DANGEROUS AND DEFECTIVE ROOF STRUCTURES:
WHY MOST CARS ARE NOT CRASHWORTHY IN ROLLOVER ACCIDENTS

By Mark P. Robinson, Jr. and Kevin F. Calcagnie

Automotive safety is dependent upon the proper function and reliability of numerous systems, including brakes, steering, airbags, belts, seats, tires and suspension. However, the overall safety of an automobile is determined in large part by structural design features which dictate its crashworthiness, or its ability to protect an occupant in the event of a collision. Crashworthiness has been defined as "the protection a passenger motor vehicle gives its passengers against personal injury or death from a motor vehicle accident."^[i] Manufacturers are required by law to evaluate the crashworthiness of their vehicles and take such steps as may be reasonable and practical to forestall particular crash injuries and mitigate the seriousness of others.^[ii]

When evaluating collision safety engineers look to how a vehicle performs in four different crash modes to determine how well the occupants are insulated from the impact forces typically encountered. These include frontal collisions, rear impact collisions, side impact collisions and rollovers.

Perhaps the most neglected of these is the rollover. Unlike the Hollywood rollover, where the car going 90 mph rolls over four or five times and the occupants walk away just before the car explodes, in many real life rollover accidents, one or more occupants will never walk again. This is because unlike bumpers, frame rails and side beams designed to mitigate the forces of impact in the other three crash modes, most automobiles have insufficient structure in their roofs and supporting pillars to withstand foreseeable rollover impacts without significant and potentially injurious passenger compartment intrusion.

It is somewhat ironic that the weakest portion of an automobile's structure is relied upon to protect what is arguably the most delicate portion of its occupant-the head and neck. However, because roof crashworthiness is governed by a weak Federal Motor Vehicle Safety Standard, most vehicles' roofs will collapse when subjected to a one-foot inverted drop, or the equivalent of a 5 mph parking lot collision.

Because of this, roof crush in rollover collisions often results in serious neck and spinal cord injuries, including quadriplegia, as well as fatal head injuries. This article will discuss the dangers in current automobile roof designs, the reason these defects are allowed to exist, and available safer alternative designs which are economically and mechanically feasible.

ROLLOVER ACCIDENTS AND INJURIES

Accidents involving rollovers occur for a variety of reasons and often combinations of several factors. These can include vehicle instability, icy or slippery roadway surfaces, tripping mechanisms, fixed objects, underride collisions and uneven terrain. Rollovers are an inevitable occurrence on the nation's highways and streets. Data from the National Accident Sampling System (NASS) and the Fatal Accident Reporting System (FARS) indicate that about 220,000 light vehicles (cars, pickups, vans, and multipurpose vehicles) are involved in rollovers every year. The number of rollover exposed occupants in these vehicles is about 350,000 per year.

Specifically there are: 135,200 occupants killed or injured in car rollovers, 52,900 in pickup rollovers, 13,300 in van rollovers, and 22,800 in mpv rollovers. Of these, there are about 224,000 per year killed or injured; 9,800 fatalities; 14,100 seriously, severely or critically injured survivors; and 200,400 moderately or lightly injured survivors. Occupants incur about 786,000 injuries every year; roughly 3.5 injuries per occupant. Except for ejection related exterior contacts (35.6% of the total comprehensive harm), the most frequently harmful contacts are: the roof, pillars, rail and headers (28.1% combined.).

The most frequently harmed body regions are the brain and other head regions which account for the most significant fractions of harm, about 55% when combined. Certain sources of injury, e.g. the roof, pillars, rail and headers, are more frequent among the most severe injuries than in all injuries.^[iii]

A study of NASS files estimated that automotive rollover impacts critically injure and kill thousands of people each year including: 12,000 head injuries per year of which 2,000 are totally disabled and 5,400 have severe partial disability, 3,100 life-threatening but surviving spinal cord injuries, 5,900 fatal cervical spine injuries and 500 injuries resulting in quadriplegia. The study, which also analyzed the results of rollover testing conducted by General Motors using roll caged and conventional roofs and anthropomorphic instrumented dummies, concluded that there is "a greatly increased risk of severe injury to occupants under a collapsing roof section."^[iv]

Rollover accidents have been called "the major contributor to serious spinal injuries resulting in paralysis." There is an established relationship between spinal cord injury and local roof crush, and the degree of danger can be dependent upon the strength of a particular vehicle's roof. In experimental rollovers, neck loads in test dummies are significantly higher in standard vehicles than those with strengthened designs.^[v]

The problem is becoming more acute because of the increasing number of light trucks and SUV's on the road, as they have a substantially higher rollover rate than cars.^[vi] SUV's height, along with other factors, contributes to a rollover rate of 98 fatalities per million registered vehicles compared to only 44 fatalities per million registered vehicles for all other light type vehicles. According to the National Highway Traffic Safety Administration (NHTSA), more than 60% of the SUV occupants killed in 1997 died in crashes in which the vehicle rolled over. Because of this the NHTSA has noted that it may propose changes to the roof crush standard, Federal Motor Safety Standard (FMVSS) 216.^[vii]

