CRASHWORTHINESS OF FORD F250 TRUCK ROLL OVER INVERTED VEHICLE DROP TEST (SAE J996)

by Pritesh K Shah B.E. Mechanical Engineering, January 2003, Pune University, INDIA A Project Report Submitted to the Faculty of Old Dominion University in Partial Fulfillment of the Requirement for the Degree of

MASTER OF ENGINEERING
in

DESIGN & MANUFACTURING MECHANICAL ENGINEERING OLD DOMINION UNIVERSITY August 2011

ME-11-0082

CRASHWORTHINESS OF FORD F250 TRUCK ROLL OVER INVERTED VEHICLE DROP TEST (SAE J996)

by Pritesh K Shah B.E. Mechanical Engineering, January 2003, Pune University, India

A Project Report Submitted to the Faculty of Old Dominion University in Partial Fulfillment of the Requirement for the Degree of

MASTER OF ENGINEERING in DESIGN & MANUFACTURING MECHANICAL ENGINEERING OLD DOMINION UNIVERSITY
August 2011

Approved By:

___________________________ (Advisor)

___________________________

ABSTRACT

A finite element analysis of roof strength for Ford F250 truck as per SAE J996 standard is performed using LS-DYNA. A detailed finite element model was taken from NCAC (National Crash Analysis Centre) at George Washington University Virginia. NCAC is one of the prominent leaders in vehicle safety research. Efforts were taken to subject the F250 FEM (finite element model) for roll over impact study with the procedure suggested by SAE J996.

After running the F250 FEM for roll over, key areas for reduced impact on the roof structure of the truck were addressed and were redesigned. Further, a detailed study is carried out to compare the results between the original F250 model with the redesigned model. A comparison chart was made to illustrate the reduced impact on the predefined nodal points on the roof structure.

TABLE OF CONTENTS
MASTER OF ENGINEERING...........................................................................................1

MECHANICAL ENGINEERING.......................................................................................1 ME-11-0082.....................................................................................................................1 OLD DOMINION UNIVERSITY...............................................................................2 CHAPTER 1........................................................................................................................8 INTRODUCTION...............................................................................................................8 ..............................................................................................................................................8 1.1. United States standards applicable to rollover.............................................................9 CHAPTER 2......................................................................................................................12 CASE SETUP AND ANALYSIS......................................................................................12 2.1. F250 Model Environment..........................................................................................12 ....................................................................................................................................13 2.2. Frontal Impact Analysis ............................................................................................14 2.3. LS- DYNA simulation...............................................................................................15 2.4. Dummy Database.......................................................................................................19 CHAPTER 3......................................................................................................................25 ROLLOVER INVERTED VEHICLE DROP TEST......................................................25 3.1. Case setup..................................................................................................................25 3.2. Results of rollover impact of Ford F250 (SAE J 996)............................................28 3.2.1. Total Energies..........................................................................................................28 3.2.2. Roof intrusion/deflections........................................................................................29 3.3.3. Force V/s Deflection................................................................................................30

3.4. Redesign of Roof structure Ford F250...................................................................31 3.5 Results: Redefined Roof Ford F250 ........................................................................35 3.5.1. Total Energies..........................................................................................................35 3.5.2. Roof Intrusion/deflection.........................................................................................36 3.5.3. Force V/s Deflection................................................................................................37 3.6. Comparison.................................................................................................................38 CHAPTER 4......................................................................................................................40 CONCLUSIONS................................................................................................................40 REFERENCES..................................................................................................................42

LIST OF FIGURES

Figure 1 Highway rollover fatalaties...................................................................................8 Figure 2.1 NCAC FORD F250 Finite Element Model...................................................13 Figure 3 Ford F250 with frontal impact with a rigid barrier (no dummy) @ 35 mph....15 Figure 4 Different stages of simulation for frontal impact (no dummy) @35mph............18 Figure 5 Hybrid III 50th percentile rigid Fe dummy.........................................................19 Figure 6 Positioned Hybrid III dummy..............................................................................20

