Limitations of Current Bumper Designs and Potential Improvements

Author(s): David Aylor, Danny L. Ramirez, Matthew Brumbelow and Joseph M. Nolan
Source: SAE Transactions , 2005, Vol. 114, SECTION 6: JOURNAL OF PASSENGER CAR:
MECHANICAL SYSTEMS JOURNAL (2005), pp. 1591-1598
Published by: SAE International

Stable URL: http://www.jstor.com/stable/44725186

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

SAE International is collaborating with JSTOR to digitize, preserve and extend access to SAE
Transactions

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
 2005-01-1337

Limitations of Current Bumper Design

and Potential Improvements
David Aylor, Danny L. Ramirez, Matthew Brumbelow and Joseph M. N
Insurance Institute for Highway Safety

Copyright © 2005 SAE International

INTRODUCTION Cars underrode minivans, pickup trucks, and sport utility

vehicles (SUVs) more frequently, and the resulting dam-
For every 100 insured passenger vehicles up to 3 age was even more expensive to repair. Cars under-
years old, about 7 insurance claims are paid each rode minivans in 30 percent of the cases, with average
year for collision damage to the insured vehicles, repair estimates of $1,584 compared with $780 for non-
costing an average of $3,721 per claim. This excludes underride crashes. Cars underrode pickups in 58 per-
costs covered by collision insurance deductibles, cent of the cases, with average claims of $1 ,543 com-
which typically are $250 or $500. It costs about $265 pared with $955 for crashes not involving underride.
per year per vehicle to pay these claims, excluding Underride crashes occurred most commonly between
administrative costs.1 It costs another $83 per year to cars and SUVs. Cars underrode SUVs in 67 percent of
pay for other property damage caused by the vehi- the cases, with average repair estimates of $1,378
cles, typically to another vehicle in a two-vehicle colli- compared with $802 for non-underride crashes.
sion (collision damage insurance covers damage to
the insured vehicle; property damage liability insur- Review of real-world crash damage indicates three com-
ance covers damage caused to other vehicles).2 Thus ponents of good bumper design that currently are lack-
the average recent model vehicle incurs crash dam- ing on many passenger vehicles: compatible geometry,
age costs of at least $348 per year. For 2003 the stability during impacts, and effective energy absorption.
costs to repair these vehicles and the damage they Bumpers must be located and sized so they engage the
caused would have totaled $16-18 billion, based on an bumper systems on other vehicles with sufficient over-
estimated 47 million vehicles. The high cost of prop- lap to account for variations in ride height due to occu-
erty damage in motor vehicle crashes also has been pant and cargo loading and braking. Once engaged,
recognized by the federal government.3 bumper systems should offer a stable interface and re-
main engaged throughout the impact. Bumper stability
Eighty-one percent of vehicle damage repair estimates can be influenced by bumper cover geometry, bumper
are for front or rear impacts, and 65 percent of these reinforcement bar shape and strength, and energy ab-
entail costs less than $2,500.4 Vehicle bumpers could be sorber design. After the geometry and stability require-
expected to play a major role in preventing or limiting ments have been met, bumpers still must have suffi-
the damage. However, regulatory requirements for car cient energy absorption capabilities to confine damage
bumpers are inadequate, and many passenger vehicles to the bumper system itself.
are not even required to have bumpers. This problem is
illustrated in an Insurance Institute for Highway Safety This paper describes the limitations of current bumper
(MHS) survey of damaged vehicles brought into five designs with regard to the necessary components of
Washington, D.C. metropolitan area drive-in claims cen- good low-speed crash protection and reviews an alter-
ters for repair estimates.5 In many cases, vehicles in- native test procedure under development that could en-
volved in front-into-rear crashes sustained significant courage much needed improvements.
damage to safety equipment (lights) and cosmetic parts
(hoods, fenders, grilles), with only minor damage to the GEOMETRY
bumper itself. Often this resulted from underride, either
because the bumpers failed to match up or because the The first limitation of many vehicle bumper systems is
bumpers did not remain engaged during the impact. In poor geometry. According to Federal Motor Vehicle
car-to-car collisions, underride occurred in 21 percent of Safety Standard (FMVSS) 581 , the purpose of a bumpe
the cases and increased the average repair estimate to system is to "reduce physical damage to the front and
$1 ,083 compared with $750 without underride. rear ends of a passenger motor vehicle from low speed

