FRONTAL CRASH SIMULATION USING RADIOSS

 

 

INTRODUCTION

 

A test that replicates an automobile striking a rigid wall at a specific speed is called a frontal crash with a rigid wall. The test is meant to assess the vehicle's safety features, like seat belts and airbags.

• Euro NCAP: Vehicles are tested at 50 km/h (31 mph) against a structural barrier. A petite female frontal collision dummy is used in the front driver's seat and the back passenger side seat of the test vehicle.
• CMVDR 294: Models a frontal collision at 48.3 km/h with a hard wall.
• IDC-Online: Models a frontal vehicle crash test at 35 mph (≈56 kmph) against a hard wall.

 

A frontal collision occurs when the front ends of two different cars collide while they are moving in opposing directions. The majority of the kinetic energy used in the collision is transformed into the car's internal energy. It is possible to transform other kinetic energy into heat, sound, etc. 

In the NHTSA's front-crash test, a car is driven at 35 mph directly into a rigid barrier, slamming the entire width of the car against the barrier. The degree of impact forces applied to the head, neck, chest, and legs are recorded using instrument-bearing, seat-belted crash test dummies in the two front seats. These measurements are correlated with injury, however the star ranking was previously based solely on the head and chest data. The front passenger and the driver each have their star rating. NHTSA's full-frontal, rigid-barrier test has drawn criticism from certain automotive experts who argue that it is unrealistic since head-on collisions with solid, flat walls are not common. Some contend that whether or not the restraint systems—mainly the safety belts and air bags—work well in the actual world, flat-barrier testing is a useful tool for evaluating their efficacy.

 

 

 

 

 

 

 

 

In contrast to NHTSA front-crash tests, Insurance Institute for Highway Safety (IIHS) front-crash tests emphasize the vehicle's structural integrity and performance concerning seatbelts. Currently, IIHS does two series of front-crash tests: one uses a larger overlap, involving 40% of the car's front, while the other, introduced in 2012, employs a lower overlap, involving only 25% of the front.

Both mimic the scenario of two cars of the same weight colliding and partially overlapping head-on. The area directly ahead of the driver is tested in the older test, which has a 40 per cent offset. The more recent test resembles a head-on collision between two automobiles travelling in opposite directions or a single vehicle colliding with a stationary object, such as a tree or utility pole. These are not at all like the full-width crash that the NHTSA employs. Only the left front of the automobile strikes the barrier in both of the IIHS front-crash scenarios, which employ an impact speed of 40 mph rather than 35 mph. Whereas a stiff barrier is used in the 25% overlap test, a deformable barrier is used in the 40% overlap test. 

As the accident progresses, the vehicle tends to twist around the point of impact due to the new small-overlap test. Because the test involves occupants moving both forward and to the side, it presents additional difficulties for certain safety belts and airbag systems. The side-impact airbags might also need to deploy in this frontal collision. Furthermore, many vehicles are not built to handle an impact that involves a larger area of the front end as well as they do a corner hit. More encroachment into the driver's footwell may result in serious injury to the legs.

Because the collision energy is focused in a smaller region and the speed is higher, the IIHS frontal tests are more rigorous than the NHTSA's. In both cases, forces to the head, neck, chest, legs, and feet are recorded by an instrument-equipped crash dummy situated in the driver's seat. Based on the pressures applied to the dummies and the state of the vehicle's construction, cars are classified as Good, Acceptable, Marginal, or Poor.

 

 

OBJECTIVES

 

Frontal crash-BIW

• Check the unit system and either follow[Mg mm s] or [Kg mm ms].

• Create an appropriate interface, friction 0.2 and recommended parameters.

• Make sure of no penetrations and intersections.

• Correct rigid bodies if any issues.

• Create a rigid wall with friction of 0.1.

• Compare the model weight with the full-scale 300k nodes model and use added masses to reach a target weight of 700kg while getting CG about the required range.

• Initial velocity 35 mph.

• Use a model checker to ensure good quality.

• Time-step:0.5 to 0.1 microseconds.

• Run 80 ms.

 

Output requests:

• Sectional force in the rails at the location of indicated node 174247.

• The axial force received on the rails from the bumper.

• Shotgun cross-sectional forces.

• A pillar cross-section.

