This is the complete transcript of NASCAR's announcement of the findings in the death of Dale Earnhardt: JIM HUNTER: Good afternoon. I'm Jim Hunter, Vice-President of Corporate Communications. As most of you know, today marks the most intensive,...
This is the complete transcript of NASCAR's announcement of the findings in the death of Dale Earnhardt:
JIM HUNTER: Good afternoon. I'm Jim Hunter, Vice-President of Corporate Communications. As most of you know, today marks the most intensive, and certainly most anticipated investigation.
After hearing today's report, I think you will better appreciate the scope of this and better understand why it took so long to complete. Before we begin today's press conference, I'll allow you a moment to explain what is going to happen in the next hour or so.
First, you are going to hear from NASCAR president, Mike Helton. Then we are going to hear from independent experts who conducted this investigation.
Their presentations will last approximately one hour. To be followed, a question and answer period.
Once the Q and A is completed, members of our staff will distribute press kits and a two-volume report on this study.
Also, after the presentation is complete, a transcription of the presentation and the Q and A session will be available in the media room.
We plan to hold the Q and A session in a very orderly manner, so your cooperation will be greatly appreciated. In the meantime, we would like to request that everyone turn off all cell phones and pagers. To open this presentation, let me introduce the President of NASCAR, Mr. Mike Helton.
MIKE HELTON: Thank you, Jim.
Nothing we do can bring back those that we've lost as part of our sport. We can, however, learn from those losses and honor them in what we do moving forward. It may seem a bit unusual to discuss moving forward at the beginning of this presentation, but in some ways that seems most appropriate because it is the one good part of this tragedy, and because I can think of no more important part.
But to do that, we need to briefly look back.
One of the reasons for creating NASCAR more than 50 years ago was Bill France Sr. desire to improve safety for drivers and spectators. There are very few in our history who have had the appreciation and energy toward making it safe, as did Dale Earnhardt.
As this sport has evolved, so have safety improvements. One can look at the cars and equipment used 50 years ago, 25 years ago, 10 years ago or two years ago for that matter and compare today's cars and equipment and see clear examples of safety enhancement evolution.
Through the collaborative efforts of many in this sport, many improvements have taken place. Three excellent examples are the implementation of fuel cells and the roof flaps and the development and continuous improvement of the roll cage.
Almost a year ago, NASCAR began working on plans for an advanced research and development center. We've hired a director who is putting together a team that will focus on safety enhancements, cost management and developing a vision for the cars for the future. The facility we expect will be fully operational by next year.
The week after the Daytona 500, NASCAR began to tell drivers (inaudible) We also began regular meetings with the manufacturers for the sole purpose of discussing safety opportunities. These meetings have already proven useful in areas of occupant restraint system, head and neck restraint systems, improving the seat designs, better padding, energy and management, all as it relates to cars and barriers.
We've continued sharing what we've learned with the drivers. At Richmond in May, we met with the drivers and owners to provide an update. We also met with them recently during testing at Indianapolis for Dr. Gramling and John Melvin, two of the most famous experts on head and neck restraints, demonstrated the benefits of those systems and encouraged their use.
Manufacturer representatives from Ford, Diamler Chrysler and General Motors also discussed progress being made on such issues as energy management, seat design and other occupant-related innovations. It's also to take note of the importance of the work that these men on stage have accomplished. You'll be hearing from them in a few minutes, and I hope you're as impressed with their thoroughness and efforts as we are.
We learned a great deal from their work and have compiled a body of knowledge that will contribute to safer races well into the future; specifically, a computer car crash model has been created to assist all of us in designing safer race cars. Their investigation provided extensive information about occupants' movement in a barrier crash at a critical angle, which will also assist race car and restraint system manufacturers in their future designs.
Their work includes significant new information on the problem of dumping, which will provide highly useful information to try to avoid another separated belt. Investigation contains an analysis of the operation of a seat belt restraint system under a maximum load created by a critical angle crash, which will assist occupant restraint manufacturers in design of their systems.
We have additional knowledge to apply now to barrier testing and analysis, which began with Dr. Sicking and his Nebraska team along with the IRL last year.
Additionally, there are a number of steps NASCAR is taking as a result of the investigation. First, as you know, NASCAR has for some time recommended the use of head and neck restraints. We've also urged drivers to talk to Dr. Melvin about the benefits of these systems. This year we increased the size of the openings of the windows to help alleviate concerns about exiting their car while wearing one of those devices. We're pleased that a majority of Winston Cup drivers now use them, but we're not completely satisfied. So we've intensified our efforts with drivers, equipment manufacturers and the outside experts with the goal of helping all drivers find a system which they feel comfortable and safe with.
Second, we've asked Dr. Melvin and Dr. Raddin to lead a study on our behalf of the occupant restraint systems to better examine belt strength, how to avoid dumping and a separated belt, optimal installation methods, belt accessories such as pull tabs and spring clips. We anticipate results by the end of this year and NASCAR is prepared to take action to approve the system as appropriate.
Third, we are committing to the installation of crash data recorders by the beginning of next season. Such devices will help drivers, owners and manufacturers and NASCAR study the impact effect of drivers in cars. In the meantime, we're working with the industry on the details to make that a reality.
And fourth, we currently work with track operators to provide the best local doctors licensed to practice medicine in their jurisdictions on the NASCAR circuits.
This is a good system. Since these doctors are intimately familiar with the local hospitals and the medical resources and have regular and extensive experience dealing with serious trauma cases.
But to enhance this system, we're in the process of hiring a qualified person to serve as a full-time liaison who will be knowledgeable about the drivers' medical histories and will coordinate with those local physicians. We expect to have someone in place by the beginning of the 2002 season.
Also, we're currently seeking a fully devoted individual to establish procedures for and oversee future accident investigations.
We will continue to aggressively pursue any legitimate safety enhancement opportunity. Importantly, as with other safety initiatives, we've implemented, we will not rush to fix one thing without fully understanding the unintended consequences in other areas.
