Static displacement_Constraints_Optistruct

Altair Forum User
Altair Forum User
Altair Employee
edited October 2020 in Community Q&A

Hello, 

 

I'm using optistruct for a topology optmization case in order to minimize the mass. I'm using as constraint 'static displacement': 

 

To applicate this constraint , i proceed like this :

    a- Define DISPLACEMENT_x i as the first reponse 

    b- Define the absmax ( DISPLACEMENT_x) as the second response 

    c- Define the constraint basing to  the second response 'absmax ( DISPLACEMENT_x) 

 

1- Is it ok to proceed like this ? 

 

2- If No, how i can proceed ? 

 

3- If Yes, why we have to use function 'absmax' to define 'static displacement ? 

 

Thanks in advance, 

 

Kind Regards, 

 

Walid Belhaj, 

 

 

 

Answers

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    Hi Walid,

     

    I am not clear with the your requirement, sorry.

     

    AbsMax takes the highest point in the entire function and uses as response value. 

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    Hi Prakash, 

     

    to simplify, is it necessary to use absmax of the displacement as respponse for displacement constraint ? 

     

    Kind Regards, 

     

    Walid Belhaj, 

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    @Walid

     

    If you just want to restrict the displacement of a node or set of nodes, static displacement will serve the purpose. 

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    you mean that i can put a value ( negative or positive)  not nessessary a positif value in static displacement to restrict the displacement ? 

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    Yes,

     

    If you are using negative displacement make sure you provide the constraint value in lower bound. (say -0.4 is your restricted value provide this in lower bound as -0.4>-0.5)

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    Thank you very much Prakash, 

     

    that's clear.

     

    I have one more question if possible : 

     

    Is it possible that i divid my part in two components ( property) and assume draw for each direction  ( +X, -X ) with option 'SINGLE' or it's better to make one component ( one property) and assume Draw ( X) with option 'SPLIT' ? 

     

    Kind Regards, 

     

    Walid Belhaj, 

     

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    Hi Walid,

     

    I saw the same from you in a different post. Is there a reason for having different draw directions?

     

    SPLIT should suit instead. There is an example on OptiStruct help which explains difference between split and single draw. Please refer to that and make your choice :)/emoticons/default_smile.png' srcset='/emoticons/smile@2x.png 2x' title=':)' width='20'> 

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited July 2017

    Dear Prakash, 

     

    The only reason is to get more precise result by dividing in two properties ( but honsetly I'm not sure  if this have an effect or not , it's the first time for me to do such thing :)/emoticons/default_smile.png' srcset='/emoticons/smile@2x.png 2x' title=':)' width='20' /> ).

     

    I already read the help and in my point of view the SPLIT is enough. Just i want to be sure and get another opinion from one who know well.

     

    I said 'SPLIT' enough because the part used is casted by HPDC ( High Pressure Die Casting) so there are two molds. Using SPLIT means that the rejection of the part will be in the two direction ( +X, -X) and optistruct will define the suitable splitting surface.

     

    Is this reasonable ? 

     

    Thank you again for your  response and your time :)/emoticons/default_smile.png' srcset='/emoticons/smile@2x.png 2x' title=':)' width='20' />.

     

    Kind Regards, 

     

    Walid Belhaj,