FEDERAL MOTOR VEHICLE SAFETY STANDARD 216

The only current standard in this country governing the crashworthiness of automobile roofs is FMVSS 216.The explicit purpose of FMVSS 216 is " to reduce deaths and injuries due to the crushing of the roof into the passenger compartment in rollover accidents."^[viii] NHTSA's initial proposal for FMVSS 216 noted the necessity of protecting restrained occupants from roof collapse:

"The strength of a vehicle's roof has an evident bearing on the integrity of the passenger compartment in a rollover-type accident and consequently on the safety of the occupants. When applied to 1969 accident data, the analysis developed in a recent study indicates that approximately 1400 motor vehicle occupants were killed in that year by impact with roof structure in rollover accidents. Roof intrusion would have been sufficient in many cases for the roof to have struck the head of a properly restrained occupant. The benefits of occupant restraint are negated if the passenger compartment collapses in this fashion, and it is therefore important that minimum roof strength requirements be established."^[ix]

Evaluation of roof intrusion is determined by a static test in which "a force of one and half times the empty weight of the vehicle or 5,000 pounds, whichever is less, is gradually applied to the roof in the vicinity of the 'A' pillar, the pillar between the windshield and the forward side windows. The force is applied by a flat test device at a 25 degree roll angle and 10 degree pitch angle to simulate the direction of forces that can be encountered in a rollover. During the test, the roof may show no more than 5 inches of intrusion, as measured by the movement of the test device, a 30" x 72" platen.

The standard became effective for passenger cars on September 1, 1973, and was extended to light trucks, multi-purpose vehicles and vans with a GVWR of 6,000 pounds or less on September 1, 1994, as a result of NHTSA's concern about the increasing number of light trucks as a percentage of total vehicle sales:

"In the case of Standard No. 216 the Agency has observed a significantly greater incidence of rollover accidents per registered vehicle involving light trucks, compared to passenger cars. NHTSA believes that rollover protection requirements are especially important for light trucks in view of traffic crash data which show that the light truck fatality rate per registered vehicle in rollovers is approximately twice that for passenger cars . . . The agency believes that the benefits from these ejection-prevention measures will be fully realized only if adequate protection is provided to guard against the collapse of the passenger compartment roof in a rollover crash . . ."^[x]

Federal Motor Vehicle Safety Standards, by definition, are only "minimum" standards for motor vehicle performance.^[xi] In the case of FMVSS 216 "minimum" is an understatement. Aside from the glaring flaw that 216 applies only to the two 'A' pillars and none of the other pillars in a vehicle, the major deficiencies with the standard are that the test itself is static instead of dynamic, and the forces involved are much lower than the energy experienced in a typical rollover above 30 miles per hour. The force applied does not even equal that incurred in dropping a vehicle onto its roof from a height of one foot.

Moreover, the 216 test does not, and cannot, simulate real world rollover conditions where the roof can be subjected to several G's and several times the weight of the vehicle. It has been known for years that foreseeable impact forces in a rollover involve loads sometimes 4 to 5 times the vehicle weight. Additionally, static rollover tests are not reflective of occupants' motions during a rollover, and will not demonstrate kinematics and potential injury to occupants in a rollover.

Although 216 is based upon a test recommended by the Society of Automotive Engineers, the SAE at one time recommended three different roof tests- a hill roll, a ramp roll, and broadside rollover. However, none of these was included in 216. Although they are not required to do so, some manufacturers have conducted dynamic rollover testing, either by inducing rollover with a ramp or by launching vehicles from a dolly or sled. Another form of dynamic testing involves dropping the vehicle on its roof from a specified height. However, because the manufacturers are not required to conduct such testing, dynamic rollover testing is not performed on a regular basis, and is the exception rather than the rule. As a consequence of a weak standard and minimal "real world" testing, most automobile roofs are incapable of maintaining their integrity in a rollover without substantial intrusion into the survival space.

RESISTANCE TO SAFE ROOFS

Automakers have long been aware that if the roof of the vehicle is not sufficiently crashworthy, serious injuries will result in the event of a rollover. As early as 1941 automotive engineers were suggesting designs for strengthening roofs and supporting components, including bracing and box structures.^[xii] A 1949 patent described a proposed roof as "a design for the top of an automobile that might 'withstand the crushing stress in case of an accident and the overturning of the car.' The roof is suitably reinforced."^[xiii]

In the 60's safety engineers stressed the concept of 'survival space' and the importance of protecting occupants from injury due to crash-induced intrusion. Along with occupant restraint systems to adequately decelerate occupants in a crash, reducing passenger compartment intrusion was recognized as vital to crashworthiness and injury prevention:

"Occupants' protection requires two design considerations: 1) Reduction of decelerations on the occupant (restrained by a belt) by intentional deformability of the structure; 2) limitation of passengers' compartment deformation to avoid passenger crushing. It is thought that the second trend should have basic importance. It is necessary to ensure a minimum space called 'survival space'. This is also necessary in the event of severe crashes."

"It can be concluded that, in the present situation, controlled structure deformability, in addition to being difficult to accomplish, proves somewhat beneficial only in a small percentage of accidents. Unfortunately, in many cases it is in fact practically ineffective.