While we may have fallen short at times in our communications, it's my strong belief that we have been responsible in the area of safety. We will continue to approach this with a firm belief that even in the sport where danger is inherent, any single death or serious injury is one too many.
We're committed to accelerating the evolution of safety within this great sport through continued collaborative efforts using the best minds and the best technology we can find whenever and wherever we can find it.
JIM HUNTER: Thank you, Mike. Before I turn the podium over to Dr. James H. Raddin, let me add that NASCAR asked two additional highly respected experts in the field of injury causation to conduct a peer review and to comment on his report. Both experts agreed his report is objective and scientifically based, and they concur with his conclusions. You will find their statements and CVs in the report.
Dr. Raddin is a director and principal consultant with San Antonio Texas based Biodynamic Research Corporation, BRC. In this position, he performs injury causation analysis in a variety of automotive and aircraft crash settings. Dr. Raddin has an extensive background in biomechanical research from years as an Air Force physician.
He is the author or co-author of 29 publications dealing with biomechanics. Before joining BRC, Dr. Raddin was Vice-commander of an Air Force school of aerospace medicine at Brooks Air Force base in Texas. He has designed and carried out numerous experimental impact tests with volunteer human subjects including himself.
Through the years, Dr. Raddin has received numerous awards and honors for his work. He is a graduate of the University of New Mexico School of Medicine and he has a bachelor of science and master of science from MIT.
Ladies and gentlemen, Dr. Raddin.
DR. JAMES H. RADDIN, JR.: Good afternoon. I will be presenting the results of an injury causation analysis today, and I need to tell you that this has not been the work of one person; it's been a team effort. It's been a collaboration with my colleague, Jim Benedict, with additional support from members of the staff at Biodynamic Research corporation in San Antonio.
Jim has participated fully in this program and would be here today except he had some heart surgery about a month ago and is still recuperating successfully, but otherwise would be here and have a part in this presentation.
What we're going to do today is look at the cause of the injuries that Dale Earnhardt suffered in the crash at Daytona. We're going to do that through the process of an injury causation analysis. That's basically a process in which we look at four steps by which the change in motion of a vehicle develops into a clinical injury. And that process happens this way.
If the vehicle's motion does not suddenly change, you don't have a problem. But when the vehicle motion changes suddenly, it results in displacements of the occupant within that vehicle, and those displacements are studied under the heading of, "Occupant kinematics."
Those displacements always result in contact with something. That something may be a restraint system, it may be a part of the vehicle's structure, it may even be the outside world. But those contacts produce stresses on human tissue which are studied under the heading of, "Biomechanics."
If those stresses exceed human tolerance at the point where they're applied to the person to whom they're applied, then you get a clinical injury. So you end up with a four-step process which has engineering disciplines involved in the first two and a half steps, and medical disciplines involved in the last step and a half.
That's the process which I'm going to go through with you today. But before I do that, I would like to address a specific issue with respect to this investigation that has brought about a lot of attention, and that issue is the issue of the separation of the seat belt.
I'd like to address that first and make that clear before we proceed to our -- the rest of this report.
I will be addressing the physical findings on the restraint system and the physical findings on the occupant.
To do that, I am going to be showing pictures throughout this presentation taken of portions of the car in which a good friend of many people, certainly someone who I respected and admired, was killed.
And I'll be talking about the injuries that were resulted from that. I do that, we're going to do that with as much sensitivity as we can, but we need to do that in order to put this on an objective footing so people can understand what, in fact, happened and not just have some opinions thrown in this direction.
So we'll be looking at the physical findings on the restraint and the physical findings on the occupant.
First of all, the restraint system is a five-point restraint system. This is a photograph taken of an exemplar vehicle, a similar vehicle, in which a five-point system is installed.
That five-point system consists of a left lap belt and a right lap belt that are buckled in the middle. The buckle allows for attachments of a crotch strap and two shoulder harnesses, each of which attaches to that central buckle. This is the area where the steering wheel would go, that's the steering column. This is the seat upon which the occupant would sit.
There were variations in the way in which the belt system was installed. There were variations in the installation in which there was a rationale behind each variation, and those variations had been successfully used for a number of years without evidence of problem.
One of the variations you can see here is the crotch strap which normally would go through a slot in the seat and attach below. In this case, the crotch strap was routed around the front of the seat system.
Additional changes included the attachment of the shoulder harnesses which more conventionally are attached to a bar behind the occupant. In this case, the harnesses were routed around a bar through a retaining hoop and attached to a bar lower on the seat.
Now, what this does is it provides a greater distance of run for that shoulder harness so that when you put a given amount of tension on something that is stretchable, the more you have that you put that tension on, the greater the displacement is allowed.
And that allowed a greater forward motion which, under some circumstances, could be beneficial. But in other circumstances, you would prefer in general to have less forward displacement and it certainly allows for the potential for head impact or impact with parts of structures should that occur.
A third difference in the installation has to do with the installation of the lap belt. Conventionally, that lap belt would come down closer to this position, but with an adjuster of this sort which is a pull-up adjuster.
The adjuster, in order to fit, as manufactured, if you installed it here, the adjuster would be up inside the slot in the seat through which that goes. And so for that reason, this was installed further back to allow the adjuster to be operated outside the seat. Those were the principal differences.
This is, in fact, a photograph of the restraint harness that was in the No. 3 car at the time of this crash. You'll see the right lap belt, you'll see the left lap belt and you'll see a separation occurring here. Normally, through this adjuster a piece of webbing is pulled so that you can tighten it.
You can see that kind of loosely lying here, that piece that should pull through is now located at this point (indicating), and is no longer in the adjuster at the point of that picture. Following the crash, however, it was in that adjuster.
This is a photograph, and I'll orient you to the photograph. It was taken by an investigator for the medical examiner's office, and it was taken at a time when the vehicle was still under the supervision and in the possession of the police officers, the police department. What we see here is a view looking down over the window. This is the ledge where the netting is hanging out the window here. You're looking alongside the seat.