1591

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
collisions."6 To accomplish this the bumpers on the front and midsize SUV from the same manufacturer: Ford
and rear of a vehicle should be located and sized to
Taurus and Explorer, Chevrolet Malibu and TrailBlazer,
Dodge Stratus and Jeep Grand Cherokee, Nissan Al-
align with and engage the bumpers on the other vehi-
cles they will encounter in a crash. tima and Murano, and Volvo S40 and XC90. Two tests
were conducted on each pair, both at 16 km/h (10 mi/h)
- the front of the sedan into the rear of the SUV and
This is not the case today. Figure 1 shows a static
then the front of the SUV into the rear of the sedan. Ve-
height comparison of the average front and rear bumper
reinforcement bars of many popular 2003-05 small andpairs had varying degrees of bumper overlap (Fig-
hicle
midsize sedans and SUVs. Although the average ures
front3a-c).
bumper bar for small cars is 496 mm from the ground at
its highest point, the average rear SUV bumper bar is
512 mm high at its lowest point. Even without the effects
of precrash braking, the bumpers on most small sedans
will not engage those of SUVs in front-into-rear colli-
sions. Similarly, when an SUV rear ends a small sedan,
the average bumpers have only 36 mm of overlap, an
amount that almost certainly will disappear if either ve-
hicle is braking at the time of impact.

Figure 2 - New Vehicle Registrations by Vehicle Type ( Source : R.L.

Polk & Company7)

Figure 1 - Average Bumper Bar Heights by Vehicle Type (SUVs,

Small Cars, and Midsize Cars)

The FMVSS 581 test does force some measure of com-

patibility among passenger car bumpers because the
bumpers are tested with pendulum impacts 16-20
inches from the ground. The result is that most car
bumpers are somewhat geometrically compatible in their
static, unloaded conditions. However SUVs, pickups,Figure 3a - Bumper Bar Comparisons, Sedan Front into SUV Rear
and minivans are exempt from this standard. In fact,
these vehicles are not required to have bumpers at all,
as is the case for the rear of the 2001-04 Toyota RAV4.
The significance of exempting light truck vehicles
(LTVs) from any bumper requirement has grown dra-
matically during the past decade as these vehicles (par-
ticularly SUVs) make up an increasing proportion of the
overall vehicle fleet. Since 1975 LTV sales have grown
from 16 to 47 percent of all new vehicles (Figure 2). As
SUV sales have grown, so has the likelihood of bumper
mismatches in everyday crashes.

In June 2004 MHS conducted a series of crash tests to

examine more closely the interactions between SUV
and passenger car bumpers in minor collisions.8 The
tests involved five pairs of vehicles, each with a sedan Figure 3b - Bumper Bar Comparisons, SUV Front into Sedan Rear

1592

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
 IIHS also conducted a 16 km/h front-into-rear test with a
pair of 2004 Toyota RAV4s. The RAV4 has no bumper
reinforcement bar in the rear. It also has a rear mounted
spare tire that makes first contact with most objects
striking it from behind. As a result, the striking RAV4
sustained almost $4,400 in damage, whereas the struck
vehicle had more than $3,200 in damage. Two vehicles
should be able to withstand a collision of this severity
with only minimal damage, even without the ideal condi-
tion of being the same model.

