New Energy-Absorbing High-Speed Safety Barrier

Transportation Research Record 1851■ 53Paper No. 03-2218

For many years, containment for errant racing vehicles traveling onoval speedways has been provided through rigid, concrete containmentwalls placed around the exterior of the track. However, accident expe-rience has shown that serious injuries and fatalities may occur throughvehicular impacts into these nondeformable barriers. Because of theseinjuries, the Indy Racing League and the Indianapolis Motor Speedway,later joined by the National Association for Stock Car Auto Racing(NASCAR), sponsored the development of a new barrier system by theMidwest Roadside Safety Facility at the University of Nebraska–Lincolnto improve the safety of drivers participating in automobile racing events.Several barrier prototypes were investigated and evaluated using bothstatic and dynamic component testing, computer simulation modelingwith LS-DYNA (a nonlinear finite element analysis code), and 20 full-scale vehicle crash tests. The full-scale crash testing program includedbogie vehicles, small cars, and a full-size sedan, as well as Indy RacingLeague open-wheeled cars and NASCAR Winston Cup cars. A combina-tion steel tube skin and foam energy-absorbing barrier system, referredto as the SAFER (steel and foam energy reduction) barrier, was success-fully developed. Subsequently, the SAFER barrier was installed at theIndianapolis Motor Speedway in advance of the running of the 2002 Indi-anapolis 500 race. From the results of the laboratory testing program aswell as analysis of the accidents into the SAFER barrier occurring duringpractice, qualification, and the race, the SAFER barrier has been shownto provide improved safety for drivers impacting the outer walls.

In recent years, automobile racing has arguably become one of themost popular sporting events in the United States as well as interna-tionally. This fact is evidenced by the number of weekend races, largefan-based support, corporate sponsorship, and around-the-clock cabletelevision coverage. Furthermore, this fact can be seen by the varietyof race series available to both drivers and spectators, including theIndy Racing League (IRL), National Association for Stock Car AutoRacing (NASCAR), Formula 1, Championship Auto Racing Teams,and the International Race of Champions (IROC) race series, to namea few.

For most of these race series, high-performance cars and noweven various classes of trucks typically travel several hundred timesaround oval tracks at very high speeds. On the larger ovals, whichinclude banked corners, the speeds of the open-wheeled-type vehiclesattain or exceed 370 km/h (230 mph).

Consequently, these higher speeds have increased the potentialfor more dramatic accidents, injuries, and fatalities. As a result, theneed for increased track safety for the drivers as well as the specta-

tors has become more apparent. Historically, some of the early raceswere conducted without safety barriers around the raceways. This isstill evidenced today on several smaller dirt tracks, where steepbanking is the only source of notable protection or containment.However, over time, the owners of most of the high-speed trackfacilities have recognized the need for safety barriers. For drivers,this has resulted in the installation of rigid concrete walls on ovalsand steel or tire barriers on other road courses. For spectators, tallchain-linked and cable combination barriers have been constructedabove or adjacent to the rigid barriers to contain the vehicle debrisduring an accident.

For the most part, these safety features have provided consid-erable protection; however, serious injuries and fatalities to thedrivers and fans are still occurring. Therefore, the need still existsto provide improved safety to those participating in the sport ofautomobile racing.

PROJECT SCOPE

The Midwest Roadside Safety Facility, sponsored by IRL and theIndianapolis Motor Speedway (IMS), which were later joined byNASCAR, was asked to develop a new high-speed crash barrier sys-tem whose crash performance was significantly improved over thoseof existing systems. Additional design criteria for the new systemincluded a self-restoration capability and modularity. After a de-formable barrier is impacted at high speeds, it is important that thesystem self-restore to a reasonable degree so that the barrier can beleft in place during the remainder of the race. Modularity would allowthe system to be somewhat easily constructed and would allow par-tially deformed sections to be replaced between races. Under someextreme circumstances, a section of the barrier could be replaced dur-ing a race if the modular units were quickly and easily disconnected,replaced, and reconnected. For example, if an energy-absorbing foamsystem was used, cartridges could be quickly replaced during a briefhalt in the race.