What we see in this area right here is the adjuster. And you see in that adjuster the pull tab that, the end of the webbing which has some tape wrapped around it, and that is still stuck through the adjuster at this point.
But as you look carefully at that adjuster, which has a blown-up view right here, you see that the piece of webbing that would normally continue on to the buckle is not present in that adjuster.
And that's present in a photograph taken by an investigator from the medical examiner's office. That piece of webbing, instead of being attached here and routed up through the seat, is lying loose on the floor separate.
The webbing was separated as found.
When we look at the characteristics of that separation, what we find is a separation that necessarily is a separation under load. It would not occur in any fashion in a cut.
That separation under load started at this end, where these fibers are shorter and closer to the same length, and then proceeded to pull fibers out through the other piece of pull tab that's not shown here with the last fibers hanging on being the ones that extend out to the end.
That's a separation consistent with a process known as, "Dumping," in which an adjuster would rotate forming ramps such that the one end would rotate up to this end, or slide up to this end, and the other would rotate to this end and in doing so would increase the stress concentration right at this point. That is, in fact, what happened in this belt system.
What we see is at the point where the separation began, it began right here, and then proceeded to pull fibers out of the other end.
Now, what's observable as we look at this is that there is clear physical marks that tell us that at the time the load began on the adjuster, the adjuster had the webbing through it in a position that was symmetrical.
In other words, there was not a misadjustment of this webbing in the adjuster prior to load. The way you can tell that is this: Here is the webbing surface, which is smooth and normal webbing surface coming across here.
We do not have teeth marks from the lock bar coming across this tab, but you find separation -- or find disturbance of the surface where the lock bar has chewed into the surface of this fabric in a position that is diametrically opposite and coming straight across that webbing.
That says that that adjuster lock bar was straight across the webbing at the time that the load began to be applied.
This was not a misadjusted webbing. The tear began here at the lower end of the lock bar and proceeded across, as we see here.
The lock bar has a knurled surface to it. It's got a -- that knurled surface for the purpose of keeping tension present. The lock bar location at that end is where the separation occurred because the other webbing was up to this side.
It left a mark that is a diagonal lock on the pull tab portion of the webbing. That diagonal mark comes across right here. And the reason it did that is because portions of this pull tab were pulled through the adjuster in the process of separation.
The mark that we see coming diagonally across it is not a tightening mark with it being this tightened because it's on the opposite end; it's on the loosening end of the pull tab, and that mark occurs because as this separates, it stops pulling this end of the webbing through and as more and more gets pulled through here, it pulls more and more in a diagonal fashion pulling portions of this pull tab through the adjuster, leaving it still in the adjuster as it was found by NASCAR officials when they first looked at it.
Let me show you an example. Here's a piece of webbing, an adjuster. This is asymmetrically positioned as you see here. It dumps and goes to this end first.
That happened quick. I'll show it to you again. Separation in the pull tab continues with the adjuster as it comes away, and that's the way it was found.
We now need to look at physical evidence on the occupant. That physical evidence is in the form of abrasions which were described in the autopsy report with the first abrasion being one which is across the left hip in a lap belt distribution.
I would get abrasions like that when we would ride crash tests from collisions. You'd typically see an abrasion like that on this hip and you'd see an abrasion like that on the other hip.
That's not what we see in the Earnhardt accident, because instead, we see an abrasion that is a longer abrasion, a little over eight and a half inches long, preceding in a diagonal fashion basically paralleling the inguinal ligament.
This is further evidence that just like the physical evidence on the belt tells us that the separation occurred under load, because after the separation occurs, if the occupant continues to move forward, he now has a different kind of restraint system that has no left lap belt but it still has a crotch strap and it still has a right lap belt and it still has shoulder harnesses.
And as you go into that system, as you move forward and to the right in this case, it leaves a diagonal abrasion consistent with the crotch strap forming a loop with the right lap belt.
Those marks tell us that Dale Earnhardt moved forward into a separated belt which could not have occurred unless the belt separated under load.
So the physical evidence is clear, both on the belt and in the injuries, that the belt separated under load. We've addressed these two.
Other parts of the report which you will receive will talk about fiber analysis showing that these were torn fibers, not fibers cut with an instrument. DNA analysis that says that the blood on the belt is the same as the blood in the car.
There was a difference in the way it was deposited. If the blood was deposited first and you cut across it, you would expect to see similar blood deposits on both sides of the cut.
What we see is different kinds of blood deposits on one side than you see on the other side, different amounts and ways in which that blood is deposited.
You've seen the medical examiner photographs and there were extensive interviews of those that had access to the belt. The conclusion is clear: There was no cutting of the belt afterwards. It's a belt that separated under load.
We'll now go into the injury causation analysis. And to begin with, we'll look at vehicle dynamics. As I mentioned, it's the province of the reconstruction.
Jim Hunter will introduce the reconstruction team so you can hear the presentation from then and then I'll come back and fill in the rest of this analysis.
JIM HUNTER: Thank you. Dr. Dean L. Sicking is a civil engineering professor at the University of Nebraska at Lincoln. He is one of the world's leading independent researchers on barrier and crash safety. His work includes the study and analysis of vehicle crashes.
Many of Dr. Sicking's recent contributions have come through his work at the Midwest Roadside Safety Facility in Lincoln. Dr. Sicking and his Nebraska team have been responsible for many patented highway safety features. Some of their developments include energy-absorbing guard rails, crash cushions and median barriers.
As part of its broader mission, Dr. Sicking's Midwest center also conducts full scale crash testing and structural testing of vehicles and safety devices. One project involves the center's ongoing work with the Indy Racing League and NASCAR on energy absorption barrier systems.
Other projects involve vehicle crash modelling and accident reconstruction, as well as computer simulation of vehicle dynamics. Dr. Sicking is personally responsible for 16 separate roadside safety patents in the United States. His designs have also been adopted in numerous foreign countries.
Dr. Sicking is an influential member of many national safety committees. He earned BS, MS and Ph.D. degrees at Texas A & M University. It's my pleasure to welcome Dr. Dean Sicking.