Bumper geometric incompatibility is worsened when

different real-world vehicle loading and braking scenar-
ios are considered. In these situations, static overlap is
insufficient to guarantee good bumper engagement.
Vehicle ride heights can change drastically under dy-
namic conditions. In the insurance claims study dis-
cussed earlier, most of the significant damage resulted
from front-into-rear collisions; the striking vehicle's front
bumper underrode the struck vehicle's rear bumper. If
both vehicles are braking in this type of crash, the dif-
ferences in bumper heights are magnified. Under heavy
braking, the rear of the struck vehicle lifts, whereas the
Figure 3c - Bumper Bar Comparisons, Four Vehicle Pairs (Arrows front of the striking vehicle dives, leading to more se-
Indicate Bumper Bar Locations)
vere underride and more expensive claims.

Combined vehicle damage for each paired test ranged To investigate these scenarios, IIHS conducted tests to
from about $1,250 to more than $6,000 (Table 1). Some evaluate the amount of vertical movement at the bump-
vehicles sustained damage causing major coolant leaks ers during vehicle braking. Front dive tests were con-
from broken radiators. In similar real-world collisions, these
ducted on SUVs and sedans at initial speeds of 32 km/h
vehicles would have to be towed from the scene, adding (20 mi/h) and 64 km/h (40 mi/h), and rear lift tests were
more expense to the already high repair costs. The Ford conducted at initial speeds of 64 km/h. At 32 km/h, the
Taurus and Explorer sustained the least amount of dam- SUVs averaged 8 cm of front dive, whereas the sedans
age, mainly because their bumpers were among the most averaged 6 cm. At 64 km/h, not all movement could be
geometrically compatible; the Explorer's bumpers com- captured in video analysis, leading to underestimates of
pletely overlapped those of the Taurus (Figures 3a-b). total front dive. Still, the SUVs averaged 9 cm of front
Conversely, the bumpers on some of the other vehicle dive and 6 cm of rear lift. The sedans averaged 7 cm of
pairs completely bypassed each other. This was the case front dive and 6 cm of rear lift. The maximum front dive
in the test of the Volvo S40 into the XC90; combined dam-
and rear lift were 11 and 8 cm, respectively, for the
age for both vehicles was more than $6,000. Similarly, SUVs and 9 cm for the sedans. If the SUV with maxi-
there was no bumper overlap in tests of the Nissan Mu- mum rear lift and the sedan with maximum front dive
rano into the Altima and the Dodge Stratus into the Jeep were to collide while braking, the total bumper move-
Grand Cherokee. Damage amounts for these tests were ment between the two vehicles would be about 17 cm,
correspondingly high.
or more than 6V2 inches. At their static ride heights,
Table 1 these two vehicles have 6 cm of bumper overlap. How-
Damage Repair Costs, 16 km/h (10 mi/h) Front-into-Rear Tests ever, when brake dive is considered the bottom of the
Sedan into SUV Sedan SUV Total
SUV bumper would be 1 1 cm above the top of the se-
Ford Taurus into Explorer $1,784 $824 $2,608
Chevrolet Malibu into Trailblazer $3,163 $937 $4,100
dan bumper, certainly leading to underride. By compari-
son,
Dodge Stratus into Jeep Grand Cherokee $3,256 $1 ,279 $4,535 the federal bumper standard test zone is only 4
Nissan Altima into Murano $4,507 $1,188 $5,695 inches tall. Compliance with this standard does not
Volvo S40 into XC90 $4,984 $1 ,096 $6,080
guarantee good bumper alignment in real-world crashes
SUV into Sedan Sedan SUV Total where braking is involved.
Ford Explorer into Taurus $701 $555 $1,256
Volvo XC90 into S40 $1 ,695 $2,361 $4,056
IIHS also reconstructed an insurance claim case involving
Chevrolet Trailblazer into Malibu $1,851 $2,316 $4,167
Nissan Murano into Altima $2,517 $2,485 $5,002 brake dive; the front of a 2001 Honda Accord underrode
Jeep Grand Cherokee into Dodge Stratus $2,848 $3,281 the rear of a 1992 Geo Storm (Figure 4). The Accord sus-
$6,129
tained extensive damage to its bumper cover, headlamps,
Note: Sedans and SUVs are 2004 models; the Altima is a 2005 model.
Repair costs reflect July 2004 parts and labor prices. hood, fenders, radiator, and air conditioning system total-

1593

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
 bumper standards, they still can allow damage costing
thousands of dollars in minor collisions, even when the
bumpers are aligned at the time of impact. One reason
is a lack of bumper stability. After vehicles first contact,
their bumper systems have to remain stable to stay en-
gaged and effectively absorb crash energy.

