Challenge

Week - 4 - Crash Box Simulation

AAsad K P28 Apr 2024

Week 4: Crash Box Simulation

Aim

To conduct a crash test simulation for a crash box, an energy-absorbing component designed using the mentioned criteria, observe and analyze how varying thicknesses affect the behaviour of the crash box and the impact of thickness variations on the crash box's performance.

The specifications and assumptions for the crash box simulation include the following

The crash box is designed as a simple rectangular channel, considered the most basic form of a crash box.
The crash box is made of steel, and the simulation will initially use an elastic material model for correctness verification. A nonlinear material card data will be provided later.
The crash box will crash into a rigid wall created using the keyword, 'RIGIDWALL'.
The initial velocity of the crash box is approximately 50 kmph.
The unit system used for the simulation is gm-mm-ms.
The simulation should satisfy one of two conditions:

a) The crash box crashes on the rigid wall and rebounds completely, or

b) The crash box buckles and folds in the axial direction, coming into self-contact.

After the first simulation is completed and the results are collected, the thickness of the crash box will be increased from 1.2mm to 1.5mm for the second simulation. The results of the two simulations will then be compared.

Output Required

The following deliveries are needed from the simulation:

Input file (Keyword File).
Output files (D3PLOT, GLSTAT, SLEOUT, RCFORC).
Final simulation animation.
The cross-sectional force, along its length, is generated in the middle of the crash box.
The acceleration plot of a node, along its length, is in the middle of the crash box.
The maximum directional stress and strain along the length of a crash box, such as X-strain and Y-strain.
The energy plots comprise Total Energy, Kinetic Energy, Internal Energy, Hourglass Energy, and Sliding Energy.
The comparison between the acceleration and stress-strain plots for the crash box with 1.2mm and 1.5mm thickness.

Procedure

Step 01: Importing And Inspecting The Model

Download the keyword file and save it in a suitable location. Open LS-Prepost and load the keyword file (.K format).

File >> Open >> LS-DYNA Keyword File >> Select The File >> Open.

Open the keyword manager and understand the model with the assistance of the keyword manager.

Step 02: Defining Rigid Wall

For the crash simulation, a rigid wall is required where the crash box will be impacted. The 'create entity' tool can be utilized to define a plane rigid wall. Please note the direction of the head points of the rigid wall.

Model And Part >> Create Entity >> Rigid Wall >> Rigid Wall >> Cre >> Planar >> GeoVector >> 1n+NL (Define vector with 1 node and normal) >> Enable Pick Node and select a node in the edge of the crash box >> Increase 10 mm to the 'x' value. >> Change the x coordinate of the head point to its negative value. >> Apply >> Done.

We can modify the dimensions of the rigid wall using the 'Mod' tab or the keyword manager. This rigid wall is placed with a gap of 10mm to the x-axis with the crash box.

Step 03: Assigning The Initial Velocity

Considering the question, the crash box has an initial velocity of 50 kmph. If we convert to the 'gm mm ms' unit system the value = `50 xx5/18=13.888 (mm)/(ms)`. Therefore, the crash box is moving at a velocity of 4.166 mm/ms in the direction of the x-axis.

Model And Part >> Create Entity >> Initial >> Velocity >> Cre >> Enable CreSet >> Using Area > Bynode Selection, Select All The Nodes In PID 1 >> Vx = 4.166, all other velocity values (Vy, Vz, Vrx, Vry and Vrz) is equal to zero >> Apply >> Done.

Step 04: Defining Sections

Creating sections will help us determine the internal forces acting on the member. Create a section, with a thickness of 1.2mm. To create the shell section with default values,

Model And Part >> Keyword Manager >> All >> Section >> Shell >> Give Suitable Title >> SECID = 1 >> T1=1.2 >> Maintain other values at their default settings >> Accept >> Done

Step 05: Defining Material Card.

The question indicates that the crash box is constructed from steel. Therefore, a material card with steel properties must be defined. For this purpose, MAT-024 will be utilized. The material properties are based on ASTM A36 Steel.

