Tension peak shouldn´t be there

Beginner93
Beginner93 Altair Community Member
edited October 2020 in Community Q&A

Good evening to everyone,

 

I calculated the stress in a beam and I think there is a tension on the end of the beam that shouldn´t be there (marked it with blue colour). I think it´s got an effect on the tension in the two points if calculated by my own, so that the results don´t match exactly. The moment of inertia that I used and the center of gravity is given by hyperworks, so that they has to be correct. Can anyone tell me why there is such a high tension, or is it correct and the difference between the results got an other reason? The beam is fixed with one side, but i´ve choose the dof that he can contract. The centre of

gravity is 16,12mm in z direction.

 

 

<?xml version="1.0" encoding="UTF-8"?>1267788379_Fz-tensionpeak.thumb.jpg.b99db46ac837df93a094746edb8bd2c7.jpg<?xml version="1.0" encoding="UTF-8"?>Fz-force.thumb.PNG.081bd76a42cd7281c84a36c7801af7f9.PNG

 

Thanks for you help,

Beginner 

Unable to find an attachment - read this blog

Tagged:

Answers

  • Mario_21478
    Mario_21478 Altair Community Member
    edited May 2018

    Hello,

     

    Of course, since you use finite shell elements for the problem, you typically get structural stresses.

    Your formulas describe the analytical solution that provides nominal stresses. If you model the problem with beam elements you would get this solution.

     

    I don't understand your chosen  boundary conditions, they evoke an additional numerical singularity.

    I have changed the boundary conditions to all fixed and get the attached results (simple averaging to get nodal stresses for plots)

     

    Here you can see the comparison between structural and nominal stress. The effect is especially enhanced by the super fine modeling.

     

    Best Regards,

    Mario

     

    <?xml version="1.0" encoding="UTF-8"?>image.thumb.png.d0477d08c171b7a3f87ac76123a8bc98.png

  • Beginner93
    Beginner93 Altair Community Member
    edited May 2018

    @Mario thanks a lot for your help. 

    Altair Forum User said:

    If you model the problem with beam elements you would get this solution.

    I´ve already used an 1D beam element but i want to visualize the results on poster. And with the 1D element you just can see a simple 'stick', or is there an other possibility to use beam elements?

     

    Best regards,

    Beginner

Categories