# Mass Scaling in Explicit Dynamics

Member

Hi,

I was running an Explicit Dynamics simulation of a cell phone drop test (as part of Ansys Innovation Course called Strain in Deformation Analysis). In the simulation mass scaling was used and after getting the results I started wondering how much did the mass scaling affect to the mass of the cell phone.

I got the following results as shown in the figures:

Original mass of the phone body: 0.48685 kg

After this I scoped MASSALL result to the phone body and got this result:

Throughout the simulation the total mass of the phone body is 0.70212 kg. I guess this means that the mass scaling has increased the total mass of the phone body to 0.70212 kg and the value does not change during the simulation since there is not so much distortion of the elements during the simulation that the values would change?

However, after this I ran the same simulation with mass scaling turned off and got this result:

Now at the end of the simulation the mass of the phone has also increased from original 0.48685 kg to 0.58612 kg even though the mass scaling should not be used? Is there something else changing the mass of the phone body or does the mass scaling still somehow affect in the background?

Thank you!

Tagged:

• Forum Coordinator

Mass from solver output is more accurate, the mass shown at the selection information is an estimation.

• Member

Hi mji,

Thank you very much for your comment!

I checked the solver output file admodel.prt for the analysis with mass scaling turned on. I found this data from the file (the unit of mass is milligrams):

The highlighted row corresponds to the phone part. I guess these are the correct values for the mass scaling that I should use when checking how much the mass scaling affected to the mass of each part.

However, I'm a little bit confused now what do those MASSALL results mean that I showed in the first message, especially the row saying "Total [kg]".

-Nikke

• Member

Can you please elaborate the steps to do the mass scaling in Explicit Dynamics

Actually i am doing turning operation Simulation in 3D and my simulation shows me 350hours to run the whole simulation

so can you explain me how to do the mass scaling , so my simulation time gets small to about 3-4 days without change in Result when compared to the Experimental Formula

• Member

Hi Kp_9079,

You turn mass scaling on in the analysis settings:

I don't know all the details about the settings but as I understand them "Minimum CFL Time Step" sets the smallest time step you allow for the solver and if the time step would go below it mass scaling will be applied to keep the time step larger than that. "Maximum Element Scaling" gives the maximum amount how much the mass of an element can be scaled and "Maximum Part Scaling" how much the mass of a part can be scaled up. I'm guessing that "Update Frequency" defines how often the solver checks the need of mass scaling and if needed updates the mass scaling values for the elements.

Also, in case if you didn't know, time step is defined by the characteristic length of the smallest element in your mesh based on the CFL condition. So by increasing the mesh size and trying to make sure that your elements in the model are as equal size as possible with each other can help to reduce the simulation time. Also using hex mesh with linear element order is highly recommended.

Hopefully this helps and others can comment also since I have just got started with learning about explicit dynamics too. :)

• Member

Thank you @Nikke

In these photo i can see min mesh metric characteristic length and the graph

so can you tell me how to locate these element with min mesh metric characteristic length from the graph

• Member
edited October 22

Click on the tall blue bar and those elements are shown in the graphics window.

There may be some elements with shorter CL but those bars are too short to see with the top of this tall blue bar visible on the scale. Watch this video to learn how to use the Controls button to zoom in on a small number of elements and the short end of the CL scale.