- Forum
- Public Forums
- Finite Element Modelling - Code_Aster
- Simple question on cylindrical support [solved]
Simple question on cylindrical support [solved]
- theCapitain
- Topic Author
- Offline
- Junior Member
-
Less
More
- Posts: 21
- Thank you received: 0
14 years 9 months ago #4455
by theCapitain
Simple question on cylindrical support [solved] was created by theCapitain
Hi guys,
I'm really new to CAElinux so please forgive me if I'm going to post an elementary question.
I need to constraint a cylindrical surface to simulate a bearing support.
To this end, using Ansys, you should rotate the nodal coordinate systems of the nodes on the cylindrical surface into a previous created cylindrical csys and then set dx=0 (as x represents R in cylindrical csys) for these nodes. I guess Code Aster provides a similar procedure but I have no idea of what it is. Could someone give me suggestions for managing this problem?
Thanks in advance<br /><br />Post edited by: theCapitain, at: 2010/07/27 10:25
I'm really new to CAElinux so please forgive me if I'm going to post an elementary question.
I need to constraint a cylindrical surface to simulate a bearing support.
To this end, using Ansys, you should rotate the nodal coordinate systems of the nodes on the cylindrical surface into a previous created cylindrical csys and then set dx=0 (as x represents R in cylindrical csys) for these nodes. I guess Code Aster provides a similar procedure but I have no idea of what it is. Could someone give me suggestions for managing this problem?
Thanks in advance<br /><br />Post edited by: theCapitain, at: 2010/07/27 10:25
- Claus
-
- Offline
- Moderator
-
Less
More
- Posts: 670
- Thank you received: 34
14 years 9 months ago #4456
by Claus
Code_Aster release : STA11.4 on OpenSUSE 12.3 64 bits - EDF/Intel version
Replied by Claus on topic Re:Simple question on cylindrical support
Not entirely sure if this is what you ask for, but for a bearing you would want to impose no movement in X and Y (and possibly Z, lengthwise) right?
Then you would simply use DDL_IMPO: X=0 and Y=0 on a node or face group. That would block the movements in X and Y as per the global csys
Regards,
Claus
Then you would simply use DDL_IMPO: X=0 and Y=0 on a node or face group. That would block the movements in X and Y as per the global csys
Regards,
Claus
Code_Aster release : STA11.4 on OpenSUSE 12.3 64 bits - EDF/Intel version
- theCapitain
- Topic Author
- Offline
- Junior Member
-
Less
More
- Posts: 21
- Thank you received: 0
14 years 9 months ago #4457
by theCapitain
Replied by theCapitain on topic Re:Simple question on cylindrical support
Thanks Claus for the prompt replay.
Maybe I haven't explained very well. I have a cylindrical surface (like a shaft) and I want to keep its rotation free, the axial displacement may be imposed or not, doesn’t matter. You must think in cylindrical coordinates: rot free; R=0; z as required.
I cannot impose X=0 and Y=0 on the face in global Cartesian csys because this prevent the shaft's rotary motion.
Maybe I haven't explained very well. I have a cylindrical surface (like a shaft) and I want to keep its rotation free, the axial displacement may be imposed or not, doesn’t matter. You must think in cylindrical coordinates: rot free; R=0; z as required.
I cannot impose X=0 and Y=0 on the face in global Cartesian csys because this prevent the shaft's rotary motion.
- Claus
-
- Offline
- Moderator
-
Less
More
- Posts: 670
- Thank you received: 34
14 years 9 months ago #4458
by Claus
Code_Aster release : STA11.4 on OpenSUSE 12.3 64 bits - EDF/Intel version
Replied by Claus on topic Re:Simple question on cylindrical support
Yes, it would prevent the rotation IF you imposed a displacement/torque , but not a rotational load - as far as I know anyway.
I have an example where I tested a radial fan - for fun - and the load command is like this:
I have an example where I tested a radial fan - for fun - and the load command is like this:
This fixes the fan so it can expand/contract in the axial direction, but is clamped down in the Y/Z direction.CHAR=AFFE_CHAR_MECA(MODELE=MODE,
ROTATION=(209.0,1.0,0.0,0.0,),
DDL_IMPO=(_F(GROUP_MA='hold1',
DX=0.0,
DY=0.0,
DZ=0.0,),
_F(GROUP_MA='hold2',
DY=0.0,
DZ=0.0,),),);
Code_Aster release : STA11.4 on OpenSUSE 12.3 64 bits - EDF/Intel version
- Claus
-
- Offline
- Moderator
-
Less
More
- Posts: 670
- Thank you received: 34
14 years 9 months ago #4459
by Claus
Code_Aster release : STA11.4 on OpenSUSE 12.3 64 bits - EDF/Intel version
Replied by Claus on topic Re:Simple question on cylindrical support
Perhaps you're not trying to apply the same type of load as I just described - in that case, a way to simulate a bearing, is to link the bearing surfaces to a single point, and control boundary conditions from there.
Take a look at a way to do this here:
www.caelinux.org/wiki/index.php/Contrib:...rque#Applying_Torque
Regards,
Claus<br /><br />Post edited by: Claus, at: 2010/07/22 02:48
Take a look at a way to do this here:
www.caelinux.org/wiki/index.php/Contrib:...rque#Applying_Torque
Regards,
Claus<br /><br />Post edited by: Claus, at: 2010/07/22 02:48
Code_Aster release : STA11.4 on OpenSUSE 12.3 64 bits - EDF/Intel version
- theCapitain
- Topic Author
- Offline
- Junior Member
-
Less
More
- Posts: 21
- Thank you received: 0
14 years 9 months ago #4460
by theCapitain
Replied by theCapitain on topic Re:Simple question on cylindrical support
Hi Calus,
you are right, that's the case, I have to impose the displacement and imposing dx=0 and dy=0 results in preventing shaft's rotation.
Linking the cylindrical surface to a single node is an excellent solution that, among other things, doesn't stiffen the shaft on the bending plane (so you are that Claus huh? Congratulations men!).
Anyway, while I was trying to solve the problem reading the documentation (with extreme difficulties: my French is very basic) I've found a way similar to which I was thinking when I started this topic:
char=AFFE_CHAR_MECA(MODELE=mode,
FACE_IMPO=_F(GROUP_MA=bearing,
DNOR=0.0,),),);
so the tangential DOF of the surface's nodes (DTAN) should be free. Am I right?
you are right, that's the case, I have to impose the displacement and imposing dx=0 and dy=0 results in preventing shaft's rotation.
Linking the cylindrical surface to a single node is an excellent solution that, among other things, doesn't stiffen the shaft on the bending plane (so you are that Claus huh? Congratulations men!).
Anyway, while I was trying to solve the problem reading the documentation (with extreme difficulties: my French is very basic) I've found a way similar to which I was thinking when I started this topic:
char=AFFE_CHAR_MECA(MODELE=mode,
FACE_IMPO=_F(GROUP_MA=bearing,
DNOR=0.0,),),);
so the tangential DOF of the surface's nodes (DTAN) should be free. Am I right?
Moderators: catux
- Forum
- Public Forums
- Finite Element Modelling - Code_Aster
- Simple question on cylindrical support [solved]
Time to create page: 0.147 seconds