• Acceleration curve received on the accelerometer at base of B pillar (on B pillar rocker).

• Intrusions on the dash wall 66695,66244.

 

 

 

PROCEDURE

 

 

•  The frontal car model is imported as (.rad) file format in the imported solver deck tab. The model comprised of around 72 components required for frontal crash analysis in this assignment. The standard unit system of kg-mm-ms-KN-MPa is employed throughout the analysis in Begin_card as shown below.

 

 

 

 

 

• To prevent a group of secondary nodes from penetrating the wall, a stiff wall is a nodal restriction given to them. The secondary node's acceleration and velocity are adjusted if contact is found. An infinite planar surface that is stiff is called an infinite wall. Two points that indicate the stiff wall normally define it. The rigid wall is developed in a YZ-plane and with an allowance of 1mm in X-direction from the front bumper of the car model. Friction of 0.1 is provided to the rigid wall as given in the assignment.

 

 

• Using the find connectivity tool we observe a lot of components are not connected. The inner and outer B-pillar components and some components in the fender area are not connected with the rest of the components of the car model. So the 2 noded spring connectors are used to deploy connections to the corresponding components.

 

 

 

 

 

 

 

 

 

 

 

 

 

• The model checker tool is used to check and rectify any errors found in the model during the pre-processing phase.

 

 

• A type 7 contact is assigned between the car model and the rigid wall with many recommended flags. This interface replicates the broadest range of interactions and effects. A master surface and a group of slave nodes come into contact. It is an unrestricted, quick search algorithm. Interfaces have a gap that controls the moment at which two segments come into touch. Sliding across contact surfaces is possible with a Type 7 interface. The model considers coulomb friction of 0.2 between the surfaces.

 

 

• Penetrations and intersections should be removed from the model with Tools>Penetration Check in Hypermesh.

 

 

• The material used for shell elements in this assignment is LAW2 or the Zerilli-Armstrong Plasticity Model. Similar to the Johnson-Cook plasticity hypothesis is this law. The work-hardening curve is defined by the same set of parameters.

 

 

• This material M21_DPRAG is used to represent materials having internal friction, such as rock-concrete, and is based on the Drücker-Prager yield requirements. The pressure within the material affects how plastically certain materials behave. The sole distinction between this law and /MAT/LAW10 (DRAGP1) is that the pressure is input as a user-defined function of volumetric strain in this law. Only solid elements are compatible with this law.

 

 

• The P14_Solid property is provided to the solid elements and Brick 8 solid formulation is applied. Brick 8 is an eight-node, reduced integration element that uses linear interpolation. integration at one point. Viscous hourglass stabilization (penalty). Typical length, comparatively cheap

 

 

• The P1_Shell property is applied to shell elements with QEPH shell formulation. With one integration point on the surface, QEPH is an enhanced under-integrated four-node element. Stabilization of the hourglass physically. Optimal balance between price and quality. In general, QEPH elements are just 15% more expensive than BT elements, and their outcomes are comparable to QBAT results.

 

 

• The P13_SPR_BEAM property is allocated to 2 noded spring connectors used to connect all components in this assignment. This property is allocated to two or three node elements. It has six degrees of freedom for each node based on the attribute. The behaviour definition for each degree of freedom (three translations and three rotations) is comparable to that of the preceding spring element. A connection exists between shearing and bending. Minimum and maximum uncoupled failure parameters for every direction or coupled failure model. Failure is defined as a function of internal energy, force, or elongation (strain). A sensor can be used to activate or deactivate a spring element.

 

 

 

• Using the solver tab many boundary conditions, and loading conditions are provided. The INVEL card employs an initial velocity of 35 mph which is equal to 15.6464 mm/ms. All the nodes of all components are allocated for the initial velocity of car model towards the rigid wall.

 

 

 

 

 

• The SECT card is created to observe sectional forces as per desired output requests. It is designated by a frame card that determines the local coordinate system of the section plane on a particular area of a component.

 

 

 

 

 

 

• The springs on the dash wall at given nodes 66695 and 66244 are developed. The P4_Spring property is assigned to both springs with very low spring stiffness and mass.