Figure 7 Ford F250 with frontal impact with 50th percentile Hybrid III dummy @ 35 mph....................................................................................................................................24 Figure 8 Different stages of deflection of roof Ford F250 (SAE J 996).........................27 Figure 9 Total Energies plot for Ford F250 FEM...........................................................28 Figure 10 Roof displacement V/s Time plot for Ford F250 FEM..................................29 Figure 11 Wall force V/s Displacement plot for original F250 FEM................................30 Figure 12 Support plate finite element model....................................................................31 Figure 13 Different stages of deflection of redesigned roof Ford F250 (SAE J996).....34 Figure 14 Total Energies plot for redesigned roof F250 FEM..........................................35 Figure 15 Roof displacement V/s Time plot for redesigned roof F250 FEM....................36 Figure 16 Wall force V/s Displacement plot for redesigned roof F250 FEM...................37 Figure 17 Comparison of deflection for original and redesigned roof F250 FEM............38 Figure 18 Comparison of Wall Force V/s Displacement for original and redesigned roof of F250 FEM......................................................................................................................39

NOMENCLATURE

FEM Finite Element Model FMVSS Federal Motor Vehicle Safety Standards SAE Society of Automotive Engineers

NHTSA National Highway Traffic Safety Administration NCAC National Crash Analysis Centre ms- Milliseconds N- Newton (Force) DOE Design of experiment

CHAPTER 1 INTRODUCTION
Every year more than 10,000 people die in rollover crashes. Rollover crashes constitute to a meager 2 percent of all collisions and yet account for 24 percent of passenger fatalities. Rollovers are one of the most dangerous forms of vehicle crashes because of the high occurrence of occupants catastrophic head injuries and fatalities.

Figure 1 Highway rollover fatalaties

In 2007, the National Highway Traffic Safety Administration (NHTSA) made it mandatory that all vehicles to be equipped with electronic stability control (ESC). By the 2008 model year, ESC was standard on 65 percent of passenger cars, 96 percent of SUVs and only 11 percent of pickups. The ESC technology helps minimize skidding, as well as maintains control when drivers swerve. ESC senses when a driver may lose control and automatically applies brakes to individual wheels (Outer front wheel to counter over steer or inner rear wheel to counter under steer) to help stabilize the vehicle and avoid a rollover [6].

Considerable research has been undertaken over the years to differentiate rollover according to severity and to develop a standard rollover test. In most cases the studies are applicable to passenger cars. However, many of the principles are generally

applicable to all vehicles. In recent years, the increase in rollover casualties from the growing population of pickups and SUVs has emphasized the need to examine these vehicles as separate classes.

1.1.

United States standards applicable to rollover

The following FMVSS (Federal Motor Vehicle Safety Standards) are practiced in

application of rollover safety features. [1] Interior Protection -FMVSS 201 Glazing Materials -FMVSS 205 Door Locks and Retention - FMVSS 206 Occupant Crash Protection - FMVSS 208 Windshield Mounting - FMVSS212 Side Impact Protection - FMVSS 214 Roof Crush Resistance - FMVSS 216 Fuel System Integrity - FMVSS 301 [1]

Rollover crashes have been differentiated by a large number of parameters that may influence rollover outcomes. Examples of such parameters are the terrain topography, roadway grade, curvature and vehicle type. Further, a basis for selecting test

procedures and relating compliance with these procedures to real world benefits is vitally needed. The problem of assessing countermeasures in rollover is perplexed by the lack of dummies and test procedures to study rollover. There is no rollover test dummy which has been validated in a manner similar to the Hybrid III in frontal crashes, or the SID in side impacts.