In 2004 IIHS described initial tests designed to isolate

Figure 4 - 2001 Honda Accord (Left) and 1992 Geo Storm (Right) the causes of bumper instability and establish an
from Drive-in Claims Case
evaluation procedure.9 The tests illustrated the ten-
dency of some bumper systems to underride or over-
ing more than $4,000; the bumper reinforcement bar was
ride after initial engagement. Two design flaws lead-
undamaged. The Storm required a new bumper system ing to instability were hypothesized. First, if the tor-
but sustained no other significant damage. sional bumper bar stiffness is not matched to the lon-
gitudinal crush resistance of the bumper mounting
The bumper reinforcement bars on these two vehiclessystem, the bumper bar may rotate under vertically
align statically (Figure 5), but IIHS brake dive experiments
offset loading, causing destabilizing forces to be ap-
found that, while braking, the rear of the Storm could lift by
plied to the bumper system. Second, in many cars
as much as 7 cm and the front of the Accord could dive by there is a significant offset between the bumper cover
the same amount. For the crash reconstruction, the vehi- and bumper bar, which often is filled with some type
cles were tested first at static height and then with theof energy-absorbing material. When two such vehicles
maximum 14 cm braking offset. Both tests were con-strike each other (or another compliant surface) the
ducted at 20 km/h (12.4 mi/h). Under static height condi-bumper bars do not load each other directly during the
tions, test damage was restricted to the bumper systems crash. Instead, the bumper covers and foam absorb-
on both vehicles. When the braking effects were applied ers interact with each other, often deforming asym-
by expanding the rear springs on the Storm and tighteningmetrically and producing vertical forces on the bumper
the front springs on the Accord, damage results were verysystems. IIHS showed that placing an energy-
similar to the actual insurance claim (Figures 4 and 6). absorbing material on a rigid barrier promotes under-
ride similar to what is seen in car-to-car crashes. Ad-
ditional unpublished research and testing have found
that the tendency of some vehicles to underride the
bumper barrier with an energy-absorbing face is not
an artifact of the specific absorber material but can be
replicated with a range of energy-absorbing materials.

In 1999 IIHS conducted vehicle-to-vehicle tests illustrat-

ing the bumper stability problem. Tests were conducted
with the front of a vehicle moving at 16 km/h striking the
Figure 5 - Static Bumper Heights for 1998 Honda Accord rear of a stationary vehicle of the same make and
and 1992 Geo Storm (Bumper Bars Shown in White)
model. In six tests, two models exhibited instability -
the 1999 Ford Windstar and 1999 Saab 9-3 (Figure 7).

Figure 6 - 1998 Honda Accord (Left) and 1992 Geo Storm (Right)
from IIHS Reconstruction Test

STABILITY

Matching geometry is a necessary first step in bumper

compatibility, but it is not the only key to good bumper
Figure 7 - Saab 9-3 (Top) and Ford Windstar (Bottom), IIHS 16 km/h
design. Although passenger cars meet minimum Stability Tests

1594

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
Another example of poor stability was seen in IIHS's
2004 test of a Chevrolet TrailBlazer into a Malibu at 16
km/h. The bumpers initially were aligned (Figures 3a-b);
during the crash, the bumpers engaged but slipped past
each other, causing the TrailBlazer to override the
Malibu (Figure 8). The result was nearly $4,200 in dam-
age to the two vehicles, even though their bumpers
were at similar heights. The TrailBlazer-into-Malibu test
demonstrates that a bumper system must be stable Figure 9 - Good Energy Absorption, 1998 Volkswagen New Beetle
enough to take advantage of matching geometry. Front (Left) and Rear (Right) Bumper Systems