`rho = 7.85xx10^-3 (gm)/(mm^3)`
`E = 200 xx10^3 (gm)/(mm ms^3)`
`PR = 0.26`
`SIGY = 250 (gm)/(mm ms^3)`
`ETAN = 25xx10^3 (gm)/(mm ms^3`

Model And Part >> Keyword Manager >> All >> MAT >> 024-Piecewise_Linear_Plasticity >> Give Suitable Title >>Give Appropriate Material Values >> Maintain other values at their default settings >> Accept >> Done.

Step 06: Defining Hourglass Card.

The hourglass card manages the formation of the hourglass energy during the simulation.

Model And Part >> Keyword Manager >> All >> Hourglass >> Give Suitable Title >> HGID=1 >> IHQ=2 (Flanagan Belytschko Viscous Form) >> Maintain other values at their default settings >> Accept >> Done.

Step 07: Assigning Section, Hourglass And Material To The Crash Box.

For the crash box component, we must allocate the shell section and designate it as MAT-024. To accomplish this,

Model And Part >> Keyword Manager >> Model >> Part >> >> MID = 1 (For MAT-24) >> SECID = 1 (For Thickness 1.2mm)>> HGID=1 (For Hourglass) >> Maintain other values at their default settings >> Accept >> Done.

Step 08: Defining Contacts

Here we need to define an 'automatic single surface' contact among the elements of the crash box. It can be done by,

Model And Part >> Keyword Manager >> All >> Contact >> Automatic_Single_Surface >> Give Suitable Title >> CID =1 >> SURFATYP = 3 (For Selecting Part ID as Slave) >> SURFBTYP = 0 (Not Applicable For Self Contact) >> SURFA = 1 (Part ID Of Crash Box As Slave) >> SURFB = 0 (Not Applicable For Self Contact) >> Maintain other values at their default settings >> Accept >> Done.

Step 09: Defining Control Cards

Initially, it is necessary to define a control card for the termination. The end time will be 5ms

Model And Part >> Keyword Manager >> All >> Control >> Termination >> ENDTIM =05 >> Maintain other values at their default settings >> Accept >> Done.

Next, following the same method, we should design a card for the hourglass control.

Model And Part >> Keyword Manager >> All >> Control >> Hourglass >> IHQ=1 >> Maintain other values at their default settings >> Accept >> Done.

Similarly, we need to assign the control card for energy. For computing the hourglass energy, we need to assign the value HGEN=2

Model And Part >> Keyword Manager >> All >> Control >> Energy >> HGEN=2 >> Maintain other values at their default settings >> Accept >> Done.

Step 10: Creating Cross Sections

Capturing the cross-sectional force is necessary. To achieve this, a cross-section should be created along the length and situated in the middle region of the crash box.

Model And Part >> Create Entity >> Database >> Cross-Section >> Cre >> > Planae >> Give title >> 1n+NL (Define vector with 1 node and normal) >> Pick a node in the mid portion of crash box >> Apply >> Done.

Step 11: Requesting Outputs

As per the objective, we need D3PLOT as a binary output. To enable that,

Model And Part >> Keyword Manager >> All >> Database >> Binary_D3PLOT >> DT=0.05 >> Maintain other values at their default settings >> Accept >> Done.

To obtain ASCII outputs, it is necessary to request the outputs for the GLSTAT, MATSUM, SECFORC, SPCFORC, SLEOUT, and RCFORC databases. Also, enable NODOUT output for finding the nodal accelerations.

Model And Part >> Keyword Manager >> All >> Database >> ASCII_Option >>Select The Database Cards >> DT=0.05 >> Maintain other values at their default settings >> Accept >> Done.

Also, we need to enable the History_Node database to access the nodal properties. Here we take the node ID 5555 for the analysis. For that,

Model And Part >> Keyword Manager >> All >> Database >> History_Node >> Enter node ID, ID1 =5555 >> Accept >> Done.