The reduction of lateral impact loads is the critical major designgoal of this project. At impact speeds between 160 and 350 km/h(100 and 220 mph) and impact angles in excess of 20 degrees, thelateral impact forces can be significant enough to cause severe struc-tural damage to the vehicle and serious injuries or fatality for thedriver. Barrier systems typically installed along highways are eval-uated at impact speeds of 100 km/h (62 mph). These systems are de-signed to meet the safety standards specified in NCHRP Report 350:Recommended Procedures for the Safety Performance Evaluationof Highway Features (1). It is fairly obvious that barrier systemsdesigned for 100-km/h (62-mph) impacts would not perform the sameif they were impacted at the higher speeds obtained on racetracks.

New Energy-Absorbing High-Speed Safety Barrier

John D. Reid, Ronald K. Faller, Jim C. Holloway, John R. Rohde, and Dean L. Sicking

J. D. Reid, N104 WSEC (0656), R. K. Faller and J. C. Holloway, 1901 Y, Build-ing C (0601), and J. R. Rohde and D. L. Sicking, W348 NH (0531), Universityof Nebraska, Lincoln, NE 68588.

54 Paper No. 03-2218 Transportation Research Record 1851

As an example, during the first phase of the research project, an820-kg (1,808-lb) small car was impacted into a concrete wall at160 km/h (100 mph) and an angle of 20°. During this crash test, thepassenger side of the vehicle crushed considerably, while the carwas redirected to travel along the side of the concrete barrier. By useof the procedures recommended in NCHRP Report 350, the lateraloccupant ride-down acceleration and occupant impact velocity werecalculated to be 26.5 g and 10.0 m/s (32.8 ft/s), respectively. Therecommended maximum allowable values for these measurementsare 20.0 g and 12.0 m/s (39.4 ft/s), respectively. However, it shouldbe noted that the recommended values are for 100-km/h (62-mph)impacts and that NCHRP Report 350 does not address crash testingat higher speeds. Thus, the design goal of this high-speed safety bar-rier development project is to develop a deflecting wall that allows asignificant reduction in the lateral impact loads but that does not allowan increase in longitudinal vehicle decelerations or the potential forvehicle snag, pocketing, or high-angle redirection away from the wall.

BACKGROUND

Over the years, former race car drivers, engineers, researchers, andspeedway owners and employees have worked to advance the stateof the art with respect to safety barriers. A simple solution has con-sisted of loosely stacking foam blocks around the exterior walls ofthe track, thus reducing the impact between the errant vehicle andthe rigid wall. Consequently, the barrier debris resulting from theuse of this option can be significant as well as costly because of theinefficient use of an energy-absorbing material. Another barrier con-cept has consisted of the placement of banded tires around selectedregions of the track. However, this configuration can result in sig-nificant vehicle override or underride and can produce hazards thatmay penetrate the vehicle or interfere with the driver.

More recently, several other advances in safety barriers have beenidentified and developed. A new compression barrier consisting oflarge-diameter, thick-walled resilient cylinders was developed forattachment to a rigid concrete racetrack wall (2). For this system,now known as the FLAG barrier, the cylinders are placed adjacentto one another, thus forming a longitudinal row with the cylinderopenings positioned parallel to the vertical axis. Smaller-diametercylinders, placed on the traffic-side face of the longitudinal barrier,are attached between the recessed regions where the larger cylindersare no longer in contact, thus reducing the potential for vehiclepocketing. In 1998, researchers at the Department of AerospaceEngineering of Politecnico di Milano in Milan, Italy, published theresults of a computer simulation effort that investigated and evalu-ated several racing barrier options by using both moving sled test-ing and computer simulation modeling with VEDYAC (3). Afterthe research study, the researchers provided several conclusionswith regard to a new barrier concept, known as GrandPrix, as wellas its use in combination with other safety barriers, such as tires andwater-filled modules.

