FID Final Publishable Report

Document Sample
FID Final Publishable Report Powered By Docstoc
					                               FINAL PUBLISHABLE REPORT

CONTRACT N° : 1999-RD.10559

PROJECT N°        : GRD1-1999-10559

ACRONYM           : FID

TITLE             : Improved Frontal Impact Protection through a World Frontal Impact Dummy.


                          TNO Automotive, Safety               (TNO)           NL

PARTNERS          :
                          BASt                                 (BASt)          DE

                          INRETS                               (INRETS)        FR

                          Transport Research Laboratory        (TRL)           UK

                          University of Heidelberg             (UoH)           DE

                          Polytechnical University of Madrid   (UPM)           E

REPORTING PERIOD                 : FROM         1/1/2000       TO       31/10/2003

PROJECT START DATE               : 1/1/2000                    DURATION : 46 months

Date of issue of this report     : 22/12/2003

                                                Project funded by the European Community
                                                under the ‘Competitive and Sustainable
                                                Growth’ Programme (1998-2002)
                                           FID project Final Publishable Report

1     Table of contents

1     Table of contents........................................................................................................................... 2

2     Executive Publishable Summary................................................................................................. 3

3     Objectives of the Project .............................................................................................................. 3

4     Scientific and Technical Description of the Results .................................................................. 5
    4.1   WP1: Accident Analysis............................................................................................................. 5
    4.2   WP2: Generating additional PMHS data.................................................................................... 7
    4.3   WP3: Definition Requirements for Frontal Dummies.............................................................. 13
    4.4   WP4: Performance of Existing Frontal Dummies.................................................................... 14
    4.5   WP5: Modification and Testing of Dummy Enhancements..................................................... 20
5     Management and Co-ordination Aspects ................................................................................. 27
    5.1   Consortium ............................................................................................................................... 27
    5.2   Description of the Participants ................................................................................................. 28
    5.3   Web site.................................................................................................................................... 30
    5.4   Harmonisation issues................................................................................................................ 30
    5.5   Dissemination........................................................................................................................... 30
    5.6   Partners – Contact Persons ....................................................................................................... 32
6     Results and Conclusions ............................................................................................................ 33

7     Acknowledgements..................................................................................................................... 34

8     References ................................................................................................................................... 35

                              FID project Final Publishable Report

2    Executive Publishable Summary

Frontal collisions produce the highest frequency in fatalities and injuries in road accidents, the costs to
the European Society alone being more than €160 billion each year. Even though the European
Frontal Directive became effective only in 1998, further research is required to better assess the
protection offered to car occupants in frontal impacts. An important motive for this is found in the
current Hybrid-III crash test dummy that has shown important deficiencies in human likeness and
injury assessment capabilities. The Improved Frontal Impact Protection through a World Frontal
Impact Dummy (FID) project aimed to contribute to a further reduction of the amount of injuries and
fatalities in frontal collisions. The most important achievement of the project is the introduction of an
improved a frontal impact crash test dummy with realistic movements and injury indicating
measurements for future automotive crash testing. The FID consortium has defined the requirements
for an advanced frontal impact dummy based on new biomechanical and accident investigations and
has recommended a design that meets the demands in terms of biofidelity, anthropometry, durability
and injury assessment capabilities. The FID consortium existed of six experienced and
multidisciplinary partners from the UK, France, Germany, Spain and the Netherlands. An important
dimension within the project was that of harmonisation of safety regulations world-wide and in
particular between Europe and the United States.

3    Objectives of the Project

In terms of the number of fatalities and the severity of injury, frontal collisions represent the most
serious accident type among various road vehicle accidents in Europe. The objective of this project
was therefore to contribute to the reduction of the amount of injured and death car occupants involved
in frontal collisions.

The FID project contributes to European Union policies in the field of “Transport” and in particular
the aspect of “Safety of Transport”. The overall societal costs of fatalities and injuries due to road
accidents in the European Community are more than € 160 billion. Fatalities and injuries due to
frontal impacts constitute an important part of the problem. This project contributes to better
protection of vehicle occupants in frontal collisions. Furthermore since injuries in frontal collisions
often lead to permanent impairments, any reduction in the frequency of these types of injuries are of
great importance from ‘quality of life’ point of view. The automotive industry has a responsibility to
reduce the individual suffering and total costs due to accidents in society. By developing products that
will reduce the risk of injuries, the benefit is reduced injury related costs and suffering. Vehicle safety
is a problem all over Europe but most critical in terms of numbers of casualties in some less-
developed countries (Greece and Portugal). Since vehicle design improvements resulting from this
project are to be introduced at a European scale, these improvements may be in particular beneficial to
countries with high number of causalities.

The FID project includes deliverables, which directly tie to the European Regulations. The broad co-
operation in this project forms a base for wide acceptance of the findings of the research performed in
the project. Furthermore the biomechanical results, which include a test dummy and its specifications,
injury criteria and tolerance levels are presented to Working Groups of the International Standard
Committee (ISO). At the international level various aspects of this project are co-ordinated with
organisations world-wide, in particular through EEVC and the International Harmonised Research
Agenda (IHRA) Biomechanics Working Group. The project results are presented to NHTSA at the
government-industry meeting in America.