DR. DEAN L. SICKING: Thank you, Jim. Today I would like to summarize the work that was done at the University of Nebraska and the Midwest Roadside Safety Facility. That work was led by myself and Dr. John Reid, who is an associate professor in mechanical engineering at the University of Nebraska.
First thing I want to do is review the crash that occurred on the last lap in the fourth turn of the Daytona 500 of this year. If you'll recall, the No. 3 car lost control toward the inside of the track, the driver immediately corrected, regained the track where he was struck, where there was a collision between the No. 36 and No. 3 car.
Thereafter, the No. 3 car slammed into the barrier.
Our job was to obtain the vehicle kinematics throughout this event. By vehicle kinematics, I mean the motions of all the vehicles throughout the crash.
Of primary interest to us was the barrier impact conditions. What was the speed, angle and orientation of the No. 3 car when it struck the barrier? And the same thing for the No. 36 car.
We also were very interested in the collision between the No. 3 and No. 36 cars, and we also needed to know what the barrier impulse imparted to the No. 3 car was. By barrier impulse, I mean the forces applied to the No. 3 car during the barrier impact.
This is important for determining or estimating the injury causation that Dr. Raddin will talk about in a minute.
Any accident reconstruction begins with collection of all available evidence. This accident, there was a great deal of evidence. At the site, there was a significant amount of tire marks and barrier marks available. There was also -- the vehicle was available for crush measurements.
Each vehicle in the race carried a GPS receiver which recorded -- which was used to record the position of the vehicle five times every second, which, again, another important source of information. And, finally, this crash event was actually captured on seven video cameras, which could be used to locate each of the vehicles throughout the event.
Once we've collected all available data and analyzed it, we were able to estimate vehicle conditions and impact scenarios throughout the event. We wanted to then verify critical impact conditions for this case, that would be the No. 3 car impact with the barrier.
To do that verification, we conducted a full scale crash test to make sure that the energy of impact that we've estimated for this crash was correct.
Finally, we did a detailed computer model of the full crash event. The only way we could reconstruct the full crash event was with computer modelling.
Because, as you saw in this crash, this vehicle was in a non-tracking mode, meaning that the rear tires were not following the front tires when it struck the barrier. Under today's technology, it is not possible to conduct a full scale crash test in a non-tracking mode at speeds up to 160 miles per hour.
So the only way to fully reconstruct this event was to reproduce a computer model which we'll talk about later.
I want to give you a couple definitions. Barrier impact is controlled by the amount of energy that has to be dissipated during the impact and the direction of the forces applied to the vehicle.
There's two parameters associated with the velocity vector. That's the speed and the direction. The speed of a vehicle, I think everyone in here can appreciate that the faster a car is traveling, the more energy has to be dissipated during the barrier impact and the more severe the event.
The other thing associated with the velocity vector is its angle relative to the barrier. We call that the trajectory angle. The trajectory angle also has a great influence on the amount of energy that has to be dissipated during a vehicle collision with the barrier. So we'll be talking about speed and trajectory angle, and those control how much energy must be dissipated. The third parameter is heading angle. That's the orientation of the vehicle relative to the barrier at the time of impact.
Heading angle controls the point at which forces are applied to the vehicle and the direction they're applied to the vehicle, which, in turn, affects how forces are applied to the occupant's seat and to the driver restraint system. Which, again, becomes an important part of the injury causation.
When we went to the track, we found a great deal of tire mark evidence. We actually found marks associated with all eight tires -- well, seven tires and one suspension -- recall that the right rear wheel of the No. 3 car was broken off during its collision with the No. 36 car.
But there were scrape marks associated with the right rear shock absorber and a u-bolt on that suspension and these are those scrape marks and clearly identified during this crash.
What we found was, when we analyzed the angles of these tire marks when they approached the barrier, was that the front tires for both cars had a significantly higher angle of incidence with the barrier than did the rear tires.
This indicates that just as shown in the video, these vehicles were rotating into the barrier at the time they struck the wall.
That when a vehicle is rotating into the barrier, the front wheels are carried closer to the barrier while the rear wheels are carried further away from the barrier.
That would make the angle of approach for the front tires to be higher than the actual trajectory angle, and the angle of approach for the rear tires to be lower than the actual trajectory angle. From tire evidence alone, we were able to conclude the trajectory angle for the No. 3 car was between 12 and 15 degrees and the trajectory angle for the No. 36 car was between 10 and 12 degrees.
There's been speculation both the 36 and 3 car were traveling at the same speed and therefore they had the same severity of impact with the barrier.
What this tire mark evidence clearly indicates is that the severity of the No. 3 car impact with the barrier was significantly different, just a two or three-degree change in trajectory angle creates a 25-percent increase in the energy that must be dissipated during the barrier impact. That's quite a substantial difference.
The GPS data that each car carried a receiver that allowed Sport Vision to record its location five times every second. From that data we were able to identify the location of this receiver throughout the event by taking the differences in location between each recording event where it will determine both the velocity and the trajectory angle of that vehicle.
However, because there was only one receiver on the vehicle, we were not able to get heading angle. So from the GPS data we were able to get velocity, in terms of speed and trajectory angle, but not heading angle.
We were primarily concerned, again, with the barrier impact and the collision between the No. 3, No. 36 car.
What we found from the GPS data was that the speed of the vehicles at time of collision with the barrier was between 156 and 161 miles per hour. And that the trajectory angle data from the GPS exactly matched the data from the tire marks. It's always good when all the evidence gives you the same answer.
The collision between the No. 3 and No. 36 car, remember, we found that this collision didn't change the speeds of the vehicles greatly, but it did change the direction. So it's the total velocity change, including the change in direction of the speed, amounted to about nine to eleven miles per hour for this impact event.
Now we go to the photogrammetric. The seven video cameras that captured the event take frames approximately 30 times every second. Each frame can be used through photogrammetric to determine the location of the vehicles.