Finally, John Pierce of Kestrel Advisors, Inc., and Kevin Forbesof IMS, in conjunction with IRL, developed a polyethylene energy-dissipating system known as the Polyethylene Energy DissipatingSystem (PEDS) barrier for use on oval racetracks in early 1998. Thenew barrier system was configured by using high-density polyeth-ylene (HDPE) cylinders that were covered with a thin HDPE coverskin on the front and upper surfaces. To expedite construction andrepair, the barrier system was designed and fabricated into modularunits that were then attached to the concrete wall with a cable

restraint system. The cover skin was used to reduce the potential forvehicle pocketing in the front face as well as to eliminate the poten-tial for driver limbs to become caught in the cylinder openings onthe upper surface.

During the IROC race at IMS in August 1998, race car driver ArieLuyendyk was involved in a crash that resulted in his car impactingrearward on the PEDS barrier installed downstream from the insidecorner of Turn 4. The estimated impact condition for this event con-sisted of a 1,633-kg (3,600-lb) car striking the barrier at a speed of209 km/h (130 mph) and an angle of 32°. Remarkably, after thissevere impact event, the driver did not sustain any serious injury,very likely because of the existence of the PEDS barrier, the rearwardimpact orientation, as well as other improvements in vehicle safety.Although this barrier system mitigated the severity of the impact,modifications to the PEDS barrier were warranted because severalmodular units became detached and scattered across the track,thereby causing potentially hazardous debris for oncoming cars. Sev-eral modifications were then made to the PEDS barrier to increase itsenergy-absorbing capabilities as well as to prevent the units frombecoming dislodged. This second generation of the PEDS barriersystem was termed the PEDS-2 barrier.

ENERGY-ABSORBING TECHNIQUES

Although many energy-absorbing techniques exist, the concentrationhas been on two specific types: (a) HDPE plates and (b) crushablefoam. HDPE plates of various shapes and sizes can be designed toabsorb significant energy through buckling and subsequent bendingof the buckled plate. In many instances, HDPE plates can be designedso that they recover their shape and energy-absorbing capability rel-atively quickly and thus might be reusable during a race. Energy-absorbing foams can be designed to provide uniform crushing loads.Polystyrene foams and recoverable polyurethane foams can recovertheir shapes considerably, but on reloading their effectiveness isgenerally unknown.

Several different variations of these concepts were analyzed duringthe project. This analysis included analytical calculations, computersimulation, static components and material testing, dynamic testingof components and subsystems of bogie vehicles, and full-scalevehicle crash testing of complete systems. Two of the initial barrierprototypes investigated, shown in Figure 1, used HDPE material asa spin-off of the PEDS-2 system.

A primary HDPE system is shown in Figure 1a. The system con-sists of energy-absorbing panel members and an HDPE front skin50 mm (2 in.) thick. The cables shown are used to provide tensionto the system during an impact. A foam-based system is depicted inFigure 1b. That system uses a thin steel plate skin and Z-shapedHDPE plates between foam cartridges. Although both of these typesof systems showed promise in regard to the absorption of energy,they lacked the desired attributes, such as the prevention of pocket-ing, low levels of friction between the vehicle and the outer skin,puncture strength, bending strength, and ability to transfer loads toa large number of cartridges at the same time.

FULL-SCALE TESTING

After examination of the available literature and documentation ofprevious real-world crashes provided by IRL and after a few initialfull-scale crash tests, the researchers narrowed the design impact

conditions down to two impact angles and two impact speeds:impact angles of 20° and 25° and impact speeds of 190 to 245 km/h(120 to 150 mph). It was believed that if safety improvements atthese speeds and angles could be demonstrated, similar safetyimprovements would be available for actual racetrack applications.Typical test impact conditions for both IRL and NASCAR vehiclesare shown in Figures 2a and 2b respectively. A total of 20 full-scalecrash tests were run to develop the system discussed in this paper.Those tests are summarized in Table 1.

The crash testing procedures and the recording and documenta-tion of the data under these unprecedented impact conditions provedto be very difficult, challenging, and dangerous. Therefore, detailsof the enormous tasks, problems, and solutions will be documentedin a separate research paper focusing on high-speed crash testing.

SIMULATION

Computer simulation with LS-DYNA (4), a nonlinear finite ele-ment analysis code, was used throughout the project. A paperdetailing the initial simulation efforts was presented at the 2000American Society of Mechanical Engineers Congress and is pub-lished elsewhere (5). In brief, that paper showed that both energy-absorbing methods (buckling and folding of HDPE plates andcrushing of foam blocks) indicated that a significant reduction of

Reid et al. Paper No. 03-2218 55

the pulse was attainable during high-speed impacts. The systemspresented showed great promise in improved high-speed safetybarrier designs.