The consortium actively promotes that the dummy will be evaluated by others (in particular the
industry) after conclusion of the project. The European automotive manufacturers and suppliers can

                              FID project Final Publishable Report

benefit through this project from new test tools in particular a new crash dummy, corresponding injury
assessment tools and a validated mathematical model of this dummy (computer aided engineering tool)
for the development of safer cars in frontal collisions. Due to this, their market penetration and the
competitiveness of the European industry is expected to increase since safer cars are a strong market
factor. This will contribute to the preservation of employment in the automotive industry and may even
result in creation of new jobs.

Loss of productivity as a result of car crashes costs is an enormous amount of money each year to all
companies in Europe. The reduction of work-related road accidents will contribute to improvement in
working conditions and add substantial savings to the European companies. Vehicle fleet owners will
gather a cost saving in the cost to manage a vehicle fleet.

Particular challenges in the field of passive safety are the trend towards smaller, lighter and more fuel-
efficient vehicles and the increased usage of electrical vehicles for environmental reasons. An optimal
combination of various technologies is required to offer passengers of these lighter vehicles a similar
level of protection as in conventional vehicles. This project results in improved protection methods in
frontal collisions and consequently indirectly will contribute to the easier introduction of
environmentally friendly vehicles in Europe.

The objectives defined for each work package within the project are defined below:

WP1 - Accident studies:
• To prioritise the injuries for different body segments during frontal impacts;
• List injury types for use in identifying dummy performance/instrumentation needs for subsequent
  tasks of the project;
• Assess the differences in injuries that can be attributed to the airbag, the seat-belt system, and the
  internal vehicle structures distinguished between vehicles of modern design in comparison to
  older vehicles.

WP2 - Biomechanical data:
The objective of WP2 to obtain new biomechanical data on the behaviour of the human body under
frontal impact conditions in order:
• To quantify the response to impact of shoulder/thorax complex, pelvis/femur and foot/ankle
• To better understand the injury mechanisms.

WP3 - Dummy requirements:
• Definition of the biofidelity performance requirements for evaluation of frontal impact crash
• Definition of anthropometry requirements for evaluation of frontal impact crash dummies;
• Review of current injury assessment values;
• Definition of requirements for evaluation of the durability, repeatability, instrumentation of
  frontal impact dummies.

WP4 - Dummy evaluation:
• To evaluate existing dummies (Hybrid-III and THOR-Alpha) in terms of biofidelity, repeatability,
  durability, sensitivity and kinematics, with respect to the requirements defined for a frontal impact
  dummy in WP 3;
• The performance of the whole dummy and its individual body parts will be evaluated
  experimentally in both component and sled based tests.
• To define areas of design that need modification will be identified for development in WP 5.

                             FID project Final Publishable Report

WP5 - Dummy development:
• To develop and update version of multibody model THOR developed within the ADRIA project;
• To design, develop and manufacture improved prototype dummy parts, systems and/or body
• Develop and verify feasibility of enhanced dummy parts.

4     Scientific and Technical Description of the Results

4.1    WP1: Accident Analysis

The objective of the frontal accident analysis was to prioritise injuries for various body regions of
belted drivers and front seat passengers, occurring in frontal accidents. The aim was to prioritise
injury types for different body regions and to attribute the injuries to the airbag, seatbelt and the
internal vehicle structures. A list of injury types of the most severe and most frequently injured body
parts has been developed for use in identifying dummy performance and instrumentation necessary
for the subsequent work packages in this project.

The accident analysis performed was based on the largest two European accident databases available:
The database of the UK Co-operative Crash Injury Study (CCIS) and the German database of the
Medical University of Hanover (MUH). The data used were from frontal accidents (11-1 o’clock),
focusing on car to car and car to obstacle impacts, EES < 80 kph and vehicles not older than 1990.
Occupants were 12 years or older and belted (driver and passenger). Table 1 shows the comparison of
the two databases, categorised for driver and passenger and airbag deployment and not fitted/no
deployment. It can be observed that the CCIS database includes more airbag cases. Further study
showed that the overall accident severity of MUH data seems less than in CCIS cases, which is due to
the different sampling strategy used for both databases. This difference added to the different size of
the databases and different EES calculation applied, made that the databases could only be analysed
Table 1:    Comparison of accident databases
                                 All EES < 80            CCIS     MUH
                              Driver – no airbag          151     747
                                Driver - airbag           152      57
                             Passenger – no airbag         87     281
                              Passenger – airbag           10      11

One part of the analysis examined the maximum injury to each body region for all occupants. This
provided information on the frequency and severity of the injuries in the samples by body region. To
identify injuries on specific body regions, the accident data distinguished between the following body
regions: head, face, neck (including cervical spine), shoulder, thorax (including thoracic spine),
abdomen (including lumbar spine) pelvis, upper arm, elbow, lower arm wrist hand, arm (not further
specified) upper leg, knee, lower leg, ankle foot, leg (not further specified). For each of the above
body regions the total number of injuries were specified, as well as a breakdown into the different
injury types. The Abbreviated Injury Scale (AIS) codes, Revision 90, was used to define the injury
types. To ensure consistency between the two databases, the injuries in each body region were
classified as skeletal, soft tissue, or surface type. For the extremities, a type of injury related to
tendons, muscles and ligaments was included. For the head, a brain injury type was used, and for the
spine, nerve and cord types were included. As an example of the type of results that were obtained in
this study, Figure 1 shows the comparison of the CCIS sample distribution for AIS2+ injuries of