"Hence, it is expedient to emphasize that another requisite is fundamental, that is, the passenger compartment deformation must be limited to avoid occupant crushing."^[xiv]

Engineers in the 70's and early 80's confirmed that even properly restrained occupants were at risk for serious injury due to roof contact, and that there is a significant association between the amount of roof crush and injury severity.^[xv] In a 1982 study of 5700 rollover fatalities, it was found that 3100 individuals were trapped inside the vehicle where roof intrusion was evident, and two large injury rate increases were observed at 4 and 16 inches of roof intrusion.^[xvi] A 1989 evaluation of the roof crush resistance of passenger cars prepared for NHTSA observed: "A number of strategies are available to reduce deaths and injuries in rollovers . . . The occupants' living space within the passenger compartment must be maintained. The roof has to be strong enough to resist severe compression when the car rolls over.^[xvii]

Not everyone agrees that roof crush resistance and roof structural integrity are important to crashworthiness and reducing injuries in rollover accidents. While acknowledging that drivers and passengers are severely injured in rollovers, manufacturers have taken the position in product liability litigation that roof crush has no role in injury causation. The automakers contend that the number and severity of injuries in rollovers are a function of the severity of the collisions themselves as opposed to the degree of roof deformation.

Support for their argument comes from researchers who have concluded that while there is a statistical relationship between increments of roof crush and average injury levels, injury severity is more likely "not a direct function of roof crush."^[xviii], and that "roof crush is probably an indicator of accident severity"^[xix] Some studies by GM employees have even concluded that adding a roll cage to a production vehicle will have no effect in reducing neck loads in a rollover and no increase in the level of protection provided.^[xx]

Citing these opinions and others, manufacturers deny that any improvement in roof structure strength or crush resistance would have any safety benefit, and are therefore opposed to any suggestion that roof crashworthiness be improved through design changes. These carmakers are steadfast in their opposition to any call for improving FMVSS 216 or adding requirements of dynamic testing similar to those run by some manufacturers over three decades ago.

Moreover, in product liability litigation involving allegations of inadequate roof structural crashworthiness, the defendant manufacturers invariably argue that the injuries were caused not by the crush of the roof, but rather, by the occupant traveling to the roof due to the centrifugal force of the roll itself. In a typical scenario the injured plaintiff will be found post accident with a broken neck and damaged spinal cord beneath a roof that has collapsed several inches into the passenger compartment. Even though there is insufficient space remaining for the plaintiff to sit upright in the seat without touching the roof, the manufacturer will contend that the injury occurred before the roof crush took place.

The claim is that before the roof crushed inward, the plaintiff moved toward the roof and struck his head there as the vehicle contacted the ground, causing cervical fractures similar to those incurred in a diving accident. This argument is advanced even in cases where there is so much crush that the injured occupant is found pinned between the roof and the seat such that the roof must be cut off at the pillars to enable emergency crews to extract him.

Certain manufacturers also contend that dynamic testing of roof strength is unnecessary and should not be required, citing non-repeatability. The argument is that because rollovers occur in a variety of circumstances, depending on the terrain, the number of rolls, and the attitude of the vehicle at roll initiation and each ground contact, there are a seemingly infinite number of ways a vehicle will roll and impact the ground. These manufacturers claim that because of this variability each test will yield different results.

It is clear there is a split of opinion among the engineering community, the NHTSA and the manufacturers as to the significance of roof crush resistance to safety and the role of roof deformation in injury causation. However, some manufacturers, including those known for their attention to safety, are in apparent disagreement with their colleagues in the industry. The divergence of views is illustrated by the response to the proposal to extend FMVSS to light trucks in 1991:

"General Motors Corporation while not opposing an extension of Standard No. 216 to light trucks with a GVWR of 85,000 pounds or less, disagreed with NHTSA's analysis of the safety need for the proposal. GM believes that studies have demonstrated the lack of a causal relationship between roof crush and occupant injury in rollover accidents. According to GM, occupant injury causation in rollovers results primarily from ejection or occupant impact with the vehicle interior. According to GM, the most effective method to mitigate injury in rollovers is for occupants to use occupant restraints properly. NHTSA agrees that the principal cause of the high fatality rate in light truck rollover crashes is occupant ejection . . . However, efforts to increase occupant safety through the increased use of safety belts can only be beneficial if those benefits are not negated by the collapse of the passenger compartment in a rollover crash. . . . Volvo Cars of North America explicitly supported the extension of the standard to light trucks up to 10,000 GVWR. In addition, Nissan Research and Development Inc., Volkswagen of America, Inc., and the Insurance Institute for Highway Safety implicitly supported such an extension since they supported the proposed rule without reservation."^[xxi]

Some manufacturers are not afraid to admit that roof strength plays a role in vehicle safety in the event of a rollover. During the 80's Volvo ran a television commercial with several Volvos stacked on top of each other to demonstrate the strength of the vehicle's design. In a 1986 Saab 900 Promotional Brochure listing "Saab 900 Engineering Features"it was stated:

"Maximum safety for occupants in the event of a collision is no longer just a matter for the legislators - today the issues concern solicitude and moral rights. In this context, Saab is often cited as a forerunner and an example for other manufacturers to follow.