Simulation results for the impact of a NASCAR Winston Cup carinto a steel tube and foam barrier design are shown in Figure 3. TheWinston Cup car model was provided by NASCAR and was origi-nally developed by Altair Engineering. One of the difficulties encoun-tered with the simulation was contact penetrations that led to codeinstabilities, which subsequently caused LS-DYNA to abort. This wasdue to the high speed of the impact. Modifications to improve the sim-ulation stability included refinement of the mesh size of both the vehi-cle and the barrier, lowering of the overall time step, and modificationof the contact penalty scale factors.

The primary benefits of simulation turned out to be evaluation ofthe various foam alternatives and determination of the amount ofenergy-absorbing foam to be used in the final barrier system.

DESIGN

After analysis of all the research completed on this project, a finaldesign was recommended to IRL for construction at IMS, known asthe Brickyard. This section provides some of the details of thatdesign. The authors intend to present the details of the developmentprocess that led to the system recommended for the Brickyard in a

(a) (b)

FIGURE 1 Two initial barrier prototypes: (a) primary HDPE system and (b) foam-based system.


most likely could accommodate barrier heights ranging between 914and 1,067 mm (36 and 42 in.) without further testing. Four internalstiffened steel splices connect each of the modules together withbolted fasteners. Bundles or cartridges of sheets of extruded poly-styrene foam 51 mm (2 in.) thick are placed between the existingconcrete wall and the steel tubing every 3 m (10 ft). Six or sevensheets of polystyrene make up each cartridge, depending on thelocation of the module. Additionally, cables 9.53 mm (3⁄8 in.) indiameter are used to anchor the steel tubing to the concrete wall andare spaced on 3-m (10-ft) centers.

SYSTEM PERFORMANCE

The results from the crash testing program clearly indicated that theperformance of the new barrier design was improved comparedwith that of rigid concrete walls in high-speed crashes. This is nottrivial, because concrete walls exhibit some very desirable featuresnot easily reproduced with deformable barriers, including no riskof snagging or pocketing and low levels of sliding friction betweenthe vehicle and the wall.

Tests IRL-17 and IRL-15 were Winston Cup vehicle tests withimpacts at 25° without and with the new barrier design, as shown inFigures 5a and 5b respectively. The vehicle in the test with the con-crete wall (Test IRL-17) impacted the wall at 225 km/h (140 mph),while the vehicle in the test with the new barrier (Test IRL-15)impacted the barrier at 238 km/h (148 mph), a significant increasein initial impact severity. A reduction in vehicle damage due to thenew barrier was evident, as shown in Figure 5.

Tests IRL-9 and IRL-16 were IRL open-wheeled vehicle testswith impacts at 21° without and with the new barrier design, asshown in Figures 6a and 6b respectively. Both tests were performedat approximately the same speed, 230 km/h (143 mph). A reductionin lateral decelerations between the two tests was significant, asshown in Figure 6.

Although space limitations prevent a complete record of the crashtesting results from being provided, Table 2 shows some of the full-scale results associated with the IRL open-wheeled vehicles and the

(a)

(b)

FIGURE 2 Two test vehicles and their impact conditions: (a) IRLvehicle and (b) NASCAR vehicle.

Test Type Barrier Description Test Quantity

HDPE Skin with HDPE Panel Absorbers

2

Bogie Vehicle

Steel Tube Skin and Spaced Foam Absorbers (Rearward)

1

Concrete Wall

1

Small Car

HDPE Skin with Continuous Foam Absorbers

2

Large Sedan

HDPE Skin with HDPE Z-Plates

1

Concrete Wall

1


1

IRL Open-Wheeled

Steel Tube Skin with Spaced Foam Absorbers

3

Concrete Wall

2


1

Steel Plate Skin with Continuous Foam Absorbers

1

Steel Tube Skin with Continuous Foam Absorbers

1

NASCAR Winston Cup

Steel Tube Skin with Spaced Foam Absorbers

3

TABLE 1 Summary of Full-Scale Testing

separate paper. However, the system described herein is very similarto several of the final full-scale crash systems tested.

