TXT

SAE 2008-01-2981

By Jonathan Smith,2014-05-27 15:44
5 views 0
SAE 2008-01-2981

     ??ÎÄÓÉtxz06??Ï×

    pdfÎĵµ?ÉÄÜÔÚWAP?Ëä?ÀÀÌåÑé???Ñ????ÒéÄúÓÅÏÈÑ?ÔñTXT???òÏÂÔØÔ?ÎÄ?þµ????ú?é????

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:36 AM

     2008-01-2981

     Vertical Impact to an Open Wheel Race Car and Development of a Crash Test to Simulate Driver Response

     Jeff Horton

     Indy Racing League

     Terry R. Trammell

     Indy Racing League Trammell Motorsports and Consulting Orthopaedics Indianapolis, Inc.

     James R. Chinni

     Center for Advanced Product Evaluation

     Copyright ? 2008 SAE International

     ABSTRACT

     The Indy Racing League (IRL) continuously strives to improve safety for drivers of open wheel racecars. As part of a comprehensive engineering effort, the IRL carefully investigates crashes that occur to understand crash causation, the vehicle dynamics involved and driver outcome. Over time, these investigations lead to improved facilities, vehicle design and restraint system performance. One of these investigations involved an open wheel racecar that struck a barrier and became airborne. Without the benefit of an intact suspension, the bottom of the vehicle struck the ground with significant vertical deceleration, leading to driver injury. The vehicle??s onboard event data recorder captured the event. Working with the IRL, the Center for Advanced Product Evaluation (CAPE) developed a dynamic sled test to recreate this specific event. The test involved design of a sled buck to simulate the key elements of the racecar??s seating system, restraint system and interior. The buck was oriented on the sled platform such that the vehicle??s vertical axis was aligned with the sled??s horizontal axis of motion. Unique facility adaptations were required to duplicate the deceleration pulse from the event data recorder. The IRL??s THOR anthropomorphic test device (ATD) was seated inside the sled buck to measure biomechanical response during the tests. Results were compared against known human response in the actual event. Several energy absorbing design alternatives were explored to mitigate injury in this vertical impact.

     INTRODUCTION

     The Indianapolis Motor Speedway and the Indy Racing League have a long history of safety research and innovation. Since the inception of the IRL in 1996, there has been continuous research done towards the improvement of driver safety. The goal is to improve the survivability for the 50+ drivers in the Indy Car and Firestone Indy Lights series. The exciting pace of open wheel racing in the IRL also makes it a dangerous occupation. Highly competitive and skilled drivers race close together at speeds exceeding 200 miles per hour, on lightly banked oval tracks, driving within inches of exterior walls. This combination of factors makes safety research particularly important for the IRL.

     FIGURE 1: INDY CAR RACING

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 1382 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:36 AM

     The IRL uses many avenues for safety research including sled testing, full size car testing, accident simulation, and crash data analysis. Some of the results from this research include design improvements to the rear attenuator and transmission stiffness to reduce the severity of rear impacts, Zylon panels on the car sides to reduce intrusion during side impacts and modifications to the head surround thickness and attachments to reduce head injury potential. When combined with the semi-reclined seating position of the driver, tight fitting six-point restraint systems, energy absorbing seats and other factors, IRL drivers can walk away from crash events that would be devastating in other vehicle types2. The IRL maintains a very thorough accident investigation process in both series. The key elements of the investigation process are as follows: Accident Scene ?C One of the on-track paramedics records skid marks, impact angle and object impacted, and initial driver health. All of this information is entered into a database. Post accident ?C A small team of people record all of the damaged parts of the race car, down load the ADR3 data, draw several graphs and post information on a secure web site. They inspect the seat and head surround for damage, and look for root cause of accidents if necessary. All of this information is then entered into a database. Driver Condition ?C The driver is evaluated upon arrival to the medical center and status is entered into a database. All of the above databases can be queried together to identify trends, target research areas and design meaningful test programs. With nearly 1,000 entries, the database indicates meaningful conclusions and research directions. Each car in both series is equipped with a crash recorder

    and each driver has instrumented ear sensors. The crash box is an ADR3, which is designed and built by Delphi motor sports. The crash box is mounted on the tub floor below and slightly forward of the dash bulkhead. The ADR3 has 3 internal accelerometers that record the chassis longitudinal, lateral, and vertical accelerations at 1000HZ. The driver ear sensors measure 3 axis of acceleration in each ear and are also recorded at 1000HZ by the ADR3. The ADR3 continuously monitors these input channels. When a crash event occurs, it records one minute of pre-crash data and 30 seconds of post-crash data. This long record time enables the ADR3 to capture pre-crash vehicle dynamics as well as secondary or tertiary collisions.