"For many years, impartial observers have regarded the Saab 99 as being one of the safest cars on the road. The Saab 900 satisfies the safety requirements by an even wider margin.... An internal stiffener, with a slightly rounded cross-section, encircles the roof. This is also well padded.... The windshield pillars are unusual sturdy steel sections and the steel members incorporated in the roof, floor, seals and other parts of the body form a frame of solid steel around the occupants.... The unusually sturdy windshield pillars and side pillars, and the profile steel member which encircles the roof, are essential components, designed to resist deformation of the steel safety cage should the car turn over.... .

A 1990 Saab promotional video entitled "200 Milliseconds of Your Life," states:

"In different types of collisions the body work absorbs the energy smoothly and successively. Other criteria apply if the car rolls over. Then it is the strength and rigidity of the body and windscreen pillars which preserve the survival space."

When it comes to the necessity of dynamic rollover testing, there is a similar difference of opinion among manufacturers. In 1992 the Insurance Institute for Highway Safety canvassed executives at 17 automobile companies to find out what kinds of additional crash tests they conduct beyond those required by Federal Motor Vehicle Safety Standards. Respondents indicating that they conducted rollover testing were BMW, GM, Hyundai, Mazda, Mercedes, Saab, Subaru, VW/Audi and Volvo. Respondents indicting that they did not conduct rollover testing were Toyota, Nissan, Jaguar, Ford and Chrysler.^[xxii]

Some of the manufacturers who have adopted the philosophy that roof crush is not injurious and that improved roof strength is unnecessary, have in the past touted the ability of their roofs to protect occupants in a roll. The following transcript from a 1965 Ford Motor Company promotional film describes a ramp induced rollover test of a Ford vehicle with instrumented dummies:

". . . [O]f particular interest is this slow motion study of the front seat dummies, both wearing seat belts. Note that both dummies hold their seated positions throughout the violent maneuver of the vehicle, and that neither dummy is thrown forcibly against the interior of the car. Note the horizon coming around. Now we're landing hard on the roof for the first time over. Coming out of the roll, note the dirt on the fractured windshield. The horizon is coming up again, and we're going into our second rollover. We're half over again, now landing on the roof the second time without any collapse of the structure or injury to the passengers. The horizon is coming around, the windshield is broken lose, and we are about to settle hard on all four wheels. Note that even after two complete rollovers, full protection for the passengers is provided by the roof structure."^[xxiii]

Likewise, Ford Motor Company brochures for the 1985 through 1988 Ranger trucks announced:-"Ford Ranger Lifeguard Design Safety Features - Occupant Protection - Safety - designed roof structure." Moreover, documents from Ford's own engineers disclose opinions completely contrary to the position that roof crush does not cause injury. For example, one Ford internal memo notes:

"The National Safety Council 1966 accident data was subdivided into accident categories by ACIR in the following manner.... This indicates that there were 605,000 rollover accidents in 1966 of which 4,530 involved at least one fatality.... The preceding is a summary of available rollover accident statistics from which some very basic conclusions can be drawn.

"1. A significant number of accidents result in roof damage.

"2. People are injured by roof collapse. The total number of nationwide deaths and injuries cannot be estimated but it is a significant number....

"It is obvious that occupants that are restrained in upright positions are more susceptible to injury from a collapsing roof than unrestrained occupants who are free to tumble about the interior of the vehicle. It seems unjust to penalize people wearing effective restraint systems by exposing them to more severe rollover injuries than they might expect with no restraints."^[xxiv]

In the early 70's General Motors also considered conducting tests to drop vehicles from a height of 2 feet and allowing no more than 3 inches of intrusion. Similarly, Toyota at one time was concerned about roof crush resistance, and even conducted dynamic rollover testing. A 1968 Toyota ramp rollover test was conducted to test "preservation of survival space." After initiating a rollover at 43.4 mph, the vehicle rolled 2 3/4 times. The report concluded:

"...Present rollover test showed small deformation of interior; survival space was adequate.

"...[D]amage to the RT 43L interior from this rollover test was extremely small, and it was ascertained that there was sufficient survival room as well."^[xxv]

Toyota also had an experimental safety vehicle program in the early 70's, which emphasized crashworthiness. Tests of modified and production cars were conducted using both drop testing from 2 feet and actual rollovers at 50 kph from moving dollies to evaluate compartment intrusion.^{[xxvi] [xxvii]}

THE ROLE OF OCCUPANT RESTRAINT SYSTEMS IN ROLLOVER PROTECTION

Despite ample documentation demonstrating contrary opinions and actions, many auto manufacturers continue to claim that roof does not cause injury. As a consequence, in roof crush litigation the "roof dive" theory is almost always the opinion of defense experts in biomechanics, occupant kinematics and accident reconstructionists. To counter this, plaintiffs' attorneys and their experts are quick to point out that even if the plaintiff traveled to the roof, such movement would not have occurred but for a defective occupant restraint system.