Figure 10 - Good Energy Absorption, 2003 Saab 9-3

trunk lid. A 1998 Volkswagen New Beetle sustained no

damage in the rear barrier test and only required re-
Figure 8 - Chevrolet TrailBlazer into Malibu, IIHS 16 km/h alignment of the headlamps after the front barrier test.
Front-into-Rear Test

The bumper assembly on the Saab 9-3 has a foam ab-

ENERGY ABSORPTION sorber and steel reinforcement bar. The reinforcement bar
attaches to box-like, integral brackets that can serve as
Once a vehicle's bumper is initially aligned and able to absorbers. In the 5 mi/h flat barrier test, the brack-
energy
remain engaged with a range of other vehicles, itets and foam absorber absorbed much of the crash en-
re-
quires sufficient energy-absorbing capabilities to ergy,
pre- preventing damage to any underlying parts. Addi-
vent damage to other parts of the vehicle. IIHStionally,
has the foam and bumper bar provided enough offset
used 5 mi/h front and rear flat barrier tests as measures to prevent damage to more expensive parts of the vehicle.
of a bumper system's ability to absorb crash energy.10 In The total repair cost for this test was only $1 81 .
many of these tests, vehicles have sustained serious
damage to underlying body structures with little damage Many examples of poor energy absorption also have
to energy-absorbing material. A bumper system's en- been observed. Since 2002 47 percent of vehicles
ergy-absorbing material, including any foam, should be evaluated in the 5 mi/h front and rear flat barrier tests,
sacrificial and prevent high forces from damaging un- arguably the easiest two tests in the series of four, re-
derlying parts. A bumper system also should provide sulted in at least 3 hours of repair for body damage be-
adequate crush depth to absorb crash energy while yond the bumper. In addition, 63 percent of these vehi-
keeping the bumper bar away from expensive parts cles had little or no damage to the bumper bar itself.
such as the headlamps, hood, or trunk lid. Such results indicate there are many vehicles with body
structures that are more compliant than the bumper sys-
IIHS has observed many examples of both good and tem. Bumper system strength must be well matched
poor energy-absorbing bumper systems. Two examples with the underlying support structure on the vehicle to
of good energy absorption are the front bumpers on the ensure damage is confined to the bumper system itself.
1998-2005 Volkswagen New Beetle (Figure 9) and
2003-05 Saab 9-3 (Figure 10). The New Beetle's front Another problem IIHS has encountered in its bumper
and rear bumper systems have a steel bumper rein- test program is that some vehicles have inadequate
forcement bar mounted on hydraulic energy absorbers. space available for the bumper to deform before more
These absorbers successfully absorbed the crash en- expensive vehicle components become damaged. Tests
ergy from both the front and rear 5 mi/h flat barrier tests. of the 2001-05 Mercedes C320 (Figure 11) illustrate the
Another important aspect of the New Beetle's design is type of damage that can occur without adequate offset
the offset between the bumper system and other vehicle of the bumper system from the rest of the vehicle. The
components. The front and rear bumpers extend 1 3 and C320 had only 6 cm of front offset (compared with 13
17 cm, respectively, from the edges of the hood and cm for the New Beetle), resulting in both headlights be-

1595

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
 NEW TEST PROCEDURE

Geometry, stability, and energy absorption are three vit

elements of bumper design. However, current bump
designs target laboratory tests that produce deformat
modes unlike those observed in many real-world cras
(Figure 12). Furthermore, limiting bumper test requi
ments to passenger cars has produced a fleet in whic
many of the vehicles have incompatible bumper heights.
test protocol is needed that can address all three bum
design issues and be applied to the entire fleet.
Figure 11 - 2001 Mercedes C320

ing broken in the 5 mi/h front flat barrier test. In the rear
barrier test, the C320 sustained damage to the trunk lid
and rear body as a result of both insufficient crush depth
and a lack of energy absorption capability.