                                FID project Final Publishable Report

restrained drivers, with and without airbag deployment. Table 2 shows the AIS 2+ injury distribution
of the MUH data set for those body regions, which will directly be addressed by frontal airbags.
Table 2:     Distribution of AIS2+ injuries per body region for restrained frontal occupants (MHH)
                    Percentages of                                     Injured Body Region
                                       Injury caused by
                    AIS 2+ Injuries                           Head        Neck    Thorax Other
                                       Airbag                   14.8                   1.9
                                       Steering Wheel                                  5.6
                                       Body Movement             3.7                   1.9
                      Frontal Airbag
                                       Belt                                            3.7
                     (n = 54 Injuries)
                                       other / unknown           3.7                   1.9
                                       TOTAL                    22.2          0.0     14.8  63.0
                                       Steering Wheel            3.1                   1.5
                                       Belt                      0.1                   2.6
                         without       Body Movement             0.7          1.2      0.4
                      Frontal Airbag Windscreen                  1.5                   0.1
                    (n = 740 Injuries) other                     3.2          0.1      1.1
                                       unknown                  24.7          2.2      7.8
                                       TOTAL                    33.4          3.5     13.5  49.6

                            Percentage of                                   No airbag (N=88)
                   30%      the number of                                   Airbag (N=96)




















       Figure 1:     Distribution of AIS2+ injuries per body region for restrained drivers (CCIS).
Despite of the small case numbers, the accident analysis shows clearly that the head and especially the
thorax are the body regions most severely and most frequently injured in frontal accidents. In
accidents with airbag deployment, the injury severity of head injuries decreases considerably, while
the percentage of severe thorax injuries remains constant. It follows from the accident data that severe
head injuries are mostly brain injuries without fractures. Severe thoracic injuries are mainly organic
injuries and fracture of the ribs. The number of foot injuries, in particular fractures, is considerably
high in accidents with airbag deployment. In accidents without airbag deployment, drivers do often
suffer face injuries. Interestingly this remains also to be the case for front seat passengers in accidents
with airbag deployment.

Regarding the requirements for an advanced frontal dummy, the accident study has pointed out that
head and thorax area are the most important part to protect. Injury assessment for these body regions,

                               FID project Final Publishable Report

as well as for the abdomen, may require different criteria for different types of contact (e.g. steering
wheel, belt or airbag) and hence the responses of these body areas should be appropriate and sensitive
to different loading conditions. Injuries to the lower extremities need more attention than given so far.

The main findings of WP1 were presented at a workshop held in conjunction with the 4th consortium
meeting, 21 February 2001 in Heidelberg.

4.2   WP2: Generating additional PMHS data
4.2.1 Activity 2.1: Human Thorax/Shoulder behaviour during frontal impact
The objective of this task was to obtain additional data on the behaviour of the human thorax/shoulder
complex under frontal impact conditions with and without air bag. Sled tests were performed by
INRETS-LBMC with fresh nonembalmed 50th percentile Post Mortem Human Subjects (PMHS).
Two series of three tests were performed corresponding to two levels of severity (low severity 30
km/h, high severity 50 km/h) and two types of restraint systems. The restraint system used for the 30
km/h test included a 4 kN force-limited shoulder belt and a static lap belt. A driver airbag was
mounted for the series at the highest severity (50 km/h). The complete restraint system originates
from state of the art French car manufacturer.

The observed injuries concern mainly the bones of the thorax (ribs, clavicle and sternum, see Table 3).
No other injuries, in particular visceral injuries are found. For all test conditions, the left ribs are more
often fractured and sternal fractures are also observed. Concerning the clavicle, only dislocations of
the sternoclavicular joint occurred on the left shoulder and are probably caused by the shoulder belt.
The injuries are a little reduced with the reduction of the velocity.
Table 3:     Injury Summary Table,

           Test           ribs fractured Left/Right       Sternal fracture       Clavicle fracture
                                                                             sternoclavicular dislocation
           FID11                    11/0                        no
           FID12                     2/4                        yes                       no
           FID13                     0/0                        no           sternoclavicular dislocation
           FID14                     2/0                        yes                       no
           FID15                     3/6                        no                        no
           FID16                     0/0                        no                        no

The number of fractures is sometimes important despite the reduction of the load on the thorax due
the belt load limiting system and the airbag. It is not surprising, regarding the great severity of the
impact. A recent study (Ydenius and Kullgren, IRCOBI 2001) has shown that the injury risk for
MAIS>2 is greater than 80% for a mean sled deceleration about15 Gs. This observation can be
correlated with a injury risk around 60% at 50 kph (MAIS>2) and 20% at 30 kph (but with a lower
deceleration). Although, the bone injuries observed on PMHS were often greater than those observed
in real car crashes, these results are in accordance with the data presented here. Nevertheless, the
reduction of injuries thanks to the airbag is not negligible. The reduction of thoracic injuries is
comparable to the reduction obtain with lessening the severity of the impact, i.e.: lower velocity,
lower energy, (See Table 3).

