Car crash analysis 2004 - 2005

Car crash analysis

This problem will evaluate the potential injuries sustained during vehicle crash
using a passive human model . The human model will be combined with an ADAMS
model of a car interior or a test sled. The model will be restrained using forces to
simulate the seat/shoulder belts. The sled is then subjected to an acceleration pulse from
the rear, and in another simulation one from the front. Injury potential will be evaluated
by examining the segment impact forces, the segment accelerations and the loads at the
anatomical joints.

Features of this example problem include:

* Creating body segments from the UK People Size Anthropometric database

* Contact elements between body segments and environment
* Hybrid III Crash Dummy strength characteristics at the joints
* Posture manipulation
* Passive simulation
* Combining human model with ADAMS environment model
* Attaching the human model to the environment with seat belt forces.
* Comparing the results from different analyses.

Sections

* Generating the Body Segments

G.V.I.T 8th Semester Mechanical Page 1

Car crash analysis 2004 - 2005

* Generating the Joints

* Posing the Human Model
* Creating the Car Interior Model
* Creating Contact Forces
* Running the Equilibrium Analysis
* Creating the Seatbelt Forces
* Creating the Crash Pulse
* Running the Simulation for the Forward Hit
* Running the Simulation for the Rearward Hit
* Interrogating the Results

This Audi A8 car-crash model contains numerous materials and structural

components modeled by 290,000 finite elements . The model predicts the extent of
deformation in the car after a crash.

The lighter the structural materials that make up a car, the less fuel it requires per
mile of travel. But, if a car is lighter than the steel cars most of us drive, will it hold up as
well in a crash? To answer this question, the Computational Material Sciences Group in
ORNL’s Computer Science and Mathematics Division has completed a computational
simulation of an all-aluminum Audi A8 car crashing against a rigid barrier at 35 miles
per hour (mph). The group, led by Srdan Simunovic, built the computer model after
disassembling an Audi A8 car and scanning its structural components into a computer.
The model contains equations and numbers representing approximately 290,000 finite
elements and 200 different material components. The model has been run on the IBM
RS/6000 SP supercomputer at DOE’s Center for Computational Sciences at ORNL.

While the ORNL supercomputer was crunching the numbers, an actual crash test
of an Audi A8 car was performed by the National Highway Traffic Safety
Administration (NHTSA). The test results were used to tune the computer model and
determine whether its predictions about the ex-tent of deformation throughout the car
were correct.

G.V.I.T 8th Semester Mechanical Page 2

Car crash analysis 2004 - 2005

The deformation predicted by the model was verified against the crash test data.
That suggests that this model can be used as a low-cost method for testing new design
concepts and materials-processing technologies without the need for building and
crashing expensive prototypes. The model can be used to test crashworthiness and
analyze stiffness of structural components.

The ORNL group also has developed computer models of vehicles whose bodies
are made of regular steel and high-strength steel. With funding from NHTSA, the group
recently developed detailed computer models of the Ford Explorer. One material model
predicts how the Explorer’s body material will behave as the vehicle collides from
different angles with a rigid barrier at 35 mph.

The group were also working on the computational analysis of a concept car
made of high-strength steel. The high-strength-steel, UltraLight Steel Auto Body
(ULSAB) design and the computational crash models were developed by Porsche
Engineering Services, Inc., for the ULSAB Consortium. They used the models to predict
the effects on new advanced materials of various collisions, such as two cars colliding
with each other. Because these new steel alloys have such high strength, less steel is
needed for the body of the car, making it lighter. They found that the ability of the high-
strength steel vehicles to hold up in a crash can be even better than that of today’s
heavier steel vehicles.

To explain what happens when cars made of metal crash, Simunovic squeezes an
empty beverage can until it folds and collapses like an accordion. This is what we want
to happen to a car during a collision with a rigid barrier or another car. Metals tend to
bend and deform as they absorb the energy of the impact. It is this simple plasticity of
metals in response to sudden impacts that we can simulate using our materials modeling
codes.