As mentioned earlier many aspects contribute to the occurrence of rollover crashes. Rollover occurrences are mainly related to the vehicle type, unsafe and irresponsible driving behaviors such as hard steering maneuvers, bad road design, etc. Certain class of vehicles, such as Sport Utility Vehicle (SUV) and pickup trucks, are more prone to rollover than other classes of vehicles. Most vehicles do not have adequate roof strength during a rollover. The quasi-static roof crush test mandated by the FMVSS 216 subjects the vehicle to a maximum force significantly less severe than would be applied to the vehicle during a multiple rollover. The Society of Automotive Engineers (SAE) recommended practice J996, Inverted Drop Test, is also a test of rollover crashworthiness, and was developed by SAE in the late 1960s. Since it is a more severe test, numerous engineers prefer it to the quasi-static FMVSS 216 test. The SAE J996 test was designed, to obtain as closely as possible deformation of a vehicle roof or roll bar structure which occurs in a vehicle rollover. In this test, the vehicle is inverted, given a roll angle, pitch angle, and drop height that are representative

of the assumed loading at rollover. The angles present ensure that the majority of potential energy is transferred directly to the A-pillar structure. This standard does not specify any crush measurement methodology, permanent or dynamic [7]. The project focuses on the roof strength of the Ford F250 in event of a rollover by testing the FEM using the Inverted Vehicle Drop Test Procedure SAE J996.

CHAPTER 2 CASE SETUP AND ANALYSIS 2.1. F250 Model Environment

A Finite Element Model (FEM) of Ford F250 was used from the National Crash Analysis Centre (NCAC) [2]. NCAC at The George Washington University's Virginia Campus is one of the nation's leading authorities in automotive and highway safety research. The finite element model of a 2006 Ford F250 pick-up truck was developed at the NCAC for the National Highway Traffic Safety Administration (NHTSA). The Ford F250 truck is a multi-purpose pickup truck. The vehicle obtained by the NCAC is an Extended-Cab, with a wheel base of 3610 mm (142.12 inches) and a maximum width of 2030mm (79.92 inches). The F250 has a total kerb weight of 3016 kilograms (6072.2 lbs) with a 5.4 litre V8 engine. The pickup truck has a 5 speed manual transmission with a four wheel drive configuration. However, several other models exist, such as higher engine capacity, automatic transmission and drive configuration, with no change in the general geometry.

Figure 2.1 NCAC FORD F250 Finite Element Model The model consists of 871 parts with 738165 nodes made up of 726759 elements. The following table illustrates different type of elements constituting the total number of elements. Table 1 Types of elements in Ford F250 FEM Shell element 698501 Beam element 2353 Solid element 25905

Specifically, the properties of each component are defined by a set of material cards with four types of materials being used in the model. Each of the components is subdivided into either shell elements, beam elements or hexahedron elements. Two types of shell elements are used in the finite element model, viz. quadrilateral shell elements and triangular shell elements. The formulation of both types of shell

elements used is based on Belytschko-Tsay theory. Table 2.1 enlists the material models used in LS- DYNA along with the number of components.

Table 2 LS-DYNA material models Material type number 7 S02 1 57 9 24 20 100 S01 S04 6 Material Type No. of Components

BlatzKo Rubber Damper Viscous Elastic Low Density Foam Null Piecewise Linear Plasticity Rigid Spotweld Spring Elastic Spring Non Linear Elastic Viscoelastic

18 4 70 12 48 598 106 1 5 5 4

2.2.

Frontal Impact Analysis

The F250 FEM (finite element model) was subjected to a frontal impact study with the parameters suggested by Federal Motor Vehicle Safety Standard (FMVSS) 208 which states 35 mph (56kph) into fixed barrier. 50th percentile Hybrid III adult male dummy in front seat.

Initially a full frontal impact without the dummy was simulated to check the accuracy of the model from the NCAC reports. After successful model verification a 50th percentile Hybrid III dummy was defined in the vehicle model and the interaction of dummy with the vehicle environment & the injury values were analyzed.