The occupant restrain systems are enforced by the requirements to reduce thoracic injuries. In order to
evaluate the capacity of the restraint systems like airbag and/or load limited belt, the thoracic injury
criterion must differentiate the behaviour of the thorax for the set of loading conditions that cover the
extent of restraint systems. Previous studies have demonstrated that the current thoracic injury
criterion (Kent et al., 2001; Shaw et al., 2001) does not represented with sufficient efficiency and
sensitivity the influences of the type of restraint systems used. The acceleration measured at the chest

                              FID project Final Publishable Report

for the dummy has been used as injury predictor, but this injury criterion has been based on the T1
measurement on cadavers. The T1 acceleration is considered to be representative of the chest
acceleration. The present study confirms the observation made for other cadaver tests, which included
accelerometers at both T1 and T8 vertebra, that this assumption is not true. For instance, for some
conditions, the T1 acceleration does not necessarily represent the chest acceleration. The maximum
value of the T1 acceleration does not, indicate the relative loading of the belt and airbag because, the
peak acceleration occurred only when the subject was load by the airbag (Figure 2). The chest (T8)
acceleration, by exhibiting a bimodal behaviour (Figure 2), seems to give more information about the
thorax loading conditions for driver sled tests and would be more predictive parameters. Kent et al.
(2001) have yet demonstrated that it is not the case for right front passenger sled tests. These results
are not conflicting with the present data. The fact that different studies with different kinematic
conditions and different restraint systems show different and contradicting observations, demonstrate
the importance of the crash conditions.

Figure 2:    Thoracic vertebra resultant acceleration in 50km/h test with airbag.
             Left: T1 level (1st vertebra in thoracic spine)
             Right: T8 level (8th vertebra in thoracic spine)
The Figure 3 shows clearly that the behaviour of the spine depends on the speed and restraint systems.
Without airbag, the belt with a 4 kN load limiter less restrains the body. The lower deceleration is
quasi-uniformly distributed along the thoracic spine and the movement of the corpse is quasi-
uniformly in the front direction of the impact since the vertical components are generally small
compared the horizontal one. Whereas, with airbag the value of the maximum deceleration along the
spine is very different. Specifically, the upper thorax (T1) is more decelerated than the lower thorax
(T12). As a result, interactions with the lower steering wheel rim and airbag can occurred. The
contribution of the vertical acceleration is in this case of loading not negligible.

Figure 3:    Comparison of the maximum resultant acceleration along the spine

                              FID project Final Publishable Report

The results showed that the restraint systems and the test conditions modify the behaviour of the
thorax. The restrained conditions are not reflected in the same way for the various body segments.
From the tests performed it was suggested that the chest acceleration (T8) could be a better predictor
than the T1 resultant acceleration in a combined parameter injury criterion.

Biomechanical research concerning the behaviour of the shoulder during frontal impact conditions is
rare, and the data collected during the FID project would contributed to reduce the lack of
biomechanical data which can serve as the basis from which biofidelity requirements for the shoulder
can be defined. The behaviour of the shoulder was studied through the acceleration of the acromion
and the humerus (upper and lower part).

Figure 4:    Left shoulder resultant acceleration at two level of severity
The bimodal behaviour, already showed for the torso behaviour, is visible for the shoulder during the
airbag tests. The resultants increase rapidly up to the activation of the load limiter. After that, the
slope changes and the increasing of the resultant is lower up to the load by the airbag (Figure 4). It is
the same for the humerus. For the test without airbag, the first part of the curves has the same
behaviour but the resultant remains constant during the impact. It could be interesting to evaluate the
contribution of each component, linear and angular, of the humerus resultant.

The response corridors of behaviour obtained from this data were used to evaluate the biofidelity of
both dummies (Vezin et al., 2002) in order to propose improvement of the THOR-Alpha.

The following publications have been based on the above mentioned scientific results:
[1] Vezin, P. Bruyere, K., Bermond, F., IRCOBI-2002

[2]   Vezin, P., Bruyere, K., Bermond, F. and Verriest, J.P., STAPP-2002

[3]   Vezin, P., Verriest, J.P., IRCOBI-2003

[13] Vezin P., Verriest J.P., International Crashworthiness and Design Symposium 2003

[14] Bermond F., Vezin P., Bruyere-Garnier K., Verriest J.P., Congrès de la Société de
     Biomécanique 2003, Archives of Physiology and Biochemistry

4.2.2 Activity 2.2: Impacts to the human pelvis/femur/knee
The objective of this task was to obtain additional data on the behaviour of the human pelvis / femur /
knee in conditions representative for loading of the pelvis/femur/knee complex during a frontal
impact. Impact experiments were conducted on 6 pairs of human knee joints with a simple rigid
pendulum consisted of a horizontal cylinder of radius 38 mm with a mass of 12 kg. The cadaver

                                                  FID project Final Publishable Report

specimens were mainly from an aged population. The samples mean weight was 62.5 kg ± 10.25 kg.
Since there was substantial weight variation of the subjects, the responses were scaled (Eppinger,
1984). The sample mean height was 167.5 cm ± 3.44 cm. Prior to testing, X-rays radiographs of the
body were taken in 2 planes (sagittal plane and frontal plane). After testing, the standard lower limb
radiographs were taken again and compared with the initial ones. In addition to the clinical
examination, each specimen was dissected.

