XANSYS Forum Index
    Register    

FAQ    Search    Memberlist    Usergroups    SmartFeedSmartFeed    Profile    Log in
AW: [Xansys] Cyclic symmetry Vs full 360 model
 
Post new topic   Reply to topic    XANSYS Forum Index -> XANSYS
Author Message
Briganti, Giovanni
Guest





PostPosted: Fri Dec 08, 2006 1:16 am  Reply with quote

Hi Batta,

I never done such a comparison, but 10% difference in the results is perhaps too much. I would try the following:

1) Displacement BC - check again if you put the proper constraints on the sector model
2) Force BC - as in FEM the forces are applied to nodes, be sure that, in spite of the difference in number of nodes/elements, you apply equivalent pressures on both models
3) if models have two different meshes, try a refinement (good also to check if locking occurs)
4) make at first a comparison between simplified models, without contacts and other "esotic" features: add them later on, when you're sure of what it's going on

I'm interested in some feedback from people who already encountered this problem too.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH

-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von NaniSrinivas, Batta (IE10)
Gesendet: Freitag, 8. Dezember 2006 07:52
An: ANSYS User Discussion List
Betreff: [Xansys] Cyclic symmetry Vs full 360 model


Hi all,



I am working on a cyclic symmetry model which is having bolted joints. I
am interested in the stresses in the nut threads. In 360 degree model we
have 8 bolts and I have used 90 degree model for cyclic symmetry (2
bolts). I have used pretension elements for establishing bolt preload
and surface to surface contacts were used between bolt & nut teeth and
all other interfaces.



When I have performed analysis using cyclic symmetry for sector model
and also performed analysis for full 360 degree model. I expected both
results must be the same, but I have observed around 15 ksi difference
in equivalent stress in nut threads (Sector model is having ~140 ksi and
full model is having ~125 ksi). The loading is only bolt preload. Also
for establishing pre load, full model taking more displacement than the
sector model (Sector model is more stiffer)



Please comment.



Thanks,

Batta Srinivas

Honeywell Technology Solutions Lab,

Bangalore

This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
Wang Xinyu
Guest





PostPosted: Fri Dec 08, 2006 1:59 am  Reply with quote

Hi Batta,

I got problems with cyclic symmetric analysis based on section defined shell and beam elements. When starting the solution run, the base sector will be copied but not the section definition based of the element attributes will be used for the copied sector but the actual SECN. A lot of error appears and the definitions of the copied elements are wrong. I reported this problem and they told me the problem will be resolved in Version 11.0.

Regards

Xinyu Wang
Paul Scherrer Institut
5232 Villigen PSI
Switzerland




-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--] Im Auftrag von Briganti, Giovanni
Gesendet: Freitag, 8. Dezember 2006 09:16
An: ANSYS User Discussion List
Betreff: AW: [Xansys] Cyclic symmetry Vs full 360 model

Hi Batta,

I never done such a comparison, but 10% difference in the results is perhaps too much. I would try the following:

1) Displacement BC - check again if you put the proper constraints on the sector model
2) Force BC - as in FEM the forces are applied to nodes, be sure that, in spite of the difference in number of nodes/elements, you apply equivalent pressures on both models
3) if models have two different meshes, try a refinement (good also to check if locking occurs)
4) make at first a comparison between simplified models, without contacts and other "esotic" features: add them later on, when you're sure of what it's going on

I'm interested in some feedback from people who already encountered this problem too.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH

-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von NaniSrinivas, Batta (IE10)
Gesendet: Freitag, 8. Dezember 2006 07:52
An: ANSYS User Discussion List
Betreff: [Xansys] Cyclic symmetry Vs full 360 model


Hi all,



I am working on a cyclic symmetry model which is having bolted joints. I
am interested in the stresses in the nut threads. In 360 degree model we
have 8 bolts and I have used 90 degree model for cyclic symmetry (2
bolts). I have used pretension elements for establishing bolt preload
and surface to surface contacts were used between bolt & nut teeth and
all other interfaces.



When I have performed analysis using cyclic symmetry for sector model
and also performed analysis for full 360 degree model. I expected both
results must be the same, but I have observed around 15 ksi difference
in equivalent stress in nut threads (Sector model is having ~140 ksi and
full model is having ~125 ksi). The loading is only bolt preload. Also
for establishing pre load, full model taking more displacement than the
sector model (Sector model is more stiffer)



Please comment.



Thanks,

Batta Srinivas

Honeywell Technology Solutions Lab,

Bangalore

This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
Briganti, Giovanni
Guest





PostPosted: Fri Dec 08, 2006 8:49 am  Reply with quote

Batta,

your question confirms exactly what Ian was arguing... hence my advice:

it's better you review your university lectures about FEA, before approaching the Ansys manual again:-)

So, hands off the keyboard, forget about commercial FE-Sw and look for a good book about FEA: an hour of reading can spare you days of bad work.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH


-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von NaniSrinivas, Batta (IE10)
Gesendet: Freitag, 8. Dezember 2006 14:48
An: ANSYS User Discussion List
Betreff: RE: [Xansys] Cyclic symmetry Vs full 360 model


Hi All,

When I was reading the cyclic symmetry ANSYS manual it says it supports only "linear static & large deflection nonlinear static solution options". Does it mean that cyclic symmetry is having limitation with contact nonlinearity?