The way that works is that we first produce a computer-aided drawing, detailed drawing of all the background features of the track. Produce a drawing from a camera view from (inaudible) that produce a camera from that drawing package that exactly replicates the photograph.
When we've completely replicated all the background information as well as the car location, we believe we've reconstructed the location of car during that point in time.
By reviewing or working with all seven of the video camera views, we were able to determine a great deal of information about this accident. What we found was -- this still was taken from a video camera just at the point where the No. 3 car is coming back on to the track.
At this point the driver has regained, or has attempted to regain control, steered the car back on to the track, and now has a velocity here of about 164 miles per hour. His trajectory angle relative to the barrier is along this path and has an angle relative to the barrier of over 17 degrees.
The rest of this group of cars is traveling at about 170 miles per hour. Even though he's a little bit ahead of them, they're gaining fast.
And from this point, this point in the accident to the next frame here, is about 300 milliseconds, a little less than 300 milliseconds or a little less than three-tenths of a second. Meaning there was really no time for the No. 36 car driver to respond and take any evasive action to avoid this collision.
The GPS data showed again at this point, the photogrammetric showed at this point this vehicle was traveling at about 162 miles per hour, and the No. 36 car was still traveling in the neighborhood of 170 miles per hour, 169, 170, along with the rest of this pack of cars here. This vehicle had an orientation of heading angle of about 26 and a half degrees, that heading angle was beginning to diminish. It was beginning to be straightened out back toward the track at this point.
Its velocity vector into the barrier was still about 17 degrees. The collision between the 36 car and the 3 car did two things. First thing it did was it slowed down the velocity of the car, of the No. 3 car, toward the barrier. Slowing that velocity down is a good thing; it reduced its effective angle, trajectory angle, relative to the barrier from about 17 degrees to about 13 or 14 degrees.
The other thing it did is it caused the No. 3 car to start to rotate clockwise into the barrier, which significantly changed its heading angle.
What we see here is a frame from just before impact. You can clearly see the heading angle into this barrier is much larger than it was in the previous view, and that controlled how the forces were delivered to the No. 3 car; meaning that it was more of a frontal hit than a side hit.
Estimated impact conditions from the photogrammetric analysis closely replicated the previous results from the GPS data that we gave a little bit closer estimate of speed, 157 to 160 miles per hour. Heading angle is a little tighter, tolerance on trajectory angle. This is our first estimate of heading angle that we were able to determine from both of these cars.
Again, the trajectory angle for the 3 car and 36 were quite different, meaning a significantly more severe hit for the No. 3 car.
We also, again, found from the photogrammetry that the total velocity change during the collision between the No. 3 and No. 36 car was between nine and eleven miles per hour; that the barrier impact occurred about four-tenths of a second after the collision with the No. 36 car and that the total velocity change during the impact for the No. 3 car was between 42 and 44 miles per hour.
Large majority of all off-road collisions that result in fatality occur at speeds well below that number. This is a severe hit.
We wanted to verify those estimated impact conditions, so we conducted a full scale crash test to try to reproduce the damage between the -- of the No. 3 car by running a controlled crash test at roughly the same energy conditions that we estimated from the previous slide.
Our goal, again, was to replicate vehicle damage. We started at the rear and compared damage from the rear axle forward and found both a similar pattern and a similar magnitude of damage for both vehicles indicating that our original estimate of the energy of this impact was very close.
We then went on to full scale computer modelling. We wanted to determine, conduct a full reconstruction of this accident and determine what happened during the barrier impact as best we could. So we developed a finite element model using state of the art software in conjunction with Altair Incorporated. What you see up here on the right is a structural frame of a Winston Cup car.
In fact, Dale Earnhardt's car. And down here is that same structural frame with all of the ancillary components including the engine, transmission, radiator, etc. the first question we had was is this model accurate.
So we replicated the full scale crash test using our computer model. In this case, we again compared damage predicted by the computer simulation to the actual damage to the No. 3 car. We also compared the timing of the accident of the crash. In other words, at what time did the radiator strike the wall?
At what time did the wheels strike the wall and the engine, etc.? We also compared the overall impulses applied to the two vehicles and the nature of those impulses, and they compared very well.
So we were very satisfied that this was a validated model and could be used to reconstruct the Earnhardt crash.
Again, the vehicle damage was quite similar, as you see here. And this wasn't just the overall damage. We were looking at damage to individual components, including the headers and the roll cage. The next thing we did was conduct a simulation of the full scale crash, of the actual crash.
This vehicle was impacting the barrier at the 158 mile per hour estimate we had for impact speed with a 13.6 degree angle of incidence to the barrier with a heading angle of about 57 degrees.
What this simulation showed us was that this vehicle, even without the No. 36 car here to influence it, didn't redirect. It wasn't steered away from the barrier. During a typical barrier impact, the front of the car would be pushed outward and the rear would come around and have a rear-end slap. That would be a typical redirection hit. That's what we would call that.
Another alternative, if they have a high angle relative to the barrier high heading angle, sometimes the vehicles will spin out and the right side will spin around and slap the barrier.
In either event, you have two impact events that take out a lot of velocity. In this situation, the vehicle ran into the barrier and kept the same heading angle. It was really no redirected force applied, no moments applied that changed the angle of the vehicle indicating that this was a critical angle hit. All of the velocity, lateral velocity, had to be taken out during the primary impact, which is a worst-case scenario.
Again, the vehicle damage predicted by our model compared very well with that observed on the No. 3 car throughout the vehicle, not just the overall that I showed here today.
In summary, we believe we very accurately estimated the actual impact conditions of the No. 3 car. 157 to 160 miles per hour; 13 to 14-degree trajectory angle; 55 to 59 degree heading angle; and an overall crash duration of about 800ths of a second. Total velocity change during the event between 42 and 44 miles per hour.
The velocity change of the No. 3 car when it collided with the 36 car was in the nine to eleven mile per hour range. The crash test and the computer modelling accurately reproduced damage to the No. 3 car. And, finally, I want to step back and look at what we've done for going forward.