The experiment consisted of impacting the knee of a subject in seated position with no back support.
The head was held by a cable that allowed a quasi-vertical position of the thorax. The arms were
folded on the abdomen. All impacts were performed by loading the femur axially at the distal end
through the patella. The tibia-femur angle was 90° to have the patella in front of the femur.

Biofidelity tests were performed at sub-injury levels at two impact velocities: 2.8m/s and 4m/s. One
impact was delivered to the knee from a given subject and the contra-lateral knee was impacted at the
second velocity to limit the test variability. The pendulum impact acceleration was measured by an
accelerometer located inside the rear face. Impact force was calculated by multiplying the impactor
mass by the measured acceleration. Tri axial accelerometers were fixed on both iliac crests and on the
femur of the impacted leg. All the accelerations measured on the PMHS were collected and processed
according to SAE J211 for Class 1000. Three high-speed cameras (1000 frames/s), 2 for general and
close-up lateral views and one for overhead perspective) were used to allow a cinematic analysis.

As expected, no injuries were produced at the knee, femur or pelvis as determined from the x-ray and
dissection. The results of the current tests were well correlated with previous data on the basis of peak
knee impact force in pendulum tests in similar experimental set-ups. The movement of each PMHS
closely matched and clearly showed backward and abduction movements of the impacted lower limb.
The mean knee flexion angle after impact was 18° at 2.8 m/s and 26° at 4 m/s. The mean abduction
rotation angle of the thigh angle was 14.9° at 2.8 m/s and 17° at 4 m/s. In both cases, there was very
little foot movement

Corridors were constructed for the full time history about the mean response curve and based on the
envelope of all experimental curves. Requirements for the impact force, the femur and iliac crest
accelerations at both impact velocities were defined, based on these obtained data (Figure 5). They
can be used for the development and the assessment of the biofidelity of the lower limb of
anthropometric dummy.

                      0,5                                                                                                     120

                        0                                                                                                     110
                             0          10              20                        t (ms)   30                                 100
  Impact force (kN)

                                                                                                     Femur Acceleration (g)

                       -2                                                                                                     60
                                                             Impact velocity = 2.8 m/s
                      -2,5                                                                                                    50

                       -3                                                                                                     40

                      -3,5                                                                                                    30                       Impact velocity = 2.8 m/s

                                                                                                                                                     t (ms)
                       -5                                                                                                           0          10                        20             30

                                                                                   -1                                                                                              -1
                        Corridor for the knee impact force at 2.8 ms                                                           Corridor for the femur acceleration at 2.8 ms

                                                           FID project Final Publishable Report

                          1                                                                                                      300
                                                                                     t (ms)
                               0         5     10     15      20                25            30                                 250

                          -1                                                                                                     225

                                                                                                        Femur Acceleration (g)
      Impact force (kN)


                          -3                                                                                                     150

                                                                                                                                                                           Impact velocity = 4 m/s
                                                                   Impact velocity = 4m/s                                        100

                          -6                                                                                                      25

                                                                                                                                       0              10              20          t (ms)             30

                               Corridor for the knee impact force at 4 ms-1                                                            Corridor for the femur acceleration at 4 ms-1
Figure 5:                                Biofidelity requirements for the femur-knee complex

The following publications have been based on the above mentioned scientific results:

[4]                                Masson, C., Cavallero C., Vinel, H., Brunet, C. Archives of Physiology and Biochemistry, 2002

[5]                                Masson C., Vinel, H., Portier, F., Ghannouchi S., Brunet, C., Congrès de l’Association des
                                   Morphologistes. Sousse-Tunisie, 2002 (Presentation)

[6]                                Masson C., Cavallero C., Vinel H., Brunet C., Congrès de la Société de Biomécanique, 2002
                                   (Presentation of [4])

[7]                                Masson C., Vinel H., Cavallero C., Brunet C., IRCOBI-2002

[8]                                C. Masson, C. Cavallero., IRCOBI-2003

[9]                                J. Cardot, C. Masson, C. Brunet., International Crashworthiness and Design symposium 2003

4.2.3 Activity 2.3 : Human lower leg/ankle/foot testing
Many accident studies have shown that lower leg injuries are common in frontal impact car accidents.
Of these, injuries to the ankle are the most serious and the most likely to lead to long-term morbidity.
The dummy currently used for regulatory and consumer testing, the Hybrid-III, is not appropriate for
determining the risk of injury to the ankle. Previous work by this group developed an understanding
of the loading regime necessary to induce the sort of foot and ankle fractures seen in frontal impacts.
These tests involved localised impacts to the plantar surface of the foot, parallel to the line of the tibia
which was fixed at the knee. However, when these impacts were reproduced with the Hybrid-III leg,
unrealistically high forces were obtained.