For Simunovic an even bigger challenge is modeling fiber-reinforced polymer

composites, a project he has been working on since 1993. These composites, which are
lighter than steel and aluminum, consist of glass or carbon fibers embedded in a polymer

G.V.I.T 8th Semester Mechanical Page 3

Car crash analysis 2004 - 2005

matrix.

They are developing constitutive models to predict how the material will behave
during an impact at 35 mph. Composites don’t act like metals and dissipate energy by
bending and deforming plastically in response to a blow. Although composites have
higher specific strength, they tend to be brittle, making them less likely to give as easily.
They are more likely to shatter like glass.

The impact could cause fibers to break away, or de-bond, from the polymer
matrix. The goal is to develop a composite that exhibits controlled progressive fracture
during impact. Such a material could dissipate a large amount of impact energy and
gradually decelerate the vehicle. We must learn how to model these effects and
accurately predict how they improve the ability of the material to resist breaking
catastrophically in a crash.

For computer simulations of crashes involving cars made of carbon-fiber

composites, the ORNL group will use data from the intermediate strain rate crush test
station, which was installed in 2002 at the National Transportation Research Center,
where ORNL and University of Tennessee researchers work. The station will compress
samples at speeds up to 15 mph, providing information on changes in the number of
small and long cracks produced as the impact velocity varies.

There goal was to provide the material models and computational tools that
designers need to develop highly efficient, low-emission, lightweight vehicles that have
improved safety features.

In the world today, computers are used in every field. Be it a major space
exploration or a small chore like cleaning our room. The use of computers has made our
lives easier but at the same time a computer failure can make our lives miserable too.
We trust computers more than we trust anything else these days. We use computers to
communicate, share personal information, buy goods online, etc. We also trust
computers with our safety. Starting from house alarm systems and other safety devices

G.V.I.T 8th Semester Mechanical Page 4

Car crash analysis 2004 - 2005

we have now moved on to using computers to Test Car Crashes.

Car crash analysis programs gained wide usage by the late 1980s but ARA
(Applied Research Associates) Personnel in the Silicon Valley Office have been engaged
in studying the crash response of vehicles, occupant safety, and right-of-way structures
since 1971. One of the major programs used for this testing is the DYNA3D which was
developed at Lawrence Livermore National Laboratory.

DYNA3D is a computer simulation program that models the interactions of

physical objects on impact such as vehicle impacts involving roadside structures such as
signs, supports, guardrails and crash cushions. DYNA3D, suitable for solving problems
involving rapid change, has had many applications in safety analysis. Laboratory analysts
have used DYNA3D to study crashworthiness in a number of vehicle safety studies,
where models of complex vehicles impact roadside safety structures and other vehicles,
deforming under the impact. The DYNA3D program uses a technique called the finite-
element method where a grid is superimposed on the frame of a car dividing the car into
a finite number of small pieces or elements. The grid is then entered into the program
along with data describing the specifications of the materials making up each element
such as density, elasticity, etc. While reading the effect of a head-on collision on the
structure of the car, the data can be initialized to represent a crash into a wall at a
specified speed. The program in return helps compute the force, acceleration, and
displacement at each grid point and the stress and strain within each element. Using
graphics programs, the simulation produces a picture of the car at intervals after impact.

Roger Bligh, manager of TTI's (Texas Transportation Institute) Highway Safety

Structures Program, says that computer program is more efficient than traditional crash
testing. The biggest advantage DYNA3D has over traditional crash testing is its cost-
effectiveness. With the program, we can look at a variety of scenarios that we typically
can't study by traditional crash testing methods. We're often technologically or
monetarily limited with what we can do in terms of real-world testing, but we're not

G.V.I.T 8th Semester Mechanical Page 5

Car crash analysis 2004 - 2005

limited by such constraints on the computer.