Figure 3 Ford F250 with frontal impact with a rigid barrier (no dummy) @ 35 mph

2.3. LS- DYNA simulation

The simulation was run for 150 milliseconds of simulation time with a time step of 9.00 E-07. The total computational time required for the run was around 32 hours. The simulation is done using a rigid barrier. Table 3 LS-DYNA simulation parameters Version Revision Precision LS- DYNA_971 7600.1224 Single (I4R4)

Feature OS level Number of processors Total elapsed time Simulation time

SMP Linux 8 32 hours 150 milliseconds

The truck model comes in contact with the barrier at roughly about 5 milliseconds; the front hood starts to deform at 20 milliseconds. Maximum crush occurs at 110 milliseconds. Figure below shows different stages of deformation of truck during the simulation time.

Simulation results for Ford F250 with frontal impact with a rigid barrier (no dummy) @ 35 mph

(a) @ 24ms

(b) @ 40 ms

(c) @ 70 ms

(d) @ 150 ms Figure 4 Different stages of simulation for frontal impact (no dummy) @35mph

2.4. Dummy Database

The Hybrid III is the most widely applied dummy for frontal impact. The series includes a 5th percentile female, a 50th percentile male, and a 95th percentile male. The 50th percentile is a five feet six inch dummy model with a weight of 170lbs, for the analysis the 50th percentile male dummy is used. The details of the dummy can be found in the LS-DYNA ftp site
[5]

which includes the calibration tests, positioning details, post

processing, Part ID details and the transformation details. The dummy model used in the analysis is shown below.

Figure 5 Hybrid III 50th percentile rigid Fe dummy

The dummy model is imported in LS-DYNA and positioned in the driver seat of the vehicle via H point operations. By doing this we can get the responses of the dummy and the interactions with vehicle environment along with the injury values.

The figure below shows the final positioned dummy on the vehicles driver seat. Contacts are defined with the help of contact ids obtained from the report from LSDYNA.

Figure 6 Positioned Hybrid III dummy Since all LSTC (Livermore Software Technology Corporation) dummies use the mmmskgkN unit system, we use the following conversion to convert to mm-s-tonne unit system.

Figure 2-6 Transform keyword file LS-DYNA simulation for Ford F250 with frontal impact with Hybrid III 50th percentile dummy @ 35 mph The simulation was run for 150 milliseconds of simulation time with a time step of 9.00 E-07. The total computational time required for the run was around 32 hours 29 minutes. The truck model comes in contact with the barrier at roughly about 5 milliseconds; the front hood starts to deform at 20 milliseconds. Maximum crush

occurred at 110 milliseconds. Figure below shows different stages of deformation of truck during the simulation time. Table 4 LS-DYNA simulation parameters Version Revision Precision Feature OS level Number of processors Total elapsed time Simulation time LS- DYNA_971 7600.1224 Single (I4R4) SMP Linux 8 32 hours 29 minutes 150 milliseconds

(a) @ 24ms

(b) @ 40 ms

(c) @ 70 ms

(d) @ 150 ms Figure 7 Ford F250 with frontal impact with 50th percentile Hybrid III dummy @ 35 mph

CHAPTER 3 ROLLOVER INVERTED VEHICLE DROP TEST 3.1. Case setup

For rollover - inverted vehicle drop test as per SAE J996 the Ford F250 truck was inverted with a roll angle defined at 25 and a pitch angle defined at 5.A rigid plate for impact was defined and the drop height was set at 12 inches (300mm) . Computation time was reduced by giving an initial velocity to the vehicle and was brought right above the rigid plate. The initial velocity was defined in LS-DYNA input deck via *INITIAL_VELOCITY_GENERATION.