A new double-impact sled rig was developed to load the leg in a more realistic, car-equivalent
manner. The sled separately reproduces the vehicle deceleration and footwell intrusion typically seen
in frontal impacts. For this programme a sled impact velocity of 14 m/s and a sled deceleration of 20 g
were used. Footwell decelerations of up to 160 g were used and up to 200 mm of footwell
displacement was permitted. The vehicle deceleration phase was reproduced as it would pre-load the
ankle complex and it was hypothesised that this would be important in generating injuries
representative of those seen in the field. In addition, a constant-force knee restraint and Achilles
tendon tension were applied such as to reproduce the emergency braking force found in simulator

Fourteen PMHS specimens were impacted at a range of foot loading severities with peak axial forces
under the foot of between 4.0 and 11.7 kN. These resulted in seven non-injured and seven injured

                                                 FID project Final Publishable Report

specimens. Only one impact was applied to each specimen, which was then examined for injury by
x-ray (Figure 6, Left) and necropsy. Detailed descriptions of the fractures together with the
mechanism for each were given. All of the injuries generated were considered to be representative of
those seen in car accidents. Tests at each severity were then replicated with a THOR-Lx lower leg. An
injury risk curve derived from the combination of the THOR-Lx results and the injury found in the
equivalent PMHS test is shown in Figure 6 (Right).

                                                        Probit Probability of Foot or Ankle Injury







                                                                                                    Probability of ankle injury
                                        0.2                                                         Upper 95% confidence limit
                                                                                                    Lower 95% confidence limit

                                           0.0    1.0   2.0   3.0      4.0      5.0     6.0      7.0         8.0          9.0     10.0
                                                              THOR-Lx Lower Tibia Axial Force Fz (kN)

Figure 6:    Results of PMHS and THOR-Lx lower leg tests
             Left: Talar neck and intra-articular calcaneus fractures
             Middle: Probability of disabling foot or ankle injury for the THOR-Lx lower tibia Fx
             Right: THOR-Lx leg test
The following publication has been based on the above mentioned scientific results:

[10] Hynd D, Willis C, Roberts A, Lowne R, Hopcroft R, Manning P and Wallace W, ESV
     conference 2003

The University of Heidelberg has performed foot-well intrusion tests. Frontal collisions with three
belted PMHS with simulation of the foot-well intrusion were reported. The impact velocity was 50
km/h., the average sled deceleration amounted to 15 g., and the foot-well intrusion was
simultaneously translation (135 mm) and rotational (30 degrees). Figure 7 shows oblique and front
view of the intrusion device. The age of the cadavers was 37, 45 and 62 years. Foot-well forces and
acceleration were obtained at the intrusion device, acceleration, forces and moments at the tibia and
acceleration at the pelvis were also obtained. For the 37 years old PMHS (break pedal simulation)
following injuries were observed for the left leg: Fracture of the malleolus lateralis close to the middle
of the process of the malleolus. Cartilage contusion of the talus with a bruise sized 5mm x10 mm. For
the right leg following injuries were found: Fracture of the malleolus medialis and lateralis at the level
of the tibia joint surface. Laceration of the spring ligament at the navicular and calcaneal Insertion.
Cartilage contusion at the front joint surface edge of the tibia joint surface sized 3mm x 40mm with
superficial spongy bone contusion. Superficial cartilage shear off of the talus, medial-distal at the near
side part of the tibia sized 10mm x 10mm. The 45 year old PMHS sustained no injuries. At the 62
year old subject the following injuries were observed: fracture of the medial malleolus left, fracture of
the vertebral cartilage edge of the tibia right, shear off the cartilage of the plantar side of the talus at
the front of the talocalcaneal joint with haemorrhage in the joint. Edge fracture of the dorsal, lateral
process of the talus at the talocalcaneal joint in the bony region, medial cartilage laceration caused by
compression. The injury mechanism was stated to be mainly dorsiflexion.

                              FID project Final Publishable Report

Figure 7:    PMHS foot-well intrusion test device (left: oblique and right front view)

The following publication has been based on the above mentioned scientific results:

[11] Kallieris D., Riedl H., Mattern R., ESV conference 2001

4.3 WP3: Definition Requirements for Frontal Dummies
The general objective of the FID project is to contribute to the reduction of the amount of injured and
death car occupants involved in frontal collisions. In WP 3 a set of requirements for the development
and evaluation of frontal impact dummies was defined. The requirements include: biofidelity
(humanlike impact response), anthropometry (human representation with respect to dimensions, mass,
mass distribution and joint range of motion), durability, instrumentation, repeatability, sensitivity and

Within the FID project a frontal impact crash test dummy (THOR-FT) (instrumented, 50th %ile
male), with realistic movements and injury indicating measurements, durability, repeatability will be
developed, together with a set of requirements for frontal impact dummies which correlate the dummy
response with injuries sustained by the human body.

An overview of the literature on which the biofidelity requirements for a frontal impact dummy are
based and the defined set of requirements have been reported. Biofidelity requirements have been
defined for the face, head, neck, shoulder, spine, thorax, abdomen, femur and lower leg. The
biofidelity document has been approved by EEVC/ WG12, presented at the ESV Conference in 2003,
and will be discussed further with IHRA.