A schematic of the recommended design is shown in Figure 4. Thebarrier is 521 mm (20.4 in.) thick and consists of modules 508 m(20 ft) long composed of four welded rectangular structural steeltubes. The steel tubing impact plate is 978 mm (38.5 in.) tall but


(a)

(b)

FIGURE 3 Simulation of impact of NASCAR Winston Cup car into IRL barrier at 120 mph: (a) overview and (b) close up to barrier.


FIGURE 4 Design schematic.

(a) (b)

FIGURE 5 NASCAR Winston Cup car impact test results: (a) without new barrier design and (b) with new barrier design.


Lateral Deceleration Comparison

120

100

80

60

40

20

20

10

0

–10

–20

0

0 0.05 0.1 0.15

0 0.05 0.1 0.15

–20

Legend

A IRL 9B IRL 16

Legend

A IRL 9B IRL 16

Dec

eler

atio

n [

g]

Dec

eler

atio

n [

g]

Time [s]

(a)

Time [s]

(b)

Longitudinal Deceleration Comparison

FIGURE 6 IRL open-wheeled vehicle test results: (a) without new barrier design and (b) with newbarrier design.

TestNo.

VehicleStyle

IRLIRLIRLIRLIRL

NASCARNASCARNASCARNASCARNASCARNASCARNASCARNASCAR

IRL-6IRL-9IRL-11IRL-14IRL-16

IRL-7IRL-8IRL-10IRL-12IRL-13IRL-15IRL-17IRL-18

TestDescription

HDPE Skin and Continuous FoamConcrete Wall

Steel Tube Skin and Spaced FoamSteel Tube Skin and Spaced FoamSteel Tube Skin and Spaced Foam

HDPE Skin and Continuous FoamConcrete Wall

Steel Plate Skin and Continuous FoamSteel Tube Skin and Spaced Foam

Steel Tube Skin and Continuous FoamSteel Tube Skin and Spaced Foam

Concrete WallSteel Tube Skin and Spaced Foam

VehicleWeight

(kg)

ImpactSpeed

(km/h)

ImpactAngle

(degrees)

ImpactSeverity(kN-m)

ExitTrajectory

Angle(degrees)

PeakLongitudinalDeceleration

(g)

PeakLateral

Deceleration(g)

DynamicWall

Deflection(mm)

102.0NA224315188

251NA69

27779

315NA208

NA137.0449.20

177.2269.97

47.5352.1829.3250.7794.41

109.6587.8146.72

NA26.5214.2117.7322.63

41.8329.6913.3319.2462.7547.5166.9525.79

small11.58.05.34.5

11.53.9

13.610.75.99.35.58.9

101234228370233

292282135355660661568325

23.720.721.325.320.7

20.020.519.321.725.625.624.921.5

136229225241229

199191140203237238225196

879926886904923

1,6321,6321,6301,6321,6321,6191,6401,630

NA = Not available or not applicable.

NOTE: Peak decelerations are just one data point and should not be used to evaluate the overall effectiveness of a system.

TABLE 2 Full-Scale Testing Results

NASCAR Winston Cup vehicles. Relatively high peak lateraldeceleration anomalies for Tests IRL-14 and IRL-15 were attrib-uted to the barrier system bottoming out at 315 mm. These werealso the highest-severity impact conditions tested. By selection ofthe proper energy-absorbing foam, the system can be optimized tolimit the system from bottoming out for a given impact severitytest condition.