     SUBJECT CRASH

     Ongoing analysis of the IRL crash /injury database identified an increased incidence of spinal fracture over the past 10 years.1 Forty-four instances of spinal fracture have been identified, five of which were the result of primarily vertical input. In two cases an object fell from a height directly onto the driver??s head resulting in fracture of the upper thoracic spine in both. The other three sustained a spinal fracture when a car became airborne and landed on the bottom of the tub (monocoque). Fracture levels in these 3 drivers were in the upper thoracic spine in two ( T3 and T5 ;T9 above a previous T10 ?C L2 fusion;) , and at thoracolumbar junction (T12) in the third. Indy Car cockpits are very tight, and drivers prefer to sit low in the vehicle. As a result, there is no energy absorbing material between the driver??s bottom and the vehicle chassis in the vertical direction, as shown in figure 2. However, a narrow gap exists in this region that could be used to reduce injury in this type of crash mode. The combination of spinal injuries and an opportunity to mitigate that type of injury led to research and testing of vertical impacts.

     FIGURE 2: INDY CAR COCKPIT WITH SEAT REMOVED A representative crash was selected from the crash database. In this event, the driver??s throttle appeared to stick when entering a turn. The car drove through the turn and approximately 75 feet of sand trap and collided with a tire barrier at a 70 angle. This initial impact destroyed the front suspension on both sides of the car. The impact lifted the car into the air several feet and rotated it 150 clockwise. The vehicle landed upright with a ??pancake?? impact between the skidpan and sand. The car left no skid marks in the sand from impact with the tire barrier to its final rest position.

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1 1383

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09,

2010 6:34:37 AM

     The driver in this representative crash sustained a compression fracture of T12 Gertzbein Type A,1,1.3 This was sub classified as a type 3 fracture (>10%<30% compression and >15?ã<30?ã angulation ) according to the severity scale in our previous publication.1 The injury is noted by increased signal intensity (bright) along the anterior superior vertebral endplate and the wedge deformity shown in figure 3.

     FIGURE 4: ADR3 OUTPUT FROM SUBJECT CRASH The spinal injury causation from this vertical input is fundamentally different than in a rear impact1. In the vertical event, the inertia from the driver??s head, helmet and upper torso apply a compressive load into the spine. Restraint systems do not engage in this direction and the occupant does not ramp up into the seat as in a rear impact.

     DYNAMIC SLED TEST

     Several test options were considered to create a test that would mimic the vertical response of the subject crash. A vertical drop test was considered since it would be visually and operationally similar to the subject crash. Ultimately, this approach was rejected due to concerns about repeatability of impact speed, deceleration pulse and cross axis signal sensitivity. To maximize consistency, a dynamic sled test format was chosen for the test. CAPE includes a full-scale vehicle barrier crash test facility, and has experience crashing open wheel racecars. One such test has been reported in prior literature1. When not used for full scale crash tests, the facility can be equipped to conduct dynamic sled testing, by connecting a moving platform to the facility propulsion system and attaching a decelerator to the face of the barrier. CAPE??s electric propulsion system features closed loop controls and excellent speed accuracy. Similarly, CAPE??s decelerator systems enable consistent duplication of crash pulses, including the high deceleration levels seen in open wheel racecar crashes. However, the sled??s motion is directed along a horizontal axis. To produce input into the vertical direction of the vehicle coordinate system, CAPE designed and fabricated the sled buck shown in figure 5 to align the vertical axis of the vehicle??s coordinate system with the sled??s horizontal direction of motion. The team agreed that the snug, six-point, non-retractable restraint system would securely position in the ATD in the seat in this

     FIGURE 3: MRI ILLUSTRATING T12 TYPE A SEVERITY TYPE 3 COMPRESSION ADR output from this event, shown in figure 4, identified the vertical deceleration from the pancake impact as the most harmful event. As a result of this crash, the IRL medical staff checks for similar spinal injuries of all drivers involved in a crash with high vertical deceleration.

