Velocity-Mobilty FRF's 1

[ansysnewbee]

Hello,

Iam trying to get a mobility plot (velocity/frequency) from my ansys model. I ran a full harmonic analysis and I have the displacements.

I would like to convert these displacements to velocities, to plot mobility FRF's.

Could anyone tell me how to do it in ANSYS?

Sorry if its a basic 101 question, new to dynamics..and ansys..

Emily

 

[MikeyP]

To get a velocity FRF (i.e. Mobility - which is velocity/force against frequency) from a displacement FRF, you need to multiply the complex value at each frequency by

2*pi*frequency*i

where i is the square root of -1

M

--
Dr Michael F Platten

 

[SrinivasAluri]

Thanks Emily for asking the question. I had a similar question how to get a accelerance from displacement from harmonic analysis.

If I may, are you trying to get these mobility FRFs from FE to compare against experimental FRFs? I ask this because I am trying to update an FE model by comparing FRFs from FE and expt FRFs.

 

[GregLocock]

accelerance is easy, just the same again, ie multiply the displacement by -omega^2, or -(2*pi*f)^2

Good luck with your correlation. If you /can/ get good correlation then it is a piece of work to be proud of.



Cheers

Greg Locock

 

[ansysnewbee]

Thanks everyone for the response..yes, srinivas, iam trying to do something like what you have described exp frf vs fe-frf's. Modal analysis of a simple frame.

I have the computed displacements using ansys, so I now probably need to write a little code to multiply my displacements by 2*pi*frequency*i

 

[ansysnewbee]

--Sorry, forgot to sign off

Emily

 

[elogesh]

Hai,

Dr Michael F platten already responded to your thread.

This is regarding your statement
"I have the computed displacements using ansys, so I now probably need to write a little code to multiply my displacements by 2*pi*frequency*i"

No need to write a code, mathematical manipulation can be carried out in ANSYS time history post processor-Post 26.


Regards,
Logesh.E

 

[ansysnewbee]

Hi Logesh,

Thank you for your response.

Yes, I need to create variables in time-history post processor and then plot the same.

Meanwhile, I was reading a book, and it said:
{v}=j*omega*{u}

I dont see the 2*Pi .. Is that to convert rad/s to Hz?

Thank you..

Sorry again, if the question was too simple..

Emily.

 

[GregLocock]

Yes, that's right

Cheers

Greg Locock

 

[ansysnewbee]

Hello Everyone,

Thanks for helping me with the Mobility FRF problem.

Once I got my displacements from my ANSYS model, I created the following variables (working in Hz.)

created variable 100 (-1*Pi)
created variable 101 (100*freq)
created variable 102 (101*disp)

Variable 102 is mobility..

Does this look corrrect? I did implement it in ANSYS, and it looks like I have got my Mobility FRF.

Emily.

 

[GregLocock]

no

100 needs to be sqrt(-1)*2*pi

Cheers

Greg Locock

 

[ansysnewbee]

Hi Greg,

But Iam already working with Hz. during my harmonic analysis.

Do I still need to convert to radians?

Instead I thought it (variable 100) needs to be just sqrt(-1)

Please advice me if Iam wrong.

Emily.

 

[GregLocock]

you still need to work in radians.

Cheers

Greg Locock