The following publication has been based on the above mentioned scientific results:

[12] B. van Don, M. van Ratingen, F. Bermond, C. Masson, P.Vezin, D. Hynd, C. Owen, L.
     Martinez, S. Knack, R. Schaefer; on behalf of EEVC WG12, ESV conference 2003

An overview of injury criteria and tolerance levels defined for a frontal impact dummy has been
reported. The injury criteria and tolerance levels have been defined for the head, face, neck, thorax,
abdomen, femur/knee and the lower leg. Requirements for durability repeatability and sensitivity have
also been defined for a frontal impact dummy as well ad the anthropometric requirements. The
anthropometric requirements are based on the literature review and the discussion, which has taken
place within WorldSID and IHRA. The consequences of choosing a 50th %ile male dummy in a
future frontal Directive has been discussed. Also the consequences of the age group used for
definition of the anthropometry of the THOR-FT dummy has been outlined.

                              FID project Final Publishable Report

4.4 WP4: Performance of Existing Frontal Dummies
The scientific and technical description of the results for activity 4.1-4.8 will be discussed by test
4.4.1 BASt
BASt undertook sled tests to evaluate the sensitivity and repeatability of the THOR-Alpha and
Hybrid-III dummies. Tests were undertaken at 30 km/h and 50 km/h and a typical deceleration of 15
to 20 g and 20 to 25 g respectively. The seat used had the same geometrical characteristics as the seat
used in the FID PMHS tests at INRETS and the dummies were restrained with a shoulder and separate
lap belt. Both dummies were found to be repeatable, but the THOR-Alpha had a better registration of
chest-belt interaction than the Hybrid-III and a qualitatively better kinematic response. Many
durability issues were found with the THOR-Alpha dummy that limited the ability to assess some
aspects of the performance of the dummy.

BASt also undertook head biofidelity tests as specified in WP3. Pendulum impact to the forehead and
chin of isolated dummy heads were performed [Ward, 1985] along with pendulum impacts to the
forehead of a whole, seated dummy [Melvin, 1985]. It was found that the repeatability of both
dummies was good, but that the biofidelity of both the THOR-Alpha and Hybrid-III heads was poor.

4.4.2 INRETS
INRETS undertook sled tests, face biofidelity tests and pelvis-femur-knee biofidelity tests. Two series
of six identical sled tests were undertaken with each of the THOR-Alpha and Hybrid-III dummies.
These reproduced the PMHS tests undertaken by INRETS in WP2. The first series of tests was
performed at 30, with a peak deceleration of 13 g, and with the dummy restrained by a lap and
shoulder belt with a 4 kN force limiter. The second series was performed at 50, with a peak
deceleration of 22 g, and with an airbag in addition to the seat-belt. The standard dummy

Figure 8:    Tests at BASt
             Left: Sensitivity and repeatability sled tests
             Middle: Head biofidelity component test on chin
             Right: Head impact full body test on forehead

instrumentation was used, with the addition of the same thorax and arm accelerometer array used in
the PMHS tests. The THOR-Alpha responses were, for the majority of parameters, closer to those of
the PMHS than the Hybrid-III. There was a more biofidelic response of the chest and spine for the
THOR-Alpha that was considered may change injury assessment for the inner organs.

The THOR-Alpha face was evaluated in the four test conditions defined in WP3 [Nyquist et al., 1986;
Melvin and Shee, 1988; ADRIA, 1998]. The tests were conducted using a horizontally guided linear
impactor with either a bar or flat disk at the front. The impacts were to the nose (frontal), whole face
(frontal), frontal bone (oblique) and zygoma (oblique). The impactor force-time response in the

                             FID project Final Publishable Report

frontal impacts was slightly greater the requirement and the initial rise time of the force was much too
long. In the oblique tests, the head acceleration was considerably in excess of the requirement,
particularly at the zygoma impact condition. This was thought to be due to compression of the edge of
the face foam allowing the impactor to load directly the metal parts of the face construction. The face
did not meet any of the biofidelity requirements.

Pelvis-femur-knee biofidelity tests were undertaken with the THOR-Alpha and Hybrid-III dummies in
comparison with biofidelity data generated in WP2. A 12kg impactor was used to load the knee of a
seated dummy at 2.8 and 4.0 m/s. It was found that the Hybrid-III and THOR-Alpha dummies
exhibited a higher mean peak force than the PMHS by a factor or 3 and 2.7 respectively. The results
indicated that dummy knee padding should be modified. The femur, struck-side iliac crest and sacrum
accelerations were within the PMHS corridors.

Figure 9:    INRETS tests -
             Left: Hybrid III; Middle: THOR-Alpha; Right: Pelvis-Femur Knee test

The following publications have been based on the above mentioned scientific results:

[2]   Vezin, P., Bruyere, K., Bermond, F. and Verriest, J.P., STAPP-2002

[13] Vezin P., Verriest J.P., International Crashworthiness and Design Symposium 2003

[14] Bermond F., Vezin P., Bruyere-Garnier K., Verriest J.P., Congrès de la Société de
     Biomécanique 2003

4.4.3 University of Heidelberg
The University of Heidelberg assessed the Hybrid-III and THOR-Alpha dummies in sled tests that
replicated PMHS tests undertaken in WP2. These tests were directed towards establishing kinematic
biofidelity requirements for the shoulder and lower leg. Twelve tests were performed at 50 km/h and
with an average sled deceleration of 15 g. The dummies were restrained with a standard three-point
seat-belt and footwell intrusion was simulated – the footwell was able to translate 135 mm and to
rotate by 30°. Tests were undertaken with a flat footwell, with a simulated brake pedal and with
pronation and supination of the left and right feet. The kinematics of the THOR-Alpha dummy was
much closer to those of the PMHS than was the Hybrid-III dummy and both dummies showed higher
footwell forces. The repeatability of both dummies at a given test condition was good, but the THOR-
Alpha was much less robust than the Hybrid-III and sustained serious damage to both legs in tests
with a simulated brake pedal.