Before the federal bumper standard was changed in 1982,

bumper systems were designed with better energy-
absorbing capabilities. At that time the National Highway
Traffic Safety Administration rolled back the 5 mi/h tests Figure 12 - Honda Accord Damage from Real-World Underride
that permitted no vehicle damage to a 2.5 mi/h standard (Left) vs. IIHS 5 mi/h Front-into-Flat Barrier Test (Right)
allowing unlimited bumper damage. In the early 1980s
IIHS conducted tests showing that vehicles had become IIHS's previous bumper test program assessed energy
much more vulnerable to costly damage in low-speed im- absorption capability and beam strength but not the
pacts as a result of the weaker standard. The tests in- sues of bumper geometry or stability. In conjunction wit
volved five 1982 and 1983 model year vehicles with 5 mi/h the Research Council for Automobile Repairs, IIHS
and 2.5 mi/h bumper systems. In 5 mi/h front and rear developing a new bumper test procedure to incorpor
barrier tests, none of the vehicles with the older bumper these concepts. The procedure is intended to encour-
systems sustained damage. Those with newer bumpers age compatible bumper heights under dynamic cond
sustained damage ranging from $66 to $299 (Table 2). tions and good bumper stability after engagement wh
Even in 10 mi/h barrier tests, damage costs to vehicles still evaluating the bumper system's ability to abs
equipped with 2.5 mi/h bumpers were 48 percent higher energy. The test also is expected to more accurat
than for vehicles with stronger bumpers.11 reproduce damage sustained in real-world low-spe
collisions and drive better bumper designs that li
There also has been an increase in the frequency of in- damage.
surance claims for vehicles with newer bumper systems.
Claim frequencies for the 1983 Honda Accord and Civic The procedure incorporates a bumper-like steel barrier
increased 13 and 14 percent, respectively, compared with fitted with an energy-absorbing face and cover. The bar-
1982 models with bumpers designed for the 5 mi/h stan- rier is curved (Figure 13), which is typical of popular
dard.12 The older bumpers provided superior protection bumper designs. The underlying steel bumper barrier
against low-speed crash damage and eliminated signifi-
cant numbers of smaller collision claims.

Table 2

Bumper Damage Repair Costs*, 1982-83 Models

Model Bumper 5 mi/h front- 5 mi/h rear-

Make and Model Year System into-barrier into-barrier
Honda Accord 1982 5 mi/h $0 $0
1983 2.5 mi/h $299 $198
Honda Civic 1982 5 mi/h $0 $0
1983 2.5 mi/h $188 $157
Plymouth Horizon 1983 5 mi/h $0 $0
1983 2.5 mi/h $66 $181
Plymouth Reliant 1983 5 mi/h $0 n/a**
1983 2.5 mi/h $172 n/a**
Volvo GL 1982 5 mi/h $0 $0

•Repair costs were based on January 1983 parts prices and a

••Plymouth Reliants had "2.5 mi/h" bumpers only on the front
in the rear.
Figure 13 - IIHS Bumper Barrier

1596

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
face is 4 inches tall and can be mounted at various heights
from the ground. For example, current MHS testing mounts
the barrier 18 inches off the ground, 2 inches higher than
the FMVSS 581 pendulum test (Figure 1). One research
question is whether this strategy could be used to address
the problem of incompatible bumper geometry by encour-
aging taller bumper designs on passenger cars and lower
designs on LTVs. Another question is whether overlapping
the FMVSS 581 test zone would promote bumper systems
designed for broader coverage.

The addition of an energy absorber and bumper cover

over the barrier face (Figures 14 and 15) is intended to
promote stable bumper designs by simulating the inter-
action between bumper faseias that occur in real-world
impacts. Vehicles that do not engage a significant por-
tion of the barrier surface likely will deform the compliant
surface asymmetrically, resulting in destabilizing forces
on the vehicle that can lead to underrlde or override.
The energy absorber and cover also serve to distribute
impact forces in a more realistic way than would a rigid
barrier, minimizing high inertial forces that arise when
striking a rigid barrier. In rigid barrier tests, these high
forces can be used to initiate buckling in bumper sup- Figure 15 - IIHS Bumper Barrier with Prototype Energy Absorber
and Bumper Cover
ports, yet such high initial forces are not likely to occur
in real-world crashes.
CONCLUSIONS