A total of 10 nodes are defined to obtain the database history to monitor the amount of roof intrusion. The nodes are defined via node set in the DATABASE_HISTORY_NODE_SET. The averages of the 10 nodes are used to calculate the amount of roof deflection. Figure 3.1 shows the deformation of the roof at different stages of simulation

(a) @ 40 ms

(b) @ 75 ms

(c) @ 100 ms

(d) @ 120 ms Figure 8 Different stages of deflection of roof Ford F250 (SAE J 996)

3.2. Results of rollover impact of Ford F250 (SAE J 996) 3.2.1. Total Energies

Figure 9 Total Energies plot for Ford F250 FEM Figure 9 shows the global internal, kinetic & the total energies of the system. It can be seen that at the start of the solution, the internal energy is zero and the kinetic energy is equal to the total energy. When the F250 truck hits the rigid wall, the kinetic energy of the system decreases and is converted to internal energy of the materials. The kinetic energy and the internal energies meet at roughly 70 milliseconds, and the kinetic energy continues to decrease whereas the internal energy continues to increase. The sudden spike in the internal energy is due to the hourglass energy which occurs due to

under integration of elements. The roof structure being made up of very thin shell elements excites the hourglass energy mode causing the internal energy to spike up.

3.2.2. Roof intrusion/deflections

Figure 10 Roof displacement V/s Time plot for Ford F250 FEM Figure 10 shows the average nodal displacement for the rollover impact analysis of Ford F250 truck. The maximum average displacement of the nodes is 267.64 mm, which means the roof structure had intruded in the driver compartment by approximately 10 inches. The amount of roof intrusion has to be reduced to avoid any interface between the driver and the roof structure to avoid catastrophic head injury or fatalities.

3.3.3. Force V/s Deflection

Figure 11 Wall force V/s Displacement plot for original F250 FEM Figure 11 shows the Force V/s Displacement curve for the Ford F250 truck. From the graph we can see that the truck experiences a maximum wall force of approximately 35Kn and yields a total intrusion/deflection of roof structure by 265 mm.

3.4. Redesign of Roof structure Ford F250

As seen above the F250 truck when subjected to a rollover impact gave a maximum roof deflection of 267.64 mm. To make the car safer for rollover impact, the roof of the vehicle had to be redesigned. This was done in three parts.

The roof material was changed from mild steel to ST 44 grade steel A support plate was added between the A and the B pillar for roof support. The thickness of the roof was changed from 0.8mm to 1.8mm

Figure 12 Support plate finite element model

Figure 3.5 shows the support plate used between the A and the B pillar to support the deformation of the roof structure. The section is defined as Belytschko Tsay shell element
[8]

. The plate has a thickness of 2 mm with holes and slots similar to the

connection plate in the basic model. The plate is spot welded at 15 points to the roof rail structure via spot weld beam sections and acts as an additional support to the roof rail. The total increase in the mass of the roof by this redesign was in the tune of 45.15 lbs (20.32 kgs) with the roof thickness change constituting to approximately 40 lbs.

(a) @ 55 ms

(b) @ 55 ms

(c) @ 100 ms

(d) @ 55 ms

Figure 13 Different stages of deflection of redesigned roof Ford F250 (SAE J996)

3.5 Results: Redefined Roof Ford F250 3.5.1. Total Energies

Figure 14 Total Energies plot for redesigned roof F250 FEM

Figure 14 shows the global internal, kinetic & the total energies of the system. It can be seen that at the start of the solution, the internal energy is zero and the kinetic energy is equal to the total energy. When the truck hits the rigid wall, the kinetic energy of the system decreases and is converted to internal energy of the materials. The kinetic energy and the internal energies meet at roughly 50 milliseconds, and the kinetic energy

continues to decrease whereas the internal energy continues to increase. The internal and kinetic energy should meet roughly at half the total energy and continue to stay at that point without crossing for the rest of the analysis. The remaining energy, other than the sum of kinetic & internal energies is the hour glass energy.

3.5.2. Roof Intrusion/deflection

Figure 15 Roof displacement V/s Time plot for redesigned roof F250 FEM Figure 15 shows the average nodal displacement for the Redesigned Roof Ford F250 truck. The maximum average displacement of the nodes is 193.73mm. There is a

significant reduction in the amount of roof displacement approximately 28 %( 73.91 mm).