Energy reduction and management occur in stages beginningwhen a vehicle hits the barrier at a high speed. The extreme forceof the impact causes local deformation of the steel tubing. Thisalmost instantaneous local deformation allows a reduction of extremepeaks due to the initial impact. However, the tubing is thick enoughto prevent significant gouging or snagging of the vehicle. As a tubedeforms locally, the welds in the vicinity of the impact, which con-nect surrounding tubes, may also begin to stretch and deform,absorbing more of the energy of the high-speed impact. As theheavy steel tubing begins to be pushed back, energy is then absorbedbecause of the inertia required to move those system components.Finally, the foam cartridges in the area of the impact begin tocompress and absorb additional energy, providing a sustainedcushion for the impacting vehicle. Because of the large structuraltube modules (i.e., those with high bending strengths), the steeltubes transfer the load upstream and downstream of the impact toseveral foam cartridges at relatively the same time. As the steeltubes bend toward the concrete wall (primarily elastically), the farupstream and the far downstream ends of the tubes are preventedfrom pulling out and away from the concrete wall by the steelcable anchors.

Because of the interaction of all the energy-absorbing compo-nents, as described above, for many of the impacts the final plasticsystem deformation is minimal and the race can continue withoutthe need for repairs or replacement of the system. As an example,the barrier damage in Test IRL-15 [at 238 km/h (148 mph) and 25°]is shown in Figure 7. In some cases, severe local deformation onsteel tubes that warrant immediate attention may take place. Thesecan be quickly repaired by welding a steel plate over the damagedarea. Additionally, if a few foam cartridges are significantly dam-aged, they can be easily replaced in only a few minutes. In the pre-dicted rare event of severe damage to the barrier, complete mod-ules of the system can be removed and replaced in approximately30 to 45 min.

BRICKYARD INSTALLATION AND REAL-WORLD RESULTS

On receipt of the design plans, IMS built the first installation of thenew barrier at the Brickyard, as shown in Figure 8. This racetrackbarrier system was named the SAFER (steel and foam energy reduc-tion) barrier. Installation of the barrier began in late April 2002, justin time for use in the practice and qualifying laps for the 2002 Indi-anapolis 500 race in the month of May, as well as for the 2002Indianapolis 500 race. The Indianapolis 500 race is thought by manyto be the worldwide premier event in motor racing.

During the first weekend of practice with the SAFER barrierinstalled, Robbie McGehee lost control of his vehicle and experi-enced a high-speed impact into the new barrier. Just a few weeksearlier at the Brickyard, Eliseo Salazar experienced impact condi-tions that were deemed to be nearly identical to those experiencedby McGehee, but into an unprotected concrete wall. The speeds ofboth impacts were estimated to be 290 km/h (180 mph). In the first


crash, Salazar went to the hospital in serious condition after experi-encing a peak crash deceleration of 115 g. Emergency surgery wasrequired to repair a torn vertebral artery in his chest. In the secondcrash, McGehee walked away from the crash after experiencing apeak crash deceleration of 43 g from the first contact with the SAFERbarrier. A comparison of the two acceleration crash pulses for theinitial impact event is shown in Figure 9.

During Lap 30 of the actual Indianapolis 500 race, Greg Ray,1999 IRL Champion, did a half-spin in Turn 1 and hit the SAFERbarrier with the rear of his car at nearly 354 km/h (220 mph), asshown in Figure 10. Heavy damage occurred to his race car, but Raywas checked and released without injury. Ray was subsequentlycleared to drive.

CONCLUSIONS

A new barrier system was designed to mitigate the severity ofhigh-energy vehicular impacts into rigid containment walls. Thenew energy-absorbing barrier system, the SAFER barrier system,was successfully developed, tested, and evaluated for high-speedoval racetrack applications by using LS-DYNA computer simula-tion modeling and full-scale vehicle crash testing. The SAFERbarrier system, which attaches to concrete walls, has been shownto reduce both peak vehicle and occupant decelerations comparedwith those observed during impacts with concrete walls, thusdecreasing the potential for serious and fatal injuries, which canoccur in high-speed racing accidents.

This new barrier system has been designed primarily for use as protection for errant vehicles at high-risk locations and foroblique-angle vehicular impacts, such as on the outside of curveson racetracks. However, because this barrier is primarily, but notexclusively, a longitudinal barrier, the technology has potentialapplication as a roadside barrier in locations with high accident fre-quencies, such as on curves in tunnels and congested roadways. Thistechnology also has application in the retrofitting of rigid bridge rail-ings and other permanent or temporary traffic barriers. Finally, thisbarrier technology may potentially be applied in situations in whichperpendicular severe, high-speed impacts into the system are antici-pated, for example, in situations in which crash cushions, end termi-nals, and truck-mounted or trailer-mounted attenuators are required.However, all future developments of this technology for use in road-side applications would require an economics review, design modi-fications or reductions, and evaluation through the use of full-scalevehicle crash testing according to the guidelines provided in NCHRPReport 350.