Energy Absorber Specifications

• Provided by NetShape Corporation During the past few years, IIHS has conducted researc
• Fit over bumper barrier at a radius of 340 cm refine its bumper test program and improve real-wo
• 152 cm long, 10 cm tall, 5 cm deep
• Vacuum formed to a thickness of 1 -1 .6 mm
bumper performance. Studies have shown that vehi
• Averages 550 kPa at initial displacement damage reported at drive-in claims centers is worse a
the patterns of deformation are different than damage
tained in low-speed laboratory tests. Results from t
drive-in claims study prompted the investigation of veh
underride and override common in real-world crashes and
associated with significantly higher repair costs.

Bumper Cover Specifications IIHS research has identified three weaknesses in cur-
• Vacuum formed to an estimated
thickness of 3 mm rent bumper designs: geometry, stability, and energy
• Fit over energy absorber at a radius absorption. Many vehicles have incompatible bumper
of 340 cm
systems that do not engage during low-speed collisions.
With the increase in SUV sales during the past decade
Figure 14 - IIHS Energy Absorber and Bumper Cover Including and their exemption from FMVSS 581, the issue of
Specifications
bumper geometry has been amplified. Occupant loading
and vehicle braking also can cause statically aligned
Finally, vehicles with good bumper geometry and stable bumpers to underride or override. With regard to
designs that remain engaged with the barrier will be bumper stability, IIHS tests show some vehicle bumpers
evaluated on the bumper system's ability to absorb that initially were aligned became unstable due to verti-
crash energy. Once the energy absorber on the barrier cal loading causing underride or override. IIHS research
has bottomed out, the underlying steel structure will re- into these limitations prompted the development of a
quire the bumper system to have sufficient energy- new test procedure that addresses bumper geometry
absorbing capacity and crush distance to prevent dam- and stability while still assessing a bumper system's en-
age to expensive parts. The slightly higher impact ergy-absorbing characteristics.
speed than previous tests (10 km/h compared with 8
km/h) accounts for the crash energy absorbed by the The purpose of the new test is to encourage vehicle
deformable portion of the barrier, which is estimated to bumper designs that perform well in a variety of real-
be about 1 .7 kJ for a typical vehicle. world low-speed crashes. In particular, good perform-

