XANSYS Message: 38956 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Subject: Re: Superelement query
Author: K.S. Raghavan
Date: 2002-06-19 13:53:00Hi Sheldon,
Sorry for the delay in replying to your response.
I have learnt now , perhaps the hard way , that ANSYS merges nodes in the
"use" phase or higher level "generation" phase solely on the basis of node
numbers irrespective of their global coordinates. This I consider a little
user-unfriendly.
The user has to take care to ensure that the master nodes that are meant to
be structurally merged from different superelements are all numbered
identically.
The user has also to ensure that there is no duplication of node number
among all the master nodes from various superelements that are not meant to
be merged , but are retained only to improve accuracy in dynamic analysis.
This become a bit tricky if the additional master nodes are generated
automatically using TOTAL command.
OTOH if the criterion for merging is global coordinates the user will not be
required to put efforts to meet the above. In fact merging on the basis of
global coordinates makes a lot of physical sense and promotes realization of
all the benefits that are associated with the use of this technique.
There is a case , I believe , for changing the criterion of merger of nodes.
BTW I was carrying out a simple problem of rectangular slab fixed at both
the ends. There were two substructures , each a cantilever slab ,modeled
with PLANE82 elements. The nodes on the mid-span formed the superelement.
This I reckon to be the bottoms-up substructuring approach. I have no
knowledge of what you refer to as top-down substructuring approach.
Regards .. raghavan
Regards .. raghavan
-----Original Message-----
Sent: Thursday, June 13, 2002 12:31 PM
To: xansys@yahoogroups.com
Hi Raghavan,
Based on what you wrote, it seems like ANSYS is behaving as
expected. When you mentioned that, in the second
situation, 'coordinates coordinates of the generated nodes 1 to 5 are
replaced by those of the superelement master nodes,' this seems to
indicate that you are using same node numbering in different
locations (non-coincident nodes). ANSYS does not let one define
two 'Node 1' nodes since each node number has to be unique.
Also, are you using bottom-up or top-down substructuring? I was not
sure how you ensured correct node numbering in both cases if you were
doing bottom-up substructuring.
Lastly, regarding your BEAM188 vs BEAM44 difference, I wouldn't be
able to comment more without knowing the details of the structure and
why you expect the answers to show a certain trend.
Regards,
Sheldon
--- In xansys@y..., "Kotur, Srinivasan Raghavan (Export)"
wrote:
> Hello all,
> I have an ANSYS job which has two element types - PLANE82 and
MATRIX50
> (Superelements). Master nodes of the superelements are numbered 1
to 5.
> .Performance of this model is dependent on the order in which the
two types
> of elements are generated.
> If the superelements are generated first and plane82 later (using
AMESH) the
> job gets executed correctly. The generated nodes are replaced by
the master
> nodes of the superelement wherever they are within the specified
tolerance
> limit and the numbers of the master nodes are retained. Please note
that no
> explicit NUMMRG command is given.
> OTOH if PLANE82 elements are generated first and the superelements
later ,
> coordinates of the generated nodes 1 to 5 are replaced by those of
the
> superelement master nodes. This results in element distortion and
topology
> errors.
> Can someone please explain the reason for this diabolical behavior.
Is it an
> ANSYS bug or an inevitable programming necessity.
> Thanks in advance .. Cheers to all .. raghavan (INFOTECH
(India) / Pratt
> & Whitney USA).