Rollover design protection must be considered from a systems approach. A strong roof should be combined with a strong seat and an effective occupant restraint system. If the restraint system's components and geometry are such that excessive excursion upward from the seat is permitted, harmful contact with the roof may result. Certain design features can allow extra inches of travel in a rollover, which may make a critical difference in whether the occupant is seriously injured. Along with a strong roof structure, a properly designed restraint system can be essential in preventing injury in a rollover. According to one study of rollover accidents in the NASS database :

"Restraint systems help to prevent ejection and prevent occupants from being tossed around in the vehicle compartment during a rollover crash. Improved belt restraints (e.g. belt pre-tensioners and integrated seatbelts) could benefit those belted occupants by preventing them from impacting various interior components; however, they could still be injured due to excessive roof intrusion, even when they are held upright in their seats.

"It was noted that vertical excursion of the occupant off the seat also increases the potential for head contacts, especially for those injured occupants who do not have significant headroom reduction, but still suffered a head injury . . . .

"To improve rollover crashworthiness of vehicles, headroom reduction and a belt system should be evaluated simultaneously to upgrade occupant protection. As more occupants use belt restraints and belt pre-tensioners, integrated seatbelts, web grabbers, etc., become more prevalent, the importance of headroom reduction is likely to move toward the forefront for rollover protection."^[xxviii]

It is important to note that a roof which is not structurally crashworthy can defeat almost any modern occupant restraint system by changing anchorage geometry. The 'B' pillar, the next pillar rearward from the 'A' pillar, is where most three point belt systems have their upper anchorage. If the "B" pillar collapses inward and downward, slack is automatically induced into the belt, giving the occupant several more inches of potential excursion toward the roof, the roof side header or the dashboard. This is one reason why automotive safety engineers have suggested the use of integrated belt systems, which are anchored to the seat itself, and not dependent upon the integrity of the B-pillar.

SAFER ALTERNATIVE DESIGNS

With nothing to encourage roof structural crashworthiness other than the relatively weak requirement of FMVSS 216, automakers have been turning out millions of vehicles with roofs which have the potential to collapse and cause catastrophic injury in the thousands of rollover accidents which will inevitably occur each year. Some roof structures, while able to withstand the minimal standard imposed by 216, are so unsound from a crashworthiness perspective that in a rollover at foreseeable highway speeds they will crush flat to the bottom of the side windows of the vehicle. These designs incorporate thin sheet metal structures in pillars and headers, which are often unreinforced except at connections between the pillar and the connecting header. Some pillar designs use open C-shaped tubes instead of stronger and more rigid boxed designs proposed by engineers years ago. Still others incorporate unnecessary cutouts, or open areas in the sheet metal, which are the weakest link and the first area where failure and bending of the pillars will occur.

Manufacturers have been developing ideas for making stronger roof structures for many years. Numerous patents by automakers disclose suggestions for improving performance in a rollover. A 1965 Mercedes Benz patent "discloses 'a motor vehicle roof construction in which a particularly high load capacity of the roof is achieved by so arranging and constructing the roof as to subject the sheet metal members thereof to torsional stresses under the load placed thereon.' The roof is curved slightly both concave and convex in places."^[xxix] Daimler-Benz, "discloses a more rigid vehicle roof with a new method of reinforcement. 'The problem leading to the present invention resides in achieving a roof rigidity as large as possible for safeguarding the vehicle passengers. The roof is secured by means of the roof frame constructed as bearer part at the upwardly extending column or pillars and wall parts of the vehicle body.'"^[xxx]

Another Mercedes Benz patent provides: "In order to protect the vehicle passengers against accident consequences, it is necessary that the vehicle's cell receives a high form or shape rigidity also within the roof area.^[xxxi] A 1982 Mazda patent refers a "pillar construction for a motor vehicle which has improved high strength with sufficient rigidity, without an increase in dimensions, particularly in the width of the pillars," and includes a "pillar inner panel and a pillar outer panel rigidly combined to each other so as to define a closed, cross section."^[xxxii]

In the early seventies both General Motors and Ford considered proposals for improving roof structural integrity which included integrated rollbars and reinforcement or redesign of roof rails, and A and B pillars to add reinforcement. In 1970, Ford's In-House Safety Car Program objectives called for roof drop testing from as high as 2 feet, and rollover testing at 60-70 mph, as well the following:

"- A 'structural cage' will be provided for the passenger compartment by the following structural additions and modifications:

"- Increased section, high strength, 'A' pillar similar to that developed for the 1973 Ford Program.

"-Integrated 'B' pillar roll bar, inboard of the glass planes, similar in section to that developed by Advanced Body for the original 1973 Ford Convertible program.

"-Structural boxed sections in "C' pillar area with heavy lateral structural tie in rear header area.

"-Heavier structural members for front header and roof side rails, with improved joints.

"-Lateral structural crossmember tie at belt height between 'B' pillar and under package tray between 'C' pillar.

