Week - 4 - crash box simulation

AIM AND OBJECTIVE:

To simulate the given crash box using LS-Dyna. The following objectives are to be satisfied:

1. To create a rigid wall using *RIGIDWALL_Keyword

2. The material and thickness to be assigned for the crash tube is steel and 1.2 mm respectively.

3. Intial velocity of crash box is 50 kmph

4. The unit system to be followed is gm-mm-ms

5. Thickness to be changed to 1.5 mm and simulation has to be run.

The following output requests are to given.

1. Input file of .k format and output files (d3plot, glstat, sleout, rcforc)

2. Animation of the final simulation

3. Cross-sectional force generated in the middle of the crashbox (along its length)

4. Acceleration plot of a node in the middle of the crashbox (along its length)

5. Maximum directional stress and strain along the length of the crashbox (X strain, Y strain etc)

6. Plot of all energies (Total, Kinetic, Internal, Hourglass and Sliding)

7. Compare the acceleration and stress/strain plots of 1.2/1.5 mm crashboxes

PROCEDURE:

1. Open LS-Dyna Project Manager>>Start LS Prepost>>Import the given model.

2. Open keyword Manager>>Part>>Define the part with proper name. This is shown in the figure below.

3. In keyword manager, go to material and select *MAT_PIECEWISE_LINEAR_PLASTICITY and assign the values as shown in the figure below. The unit system to be followed is gm-mm-ms as stated in the objective.

4. Now go to Create Entity>>Set Data>>Create Node set and segment set for the part present in the model. This is done to define the velocity and cross-section respectively.

5. Go to Keyword Manager>>Initial Velocity>>Select the defined nodeset and enter a velocity of 13.889 mm/ms. This is shown in the figure below.

6. Go to section>>Shell>>Create a section with 1.2 and 1.5 mm thickness respectively. This is shown in the figure below.

7. Go to create Entity>>Rigidwall>>Type:Planar>>Select the edge node and define the rigid wall. This is shown in the reference image given below.

8. Go to Keyword Manager>>Control>>Energy. Assign the proper values. This is done to obtain the hourglass energy in the plot.

9. Define a control termination card with a total run time of 5 ms.

10. Go to DATABASE>>BINARY_EXTENT>>Assign the values as shown in the figure below. This is done to get the values of strain tensors as a plot.

11. Create a contact interface using the option AUTOMATIC_SINGLE_SURFACE. This is shown in the figure below. The slave set nodes are the nodes corresponding to the crash box.

12. The hourglass energy card is defined below. To reduce the hourglass energy the improved element formulation of  IHQ 4 is assigned.

13. Go to Keyword Manager>>DATABASE>>ASCII_OPTION>>Define the necessary output files with a proper timestep values. This is shown in the figure below.

14. Create a nodeset and element set in the middle of a crashbox to compute the necessary forces. This is shown in the figure below.

15. Go to DATABASE>>CROSS_SECTION_SET. Select the necessary nodeset and element set. This is shown in the figure below.

16. Now go to the PART ID card and assign the material and cross-section accordingly and save the model with a proper extension of .k

17. Go to LS Dyna Manager, run the simulation.

RESULTS AND DISCUSSION:

1. The crash box animation obtained is shown below.

2. The Von-Mises and effective strain contours obtained are shown below.

From the effective strain contour, it is clear that the strain at the sharp corners is higher compared to the horizontal edges.

The elastic strain contours along different axes for the two cases are given below.

Elastic Strain along X-Axis:

The elastic strain along X-axis is slightly higher for 1.5 mm thick crashbox when compared to 1.2 mm thick crashbox.

Elastic Strain along Y-Axis:

The elastic strain along Y-axis is slightly higher for 1.5mm thick crashbox in comparison with 1.2mm thick crashbox.

Elastic Strain along Z-Axis:

For the two cases, the strain along z-axis is negligible when compared to the other two axes. This is because the impact is happening along the x-axis.

3. The energy plot for shell thickness of 1.2mm and 1.5mm are shown in the figures below.

From the above plots, it can be inferred that during the start of the simumation the internal energy remains zero. As the simulation progresses, due to impact in the wall the kinetic energy reduces which causes an increase in internal energy of the crash box. (This is due to conversion of kinetic energy to internal energy). 

The total energy for 1.5 mm thick crash box is higher when compared to 1.2 mm thick crash box. The values obtained for 1.2 mm thick and 1.5 mm thick crash box are 107000 Nm and 134000 Nm respectively.

4. The acceleration plot for node ID 5339 for both the cases are shown in the figure below.

 From the acceleration plots, it can be inferred that till 0.6 ms there was no change in velocity since the crash box was moving towards the wall for the crash. After 0.6 ms, due to impact the acceleration of the plot increases and the crash box rebounds after hitting the rigid wall with a particular velocity. In an overall scheme, there is no change in values between both the cases.

5. The velocity plot for node ID 5339 both the cases are shown in the figure below.

Before the impact, the crash box moves with a constant velocity of 13.88 mm/ms. At 0.6 ms, the impact to the rigid wall happens. Due to this there is a reduction in velocity from 0.6 ms after which the velocity oscillates within 6-7 mm/ms. The same trend is observed in both the cases.

6. The sectional forces for the two cases are shown in the figures below.

From the above plots, it can be inferred that sectional force is higher for 1.2 mm crash box when compared to 1.5 mm thick crash box. The sectional forces are zero till 0.5 ms. After the impact the sectional forces rise to a peak value and after which it gradually reduces.

7. The effective stress plots for the two cases are shown in the figures below.

The plots for effective stress fairly remains the same for the two cases. As discussed in the earlier plots, at 0.6 ms due to impact the stress value rises and after the impact it decreases as shown in the plot.

8. The effective strain plots for the two cases are shown in the figures below.

From the above plots, the strain value increases from 0.6 ms and reaches a maximum value at 0.9 ms after which it remains constant. The peak value of strain remains same for the two cases.

The directional stress and strain obtained along different axes is tabulated below:

 NOTE:

 1. All the time units in the plots are in milliseconds

2. To achieve a crash in the tube, the material properties can be altered. Also, the velocity of impact can be increased to exactly replicate a crash scenario.

3. The above tabulated values gives only the magnitude of stress and strain. It does not state whether it is compressive or tensile forces. These values are obtained by animating the simulation and analysing it frame by frame. The strain values shown here are elastic strain.

4. It is not possible to get accurate maximum directional stress and strain for the overall crash. This is because only an element can be defined and analysed in LS-Dyna. Also, global values are not a major consideration since it does not predict the failure. It simply gives the maximum value. But the purpose of the simulation is to know the deforming region.

5. The sleout data is zero since wall is the only contact interface present. This is shown in the figure below.

6. The rcforce output file is shown below. It remains zero since no self contact is made.

CONCLUSION:

The case setup was done as per the objective and the corresponding output requests are obtained. The two cases were discussed and the plots and results were compared accordingly as given in the problem statement.

Drive Link: https://drive.google.com/drive/folders/15FOZ4NnK5Ebh6KDyM5fJ8myc5cFm_j0c?usp=sharing