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 1384 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:37 AM

     orientation, as shown in figure 6. Pre-test inspection for gaps confirmed proper ATD position in the seat in this orientation. Tethers were added to position the ATD??s feet and legs during the test, without impeding ATD motion during the test. Gravitational acceleration is small in comparison to the vehicle deceleration.

     adjusting the size, stance, stagger, length and metallurgy of the steel bands, CAPE can duplicate a wide range of deceleration pulses. Computer simulation of the steel band decelerator enables efficient pulse development. However, this vertical crash pulse featured a unique combination of a high peak deceleration of 70g, coupled with a relatively slow velocity change of 10.5 mph. The steel bands would deform only in the linear-elastic response region, returning the majority of the energy back into the sled. This required development of a different decelerator medium. CAPE??s engineering team devised a solution using a specially shaped impactor attached to the sled, which engages aluminum honeycomb material attached to the face of the impact barrier. The arrangement is shown in figure 7. Computer modeling was used to design the shape of the impactor, defining the various areas that would engage the honeycomb material to produce the proper deceleration shape and time.

     FIGURE 5: DYNAMIC SLED BUCK The sled buck was designed to incorporate important geometric features of the Indy Car occupant environment, including seat, restraint system and head rest orientation. The buck includes transparent sidewalls, to enable high speed imaging of ATD kinematics and significant reinforcements to assure rigidity and pulse transmission. The buck features an adjustable base plate to enable insertion of variable thicknesses of energy absorbing materials underneath the seat structure. Lower restraint system anchorages and instrumentation are attached to this adjustable base. FIGURE 7: SHAPED IMPACTOR AND HONEYCOMB The dynamic sled test series used the IRL??s 50th percentile THOR ATD. THOR is particularly well suited to race car testing, because its spine and pelvic design enable it to sit properly in the semi-reclined posture of an open wheel race car driver. Because of extensive research into spinal injury, the IRL??s THOR has been specially adapted to include instrumentation at the T1, T8 and T12 vertebral location. To properly position THOR for testing, the seating positions of all IRL drivers were scanned and averaged. As with each IRL driver, the seat used in the dynamic test series is uniquely molded

    to the THOR ATD, providing the same close fit that drivers experience. The seat positions THOR to represent the average seating posture and fore-aft adjustment from the driver survey.

     FIGURE 6: PRE-TEST ATD POSITION Duplicating the deceleration pulse from the vertical crash input required development of a unique sled decelerator. The CAPE barrier facility typically uses a probe that is attached to the moving sled platform. The probe enters a decelerator structure and bends steel bars. By

     SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 1385

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:37 AM

     TEST RESULTS

     With all of the pre-test preparations in place, a series of dynamic sled tests was conducted at CAPE. The tests were captured using on-board and off-board high-speed digital imagers recording at 1000 frames per second. Additionally, data acquisition systems captured results at 20,000 Hz. Table 1 shows the energy absorbing materials used in the test series.

     Test Name: CTR02688-014

     Graph Name: Test Date: PULSE OVERLAY 05-20-2008

     Sled Accel: Target Pulse:

     10 0 -10 -20

     Max value: 3.34 [g] at 0.016 [s] Max value: 3.40 [g] at 0.063 [s]

     Min value: -70.29 [g] at 0.007 [s] Min value: -70.11 [g] at 0.007 [s]

     SLED ACCEL

     TARGET PULSE

     Acceleration [g]

     -30 -40 -50 -60 -70

     Test 1 2 3 4 5 6 7 8 9 10 11 12

     Energy Absorbing Material Baseline: None MK-1 EPS, 1.5#, ? ?? thick BSS EPP, 1.7#, ??? thick Dow Impaxx 300, ??? thick MK-1 EPS, 1.5#, ??? thick, 0.050 carbon sheet

     -80 -0.05

     0

     0.05

     Time (s)

     0.1

     0.15

     0.2

     FIGURE 8: DECELERATION PULSE OVERLAY

     Test Name: CTR02688-014

     Channel Name: Test Date: Filter Class: SLED VELOCITY 05-20-2008 CFC 180

     Delta V: 18.38 kph (11.42 mph)

