM I found an example at http://www.hlrs.de/v6.6/books/eif/pressfueltank_uniformthick.inp

There I saw the following string of commands

*NODE PRINT,FREQUENCY=0
*EL PRINT,FREQUENCY=0
*EL FILE,ELSET=SAMPLE
STH,
SINV,
*OUTPUT,FIELD,VAR=PRESELECT,FREQ=10
*ELEMENT OUTPUT
STH,
*OUTPUT,FIELD
*ELEMENT OUTPUT,ELSET=SAMPLE
STH,
SINV,
*OUTPUT,HISTORY
*ELEMENT OUTPUT,ELSET=SAMPLE
STH,
SINV,

Then I tracked down the Abaqus keywords manual (at http://www.hlrs.de/v6.6/books/key/default.htm) and searched TFM for *element output. The help page said that this command determined which variables were to be saved. A little more of searching pointed out that STH means section thickness and SINV means stress invariants.

I presume that if you save these quantities you should be able to plot the von Mises stresses. The exercise took me less than 15 minutes.

-- Biswajit

So nobody knows the

PJtree23 — Wed, 22 Oct 2008 11:15:43 -0400

So nobody knows the answer? I am guess it is impossible for ABAQUS to report the stress or strains of a shell element....can anyone please tell me what is going on here?