"-Increased strength and rigidity floor pan, with a heavier mid-vehicle structural tie in 'B' pillar area. Also, 'welded' structural center console from dash panel to 'B' pillar floor lateral tie.^[xxxiii]

Another memorandum referring to the same program discusses the type of crash protection which had already been achieved in racing designs:

"Just to show what Ford is up against, clips of six films were selected for reviewing.... Next was an actual crash sequence showing Richard Petty in his 1970 Plymouth stock car striking the Darlington Raceway outer wall at approximately 140 mph, caroming across the track, striking the inside wall and then flipping violently down the race track. These were not lazy rolling type flips; instead they were violent as the car 'slapped' the surface each time it came time. The sheet metal was partially ripped away but the inside 'cockpit dimensions' were not altered. This shows what added front structure and the roll cage concept can sustain and remain intact. Petty's injuries were a separated shoulder and assorted bruises. He was racing again one month later, winning several events in the year and most recently the 1971 Daytona 500.^[xxxiv]

The design alternatives considered are not only technologically feasible, but they are inexpensive measures that will have negligible impact on cost and performance. By contrast, the reduction in the number of severe injuries would be tremendous. Eliminating the deformation of vehicle roofs by lightweight structural changes and simple and inexpensive force limiting, energy absorbing interior surface modifications, has been demonstrated to significantly reduce the risk of severe injuries.

For example, replacing open or 'c' shaped members with boxed sections can result in a significant increase in the stiffness of the roof structure and its ultimate crashworthiness in a rollover. One study concluded that alternative rollover safety design improvements, adding approximately 50 pounds to the vehicle weight and $250.00 in costs, could save as many as 5,000 lives per year and 5,000 critical neck injuries in all accident modes.^[xxxv]

Another engineering study of rollover accidents made the following recommendations:

"Experience from road and track racing indicates that roll cages have been effective in injury prevention in severe rollovers. . . .

"· Recommendations: The following vehicle design changes are recommended to reduce both the risk and severity of serious injuries arising in rollover crashes.

"1. Maintain side window integrity (by plastic glazing) to prevent head excursions outside the vehicle.

"2. Increase roof framing and A and B-pillar strength, for axial loading and side-sway loading, and require a minimum standard for roof integrity.

"3. Provide interior energy absorbing padding to head contact surfaces-the roof itself, and the framing above the door.

"4. Modify the design of door/roof framing to reduce occupant's head being able to 'lock in' against the framing and hence result in excessive spinal loading.

"5. Improve the performance of seatbelts to reduce vertical movements of occupants in rollovers.

"6. Improve door integrity and add energy absorbing side padding."^[xxxvi]

Other proposals have included use of rigid foam in the pillars and headers, which can double or triple bending and compressive strength. More advanced forms of rollover protection have moved from the drawing board to the road. In 1998 Delphi Automotive Systems of Troy, Michigan announced that it will take occupant protection to a new level with its Advanced Safety Interior unveiled at the 1998 SAE International Congress and Exposition. The concept included a Rollover Sensing System designed to detect an impending vehicle rollover using inertial sensor technologies developed for aircraft guidance control. When the system detects an impending rollover, it will command deployment of occupant restraint devices such as seatbelt pre-tensioners, retractor locks and hypertensioners, as well as inflatable head restraints and other enhanced support structures. The combined effect of these technologies will be to minimize rollover-type injuries.^[xxxvii]

More recently Ford Motor Company announced that it will feature rollover sensors and special curtain rollover air bags on its sport utility vehicles during the 2001 model year. All of this is made possible by inflatable curtain technologies as well as advanced sensors that measure the amount of horizontal vehicle 'roll' or tilt." The devices are supposed to inflate from the SUV's headliner trim, to protect passengers in both the front- and second-row seats.^[xxxviii]

ROOF CRUSH LITIGATION

Although roof crush litigation has been around for many years, there are relatively few reported appellate decisions involving product liability cases where roof crush was alleged as a cause of injury. However, some of the successful cases illustrate the types of issues which arise and the probable arguments of the defendant manufacturers. In Doupnik v. General Motors Corporation^[xxxix], a man who was rendered quadriplegic when his car left the road, plunged down a rocky embankment and overturned brought an action against the manufacturer contending that a defect in the welds of a roof pillar of the car caused the roof to collapse during the course of the accident. The plaintiff contended that the driver's side of the roof collapsed from the forces imposed on the defective welds, forcing his head backward over the top of the seat, tearing ligaments in his neck and damaging his spinal cord. Following a jury verdict for the plaintiff the manufacturer appealed, contending that the evidence plaintiff submitted was deficient because the plaintiff's expert witnesses failed to analyze the forces acting on the A-pillar. The appellate court affirmed the verdict and rejected the manufacturer's argument, stating:

"General Motors claims that the subject of roof collapse is esoteric and only amenable to proof by an analysis of the physical forces acting upon the roof, e.g., by means of expert testimony involving a calculation of the forces acting upon the driver's side roof and the impact of such forces upon the vehicle in its defective condition. We see no reason why the mode of proof must be so constrained. There is a saying that, according to the science of aeronautical engineering, a bumblebee can not fly. But if that were an issuable fact, the defects in the theory could be shown by the observations of a beekeeper or an entomologist."^[xl]

