End Time in Explicit Dynamics

- May 20, 2020 at 12:01 am
  
  Vahidebrahimi
  Subscriber
  
  Hi every one,
  
  Im using Explicit Dynamics to model a needle penetration into the skin that takes 0.7 sec in real world, does this mean that I need to set End time as 0.7 sec as well?
  
  To give you more details, using tabular displacement, I am moving the needle toward skin in 100 steps until the end time is reached.
  
  The issue is that, as 0.7 sec is relatively big for explicit dynamics, the run time is pretty big (as the elements are small), so I need to run in 0.007 sec instead, that affects the results to some extent.
  
  What is your recommendation to reduce the run time while having a reliable result?
  
  Regards
- May 20, 2020 at 2:34 pm
  Wenlong
  Ansys Employee
  Hi Vahidebrahimi,
  
  Yes, you do need to set the End time to be 0.7s. This is because if you change the end time but keep the displacement the same, you are increasing the needle's moving speed (in your case, you are increasing the speed for 100 times), which will give you a completely different result.
  
  In quasi-static problems, mass scaling is an option to reduce the run time. It artificially increases the mass of elements to increase the time step. If the dynamic effect is negligible, you can use that with no problem. But if the dynamic effect is important (I think it is your case), you cannot use mass scaling to increase the mass of the whole model.
  
  However, mass scaling may still be used if you have some very small elements that control the time step (See: https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v201/en/exd_ag/exp_dyn_theory_time_int.html?q=explicit%20dynamics%20time%20step). You can use mass scaling to increase the mass of only these small elements (the solver will find the small elements for you). You can do that by defining the minimum time step. And you should make sure the total mass percentage increase is small (say 2%).
  
  Another thing to try is to use parallel computing (distributed mode) to run the simulation using multiple cores. Explicit dynamics is highly parallelizable and you can take advantage of that if your computer is powerful. You may also try HPC (such as Ansys RSM) is you have access to one.
  
  Regards,
  
  Wenlong
  
  ================ Note ====================
  
  If you have trouble opening the links I attached, please see the first useful link below
  
  Useful Links
- May 21, 2020 at 11:38 am
  
  Vahidebrahimi
  Subscriber
  
  Hi dear Wenlong
  
  Very helpful comment, Appreciate that,
  
  Planning to use Uni HPCs,
  
  Vahid