Using the DYNA3D program and models, users have the flexibility to simulate a
crash from different angles and speeds without the cost of repairing or replacing the
structure or vehicle for each test. The program can be used to simulate various real-world
conditions like roadside terrain. Researchers can also graphically view these impacts
from different angles to more fully study how the structure or vehicle reacts. They can
delete elements, such as the roof or hood of a car, to see how internal elements are
affected during the impact—an effect that would be difficult to study in an actual crash
test. Although it's not feasible to test all of these combinations in actual tests,
simulations allow more design variations to be analyzed without significant additional
costs. The cost of a real crash test can range from $50,000 to $800,000. The high figure
is for building and testing a unique prototype for a new car design . With these results,
researchers can make adjustments to existing structures or develop improved structures.
As safety standards in the transportation industry continually change to reflect new
vehicle designs and operating speeds, the simulation program can help find a means for
developing improved safety features or upgrading existing roadside structures to meet the
new standards. But still there is a long way to go. The sophistication and accuracy of
these models have to be increased so that they better generate what really happens during
a crash.

Now let’s look at the program from a skeptic’s point of view and see how
accurate and complete the data are that the program inserts. Force and acceleration are
basic principles so they are pretty accurate. As for the properties of the materials such as
steel, glass, aluminum, etc that is well known too. But the behavior of some of these
materials under abrupt conditions and at a high speed impact is less understood. As for
the making up of the car in blocks that can cause some trouble too. As a car is smooth
and not made up of blocks so the impact of a smooth object may be different than an
impact of an object made up of blocks or grids. These are the only problems that the
DYNA3D program is facing right now. If they can fix up these problems then more
people would be able to trust their Car Crash Analysis.

G.V.I.T 8th Semester Mechanical Page 6

Car crash analysis 2004 - 2005

As for the computed results that this program generates in comparison to actual
crash tests on real cars, the results from the program correspond very closely to data
collected in actual crash tests. That is why car crash analysis programs are replacing
physical crash testing as a design tool for new cars, but this does not mean that physical
crashes are being eliminated completely. Crash design programs are excellent design
tools that enable increases in safety with far less development cost. We are putting our
trusts in computers to help us make safer cars and at the moment it is doing a really good
job.

Automobile CAE Software is mainly used to assess the performance quality of

vehicles. As the automobile is a product of technology -intensive complexity, its design
analysis involves a broad range of CAE simulation technique. An integrate CAE solution
of automobiles can include comfort analysis (vibration and noise analysis), safety
analysis (car body collision analysis), process-cycle analysis, structural analysis, fatigue
analysis, fluid dynamics analysis, test analysis, material data information system and
system integration. We put an emphasis on simulation of a whole automobile collision
process, which will bring a breakthrough to the techniques of CAE simulation based on
high performance computing. In addition, we carry out simulation for a finite-element
car model in a distributed computation environment and accomplish coding-and-
programming of DYNA3D is accomplished. We also provide computational examples
and user handbook. Our research collects almost ten numerical automobile models such
as Honda, Ford and etc. Moreover, we also deal with different computational scales for
the same auto model and some numerical model of air bag is included. Based on the
numerical auto model, referring to different physical parameters and work conditions of
auto model, we can control physical parameters for the numerical bump simulation and
analyze the work condition. The result of our attempt conduces to the development of
new auto models.

High Speed Race Car Collision with a Solid Wall

The results are typical of a high-speed Indy car crash into a solid barrier. The

G.V.I.T 8th Semester Mechanical Page 7

Car crash analysis 2004 - 2005

driver sustained no injuries. The analysis is based on a video film of the general vehicle
path and crash dynamics of the on-board IST data recorder.

The data shown is a resultant of the combination of longitudinal, lateral and vertical
accelerations of the car produced by its interaction w ith the wall. The acceleration is
plotted in g versus time (in milliseconds). The sequence of events was as follows:

Event A : The rear of the gearbox strikes the wall, producing a longitudinal acceleration
of short duration (10 ms) with a 56g peak. The right rear wheel strikes the wall and the
suspension crushes, transmitting the impact to the chassis, producing the 10-20G
acceleration until...

Event B : The suspension (tire) bottoms out against the chassis and produces a peak

G.V.I.T 8th Semester Mechanical Page 8

Car crash analysis 2004 - 2005

acceleration of 39g, which is followed by a rebound of the car from the wall and a
clockwise rotation of the car. The sequence from A to B takes 0.05 seconds.

