Challenge

Crash Box Simulation using LS-DYNA

OBJECTIVE:

To create simulation results for a crash box under load and compare the plots for the obtained simulation results. The keywords are loaded to the model and saved in a new directory. The thickness of the box is changed and compared the results with the old thickness. Then the energy difference, stress, strain, von-mises for all the cases are plotted.

PROCEDURE:

There are three steps to be followed to achieve the simulation of the drop test, they are

Model setup,

Output requests,

Results.

MODEL SETUP:

The model setup is nothing but, the input model(keyword file) is studied and inputted with the suitable parameters in the LS-DYNA PREPOST deck, they are as follows

Title,

Assigning material cards,

Section creation,

Creating Rigidwall,

Contact interface,

Initial velocity,

Sectional_Force for nodes and elements,

Creating Control cards.

The model is opened through the command File--->open--->LS-DYNA Keyword file. or simply use the shortcut key ctrl+k. The unit system followed here is gm mm ms.

When the model is opened, it consists of only nodes and elements. The title of this test is saved as Crashbox Test.

Assigning Section card:

The material and the section for the shell are linked with the crash box through part creation. The command for the part is the Right-side toolbar--->model--->Keyword manager--->part command is used.

The section card is inserted through the command

Right-side toolbar--->model & part--->Keyword manager--->section--->shell. The thickness of the shell component is assigned as 1.2mm.

The other thickness of the shell component for the other case is assigned as 1.5mm.

Then based on each case respectively the section is loaded to the input file.

Assigning Material card:

To assign the material to the parts separately, the command used is

Right-side toolbar>model>Keyword manager>Materials >24_piecewise_linear_plasticity card is used.

*Mat_024_Piece_Wise_Linear_Plasticity:

This is Material Type 24, which is an elastoplastic material with arbitrary stress as a function of strain curve and arbitrary strain rate dependency can be defined. Failure based on the plastic strain or a minimum time step size can be defined. The input values like density, young's modulus, Poisson's ratio, yield stress, and tangent modulus are inputted to the system.

The material used for the crash box is steel and the mechanical properties of the steel is listed below,

  1. Density= 0.00783 `(gm)/(mm^3)`
  2. Young's modulus=2.10 e+05 `(gm)/(mm*ms^2)`
  3. Poissions ratio=0.33
  4. Yield strength =310 `(gm)/(mm*ms^2)`
  5. Tangent modulus = 2100 `(gm)/(mm*ms^2)`

The tangent modulus is taken as 1% of Young's modulus.

Rigid wall:

The next step is to create a Rigid wall in front of the crash box 10 mm away by using the *RIGIDWALL_PLANAR card

This is done by selecting Entity Creation > RigidWall > Create. The rigid wall type is selected as a Planar and defined vector with one point and a normal ie, 1n+NL here selected the one node, and the normal is created as a rigid wall. This rigid wall is translated/offset by 10mm from the crash box component.

The white box in the window is a rigid wall and the normal is acting opposite to the global x-axis and it is

denoted in blue color.

The rigid wall is placed correctly in the space where the x coordinates value is entered in a negative direction shows that the normal is acting in opposition to the crash box.

Contact Interface:

The objective of inputting the contact is to see how the nodes are reacting to external forces acting on the body. This gives us the exact solution to the problem. The surface-to-surface contact interface is used for the simulation. The command used for creating the contact interface

Right-side toolbar--->model--->Keyword manager--->Contact--->Automatic-surface to surface type is used.

Create a Contact using

Model and Part--->Keyword Manager--->Contact--->Automatic Single Surface. The slave segment type (SSTYP) is defined by the Part ID, here there is no need to define a master surface so only the slave surface definition is sufficient.

After the automatic contact was created as the crash box elements are getting self contact during crashing ie, Automatic single surface here it was not needed to define the master surface because the crash box itself is master and slave.

Initial Velocity:

The next step is to define a velocity of 50kmph (13.8889 mm/ms) and assigned it to each node of the selected crash box component, under

Right-side toolbar--->model--->Keyword manager--->Initial--->velocity command is used. The Crash box is selected by pick option with respect to part as shown below.

The initial velocity for the assembly is inserted in the x-axis tab in a negative direction shows that it is acting opposite to the global x-axis.

Sectional Force:

In the question, the sectional force for the nodes and elements at the centre of the crash tube along its length is to be recorded and plotted in LS-DYNA. To do that the command used here is

Model--->keyword manager--->database--->Cross_section_Set. command is used. The node set and the shell set is created using the command model--->create entity--->set_data--->set_node/elem. is used.