In the future, it is recommended that the in-service safety per-formance of the SAFER barrier system be continuously moni-tored. If actual in-service performance shows a need for potentialmodifications on the basis of real-world accident experience,those barrier modifications would be studied and evaluated by useof a combination of computer simulation modeling and full-scalevehicle crash testing. Finally, the SAFER barrier system wasdeveloped on level terrain, was configured with a barrier systeminstalled parallel to the roadway (i.e., a configuration represent-ing a corner with a large radius), and included oblique-angleimpacts with vehicles tracking forward into the barrier system.However, it is widely known that many of the oval speedwayshave high-banking corners, smaller-radius corners, or combina-tions of both corner types. In addition, many of the race car acci-dents occur as impacts that are rearward tracking or nontracking


(a)

(b) (c)

FIGURE 7 Three areas of barrier damage in Test IRL-15.


(b)(a)

100

80

60

40

20

0

0 0.02 0.04 0.06 0.08 0.1 0.12–20

Legend

A SalazarB McGehee

Acc

eler

atio

n [

g]

Time [s]

FIGURE 8 Brickyard installation of SAFER barrier: (a) track view and (b) close-up.

FIGURE 9 Acceleration crash pulses for impact events experienced by McGehee into the SAFER barrierand Salazar into a concrete wall.

events. Therefore, future research is recommended to allow engi-neers to better understand the dynamic behavior of the SAFERbarrier system when it is installed on high-banking, tight-radiuscorners or when it is subjected to rearward tracking or nontrackingimpact events.

ACKNOWLEDGMENTS

The authors gratefully acknowledge IRL, IMS, and NASCAR forsponsoring the project; John Pierce of Kestrel Advisors; IRL andNASCAR race teams and Ford Motor Sports for donating vehi-cles; Wayne State University and Delphi Automotive Systems forproviding dummies and various crash test sensors; and the Mid-west Roadside Safety Facility personnel. Additionally, the authorsthank Livermore Software Technology Corporation for its contin-ued support of LS-DYNA. The simulation work performed duringthis project was completed by using the Research ComputingFacility of the University of Nebraska–Lincoln.

REFERENCES

1. Ross, H. E., Jr., D. L. Sicking, R. A. Zimmer, and J. D. Michie. NCHRPReport 350: Recommended Procedures for the Safety PerformanceEvaluation of Highway Features. TRB, National Research Council,Washington, D.C., 1993.

2. Fitch, J., C. Goodwin, D. C. Alberson, and D. L. Bullard, Jr. Impact Atten-uation Devices for Racing. Paper 983063. Proc., 1998 Motorsports Engi-neering Conference, Vol. 1. Vehicle Design and Safety Issues, P-340/1.SAE, Warrendale, Pa., Nov. 1998, pp. 199–208.

3. Giavotto, V., and P. Astori. Design of Safer Barriers for Race Circuits.Paper 983064. Proc., 1998 Motorsports Engineering Conference, Vol. 1.Vehicle Design and Safety Issues, P-340/1. SAE, Warrendale, Pa.,Nov. 1998.

4. Hallquist, J. O. LS-DYNA Keyword User’s Manual, Version 960. Livermore Software Technology Corporation, Livermore, Calif., March2001.

5. Reid, J. D., R. K. Faller, and D. L. Sicking. High Speed Crash BarrierInvestigation Using Simulation. In Crashworthiness, Occupant Protectionand Biomechanics in Transportation Systems—2000, American Societyof Mechanical Engineers, AMD-Vol. 246, Nov. 2000, pp. 111–127.

Publication of this paper sponsored by Committee on Roadside Safety Features.


(a)

(b)

(c)

FIGURE 10 Three photos of impact of half-spin turn into SAFERbarrier by car driven by Ray during the Indianapolis 500 race.

Documents

New Energy-Absorbing High-Speed Safety Barrier