3.5.3. Force V/s Deflection

Figure 16 Wall force V/s Displacement plot for redesigned roof F250 FEM The redesigned roof structure gives a maximum intrusion of 193mm for approximately 25 Kn of wall force.

3.6.Comparison

Figure 17 Comparison of deflection for original and redesigned roof F250 FEM Figure 17 shows the comparison of deflection for the basic model (Line B) and the redesigned roof (Line A). From the graph we can see that at the end of 120 ms the total deformation of roof for the redesigned roof F250 model was 194 mm as compared to the original F250 model which has a roof deformation of 264 mm.

Figure 18 Comparison of Wall Force V/s Displacement for original and redesigned roof of F250 FEM Figure 18 shows the wall force versus displacement for the original Ford F250 and the redesigned roof FEM. From the graph we can see that the force acting upon the roof has reduced from approximately 35 Kn to 25 Kn (28.57 %). The maximum deflection of the roof has been minimized to the tune of 70mm

CHAPTER 4 CONCLUSIONS
The rollover impact test procedure as suggested by SAE J996 was performed on finite element model of a Ford F250 truck and the redesigned roof of Ford F250 and the results were compared. During the study of typical rollover, the goal is to achieve minimum roof intrusion to avoid contact between the roof structure and the passenger head to avoid and catastrophic head injuries or fatalities. The parameters that affect the outcome of a rollover impact are stiffness and geometry of roof structure, A and B pillar reinforcement, occupant restraint system characteristics and stiffness and properties of interior components. With this model all of the above mentioned parameters can be varied to have a DOE study for various rollover scenarios including varying velocities and angle of impact. To protect driver in the driver compartment following things should be considered. Defining certain accident situations (standard accidents), providing a welldefined safe zone space into which no structural elements penetrate during the collision and during the structural deformation and damage, keeping the driver in the safe zone space during the collision. In current regulation, driver safety is not adequately considered so this issue needs to be addressed during specifying the rollover impact regulations. Most of the fatal injuries in the rollover impact are caused between the passenger head and the roof structure so it is recommended that more importance should be given to the stiffness, geometry and the area of impact locations. This developed LS-DYNA

model is not validated and it just gives the estimate of the roof structure intrusion. In future by validating this model by incorporating a certified rollover test dummy actual injury values in case of a rollover impact can be accurately predicted. A detailed study can be performed to assess more severe test conditions to develop complete strong regulations. Furthermore, various other occupant positions like passenger side, rear side can be considered for study.

REFERENCES
[1] National Highway Traffic Safety Administration, "Federal Motor Vehicle Safety Standards and Regulations," in Crashworthiness, ed. Washington, DC, 1999. [2] National Crash Analysis Centre. (2007). Finite Element Model Archive. Available: http://www.ncac.gwu.edu/vml/models.html [3] Livermore Software Technology Corp. (2007). LS-DYNA Theory 2006. Available: http://www.lstc.com/manuals.htm [4] Livermore Software Technology Corp. (2007). Keyword 971 2 volumes Available: http://www.lstc.com/manuals.htm [5] Livermore Software Technology Corp. (2008). Index of /user/lstc-dummies. Available: http://ftp.lstc.com/user/lstc-dummies/ [6] D. Shepardson, "Newer SUVs now safer than cars in a crash," in Detroit News, ed. Detroit, 2011. [7] P. D. Stephen A. Batzer, P.E. & Robert M. Hooker, "DYNAMIC ROOF CRUSH INTRUSION IN INVERTED DROP TESTING." [8] Belytschko & Tsay, "Explicit Algorithms for Nonlinear Dynamics of Shells"," vol. AMD-Vol.48, ASME, 209-231, 1981. [9] K. Elitok, "Explicit Dynamic Analysis of Vehicle Roll-Over Crashworthiness Using LS-DYNA ", Mechanical Engineering, Istanbul Technical University, 2006.