Natural frequency constraint

Hi Rowan,

Can you share the error message?

* ERROR #  24 ***
 Eigenvalue SUBCASES or acceleration loads exist,
 and all referenced material densities are zero.

Hi Rowan,

Can you input density in the material cards along with E, G/Nu?

Hi again, i have put density in. Now the error has changed to :

*** ERROR # 1861 ***
 Equality constraints can only be defined when the SQP or BIGOPT
 optimization method is selected on the DOPTPRM OPTMETH card.

i don't know where to find these cards.

Thanks 

Hi Rowan,

I am not sure what the responses are but from the error message I can suggest the following:

Go to >> Analysis>> Optimization>> OptiConrtol>> OPTMETH and change it to SQP from MFD

<?xml version="1.0" encoding="UTF-8"?>

OKay, thank you this has worked and gained some results, just i have one other problem now: I have a panel with two thin aluminium sheets either side of the aluminium panel core, i want to optimise the cores volume fraction, but when i got to the results it removes the whole core, which i need some to still be there. Is there any way i can tell hyper mesh not to remove the whole core?

Thank You

Rowan,

What are the design constraints you have?

The constraint is natural frequency, I want to keep it within a certain value, while reducing the volume. Yet I want to keep it's overall dimensions, just change the structure of the core.

Hi Rowan,

Have you tried changing the ISO contour threshold value in HyperView?

Can you share the model if possible?

I have changed the ISO contour values - and there is only either full panel core with no volume removed ( the first picture shows this, the top non designable sheet has been removed from view ) it shows no panel core which should be in blue. The second picture is the meshed model in HyperWorks.

Thank you, and sorry about late reply! 

<?xml version="1.0" encoding="UTF-8"?>

<?xml version="1.0" encoding="UTF-8"?>

Hi Rowan,

sorry for a late reply.

Can you share the model with me?

You can use the dropbox link in my signature below to share the model.

Okay, Also i made a mistake in describing my problem - meant to say I want to minimise natural frequency.

I have shared the model

Hi Rowan,

Can you share the .hm or .fem file instead of .mvw file?

Yes, sorry.I have included the .hm one

Rowan,

Do you want to reduce the natural frequency of the entire structure?

yes, reduce natural frequency of the whole structure, with upper and lower boundary of the volume fraction as around 0.2 and 0.3 respectively 

you can add ribs to reduce NF.Since it is a simple model.you can easily add ribs. See attached screenshot.

Unable to find an attachment - read this blog