 

 

 

• The non-physical mass (lumped mass) up to a total of 700 kg is allotted to all components uniformly corresponding to the C.O.G. of the car model. The original mass of the model is 187 kg and an additional 513 kg is added to it uniformally. 

 

 

 

 

 

 

 

 

 

 

• In the simulation, acceleration is measured using accelerometer entities. Accelerometer entities in the Radioss interface are specified by a local coordinate system of type SKEW and a node. It is not required to use this coordinate system. Accelerations are measured in the global coordinate system if it is not defined, and in the defined local coordinate system otherwise.

 

 

• TH card describes the request for the time history output. The specific time history plots required for this assignment are displayed below. 

 

 

• In the DT_BRICK, DT-NODA and DT_INTER engine cards, the time step is provided as 0.0005 ms.

 

 

 

 

• ENG_RUN engine card is provided as 80 ms for the termination time.

 

 

• Eventually, the RADIOSS solver is started and the job is completed in 1 hour 18 minutes with effective results.

 

 

 

 

 

 

RESULTS

 

 

 

 

• The von Mises equivalent stress observed during post-processing is 0.31 MPa which exceeds the yield stress i.e 0.19 MPa of the material designated to the car components.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sectional force in the rails at the location of indicated node 174247

 

•  The sectional force of 6.13 KN is acquired on the frontal rail of node 174247 which shows a sudden rise and fall in the curve during crushing at two instantaneous run times.

 

 

 

 

 

 

The axial force received on the rails from the bumper

 

 

 

 

 

 

 

 

 

Shotgun cross-sectional forces.

 

• The maximum right shotgun sectional force is 2.25 KN which is lesser than the left shotgun sectional force i.e 2.61 KN.

 

 

 

 

 

 

 

A pillar cross-section.

 

•  The maximum left A-pillar section force is 0.38 KN which is less than the right A-pillar section force of 0.62 KN.

 

 

 

 

 

 

 

 

Acceleration curve received on the accelerometer at base of B pillar (on B pillar rocker)

 

• The maximum acceleration acquired on the left accelerometer is 3.33 mm/ms² which is less than the right accelerometer of 4.88 mm/ms²

 

 

 

 

 

 

 

 

Intrusions on the dash wall 66695,66244.

 

• The spring element 13776 comprises node 66244 and the spring element 13775 comprises node 66695. The spring elongation of the 13776th element during crash analysis is less than the 13775th element. 

 

 

 

 

 

 

 

 

COMPARISON OF FORCE VS DISPLACEMENT GRAPHS

 

• For additional research, we try to study the force vs displacement of two nodes randomly on the car model during frontal crash analysis. So the total contact force of all car components is treated as the force examined on both nodes. Node 174247 is taken from the inner left front rail and node 41164 is taken from the front solid bumper.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ERRORS

 

The two error cases were examined during solving using a blend of parameters. The errors may vary concerning the design, configuration of the model, type of elements, materials, boundary conditions, loading conditions, analysis type, model size so forth and so on. 

 

FIRST CASE

 

• In this case, the time step for solid elements was too high and the solution and results were not obtained. So the time step was optimized for the given cards within 0.1-0.5 microns.

 

 

 

 

 

SECOND CASE

 

• In this case, the engine cards were duplicated with no values assigned to new cards. Some unnecessary cards were also found that were removed further ahead. It usually happens by frequently importing and exporting the solver decks in the same folder or the model.

 

 

 

 

CONCLUSION

 

1. The output requests obtained for the left and right sections of the same parts had different results due to the unsymmetrical nature of the car model. The C.O.G. evaluated in the car model didn't have the same symmetrical components on all sides. So the mass distribution also varies corresponding to the type of component used in the car model.
2. The force vs displacement graphs give an overview of the magnitude of force and displacement respective to the position, shape size and other factors of the component during a frontal crash. 
3. To get effective results, the nodes of the identical sorts of components for output requests on the left and right portions were aligned perpendicularly.
4. The 700 kg target for lumped mass addition was established by distributing it evenly among all components of the car model.
5.  The initial velocity drop from 15.64 mm/ms to 10.2 mm/ms was examined at the end of the simulation for the car model.
6. The kinetic energy and internal energy showed signs of convergence if the run time was increased to 100 ms.
7. A negligible amount of hourglass energy was monitored post-simulation.