Thanks,
Batta Srinivas

-----Original Message-----
From: --email address suppressed-- [mailto:--email address suppressed--] On Behalf Of Wang Xinyu
Sent: Friday, December 08, 2006 2:29 PM
To: ANSYS User Discussion List
Subject: AW: [Xansys] Cyclic symmetry Vs full 360 model

Hi Batta,

I got problems with cyclic symmetric analysis based on section defined shell and beam elements. When starting the solution run, the base sector will be copied but not the section definition based of the element attributes will be used for the copied sector but the actual SECN. A lot of error appears and the definitions of the copied elements are wrong. I reported this problem and they told me the problem will be resolved in Version 11.0.

Regards

Xinyu Wang
Paul Scherrer Institut
5232 Villigen PSI
Switzerland




-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--] Im Auftrag von Briganti, Giovanni
Gesendet: Freitag, 8. Dezember 2006 09:16
An: ANSYS User Discussion List
Betreff: AW: [Xansys] Cyclic symmetry Vs full 360 model

Hi Batta,

I never done such a comparison, but 10% difference in the results is perhaps too much. I would try the following:

1) Displacement BC - check again if you put the proper constraints on the sector model
2) Force BC - as in FEM the forces are applied to nodes, be sure that, in spite of the difference in number of nodes/elements, you apply equivalent pressures on both models
3) if models have two different meshes, try a refinement (good also to check if locking occurs)
4) make at first a comparison between simplified models, without contacts and other "esotic" features: add them later on, when you're sure of what it's going on

I'm interested in some feedback from people who already encountered this problem too.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH

-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von NaniSrinivas, Batta (IE10)
Gesendet: Freitag, 8. Dezember 2006 07:52
An: ANSYS User Discussion List
Betreff: [Xansys] Cyclic symmetry Vs full 360 model


Hi all,



I am working on a cyclic symmetry model which is having bolted joints. I
am interested in the stresses in the nut threads. In 360 degree model we
have 8 bolts and I have used 90 degree model for cyclic symmetry (2
bolts). I have used pretension elements for establishing bolt preload
and surface to surface contacts were used between bolt & nut teeth and
all other interfaces.



When I have performed analysis using cyclic symmetry for sector model
and also performed analysis for full 360 degree model. I expected both
results must be the same, but I have observed around 15 ksi difference
in equivalent stress in nut threads (Sector model is having ~140 ksi and
full model is having ~125 ksi). The loading is only bolt preload. Also
for establishing pre load, full model taking more displacement than the
sector model (Sector model is more stiffer)



Please comment.



Thanks,

Batta Srinivas

Honeywell Technology Solutions Lab,

Bangalore

This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
Briganti, Giovanni
Guest





PostPosted: Fri Dec 08, 2006 10:15 am  Reply with quote

Dear Mark,

I'm not knocking down anything, least an exercise I'm also very interested into (just read my 1st post about it).
In my place, Tuscany, irony can be harder than what I showed, but nobody gets offended if the joke gets to the point. Moreover, my irony assumes Batta isn't a kid and can well respond, if he finds my remarks were unjust. Or even better, rebound my joke with a more clever one.

Now to the thechnical point:

An axisymmetric problem is such, if and only if geometry AND Boundary Conditions are axisymmetric.

Under this respect, Ansys isn't missing anything compared to more sophysticated tools, hence your remark adds nothing to the discussion.

Based on the quote (assumed it is right, I didn't check) Batta does from the Ansys' manual, I would rather say that Ansys is able to treat geometrical nonlinearities in axisymmetric problems but not others, like those due to the material formulation or to friction: probably the formulation of Ansys'axisymmetric elements or friction elements can't avoid numerical instabilities when nonlinearities occur in an axisymmetric problem. I hope someone will clarify this point.

>From how he's approaching the problem, it seems Batta is very new to FEA, so can't do anything but repeat the advices I got when I was a newbie as well:

1 - sound understanding of FE-Theory
2 - start with very simple, linear models and add nonlinearities step-by-step
3 - obviously read throughly the manual every time a detail is added to the model: but this is quite useless if step 1 is not yet accomplished.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH


-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von Troscinski, Mark A.
Gesendet: Freitag, 8. Dezember 2006 17:27
An: ANSYS User Discussion List
Betreff: RE: [Xansys] Cyclic symmetry Vs full 360 model


Aren't the restrictions for cyclic symmetry in ANSYS just a function of the software's limitations? For example, Adina (and perhaps other products as well) can handle cyclic symmetry problems with contact, provided the loading is assumed to be cyclically symmetric.