The computer model that has been developed in this effort can be extremely valuable and is currently being used to design a better barrier at our facility, and also can be used to design better vehicles and better restraint systems in the future.
With that, I will turn it over to Dr. Raddin, who will continue with the injury causation analysis discussion.
DR. JAMES H. RADDIN, JR.: Thank you, Dean.
We're going to talk now about the other end of this process. Dr. Sicking has given you the vehicle dynamics part. We're going to now look at some of the clinical injuries that were involved. What we find is that there were some injury to the head, and clearly it was a head injury that produced the death.
What we find first is that there's an area of contusion to the occipital region, which is the back portion, back lower portion of the head. That contusion was measured to be approximately 8 by 5 and a half centimeters. Now, what that means is about the size of my Texas driver's license.
And if you put that here, that's about the size of that area of contusion. Now, a contusion is a bruise, and a bruise is an area of hemorrhage in the soft tissue without a break in the skin, without a laceration. That indicates blunt trauma, blunt impact to that portion of the head.
Similar findings were noted on the right, but the only measured finding was noted on the left. And there was a notation that some more scattered hemorrhages were present and apparently smaller areas of hemorrhage present more to the right side of the scalp and towards the top.
The injury itself is a fracture to the base of the skull. Now, the base of the skull, the slide that you have up there, is looking at the base of the skull from the inside down. Let me just tell you, the base or the floor of the skull is the part of your skull that you can't put your hand on. You know, you can touch the upper part of your skull. But the base of the skull is a location where your neck attaches to your head.
But there is a floor, a bone across there, upon which the brain rests. If you look at it from the inside, as we have in that particular diagram, it says that we have a ring fracture at the base of the skull. Now, a ring fracture, if it's a complete ring fracture, is one which forms a circle typically around this foramen magnum, which is a hole in the bottom of the skull through which your spinal cord comes out. The ring fracture involved fracturing through several of the bones in the base of the skull. It's not a perfect circle. It's a ring fracture; it's an irregular ring. When it's a complete ring fracture, as this was, that ring fracture is a closed figure around the foramen magnum.
Now, there was wider separation of the bone fragments in the front, but it was described as a complete ring fracture.
There was also noted an abrasion about an inch by six-tenths of an inch or so over the right side of the chin. There were other marks on the body, including abrasions on the torso, one which was over the left clavical, which is the collar bone. That's in a distribution of the shoulder harness. There were other abrasions on the torso and a contusion, another area of bruising, to the mid abdomen.
There were multiple rib fractures. There were rib fractures that for the most part, in fact all of them, were on the left side of the chest. These were, for the most part, anterior rib fractures. They were rib fractures which showed very little bleeding around them. And there was also a fracture of the breast bone, or sternum, which was at about the location where the third rib comes in.
You'll note there's a prominence of injury to the left side of the body. There's left rib fractures, the left mark across here, that continues in other places because we've got a left fracture dislocation at the ankle.
And the left part of the body, as it comes -- is exposed to the greater amount of trauma here is consistent with some of that occurring as a result of the left side of the body moving forward more than the right. And that's consistent with what you would expect to find if a left lap belt separates under load before the forward motion is completed.
The cause of death, as listed by the pathologist who did the examination, is blunt force injuries of the head. That is the way that the cause of death was determined.
And the findings that he had there are certainly consistent with that. That gives you an idea of the injuries that we are trying to establish the cause of, in the context that we have in this crash. Dr. Sicking has talked about the vehicle dynamics. Let me just review some of the things that I take from that.
One, he said that this was a pretty severe crash. A 43 mile per hour velocity change is about what you would expect if you were sitting in a parked car and a car of similar size, similar mass, approached you at about a 22-degree angle from the front at a speed of 75 to 80 miles an hour. That's not a minor crash; that's a severe crash and it's one in which you would expect generally poor results in passenger car settings.
But even with the more advanced protection systems of a race car, it represents some challenges. Things have to work right if that's going to be protected against. The principal direction of force, or the line of action of this, involves a line of action from two different crashes. And it makes it a much more complex analysis than perhaps has been appreciated.
First of all, we had a lateral force that derived from the contact from the 36 car, and that came in at about 8 degrees off of straight lateral. The second was the impact with the wall, which when you resolve those numbers, comes out to about 22 degrees to the right forward. It's where that force is coming from and occupants tend to move opposite and parallel to those principal directions of force.
Now, as we move from vehicle dynamics to occupant kinematics, we try to understand how the occupant moved. One of the pieces of data that we have for that is looking inside the vehicle. When we look inside the vehicle, we find a number of findings. One is that there was contact with the steering wheel.
The steering wheel is deformed. This is the steering wheel, this is the hub where it mounts. These are spokes which are bent. There are abrasions in several areas around the rim, and when you put that hub on a flat surface, part of that steering wheel is pushed significantly in by a distance of roughly five inches.
And part of it is actually pulled out a little bit by a distance of about two inches. It would normally operate at about five inches straight across, if you put that together.
There were abrasions on several areas with some imprints, scuff marks on the edge of the wheel. This is a close-up looking at the edge of the steering wheel from the side.
There were marks on the seat. The seat surface is covered with a velour cushion, or pad, over the top of it.
A portion of that velour -- and I'm talking about this area from right around here and this area right around here - actually was melted from motion of the occupant. And you note that that motion was only on the right side of the seat. These are the areas of melting right here. And they go in the angle that Dr. Sicking's reconstruction tell us.
This was at about a 22-degree angle. When I put a light on those, you can see there's just a scuff with melting of portions of that fabric coming this way. There's a support structure that comes across here, so it forms a little ridge, and it was contact here and then right over the ridge, where we see those abrasions.
That says the occupant moved forward and moved forward considerably under force during the event.
There was also findings on other portions of the seat. As you look down into the seat cushion, there is a support wing that would normally be to the right of the torso.
Your arm would typically go over that. This is a support structure. And it's not -- it's just kind of a positioning rest; it's not something that is designed to force you into that position. The underlying metal of that support structure is bent to the right. When you look right at the top of that support structure where the velour transitions to what looks like a vinyl, there is a scuff with some fabric that is bunched over to the right side, and there's a scuff that goes to the right and then a subsequent scuff, a higher speed scuff, going forward.