The approximate centre is considered and the nodes and elements are selected.

Control cards:

The control cards are used to get the output plots and simulation from the

Model--->keyword manager--->database--->ASCII/binary d3plot are used. The list of control cards for the output are listed below,

DATABASE_BINARY_D3PLOT --> Used to find out the animation of the process,

DATABASE_ASCII_NODOUT --> To find the Nodal motions,

DATABASE_ASCII_GLSTAT --> To find the global statistics and energies,

DATABASE_ASCII_MATSUM --> To find the material energies,

DATABASE_ASCII_RCFORC --> To find the Resultant contact interface forces,

DATABASE_ASCII_SLEOUT --> To find the Contact interface energies,

DATABASE_ASCII_SECFORC --> To find the Cross-section forces data.

For generating section forces creating a specific node and element/shell list which defines the desired section of component is necessary. The selection of nodes and elements/shell lists is done by picking each node and element through Creating Entity--->Set Data--->*SET_NODE for node-set and

Creating Entity--->Set Data--->*SET_SHELL for element/shell-set.

The acceleration plot of a node and element of the crash box can be done by picking the desired node and element by using *DATABASE_HISTORY_NODE and *DATABASE_HISTORY_SHELL card.

The termination time of the simulation is set to 3ms. It is set from the command

Model--->keyword manager--->control--->termination.

The Hourglass and the Sliding energy both are recorded and plotted using the command

Model--->keyword manager--->control--->energy.

Then finally the model is checked using the command Model--->keyword manager---> model check command is used.

After assigning all the above-mentioned keywords, the model is ready for the simulation. The keyword file (.k) can be edited in notepad++ so that the comparison study can be done for the part which is having 1.2mm and 1.5mm thickness. Save the keyword for both files and opened in the LS-DYNA manager which solves the problem.

Output requests:

For the output request following files are loaded and displayed below through LS-DYNA.

Energies plot
Cross-sectional force plot
Acceleration plot at the center node
Directional Stress, Strain
Final animation of the simulation.
All the output files are compared with both the 1.2mm and 1.5mm thickness files simultaneously.
OUTPUT FOR 1.2mm THICKNESS:
ENERGY PLOT:
Cross-sectional Force:
Acceleration plot at the center node of the crash box:
Effective stress(von mises):
Effective plastic Strain:
Final animation of the simulation:
OUTPUT FOR 1.5mm THICKNESS:
ENERGY PLOT:
Cross-sectional Force:
Acceleration plot at the center node of the crash box:
Effective stress(von mises):
Effective plastic Strain:
Final animation of the simulation:
OBSERVATION FROM THE RESULTS:
we can able to see that the nodal acceleration in the X-direction is negative since the accelerated body when came to impact there a deceleration happened. Here the value of deceleration in X-direction is nearly the same in both cases and the resultant acceleration is a magnitude and doesn't depend on the direction.
The value of stress in the particular elements that we took for the study exceeds the yield strength in the 1.2mm thickness crash tube than the other one, so there will a permanent deformation occurs in the crash tube with the 1.2mm thickness tube, since the 1.5mm thickness tube resists more load.
The sliding and the Hourglass energy both are zero because due to the contact that we given to the model is automatic single surface contact,(only slave part) is mentioned. Though the master ID is the rigid wall there have no results for the simulation. Hence there is no sliding and Hourglass energy is plotted for both cases.
When the tube hits the rigid wall, the momentum is higher in the 1.5mm thick tube than the other one. In both cases, the law of conservation is satisfied that when the K.E. decreases with an increase in I.E., the TOTAL ENERGY remains constant. But there is small deformation occurs in the system and it is not visible to the system.
CONCLUSION:
The crash tube with a velocity of 50kmph is achieved and the output plots for different cases are analyzed through LS-DYNA. The comparison for the 1.2mm thickness tube and 1.5mm thickness tube are studied and the results are updated above.
LINK FOR THE OUTPUT FILES:
CRASH BOX TEST

Your portfolio, next

Ship work like this
in your own course.

Talk to a career counsellor

Free 20-minute call. We'll map the right course to where you want to work.

By submitting you agree to our Terms and Privacy Policy.

Talk to a career counsellor

Free 20-minute call. We'll map the right course to where you want to work.

By submitting you agree to our Terms and Privacy Policy.