Event C : While the car rotates, a longitudinal acceleration of about 10g is indicated by
the spinning of the car about its center of mass. This acceleration, which appears as a
steady state value for about 0.125 second, is probably not longitudinal, but centrifugal
acceleration due to the location of the transducer not being at the center of the rotation. If
we can infer that the car rotated about 90 degrees in that time, then the angular rate
would be 0.5 seconds per revolution. Using the equation: g = 1.22 x radius/time squared,
the radius of rotation would be about 2 feet.

Event D: The care strikes its right front wheel against the wall resulting in a peak lateral
acceleration of 55g and a duration of 0.015 second. The car continues to rotate until...

Event E : The nose and side structure impact the wall producing a peak lateral
acceleration of 37G with a duration of 0.030 seconds. The front of the car then rebounds
off the wall and ...

Event F : The car continues the spinning motion down the track.

Crush simulation of cars with FEA

Finite-element analysis procedures are now used abundantly in the automotive

industry. Linear static and dynamic analyses are conducted in a routine manner, and
nonlinear analyses are increasingly pursued.

Two analysis fields in which highly nonlinear conditions are simulated are the
crash and crush analyses of complete motorcar models. The purpose of a crash analysis is
to see how the car will behave in a frontal or sideways collision. In a crash analysis, the
crashing of a car at about 30 mph into a rigid wall is simulated. Various crash codes,
such as LSDYNA, PAMCRASH, and RADIOSS, are used. The codes have been

G.V.I.T 8th Semester Mechanical Page 9

Car crash analysis 2004 - 2005

developed based on explicit time integration, special shell elements for this specific
analysis, and modeling assumptions regarding the dynamic behavior involved. The
analysis results have been compared with laboratory test data, and the simulations have
proved very valuable. Physical speed of crushing is 10 mm/sec. (about 0.022 mph).
ADINA implicit dynamic results at 10 mm/sec. and LSDYNA explicit dynamic results.

In a crush analysis, a quite different physical phenomenon is considered. Here the

purpose is to establish the ultimate strength of the car body in a static situation. The
ultimate strength affects the behavior of the car under various operating conditions, such
as when the car overturns in an accident. The laboratory experiment to identify crush
behavior is performed by crushing the car slowly (at about 0.02 mph), using a device to
push a thick steel plate onto the car roof and measuring the load-deformation relation.

While crash analyses of cars have been carried out with much success, a crush
analysis is much more difficult to achieve. The reasons for this greater difficulty lie in
the fact that a slow-speed, almost static analysis requires increased robustness and
efficiency in the solution algorithms. Specifically, for the crush analysis, the shell
elements must be of high predictive capability, and be robust and computationally
efficient for static analysis. The contact algorithm must allow three-dimensional
multiple-body and self-body contact on the outer and inner surfaces of the car shell, be
robust, and give fast convergence in the iterations for static equilibrium at the different
deformation states.

In a crash analysis, the inertia effects "smooth out" the nonlinearities and
deficiencies in the solution algorithms. In addition, explicit time integration is usually
employed, which means that no iteration is used in the step-by-step solution (as is
required in a static nonlinear analysis). A simple time-marching-forward solution is
produced. This analysis procedure is attractive because difficulties with respect to
convergence in equilibrium iterations do not exist.

G.V.I.T 8th Semester Mechanical Page 10

Car crash analysis 2004 - 2005

In a static nonlinear analysis, iteration for equilibrium is required in each load

step. In a dynamic nonlinear analysis using implicit time integration, which reduces to a
static analysis if the time scale is long, the same iterations are performed, but inertia
forces for each time step are included. While no iterations are performed in an explicit
time integration, the time step used in the solution must be smaller than a critical time
step, for the solution to be stable. The critical time step varies during the solution
process, because it depends on the geometry and material conditions that change during
the analysis history. If a step larger than the critical time step is used for only a few
solution steps, "merely" a significant error is accumulated in the analysis. However, if the
time step size continues for more than just a few steps to be larger than the critical time
step, the solution errors grow to become extremely large, indicating the numerical
instability of the solution.

For a crush analysis, it may appear that a natural way to obtain the solution is to
use the available crash analysis codes. However, there are difficulties with respect to the
solution cost and the accuracy of the analysis results.