When you look at that up close, you can see bunched material over to the right. And there's evidence that the occupant went to the right, as well as going forward.
We have a more complex crash here.
As you look at the toe pan, this is the pedals. Here is the steering column penetrating through the toe pan. That's the floor board area in the front. There is a scuff just to the left. And as you recall, there was a fracture dislocation of the left ankle, indicating further forward motion of the left side in towards the floor.
Those marks help us understand the occupant kinematics, and they are consistent with what we see in the vehicle dynamics.
We know that occupant kinematics follow from Newton's first law, because Newton's 1st Law says an object in motion remains in motion until acted upon by some force.
The force typically begins by being applied to the car. The car gets changed and the occupant in motion is the occupant inside and continues parallel and opposite to the principal direction of force. And when that happens, we know how that looks.
This happens to be a picture of me in one of my crash tests back in the Air Force. It's a lateral hit. It's in a direction within eight degrees of the lateral motion of that other hit that we talked about, from the 36 car.
It's in the opposite direction. Here, I'm moving from a hit that was on this side, and I'm moving to my left. As you recall in the 36 impact, it was on the right side. That occupant would move to the right.
I'm in this harness wearing a harness that is both stiffer and offers greater lateral support because of the way the harness is implemented with some double straps up here.
But it is a five-point harness with a central buckle, a crotch strap, two lap belts, and I'm wearing a helmet.
The velocity change is probably -- it is a little higher, it is a little higher than what was experienced in the 36 car but the acceleration level was pretty similar. The motion you would expect from the occupant of the 36 car would, in fact, be greater than what you see here because of the differences in the restraint system.
The other thing that you can see as you look at this is that you can also understand that helmets obey Newton's 1st Law, just like people do. And as the occupant tends to go this way, the helmet starts rotating over my head. It does so less in this particular case because I've got a nape strap that comes across under the kind of -- on the back of my neck that helps retain that helmet but it still opens up the area in the occipital region, even with that helmet. You would expect greater rotation than in Dale Earnhardt's helmet.
So helmets obey the same principles, and I've looked at that helmet and it does show evidence of having moved with respect to the head. The microphone attachment, for example, has been pushed up and imprinted in to the edge of the helmet on the left side.
So the conclusion is that not only did the occupant move to the right, but the helmet tried to continue going further and rotated on the head. I would like to show you an example of the type of motion that I'm talking about. This is not a great quality video, but I'm going to show it to you anyway.
It's the best copy we have. It's a copy of a video that was made in which Johnny Benson is driving a car. We thank him for providing this. It makes an impact to the wall, not a big impact to the wall. He ended up driving the car away. It does not have the kind of damage.
But here he is. Here's his helmet. Here's a side support here. Hands on the wheel. Here comes the hit. That's not a big hit. But did you see that he moved in response to that motion, or in response to those forces on the car?
This is a kind of a still frame from the video, but you can kind of get a sense of the helmet. If you can see, he's kind of doing what I did in that crash. Your head doesn't just rotate this way; it rotates this way in a roll but it always -- you end up looking towards the floor and the helmet is moving further than he is. His hand is off the steering wheel.
As he comes back, the helmet, even in a full-face helmet here, is displaced, even still as he comes back, at a point that is still displaced with respect to his head and not lined up with it.
What we understand from this impact analysis and from looking at the evidence in the vehicle is that a complex set of kinematics occurred because in these kinematics, the occupant started from generally in this position. This would be basically a normal driving position for Dale Earnhardt. He tended to drive with his head to the left, more over towards the window.
I don't know exactly where his head was at the time this crash took place, because as he's starting to steer, he may have looked to the right, he may have moved his head off the headrest; I don't know those things.
But I'm going to give you just a representative scenario in drawings that are not intended to depict precise motions. I can't tell you that his fingers were just this way or his arms were just this way. But we're talking about a set of complex kinematics which I'm going to try to represent with these drawings.
During the impact with the 36 car, he would have necessarily responded in a way at least as much and potentially significantly more than did Johnny Benson in the video clip you just saw, or in my crash test. The helmet would rotate over and the area where we're talking about finding evidence of contusion on the head is located right here.
That puts him in a position where the left side of his head is leading instead of the front portion of his head. I would expect him to be in the process of that response. He is not likely to be at the full amount of that response.
He's probably coming back some from that response, but the 36 car impact occurred approximately 400 milliseconds before roll impact. That's approximately twice the duration of an eye blink that you just did.
The 36 car hits, the response is here, and then as an occupant would move forward into the right at about a 22-degree angle, the head would be trying to go to the right but as the torso belt arrests the torso motion, the head is going to begin to swing back towards the left after having displaced forward.
Now, we don't know where in the process the seat belt separated. If we knew exactly when that was, we could be much more specific about some of these things. But we do know that the seat belt separated under load at a time that allowed further excursion forward and we know that when the belt separates, the buckle moves further forward and that's where the torso straps are attached so the torso can move further forward and that means the head can move further forward.
That occurs with the left side of the head leading, and as it swings back around, a greater displacement occurs and there are two opportunities for head contact to produce blunt force injuries of the head.
One is in conjunction with contact with the steering wheel rim, with the helmet displaced. And the other one would be on rebound. Let's look at the steering wheel rim potential first. With the helmet displaced, contact could occur directly to that portion of the head. It could occur, and would be expected to occur, in a fashion not with the head pushing forward on the steering wheel.
These steering wheels are designed to be hit and pushed forward under impact, and they move forward with a velocity with a force applied somewhere in the neighborhood of 200 to 300 pounds in that neighborhood as you push forward. That provides some ride down of the head against it. But if you hit a ring structure more radially, it's stiffer. When you measure it, it's about 1300 pounds.