In Jordan v. Paccar, Inc.^[xli] a wrongful death action alleging that a heavy truck cab roof was defectively designed without sufficient "occupant crash protection generally and truck rollover protection", the manufacturer contended that because NHTSA declined to extend the rollover protection requirements of 49 CFR 571.216 to heavy trucks, the plaintiff should be precluded from maintaining a state cause of action for defective roof design because of federal preemption. The court denied the defendant's motion to exclude evidence in support of the claim, holding that claims of defective truck cab-roof design for failure to install rollover protection devices are not preempted by the Safety Act or regulations promulgated thereunder.^[xlii]

In Shipp v. General Motors,^[xliii] a young woman who was rendered paraplegic when she sustained a broken back in rollover accident, brought suit against the manufacturer, asserting that the car's roof had collapsed because it was defectively designed, and that the roof's impact upon her shoulders broke her back. On appeal from a jury verdict in favor of the plaintiff the manufacturer contended that the plaintiff "never established that a stronger roof is a safer roof", and that although a stronger roof could have been made, such increased strength would reduce the energy-absorbing characteristics of the design and force the energy to be absorbed in the lateral plane of multiple rollovers. In affirming the verdict the appellate court noted:

"Despite the force of GM's argument, we are persuaded that the risk to passengers of roof deformation and trade-offs involved in a more rigid roof design were questions for the jury. Holly Shipp offered evidence of the dangerousness of the Trans Am roof and the feasibility and availability of a safer alternative. That no other passenger car employed this design will not alone defeat recovery (Citations). The jury had before it evidence enabling it to engage in the necessary balancing and supporting its finding of defective roof design."^[xliv]

Finally, in Compton v. Subaru of America, Inc.,^[xlv] the unbelted rear seat occupant of a Subaru GL Station Wagon suffered a spinal cord injury resulting in quadriplegia when the vehicle rolled over. In a product liability action against the manufacturer the plaintiff contended that he would have avoided serious injury during the rollover if the roof had not collapsed rearward onto his head. Appealing from a jury verdict in favor of the plaintiff the manufacturer contended that because the vehicle complied with Federal Motor Vehicle Safety Standard 216 there was a presumption of nondefectiveness under the Kansas Product Liability Act. However, the appellate court noted that because FMVSS 216 appears only to apply to the A-pillar, and Mr. Compton's injuries involved the C- and D-pillars, the statute in question had no applicability:

"The district court found FMVSS 216 applied solely to the front pillars on either side of the windshield, known as the A-pillars. Because Mr. Compton's injury occurred when the roof's support pillars behind the rear doors (the C-pillars) and the pillars behind the cargo deck (the D-pillars) collapsed, the district court concluded Kan. Stat. Ann. Section 60-3304(a) was inapplicable. . . . In view of FMVSS 216's emphasis on the A-pillars, we cannot hold it explicitly applies to C- or D-pillar crush. Accordingly, because there appears to be no applicable regulation governing rear seat roof crush, we agree Kan. Stat. Ann. 60-3304(a) has no application on these special facts."^[xlvi]

CONCLUSION

There remains disagreement among automobile manufacturers as to the reasonable level of roof crashworthiness necessary to protect victims of foreseeable rollover accidents from serious harm. However, there is no dispute that some manufacturers have made substantial efforts to reduce injuries in rollover collisions by incorporating stronger, safer roofs and supporting structures, with mechanically and economically feasible design alternatives. There is also no disagreement that rollovers will continue to occur. As long as FMVSS 216 remains the only standard for roof crashworthiness, product liability actions may be the only means of encouraging change on the part of those automakers who continue to bury their heads in the sand and ignore the thousands of injuries and deaths caused by defectively designed roofs.

ENDNOTES

[i]. 49 U.S.C.A Section 3201 (1)

[ii]. Self v. General Motors (1972) 42 Cal.App.3d 1, 7, 116 Cal.Rptr. 575

[iii]. Injuring Contacts in Light Vehicle Rollovers, Data Link, Inc., prepared for the National Highway Traffic Safety Administration Crashworthiness Staff (1993)

[iv]. Friedman, Donald and Friedman, Keith D., Roof Collapse and the Risk of Severe Head and Neck Injury, 13th Experimental Safety Vehicle Conference, Paris, France, November 4-7, 1991, Document No. 91-S6-0-11

[v]. Rechnitzer, George and Lane, John, Rollover Crash Study Vehicle Design and Occupant Injuries, Monash University Accident Research Center, Australia (1994) ISBN Report No. 0 7326 0064 2

[vi]. Contributions of Vehicle Factors and Roadside Features to Rollover in Single-Vehicle Crashes-Tast 2 of Project, U.S. Department of National Highway Traffic Safety Administration (March 1991) Final Report DOT HS HO7 735

[vii]. DOT Press Release, March 5, 1999

[viii]. 49 CFR 571.216

[ix]. Federal Register Vol. 36 No. 3 p.166 (January 6, 1971)

[x]. Federal Register, Volume 54, No. 211, Page 46276 (November 2, 1989)

[xi]. 15 U.S.C. §1391(2)

[xii]. The Murray Corporation of America (Detroit)," Roof Rail Construction", Patent No. 2,247,457 (July 1, 1941)

