Extract m,k,c

101 views
Skip to first unread message

vinu

unread,
Jun 21, 2010, 12:26:52 AM6/21/10
to mor4ansys
Hello

I am a graduate student from NTU Singapore. Currently I am working on
substructure coupling of a milling machine. The machine has been
assumed to be two parts. One till the spindle and the other part is
the tool. I am planning to couple the FRF of the milling machine to
the FRF of the tool by substructure method. The tool has been modelled
in ANSYS as a free free solid structure. I have done the modal
analysis to find out the natural frequencies. Now I need to get the
mass, stiffness and damping of this free free solid model to continue
the simulation. I am totally new to this community. I wish I could get
some suggestions regarding this issue. It would be a great help.

Thanks
Vinay

Evgenii Rudnyi

unread,
Jun 21, 2010, 3:53:58 PM6/21/10
to mor4ansys
Hi Vinay,

You can get matrices with HBMAT. However you can make substructuring
or CMS directly in ANSYS. Hence it is not quite clear what do you want
to do.

Best wishes,

Evgenii

vinu

unread,
Jun 21, 2010, 9:56:56 PM6/21/10
to mor4ansys
Dear Evgenii

Thank you for your reply. Actually I am trying to do the substructure
coupling using an experimental FRF and simulated FRF. The experimental
FRF is obtained by direct measurement from them milling machine and
the theoretical FRF, I expect to obtain after doing a simulation from
the m,k,c extracted from the FEA of the milling tool. I saw the option
of doing substructuring /CMS in ANSYS but I am not sure whether it can
take experimental FRF and couple with the simulated FRF.
Do you have any suggestions in my case study?

Regards
Vinay Ravi

Evgenii Rudnyi

unread,
Jun 22, 2010, 4:06:15 PM6/22/10
to mor4...@googlegroups.com
Dear Vinay,

When you have a model in ANSYS, you can run FRF directly in ANSYS,
either with full matrices (but you do not have to extract them,
everything will be done internally) or by means mode superposition.

I am not sure if I understood you plan of using an experimental FRF. You
cannot use it with a finite element code, you can just compare what you
receive from FEM with experiments to check if everything was modeled
correctly.

Evgenii

vinu

unread,
Jun 30, 2010, 1:51:12 PM6/30/10
to mor4ansys
Dear Evgenii

Thanks for your reply. Its disappointing to know that ANSYS cannot
take in experimental FRF. May be i should code something in MATLAB I
guess. Thanks a lot for your suggestions.

Regards
Vinay

Evgenii Rudnyi

unread,
Jun 30, 2010, 2:23:16 PM6/30/10
to mor4...@googlegroups.com
Dear Vinay,

What do you mean by "take in experimental FRF". What an operation you
would like to achieve? You have some frequency response measured, what
do you want else?

Could you please give some example where people do something like this?

Evgenii

Reply all
Reply to author
Forward
0 new messages