That provides a significant opportunity. When you go through the kinetic analysis, you find that the velocity between the head and the steering wheel that could develop in the distance it would take to get there would put this well above 30 miles an hour and sufficiently powerful an impact to produce a basilar skull fracture in conjunction with the head tension -- or neck tension that would also be present. Rebound provides another opportunity, because as you rebound, the head is pulled back by the torso. The helmet wants to continue to stay forward, and so there's an even greater displacing characteristic of the helmet as it comes back.
So we see two specific opportunities, and I can't tell you which one of those occurred. But the kinematics are consistent with that kind of motion, and then as the person makes a secondary rebound, additional modest contacts would take place that would not expect to be -- you would not expect them to be injurious but would certainly move the helmet back into a normal position.
Well, we need to talk a little bit about biomechanics and we're done. We need to know a little bit about ring fractures. I won't try to educate you entirely about that, but I'll make some comments about it. These are part of the evidence for what we conclude as being a blunt force impact to the head, as the pathologist stated was the cause of death in conjunction with some neck tension.
We'll talk about our basis for that. We'll look at some alternate theories that have been talked about around the country.
And we'll look, I'll at least mention to you, that I went on and did sled testing in this case and I put a hybrid three anthropromorphic test on me, balanced it and put it in a size of Dale Earnhardt, put this in a position deviated to the right and got motion forward and into the steering wheel and then contact on rebound.
The rebound contact's a little different in the sled test, but taking into the fact that the rebound should occur more to the left, it certainly provides a basis for significant contacts in either of those two positions.
Let's talk a little bit about ring fractures. First of all, they're relatively common fatal injury in car crashes. We see them frequently. We tend to see them as a result of impact to the head. You don't have to impact the floor of the skull. You typically can't access that because the neck is attached. But you can impact the head in a number of different locations and produce ring fractures to the base of the skull as a result of either tension, compression, or torsion, that's twist, on the skull base.
The same kind of fracture can be produced from a number of different kinds of impact. You can get the same kind of fracture from a head impact that you get from straight torsion or from hyper extension type of impact.
Jim Benedict and I have concluded that a head impact with neck tension likely was the cause of the ring fracture for Dale Earnhardt. We base that on the autopsy finding which says there was a contusion and says it was a blunt trauma. You'll see that much of these bases are portions that were developed during this investigation. They were not present and not available to everybody to start out with.
We looked at an occupant kinetic analysis of a multicollision sequence, both the 36 car and the wall. Helmet kinetic analysis, physical findings in the vehicle and on the restraint. We looked not just at the helmet kinetic analysis, we looked at the helmet and did a CT scan on it to see if it had been impacted higher up. It was not.
We've analyzed the belt separation kinematics and how that allows a greater displacement. We've looked at the occupant injury patterns. I've shown you some of that. Physical findings on the helmet, both from the CT scans and looking at it.
The data in the scientific literature is consistent with impact causation to produce the ring fracture. And a biomechanical and kinetic analysis of the impact opportunities.
There have been other theories discussed. A number of people have talked about simple head whip. People have talked about mandibular fracture.
The previous analysis by Dr. Myers was one which basically talked about a combination of the two; didn't talk specifically about head whip alone; said that that could do that. But indicated neck tension in conjunction with a head impact, and based upon his data, he looked at that head impact as if occurring to the mandibular. So he saw that as a combination of the two.
It's been reported in various ways around the country.
With neck tension alone, I looked to see if there was something that would tend to give us conclusions with neck tension alone. I looked for the kind of torso contact that would produce the violent head whip that could produce that. I did not see dramatic injuries to the torso. We saw the rib fractures which we discussed, but not internal injuries to the torso.
I looked to see if there was evidence of stretching in the back of the neck from a head whip. That was looked at in the pathology findings, and that was not found. The back of the neck did not show damage to the ligaments or to the bones.
And simple head whip does not explain a death caused by blunt force impacts to the head.
So I don't think we're looking at something. I think it's possible that you can get and you can certainly get basilar skull fractures that way. I think it's relatively unlikely that that is the explanation in this case.
You would not expect to see evidence of -- if you were looking at mandibular impact or impact to the chin, you would look for some findings on the chin. The findings that we see on the abrasion that's described to the right side of the chin from the data I have, would appear more consistent with the chin strap of the helmet as the helmet is attempting to rotate off the head. Particularly with contact to the head.
You can certainly get a basilar skull fracture with mandibular contact. But I would expect a kinetic trajectory consistent with that. If I am forward with the left side of my head leading, I would not expect to hit the right side with my mandibular. In particular, if the left side of my head is leading and I'm to the -- already to the right of the steering wheel and moving to the right, I would not expect to hit the right side of the mandibular, nor would I expect to see evidence of contusion to another part of the head.
Because of the findings we have developed and the analysis of the kinematics, the earlier finding basically said neck tension with some head impact, which based upon that data was from the mandibular. I'm saying there's some neck tension involved but the head impact is more likely to be here. And it's in neither case likely to be pure head whip or head stretch. Our conclusions are these:
It's consistent with the analysis of things that were problems developed in something that normally works pretty well. There's a tendency to want to have a single finding that you can say, "Ahh, that did it, and that's what it is and that's what's at fault." That makes it a lot easier to report, I suppose.
But typically, when something that normally works pretty well, if just one thing goes wrong, it tends to still work pretty well. When multiple things come together and go wrong together, that's when you have problems.
And that's what has happened here. There were multiple events, each of which provided factors which contributed or potentially contributed to the injury. What we see is one of the factors was a very severe collision at a critical angle to the wall.
We see a significant car-to-car collision that occurred very shortly, twice the duration of an eye blink, before that occurred, that pre-positioned the occupant and in that pre-positioning impact and in the wall impact and in the rebound, all of those provide a basis for further displacement of the helmet with respect to the head.
The seat belt separated. It separated under load and it allowed additional forward motion.
On the basis of our analysis, I cannot give you a relative contribution; that it was 30 percent and 40 percent that. But what I cannot say is that there was no potential contribution when you have a head impact in a setting in which a belt separates and allows greater displacement.