     Recorded Impact Speed: 14.54 kph (9.03 mph) Min. Value: -3.82 kph (-2.37 mph) at 0.015 s

     Blue Confor, ??? thick, 0.050 carbon sheet Blue Confor, ??? thick, 0.015 thick

     Velocity (kph)

     Max. Value: 14.56 kph (9.05 mph) at 0.001 s

     15 12.5 10 7.5 5 2.5 0 -2.5

     BSS EPP, 1.7#, 1.5?? thick Dow Impaxx 300, 1.5?? thick BSS EPP, 1.7# 3.0?? thick Blue EPP, 1.0#, 3.0?? thick E175, ??? thick, 0.015?? Kevlar sheet

     -5 -0.06

     -0.04

     -0.02

     0 Time (s)

     0.02

     0.04

     0.06

     FIGURE 9: SLED VELOCITY CHANGE

     TABLE 1: TEST MATRIX The sled dynamics during the test demonstrated good repeatability, as anticipated during the test setup. The deceleration pulse closely matched the target pulse from the subject crash. A pulse comparison from a typical run is shown in figure 8. Peak deceleration occurred at 7 milliseconds for every test run and exhibited 3% or less variation in magnitude. Sled impact speeds had a 0.02 mph standard deviation and velocity change exhibited a 0.06 mph standard deviation. Typical sled impact speed is shown in figure 9. The post-test deformed condition of the honeycomb is shown in figure 10. FIGURE 10: POST TEST DECELERATOR CONDITION

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 1386 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:37 AM

     Among the many injury parameters recorded by THOR, spinal compressive forces (Fz) are the most significant predictor of spinal fracture. Previous research indicates that a threshold compressive force of 6,000N can be used to assess probability of burst fractures1. As can be seen from table 2 in the Appendix, measurements were more than twice this value for both the T8 and T12 locations during the

    baseline test. Subsequent tests with energy absorbing material underneath the seat reduced spinal compressive forces. Three inches of energy absorbing material were needed to mitigate spinal compressive forces below the 6,000N threshold. While spinal injuries in these thoracolumbar locations are observed in drivers during high vertical deceleration collisions, the high magnitudes observed during the baseline test would likely predict more extreme injury than found with drivers. Two potential causes may explain these observations. First, IRL drivers are typically young and athletic, providing them a higher injury tolerance than the general population. Second THOR??s lower anatomy differs from drivers. THOR??s posterior surfaces are narrower and stiffer than live subjects. This likely elevates the pelvic acceleration input into THOR and provides the ATD a narrow contact patch that penetrates the energy absorbing materials. High pelvic acceleration input is reported in table 3 in the Appendix. Moreover, witness marks in the energy absorbing materials indicate penetration by THOR??s lower spine box, a feature not seen in drivers. Refer to figure 11.

     As seen from the results in the Appendix, energy absorbing materials reduced both spinal compression loading and acceleration values in all body regions. Approximately ? inch of vertical space exists underneath the seats of current design cars. Even though the dynamic test result did not demonstrate acceptable results with ??? thickness, this gap underneath the seat is now being filled with energy absorbing material to provide incremental improvement for drivers subjected to vertical impacts. Additionally, results from the test series are being used to customize new energy absorbing materials to optimize protection in the available space. Specifications for the next generation IRL chassis will include additional vertical space to package the new materials underneath the seat.

     CONCLUSIONS

     The IRL and CAPE successfully created a dynamic sled test to replicate the dynamics seen during IRL crashes involving high vertical decelerations. A dynamic test program including twelve combinations of energy absorbing materials underneath the seat evaluated the potential to mitigate spinal injures from high vertical decelerations. All energy absorbing materials provided some level of injury mitigation. Three inches of energy absorbing materials were needed to mitigate spinal compression forces to acceptable levels. Seat designs in current design IRL racecars now include ? inch of energy absorbing materials underneath the seat, to make incremental safety improvements for drivers. Optimized energy absorbing materials are now under development, to best reduce vertical input in the available space.

     Spine Box Impression

     FIGURE 11: ATD IMPRESSION IN ENERGY ABSORBING TEST MATERIAL An improvement to THOR is being investigated, to make its posterior surfaces more biofidelic for future seat bottom testing.

     Specifications for the next generation IRL chassis will include additional package space for energy absorbing materials underneath the seat.