In this instance, suggesting that Batta expend additional energy reviewing university textbooks is a misdirected recommendation. Reviewing the capabilities/restrictions of the software tool's documentation is a wiser choice to more efficiently achieve the desired solution. And his comparison exercise between the two models shouldn't be knocked so prematurely. It appears to have opened his eyes (and mine as well) to a fundamental ANSYS product limitation.

Mark Troscinski
ANSYS Application Engineer
Belcan Corporation
209 Camelford Road
McMurray, PA 15317
--email address suppressed--
724-747-7476

________________________________

From: --email address suppressed-- on behalf of Briganti, Giovanni
Sent: Fri 12/8/2006 10:42 AM
To: ANSYS User Discussion List
Subject: AW: [Xansys] Cyclic symmetry Vs full 360 model



Batta,

your question confirms exactly what Ian was arguing... hence my advice:

it's better you review your university lectures about FEA, before approaching the Ansys manual again:-)

So, hands off the keyboard, forget about commercial FE-Sw and look for a good book about FEA: an hour of reading can spare you days of bad work.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH


-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von NaniSrinivas, Batta (IE10)
Gesendet: Freitag, 8. Dezember 2006 14:48
An: ANSYS User Discussion List
Betreff: RE: [Xansys] Cyclic symmetry Vs full 360 model


Hi All,

When I was reading the cyclic symmetry ANSYS manual it says it supports only "linear static & large deflection nonlinear static solution options". Does it mean that cyclic symmetry is having limitation with contact nonlinearity?

Thanks,
Batta Srinivas

-----Original Message-----
From: --email address suppressed-- [mailto:--email address suppressed--] On Behalf Of Wang Xinyu
Sent: Friday, December 08, 2006 2:29 PM
To: ANSYS User Discussion List
Subject: AW: [Xansys] Cyclic symmetry Vs full 360 model

Hi Batta,

I got problems with cyclic symmetric analysis based on section defined shell and beam elements. When starting the solution run, the base sector will be copied but not the section definition based of the element attributes will be used for the copied sector but the actual SECN. A lot of error appears and the definitions of the copied elements are wrong. I reported this problem and they told me the problem will be resolved in Version 11.0.

Regards

Xinyu Wang
Paul Scherrer Institut
5232 Villigen PSI
Switzerland




-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--] Im Auftrag von Briganti, Giovanni
Gesendet: Freitag, 8. Dezember 2006 09:16
An: ANSYS User Discussion List
Betreff: AW: [Xansys] Cyclic symmetry Vs full 360 model

Hi Batta,

I never done such a comparison, but 10% difference in the results is perhaps too much. I would try the following:

1) Displacement BC - check again if you put the proper constraints on the sector model
2) Force BC - as in FEM the forces are applied to nodes, be sure that, in spite of the difference in number of nodes/elements, you apply equivalent pressures on both models
3) if models have two different meshes, try a refinement (good also to check if locking occurs)
4) make at first a comparison between simplified models, without contacts and other "esotic" features: add them later on, when you're sure of what it's going on

I'm interested in some feedback from people who already encountered this problem too.

Regards,

Giovanni Briganti
AIRBUS Deutschland GmbH

-----Ursprüngliche Nachricht-----
Von: --email address suppressed-- [mailto:--email address suppressed--]Im
Auftrag von NaniSrinivas, Batta (IE10)
Gesendet: Freitag, 8. Dezember 2006 07:52
An: ANSYS User Discussion List
Betreff: [Xansys] Cyclic symmetry Vs full 360 model


Hi all,



I am working on a cyclic symmetry model which is having bolted joints. I
am interested in the stresses in the nut threads. In 360 degree model we
have 8 bolts and I have used 90 degree model for cyclic symmetry (2
bolts). I have used pretension elements for establishing bolt preload
and surface to surface contacts were used between bolt & nut teeth and
all other interfaces.



When I have performed analysis using cyclic symmetry for sector model
and also performed analysis for full 360 degree model. I expected both
results must be the same, but I have observed around 15 ksi difference
in equivalent stress in nut threads (Sector model is having ~140 ksi and
full model is having ~125 ksi). The loading is only bolt preload. Also
for establishing pre load, full model taking more displacement than the
sector model (Sector model is more stiffer)



Please comment.



Thanks,

Batta Srinivas

Honeywell Technology Solutions Lab,

Bangalore

This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.



This e-mail transmission contains information that is confidential and may be privileged. It is intended only for the addressee(s) named above. If you receive this e-mail in error, please do not read, copy or disseminate it in any manner. If you are not the intended recipient, any disclosure, copying, distribution or use of the contents of this information is prohibited. Please reply to the message immediately by informing the sender that the message was misdirected. After replying, please erase it from your computer system. Your assistance in correcting this error is appreciated.



This mail has originated outside your organization, either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
Display posts from previous:   
Post new topic   Reply to topic    XANSYS Forum Index -> XANSYS
All times are GMT - 7 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You cannot download files in this forum


Powered by phpBB © 2001, 2005 phpBB Group

sleek template created by Andrew Charron