Learning Forum

[RoyalFlowers]

Hello everyone,

I would be grateful if someone could explain me the difference between initial, minimum and maximum sub-steps for static models including contacts.

As I know increasing the sub-steps could solve convergence problems of contact models because it can divide the load level to small values and load is increased gradually during solving. But I am confused between three different sub-steps that can be defined in Ansys.

Best regards, Shabnam Samsami

[Sandeep Medikonda]

Help 1

Help 2

Help 3

How to access the ANSYS Online Help

 

[RoyalFlowers]

Dear Medikonda,

Thanks a million for your answer.

I already read the Ansys Help, but wanted to be sure about my understanding.

To facilitate convergence we consider sub-steps to apply load gradually. For example if we have only one step with 250 N static load, and  initial sub-steps=50, Min sub-steps= 40, as well as Max sub-steps 200, the load is divided by 50 and in each sub-step 5 N is applied. Then model is solved until convergence, and max. number of equilibrium iterations are 25 per step. If model can be converged with initial sub-steps, Ansys can increase sub-step increments to min value (here to 1/40) to apply higher loads in each sub-steps, but if model can not be converged by applying 5 N load in each sub-step, the sub-step numbers are increased until Max value (here 200).

I really would be grateful if you could advise me whether my understanding is correct or not.

Sincerely, Shabnam

 

[Sandeep Medikonda]

Yes, In the analysis settings you can Define Time stepping either by Substeps or by Time.

So, when you use initial sub-steps of 50 for let's say a total time of 1 sec (meaningless in static) and 250 N force which is being ramped linearly in that time frame. Your initial time step is 1/50 = 2e-02 sec. So, initially, the code is trying to apply 250/50 = 5 N in the first substep.

From there you are telling the code, either use a min. time increment of 1/200=5e-03 sec or something in between with a max. limit on time increment of 1/40=2.5e-02 sec. If convergence is being easily achieved then the code will try to use the max. value if not it will try to use the min value.

The best way to understand such concepts is to implement them on simple models.

[Sandeep Medikonda]

Also, In another discussion, jj77 posted this very good article:

http://www.padtinc.com/blog/you-dont-wanna-step-to-this-breaking-down-loadsteps-and-substeps-in-ansys-mechanical/

/forum/forums/topic/what-exactly-mean-by-initial-substeps/

[RoyalFlowers]

Dear Medikonda,

Thanks a million for your illuminating answers. Wishing you a great weekend.

Sincerely, Shabnam