     ACKNOWLEDGMENTS

     The authors would like to acknowledge the extensive contributions of Mr. Jeffrey Mowins of Mark One Composites to this project.

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1 1387

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:37 AM

     REFERENCES

     1. T.R. Trammell, C.S. Weaver and H. Bock, Spine Fracture in Open Cockpit Open Wheel Race Car Drivers, SAE 2006-01-3630 2. J.W. Melvin, P.C. Begeman et al, Crash Protection of Stock Car Racing Drivers ?C Application of Biomechanical Analysis of Indy Car Crash Research, Stapp Car Crash Journal, Vol. 50, November 2006 3. Gertzbein, SD ed. Fractures of the Thoracic and Lumbar Spine. Baltimore: Williams & Wilkins 1992,pp25 -57

     CONTACTS

     Terry R. Trammell, MD Trammell Motorsports and Consulting 5708 N. County Road 550E Pittsboro, IN 46278 Fax: 317-852-2262

    trt@motorsportsmd.com James R. Chinni, P.E. Center for Advanced Product Evaluation 18881 US 31 North Westfield, IN 46074 Phone: 317-867-8225 Jim.Chinni@imminet.com

     DEFINITIONS, ACRONYMS, ABBREVIATIONS

     ADR: Accident Data Recorder ATD: Anthropomorphic Test Device (crash test dummy) CAPE: Center for Advanced Product Evaluation IRL: Indy Racing League

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 1388 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:37 AM

     APPENDIX

     Representative results

     Location Upper Neck T8 T12

     Baseline -4,261 -12,859 -15,243

     ?? Thick -4,110 -11,715 -14,406

     1.5?? Thick -3,196 -6,702 -7,147

     3.0?? Thick -2,886 -5,678 -5,622

     TABLE 2: MAXIMUM SPINAL COMPRESSIVE FORCES, Fz (N)

     Location Head Chest (3ms) Pelvis

     Baseline 98 81 330

     ?? Thick 89 74 263

     1.5?? Thick 65 54 52

     3.0?? Thick 63 53 52

     TABLE 3: PEAK RESULTANT ACCELERATIONS (g)

     Test Name: CTR02688-001

     Channel Name: ATD Position: Test Date: Filter Class: Max. Value: 2371.97 [N] (533.24 lbf) at 0.017 s A19 T12 Spine Force XYZ Baseline 05-19-2008 CFC 1000

     2500 2000 1500 1000 500 0 -500 400 300 200 100 0 -100 -200 -300 2500 0 -2500 -5000 -7500 -10000 -12500 -15000 -17500

     Min. Value: -230.24 [N] (-51.76 lbf) at 0.061 s

     T12 Spine Fx [N]

     Max. Value: 327.63 [N] (73.65 lbf) at 0.018 s

     Min. Value: -223.48 [N] (-50.24 lbf) at 0.126 s

     T12 Spine Fy [N]

     Max. Value: 481.15 [N] (108.17 lbf) at 0.034 s

     Min. Value: -15243.34 [N] (-3426.86 lbf) at 0.016 s

     T12 Spine Fz [N]

     -0.05

     0

     0.05

     Time [s]

     0.1

     0.15

     0.2

     FIGURE 12: T12 SPINE FORCES, BASELINE

     Author:Gilligan-SID:4057-GUID:39553490-128.104.1.220 SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 1 | Issue 1 1389

     Licensed to University of Wisconsin - Madison Licensed from the SAE Digital Library Copyright 2010 SAE International E-mailing, copying and internet posting are prohibited Downloaded Tuesday, November 09, 2010 6:34:37 AM

     Test Name: CTR02688-003

     Channel Name: ATD Position: Test Date: Filter Class: Max. Value: 1952.90 [N] (439.03 lbf) at 0.019 s A19 T12 Spine Force XYZ 0.5 in, BSS 1.7lb EPP 05-19-2008 CFC 1000

     T1 2000 2 1500 Spi 1000 ne 500 Fx [N 0 ] -500 300 200 100 0 -100 -200 -300 2500 0 -2500 -5000 -7500 -10000 -12500 -15000

     Min. Value: -374.30 [N] (-84.15 lbf) at 0.068 s

Report this document

For any questions or suggestions please email
cust-service@docsford.com