[xiii]. George Phillip (Detroit), "Vehicle Body Top", Patent No. 2,481,868 (September 13, 1949)

[xiv]. E. Franchini, "The Crash Survival Space", (January 13, 1969) Society of Automotive Engineers, 69005

[xv]. Huelke, "Injury Causation in Rollover Accidents" (1976) , Highway Safety Research Institute, The University of Michigan Proceedings of 17th Conference of the American Association for Automotive Medicine; Fan and Jettner, "Light Vehicle Occupant Protection - Top and Rear Structures and Interiors, (June 1982) Society of Automotive Engineers, 820244

[xvi]. Fan and Jettner, "Light Vehicle Occupant Protection - Top and Rear Structures and Interiors, (June 1982) Society of Automotive Engineers, 820244

[xvii]. Kahane, Charles J., An Evaluation of Door Locks and Roof Crush Resistance of Passenger Cars, NHTSA Report No. DOT HS 807 849 (1989)

[xviii]. Hight, Philip; Siegel, Arnold; and Nahum, Alan, Injury Mechanisms in Rollover Collisions, SAE Document 720966 (1972)

[xix]. Huelke, Donald and Marsh, Joseph, Analysis of Rollover Accident Factors and Injury Causation, 16th Conference of the American Association for Automotive Medicine (1972)

[xx]. Baihling, G. S., et al., General Motors Corporation, Rollover and Drop Tests-The Influence of Roof Strength on Injury Mechanics Using Belted Dummies, (1990) SAE 902314

[xxi]. Federal Register, Volume 56, No. 74, Page 15512 (April 17, 1991)

[xxii]. Ralph Hoar &Associates Brief Notes (December 1993)

[xxiii]. Briggs Safety Show with Comet Rollover

[xxiv]. Ford Automotive Safety Research Office Safety Engineering Evaluation. Intracompany memo dated July 8, 1968 from J. R. Weaver, Safety Engineering to Mr. H. G. Brilmyer. Subject: Roof Strength Study

[xxv]. Toyota 2nd Technical Department Safety Laboratory Test Report RT 43L Ramp Rollover Test. (December 20, 1968)

[xxvi]. Report of the Third International Technical Conference on Experimental Safety Vehicles, U.S. Department of Transportation, National Highway Traffic Safety Administration, "Outline of Toyota ESV Program" (June 1972)

[xxvii]. Report on the Fourth International Technical Conference on Experimental Safety Vehicles, Kyoto, Japan, U.S. Department of Transportation, National Highway Traffic Safety Administration (March 13, 1973) "The Japanese Technical Presentation, Section 2, Energy Management System", Akihiro Wada, Manager, Body Development Section, Body Designing Department, Toyota Motor Co., Ltd.

[xxviii]. Rains, Glen and Kanianthra, Joseph, National Highway Traffic Safety Administration, Determination of Significance of Roof Crush on Head and Neck Injury to Passenger Vehicle Occupants in Rollover Crashes, SAE Technical Paper 950655 (1995)

[xxix]. Daimler-Benz, "Roof Construction for Vehicles", Patent No. 3,197,252 (July 27, 1965)

[xxx]. Motor Vehicle Roof", Patent No. 3,833,254 (September 3, 1974)

[xxxi]. (Daimler-Benz, "Roof for Motor Vehicles", (June 24, 1975) Patent No. 3,891,266)

[xxxii]. Toyo Kogyo (Mazda), "Pillar Construction for Motor Vehicle", (10/26/82) Patent No. 4,355,843

[xxxiii]. Ford Memorandum, In-House Safety Car Program (December 27, 1970)

[xxxiv]. Ford Memorandum, In-House Safety Car Program (February 26, 1971)

[xxxv]. Friedman, Donald and Friedman, Keith D., Roof Collapse and the Risk of Severe Head and Neck Injury, 13th Experimental Safety Vehicle Conference, Paris, France, November 4-7, 1991 Document No. 91-S6-0-11

[xxxvi]. Rechnitzer, George and Lane, John, Rollover Crash Study Vehicle Design and Occupant Injuries, (December 1994) Monash University Accident Research Center, Australia. ISBN Report No. 0 7326 0064 2

[xxxvii]. http://www.theautochannel.com/press/press/date/19980223/press010194.html

[xxxviii]. InSane, Internet Auto Safety News Bulletin, Ralph Hoar & Associates (January 12, 2000)

[xxxix]. Doupnik v. General Motors (1990) 225 Cal.App.3d 849, 275 Cal.Rptr. 715

[xl]. 225 Cal.App.3d at 869

[xli]. Jordan v. Paccar, Inc. (1992) 792 F.Supp. 545 (N.D. Ohio)

[xlii]. National Traffic and Motor Vehicle Safety Act, 15 USC 1381, 1391-1431

[xliii]. In Shipp v. General Motors 750 F.2d 418 (5th Cir. 1985)

[xliv]. 750 F.2d at 423

[xlv]. Compton v. Subaru of America, Inc. 82 F.3d 1513 (10th Cir. 1996)

[xlvi]. 82 F.3d at 1520