Step 12: Model Checking And Simulation For Each Case

Model checking should be finalized, and any errors identified within the model must be resolved.

The model features a thickness of 1.2mm. Proceed to save the file using an appropriate name, including the '.K' extension. Then Run the simulation.

Similarly, change the thickness of the shell section to 1.5mm. Then save the file with a different name with the '.K' extension in a different location and rerun the simulation.

Results

We need to evaluate Model 1 with a 1.2mm specification against Model 2 with a 1.5mm specification using simulation results, which may include animations or plots.

1) Model 1- Thickness 1.2mm

1. Directional Stresses Along X Axes, Y Axes And Von-Mises Stress

The stress acting on the x-axis is depicted in the images below.

The simulation reveals that at the moment of contact between the rigid wall and the crash box, the elements closer to the wall experience a peak compressive stress of 409.75 MPa at 0.79947 seconds. The first figure illustrates that the impacted elements are dispersed mainly across the corners of the crash box.
Subsequently, the area located just above the middle section of the crash box experiences a peak tensile stress of 196.91 MPa at 0.84977. The second figure illustrates the affected elements distributed widthwise across the crash box, which possesses a narrow width.

The stress acting on the y-axis is illustrated in the images below.

The maximum value compressive stress occurs on y axes 131.56 MPa at 0.94978 ms and tensile stress of 109.91 MPa at 1.0497 ms. Both events take place after the rebound.
The regions experiencing the maximum compressive stress are the upper sections of the larger sides of the crash box, as well as a small area where contact occurred. Similarly, maximum tensile stress acted at the upper sections of the larger sides of the crash box.

The effective stress can be represented using the Von Mises stress.

The value of maximum Von Mises stress is equal to 399.98 MPa. The stress distribution from the crash is observable throughout the entire model, extending from the bottom to the upper section of the box.

2. Strain Along X Axes, Y Axes And Von Mises Strain.

The strain along the X and Y axes, as well as the effective strain (Von Mises), are depicted below.

The simulation indicates that the Von Mises strain value remains at zero, suggesting no deformation in the crash box during the impact. A zero deformation implies that there is no strain.

3. Global Energies

The various energy plots and their observations are mentioned below.

The KE reaches a maximum value of 1.08e5 within the first 0.699 seconds, which is attributable to the initial velocity of the crash box. When the contact occurs, KE is converted to PE, so there is a rapid decrease in KE down to (1.45e3) due to this impact. Then it rebounds, and that velocity results in a slight increase in KE (1.92e6), after which it attempts to maintain that energy throughout the simulation.
In the IE, the opposite of what occurs above takes place. The IE remains zero until the impact begins. At the time of the collision, the IE instantly increased its value to the maximum value (1.076e5) and The value then rapidly decreased to 8.82e4, and the IE continued up to the end of the simulation.
Hourglass energy (HE) is not produced until the onset of the crash. The impact initiates the HE and gradually increases the value (more of a parabolic increase) (0.002e5) until the end of the simulation.
The entire crash's Total Energy (TE) initiates at a value of 1.08e5. As the crash commences, there is a slight increase in energy, followed by a gradual decline.
The sliding energy is zero, indicating that no sliding or friction occurs.

4. Cross-Sectional Force

The sectional force for the mid-section of the crash box is illustrated below.

The cross-sectional force in the x-direction is significantly higher than the forces on the y and z axes. The resultant forces remain at zero until the impact occurs. At the moment of impact, they spike sharply before declining abruptly and then fluctuating continuously.

5. Nodal Accelerations

The nodal accelerations for the mid-section of node ID 5555 in the crash box are illustrated below.

The nodal accelerations in the x-direction are significantly higher than on the y and z axes. The resultant accelerations stay at zero until the impact takes place. There is a sharp spike to approximately 1.2e3 at the instant of impact, followed by a sudden decline and then continuous fluctuations. The peak acceleration happens at about 1.75ms, reaching up to 1.4e3 mm/ms².

1) Model 2- Thickness 1.5mm