In a crush test, it takes about 10 to 30 seconds to crush the car to the required
maximum displacement of the steel plate. Since the critical time step for explicit time
integration in a crash code is on the order of microseconds, millions of time steps must
be used to perform the analysis in a physically correct manner. The computational time
for such a solution is very high, and ways have been sought to reduce the required
number of time steps for analysis. In one approach, the speed at which the steel plate is
applied is artificially increased. However, inertia effects then become important,
resulting in an artificial increase in the computed crushing force. In another somewhat
equivalent approach, the density of the material is artificially increased, again resulting in
a higher computed crushing force. Given such results, numerical experimentation is
required to assess the effect of changing the model and the load application.

All in all, these approaches can hardly be recommended. Instead, if explicit time
integration is used and reliable results are required, the actual physical conditions should

G.V.I.T 8th Semester Mechanical Page 11

Car crash analysis 2004 - 2005

be represented. This requires a huge number of solution steps and hence very large
computation times.

The actual physical condition is that the speed of load application (through the
rigid plate crushing the car) is 10 mm/second (about 0.022 mph). The ADINA solution
was obtained at that speed of load application using implicit time integration, which
corresponds in essence to an incremental static solution.

Crush analysis of Taurus car model. Comparison of ADINA implicit

dynamic results and laboratory test data (all at 10 mm/sec., i.e. about 0.022
mph).

The LSDYNA explicit time integration for the 0.05-mph speed required much
more computer time, a computer run of weeks instead of the overnight run with adina.
And the LSDYNA results are questionable because of the artificial oscillations in the
computed response—the response should be static. If the load application speed is
increased, the inertia effects reduce the response oscillations and increase the predicted
collapse load, but it is virtually impossible to predict prior to the analysis which artificial
speed of load application should be used. For example, at a speed of 10 mph, due to the
inertia forces, the predicted crushing force is much too large. With sufficient numerical
experimentation, involving changes to the load application speed and perhaps to other
parameters, LSDYNA results can be obtained that would match laboratory test results
(which, unfortunately, are not available for this car), but such experimentation requires a
lot of time and computational effort.

It is quite obvious that a solution that corresponds to the actual physical

conditions and is computationally efficient is much more desirable. Such a computed
solution is given above for a Ford Taurus model. The calculated crush results obtained
with ADINA using implicit integration compare favorably with the laboratory test

G.V.I.T 8th Semester Mechanical Page 12

Car crash analysis 2004 - 2005

results.

Completed Projects

Some of the work of the Vehicle Safety Research Centre results in a final report
and paper published in international journals and presented at renowned international
conferences. A list of completed work currently available in the public domain follows:

1. Child Restraint Systems for Cars (CREST)

2. Enhanced Coach and Bus Occupant Safety (ECBOS)
3. Ford Safety Data Analysis Project
4. Lower Leg Injury Mechanism Project (LLIMP)
5. Ricability
6. Rollover case study on behalf of Autoliv
7. STandardisation of Accident and Injury Registration Systems
8. STATS19
9. Validation of CRASH3 Accuracy for Crashes

Child Restraint Systems for Cars (CREST)

The objectives of the project were to develop new test procedures to determine
the effectiveness of child restraint systems for cars using improved instrumented
dummies. The ultimate aim of the project was to ultimately improve the effectiveness of
child restraint systems.

G.V.I.T 8th Semester Mechanical Page 13

Car crash analysis 2004 - 2005

Enhanced Coach and Bus Occupant Safety (ECBOS)

The objectives of the research were to:

1. Evaluate the real world bus accident situation and to generate an integrated
database for bus accidents
2. Determine injury mechanisms through component tests, full scale reconstruction
and numerical simulation
3. Develop numerical, component and full scale test methods with special test
procedures for city buses. Also to evaluate the influence of occupant size
4. Summarise suggestions for new regulations and written standards, and to develop
a numerical demonstration of improved interior design.

Ford Safety Data Analysis Project

In January 1995 the Vehicle Safety Research Centre commenced an on-going

consultancy service with Ford Motor Company to advise on the secondary safety
requirements of their car and light truck models. These requirements necessitated expert
input from the VSRC on two broad fronts - statistical analysis of accident data and
individual crash analysis.

