Assignment 6-Frontal Crash Simulation Challenge

Objective:

To simulate a frontal crash of the given BIW and analyse the results for the various important result points of the front part of the car.

Step 1: Setting up the Simulation - Hypermesh

The given file is imported into Hypermesh and the various components of the model can be viewed as shown below.

The unit system used in the model can be viewed using the BEGIN card. The units are kg-mm-ms.

Interface creation for self contact - Type 7

The type 7 contact is defined for all surfaces as shown below.

The type 7 card is defined with all recommended parameters as seen earlier.

Istf = 4

Igap = 2

Idel = 2

Stmin = 1.0

Fric = 0.2

Gapmin = 0.5

Inacti = 6

Iform = 2

Rigid wall creation

A rigid wall simulating the crash barrier is created normal to the frontal plane of the car with the given parameters as shown below.

The search distance Dsearch = 1000 and the friction value FRIC is specified as 0.1, as given in the problem statement.

Mass and centre of gravity

The initial mass and the cg location of the model can be viewed using Tools --> Mass details.

The initial mass is 188.42 kg and the CG location can be viewed according to the co-ordinates specified as shown below.

The node specifies the initial CG location. This needs to be moved rearward to the point (approx) specified by the red circle, and the mass needs to be increased to meet the target mass of 700 kg.

This can be done by using added mass to increase the mass at the rearwards portion of the model as shown below.

A mass of 10 kg is added to each node for a total of 51 nodes symmetrically on either side of the longitudinal vehicle axis as shown above. Therefore the total mass added is 510 kg.

The final mass and the CG location of the model after the added mass of 510 kg is 698.42 kg as shown below.

The CG is now at the required location.

Initial velocity

An initial velocity of 35 mph is assigned to all nodes of the model as shown below. Since the unit system used is kg-mm-ms, the velocity is 15.43 mm/ms.

Cross - sections

In order to view the forces at important points across the body, cross sectional planes are created at 8 locations as shown below. These are

A pillar - Left and right

Rail - Left and right

HL bracket - Left and right

Shotgun- Left and right

The local frame of all the cross-sections, located at the centre of gravity of each section can be seen in the figure above.

Intrusion - Dash wall 

The intrusion on the two given nodes of the dash wall with ID 66695 and 66244 can be seen, by measuring their displacement relative to a skew system as shown in the figures below.

Accelerometer - B pillar

The acceleration at the base of the B-pillar can be measured by placing accelerometers on either side as shown below.

The model setup is now complete and it can now be simulated to view the results.

The runtime is 80 ms and the frequency of animation plots is 5 ms.

Step 2: Simulation results

The simulation is completed and the key run parameters such as number of cycles and simulation time can be viewed in the 0001.out file.

Mass and energy error

The final mass error and energy error as given in the 0001.out file is shown below.

The final energy error is -1.1% and the final mass error is 0.3028e-1 or 3%, which is less than the error threshold of 5%.

Therefore, both the mass and the energy error are within acceptable limits.

Displacement and Von mises stress

The displacement and von mises stress values can be observed in Hyperview using the h3d file as shown below.

Energy plot

The energy conservation can be checked by plotting data from the time history T01 file in Hypergraph 2D as shown below.

As the simulation proceeds, the kinetic energy decreases and the internal energy, which is representative of the energy absorbed during the crash, increases. The total translational energy TTE, which is the sum of all the energies has a peak value of 85171.89.

Sectional forces

  Rail

The forces transmitted to the rail are higher on the right hand side initially, since that side has a more rapid deformation compared to the left side. The left side deforms more gradually and eventually has more force after a time of 45 ms.

The peak rail  forces are

Left - 13949 N

Right - 7316.5 N

Bumper - Rail contact

The force on the right side reaches a peak at around 6 ms, since the deformation is initially rapid on that side. The force on the left increases more gradually and shows lesser fluctuations, compared to the force on the other side.

The peak bumper-rail contact forces are

Left - 5128.5 N

Right - 6860.03 N

Shotgun

The force on the left fender reacher a peak at around 38 ms, after which it falls away rapidly as it has deformed a lot. The right side has a more gradual increase, reaching a peak value at around the same time, before falling off gradually.

The peak shotgun forces are

Left - 8255.2 N

Right - 7025.6 N

A pillar

The A pillar on the left side has more force than the right, which is also evident from the displacement and stress animations, which illustrate that the components on the left side in front of the A pillar (rail, fender etc) deform lesser than on the right side.

The drop-off in the left side can be very well correlated to the force plots from the rail and the bumper cases, where most of the energy on the left side has already been absorbed, which leads to the reduced forces being transmitted to the A-pilllar on the left.

The peak A pillar forces are

Left - 5189.7 N

Right - 2075.8 N

Intrusion - Dashboard wall

The intrusions on the given nodes can be viewed in the plots shown below.

The displacements are

Node 66244 - 834 mm

Node 66695 - 845 mm

The velocity plot shows the resultant velocity decreasing from an initial value of 15.43 mm/ms to around 7m/ms at the end of the crash event.

Acceleration - Accelerometer

The accelerometers placed on either side at the base of the B pillar have the following accelerations.

The peak accelerations are

Left - 1.9 m/s2

Right - 2.05 m/s2

Conclusion

The model has been simulated for a frontal crash against a rigid wall and the key results have been discussed. It can be observed that there is never a uniform deformation in the left and right sides of the vehicle, since the masses are slightly different and the component placement plays a huge role in deciding the extent of stress, deformation and forces and energy absorbed in the crash.

Simulation of the original (300K) model would result in more accurate simulation results.