Quantcast

3D shell vs continuum shell

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

3D shell vs continuum shell

Jose Fernado-2
Hi,

I've a structure consists of lots of parts (made of formed sheet metal) bonded together.  If I use the conventional 3D shell elements, I have to use tie constraints  to bond them.  Using continuum shell can capture the geometry exactly and have a good mesh across different parts.  My problem is when dealing with bending problems, how good is the performance of continuum shell element when compared with 3D shell? Do I still need at least 3 to 4 layers of element across the thickness (like solid elements) in order to capture accurate bending behaviours.  If so, the aspect ratio will be a problem.

Thanks.

Jose


     

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

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: 3D shell vs continuum shell

Ranil Banneyake
Jose,

In case of continuum shell elements no rotational DOFs are available and
hence they are not good in capturing bending properly. Usually the purpose
of increasing the number of elements across thickness with first order
elements is because, bending mode is absent in first order elements, not the
rotational DOFs. With second order elements you can go with a single element
layer and get good results. However, increasing number of elements across
thickness shall be used to improve the solution with continuum shell
elements too. It is safe to do a sensitivity study for the particular model
you have.

Ranil

On Fri, Dec 4, 2009 at 7:16 AM, Jose Fernado <[hidden email]>wrote:

>
>
> Hi,
>
> I've a structure consists of lots of parts (made of formed sheet metal)
> bonded together.  If I use the conventional 3D shell elements, I have to use
> tie constraints  to bond them.  Using continuum shell can capture the
> geometry exactly and have a good mesh across different parts.  My problem is
> when dealing with bending problems, how good is the performance of continuum
> shell element when compared with 3D shell? Do I still need at least 3 to 4
> layers of element across the thickness (like solid elements) in order to
> capture accurate bending behaviours.  If so, the aspect ratio will be a
> problem.
>
> Thanks.
>
> Jose
>
> [Non-text portions of this message have been removed]
>
>  
>


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

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RES: 3D shell vs continuum shell

Daniel B. Ferreira
I’m trying to model the reeling process of a composite oil riser. Of course,
it includes bending.

What kind of elements should I use?

Commenting you last answer, I think (please, check this information) that
every element have all plies you assign to the shell section with the
relative thickness. So, we must be careful when increasing the number of
elements trough the thickness.

Also, there’s another thing I noticed. I’ve tried to model the filament
winding trough setting many alternate plies with angles -+angle (with
proportionally reduced thickness, so that, the sum of the thickness of the
plies were constant). And I got exactly the same results for only one ply
+Angle and one ply +angle. Kind of, prepreg = filament winding.

Any comments and/or advices about that?

Regards,

 

--

 

Daniel B. Ferreira

 

Graduation Student on Materials Engineering

Federal University of Rio de Janeiro - UFRJ - Brazil

 

De: [hidden email] [mailto:[hidden email]] Em nome de Ranil
Banneyake
Enviada em: sábado, 5 de dezembro de 2009 00:26
Para: [hidden email]
Assunto: Re: [Abaqus] 3D shell vs continuum shell

 

 

Jose,

In case of continuum shell elements no rotational DOFs are available and
hence they are not good in capturing bending properly. Usually the purpose
of increasing the number of elements across thickness with first order
elements is because, bending mode is absent in first order elements, not the
rotational DOFs. With second order elements you can go with a single element
layer and get good results. However, increasing number of elements across
thickness shall be used to improve the solution with continuum shell
elements too. It is safe to do a sensitivity study for the particular model
you have.

Ranil

On Fri, Dec 4, 2009 at 7:16 AM, Jose Fernado <[hidden email]
<mailto:jose_fernado123%40yahoo.com> >wrote:

>
>
> Hi,
>
> I've a structure consists of lots of parts (made of formed sheet metal)
> bonded together. If I use the conventional 3D shell elements, I have to
use
> tie constraints to bond them. Using continuum shell can capture the
> geometry exactly and have a good mesh across different parts. My problem
is
> when dealing with bending problems, how good is the performance of
continuum

> shell element when compared with 3D shell? Do I still need at least 3 to 4
> layers of element across the thickness (like solid elements) in order to
> capture accurate bending behaviours. If so, the aspect ratio will be a
> problem.
>
> Thanks.
>
> Jose
>
> [Non-text portions of this message have been removed]
>
>
>

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





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

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: 3D shell vs continuum shell

sridharan.venkataramanan
In reply to this post by Jose Fernado-2
The continuum shells are better than solid elements in capturing bending
since they follow the same  kinematic and constitutive behavior as shell
elements. You may not need 4 elements through thickness with continuum
shells. I found them quite useful in modeling contact on either
directions. For your case I would suggest use the shell elements with
shell to solid coupling which is not very difficult to define.

 

 

Regards,

V.Sridharan

 

From: [hidden email] [mailto:[hidden email]] On Behalf
Of Jose Fernado
Sent: Friday, December 04, 2009 6:46 PM
To: [hidden email]
Subject: [Abaqus] 3D shell vs continuum shell

 

 

Hi,

I've a structure consists of lots of parts (made of formed sheet metal)
bonded together.  If I use the conventional 3D shell elements, I have to
use tie constraints  to bond them.  Using continuum shell can capture
the geometry exactly and have a good mesh across different parts.  My
problem is when dealing with bending problems, how good is the
performance of continuum shell element when compared with 3D shell? Do I
still need at least 3 to 4 layers of element across the thickness (like
solid elements) in order to capture accurate bending behaviours.  If so,
the aspect ratio will be a problem.

Thanks.

Jose

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





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

Loading...