Accident Data Analysis: Major areas of activity were analyses of US, UK and German
crash injury data to assess the effectiveness of current safety design and to provide
guidelines for improvements at the design stage of new models. The analytic methods
employed gave results representative of sections of the European crash injury population.

Individual Crash Analysis: Case studies were examined in detail to assess the

G.V.I.T 8th Semester Mechanical Page 14

Car crash analysis 2004 - 2005

mechanisms of injury and the influence of the design aspects of the vehicle. These
provided information on the relative performance of all sections of the European car
fleet.

The consultancy continued to develop both formal and informal links with
regulators and researchers by participating in relevant decision-making groups and
attendance at international conferences and working groups. In this way it was able to
monitor and quickly pass on information on safety research, future legislative
requirements and emerging safety issues.

The VSRC works in close collaboration with Ford Motor Company, both in
Europe and the USA, in order that safety information can be effectively disseminated and
discussed within the relevant engineering communities.

Lower Leg Injury Mechanism Project (LLIMP)

A collaborative project between VSRC and University Hospital Nottingham

(QMC) conducted in 1997. The research objective was to identify the loading conditions
causing below knee injuries in frontal collisions. Vehicle and injury data were combined
to determine the movements of the limb and the direction of the applied forces as the
injury was caused.

Ricability

A pilot study was undertaken during June 1999 on behalf of Ricability. The aim
of this pilot study was to develop a procedure for the measurement of vehicles converted
for use as wheelchair accessible passenger vehicles.

G.V.I.T 8th Semester Mechanical Page 15

Car crash analysis 2004 - 2005

Rollover Case Study on behalf of Autoliv

The VSRC accident investigation team have worked closely with Autoliv to build
an enhanced rollover accident database. Each case from the CCIS data that involved a
rollover has been re-assessed and additional variables relating to the rollover part of the
crash event have been coded. These variables were designed to build a picture of the
rollover sequence including causatory factors and the rollover mechanism. Autoliv are
using the data to aid the development of appropriate crash tests that will correctly
simulate real world crash conditions and much of the additional information will help
with sensor development for advanced occupant protection systems. Since beginning this
work, it’s importance has been acknowledged by the CCIS consortium and the additional
variables are now collected as part of the CCIS investigations.

Standardisation of Accident and Injury Registration Systems (STAIRS)

The aim of the STAIRS project was to try to develop a harmonised procedure for
the in-depth investigation of crashes for the purposes of improving crashworthiness and
safety regulations. The European Commission recognised the need for detailed crash
injury data to support its decision making. The data needed to be sufficiently detailed to
relate to current regulations; and analysed to reflect crashes across Europe. Three main
countries were involved in this research - the United Kingdom, France and Germany.
The key areas addressed included a core dataset, national crash population, data quality,
data processing and exchange, statistical methods for data analysis, terminology and
ethical considerations.

Each year police officers from the 51 police forces in Great Britain complete
235,000 STATS 19 road accident report forms about road accidents on the public
highway which include human injury or death. All the records are collated and the
resulting database is potentially a valuable resource for researchers concerned with road
user safety and injury reduction. One limitation of the STATS 19 database is that the
records are completed at the time of the accident, and therefore do not contain detailed
medical information concerning the casualties. However, from the records it is possible

G.V.I.T 8th Semester Mechanical Page 16

Car crash analysis 2004 - 2005

to trace the individuals involved in the reported accidents, to determine if they attended
hospital, and to request their permission to access their medical records, thereby building
a more complete picture of the accident for research purposes. The study carried out by
the VSRC assessed the feasibility of using STATS 19 records to target specific accident
types that were infrequent or non-occurring events within the CCIS crash population, and
to gather medical information from hospitals throughout central England. The
methodology was applied in particular to car drivers involved in frontal crashes in order
to increase the data available for determining any benefits of a driver's air bag.

Validation of CRASH3 Accuracy for Crashes

This project was to determine the accuracy of Crash3 software, it's relevance to
modern road vehicles and the potential for improvement. As part of this work other
software on the market was reviewed to find the most suitable software for Delta V
calculation of modern vehicle crashes

G.V.I.T 8th Semester Mechanical Page 17