1597

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms
ance in this test will require a bumper system to initially ment in low-speed crashes. SAE Technical Paper
align with other vehicles, remain stable for the duration Series 2004-01-1319. Warrendale, PA: Society of
of a low-speed collision, and be capable of absorbing Automotive Engineers.
sufficient energy without damage to expensive underly- 10. Insurance Institute for Highway Safety. 2002. Low-
ing parts. Bumper systems that meet these three re- speed crash test protocol (version V). Arlington, VA.
quirements will reduce the costly damage currently as- 11. Insurance Institute for Highway Safety. 1983.
sociated with many low-speed crashes. With some addi- Weaker bumpers allow heavy damage. Status Re-
tional research, IIHS's new test will help drive future port (18)3. Arlington, VA.
bumper designs that are compatible, stable, and limit 12. Insurance Institute for Highway Safety. 1983. Colli-
expensive damage. sion claims climb for '83 models with weaker bump-
ers. Status Report ( 18)10. Arlington, VA.
ACKNOWLEDGMENT
APPENDIX
This work was supported by the Insurance Institute for
Highway Safety. The following tables identify the vehicles used for da
in Figurei.
REFERENCES
Sport Utility Vehicles
Front Bumper Bar Rear Bumper Bar
1 . Highway Loss Data Institute. 2004. Insurance col-
Bottom Top Bottom Top
lision report (R-04): 2002-04 passenger cars, Model (cm) (cm) (cm) (cm)
pickups, utility vehicles, and large/cargo vans. Ar- 2004 Chevrolet TrailBlazer 4Ž0 Šā0 4āē 54.6
lington, VA. 2004 Ford Explorer 44.0 48.0 45.6 53.3
2004 Jeep Grand Cherokee 50.5 63.5 65.4 79.4
2. Highway Loss Data Institute. 2004. Insurance prop-
2003 Toyota 4Runner 43.0 56.0 49.2 60.0
erty damage liability report (P-04): 2002-04 passen- 2004 Honda Pilot 48.0 58.0 50.5 61.1
ger cars, pickups, utility vehicles, and large/cargo 2004 Nissan Murano 60.0 70.0 47.6 59.7
2004 Kia Sorento 40.0 51.0 49.5 59.7
vans. Arlington, VA. 2003 Volvo XC90 49.0 62.0 52.7 64.8
3. Blincoe, L.; Seay, A.; Zaloshnja, E.; Miller, T.;
2004 BMW X5
Romano, E.; Luchter, S.; and Spicer, R. 2002 The
economic impact of motor vehicle crashes, 2000. Midsize Cars
Report no. DOT-HS-809-446. Washington, DC: Front Bumper Bar Rear Bumper Bar
U.S. Department of Transportation. Bottom Top Bottom Top
4. Highway Loss Data Institute. Unpublished data. Ar- Model (cm) (cm) (cm) (cm)
2004 Chevrolet Malibu 41 .5 54.0 41 .0 54.0
lington, VA. Data from CCC Information Services 2004 Ford Taurus 41.9 56.5 39.4 53.3
Inc., a company that provides insurers and vehicle 2004 Dodge Stratus 43.5 53.0 45.5 57.0
repair facilities with an automated system for esti- 2004 Toyota Cam ry 43.0 54.0 45.5 57.5
2004 Honda Accord 44.0 53.0 45.5 55.0
mating the costs of repairing vehicle damage. 2004 Nissan Altima 41.0 51.0 37.0 49.0
5. Insurance Institute for Highway Safety. Unpublished 2004 Hyundai Sonata 46.0 57.0 45.5 56.0
study. Types and extent of damage to passenger 2004 Volvo S40 38.7 48.9 40.6 49.5
2004 BMW 3 Series 41.0 52.0 42.5 59.7
vehicles in low-speed front and rear crashes. Arling- 2004 Saturn L300 41.5 55.5 40.0 56.0
ton, VA. 2004 Suzuki Verona 41 .5 52.5 44.0 55.0
6. National Highway Traffic Safety Administration. 2004 Mazda 6 42.5 54.0 42.5 54.0
2004 Mitsubishi Galant 43.5 54.5 42.0 53.5
1 977. Tile 49 Code of Federal Regulations Part 581 ,
Bumper Standard. Washington, DC: National Ar-
Small Cars
chives and Records Administration.
Front Bumper Bar Rear Bumper Bar
7. R.L. Polk & Company. 2003. National Vehicle Popu-
Bottom Top Bottom Top
lation Profile. Southfield, Ml. Model (cm) (cm) (cm) (cm)
8. Insurance Institute for Highway Safety. 2004. Huge 2005 Saturn Ion 38Ā 5Õ4 4Õ3 51.5
cost of bumper mismatch. Status Report 39(9). Ar- 2005 Dodge Neon 39.4 52.3 40.0 51 .3
2004 Kia Spectra 42.0 53.4 43.6 52.5
lington, VA.
2004 Mini Cooper 32.6 43.8 43.2 49.9
9. Nolan, Ü.M.; Brumbelow, M.; Zuby, D.S.; and Avery, 2004 New Beetle 34.4 44.0 39.2 48.3
M. 2004. Important considerations in the develop- 2005 Toyota Corolla 43.1 54.5 42.6 53.2
2005 Mitsubishi Lancer 41 .7 48.5 37.3 49.3
ment of a test to promote stable bumper engage-

1598

This content downloaded from

146.141.14.142 on Thu, 14 Mar 2024 13:14:14 +00:00
All use subject to https://about.jstor.org/terms