                             FID project Final Publishable Report

Figure 10:   University of Heidelberg footwell intrusion sled tests - Left: Hybrid III; Right THOR-Alpha

4.4.4 Polytechnical University of Madrid
INSIA undertook a series of tests to evaluate the repeatability and biofidelity of the thorax, abdomen
and femur of the THOR-Alpha dummy. Thorax biofidelity [Kroell et al., 1971], upper abdomen
biofidelity [Nusholtz et al., 1994], lower abdomen biofidelity [Cavanaugh et al., 1986; Miller, 1989]
and femur [Haut and Atkinson, 1995] tests were undertaken as defined in WP3, along with six sled
tests at 40 and 50 km/h, to evaluate the global performance of the dummy. The aspects assessed were
biofidelity, repeatability, durability and handling.

The thorax biofidelity tests included Part 572 guided impactor tests at 4.3 and 6.7 m/s. Dummy
repeatability was found to be low at low thorax deflection, but this may be partly attributed to
mechanical failures of the dummy and repeated disassembly and assembly for repairs. In tests at 4.3
m/s, at low deflections the thorax deflection was less than the requirements, but was good at high
deflection. At 6.7 m/s, the dummy response was good at low and high deflections, but was too soft in
the middle of the deflection range.

In the upper abdomen biofidelity tests, the upper abdomen was struck horizontally at 8.0 m/s by an
18kg pendulum impactor with a simulated steering wheel section attached to the front. In addition to
the dummy measurements, the external displacement of the posterior abdomen was measured with a
string potentiometer. The dummy abdomen compression did not meet the requirements defined by
NHTSA. However, the abdomen compression measured using external marker tracking met the
PMHS corridor up to 80 mm of compression, but with a large increase in stiffness between 80 and
100 mm of compression. Biofidelity was therefore found to be good up to 80 mm of compression, but
could be improved at higher compression by redesigning the lower part of the abdomen rear part to
allow more compression. Dummy failure prevented a full assessment of repeatability in these tests.

In the Cavanaugh [1986] lower abdomen biofidelity tests, the abdomen is struck at 6.1 m/s by a
25 mm diameter bar attached to the front of a 32 kg pendulum impactor. In addition to the dummy
measurements, the external displacement of the posterior abdomen was measured with a string
potentiometer. Only one of the four tests met the NHTSA requirement, but all met the Cavanaugh
corridor up to about 80 mm of compression. Again, the abdomen was very much too stiff above this
level of compression. An additional six tests were undertaken to check the response of the abdomen at
different impact heights. Peak measured abdomen compression ranged from 45 mm at an impact point
of 30 mm above the abdomen instrumentation mounting point to 75 mm at a point 20 mm below the
instrumentation. It was also found that the abdomen instrumentation did not recover fully between test
and over the series of ten tests the initial length was reduced by up to 33 %.

In the Miller [1989] lower abdomen tests, the abdomen of a supine dummy was loaded in a controlled
manner via seat-belt webbing. Loading rates were varied from 1.1 to 3.2 m.s-1. It was found that the

                             FID project Final Publishable Report

dummy was very insensitive to the rate of loading, in contrast to the biomechanical data, and the
abdomen response was too stiff.

An isolated femur, knee and lower leg (at 90° to the femur) was rigidly attached to a mounting frame.
The knee was impacted with a 4.5 kg pendulum impactor at velocities between 1.15 and 6.25 m.s-1.
Knee and femur biofidelity was found to be very good.

The sled tests showed that the dummy’s internal tilt sensors were valuable in placing the dummy
accurately in each test. It was found that the head accelerometer signals were very noisy, possibly due
to low rigidity of the mounting block. However, in these and the biofidelity tests, dummy durability
was unacceptably poor and prevented a proper assessment of the repeatability of the dummy.

Figure 11:   UPM-INSIA sled tests with THOR-Alpha
             Left: Before test dummy position
             Middle: Post-test dummy position
             Right: Upper thorax guided impactor test (ref. Kroell)

The following publication has been based on the above mentioned scientific results:

[15] Martinez L, Ferichola G, Guerra LJ, van Ratingen M, Hynd D., IRCOBI, 2003.

4.4.5 TNO
TNO evaluated the neck of the Hybrid-III and THOR-Alpha dummies in sled tests based on data
obtained at the Naval BioDynamics Laboratory (NBDL) in the US [Ewing and Thomas, 1973] (see
Figure , left an middle). These tests specify a sled pulse and T1 (lower neck) position and acceleration
as test condition requirements. If these are met, the requirements are for head accelerations and
rotations. The dummy was placed on a rigid seat and is restrained by a five-point harness. Adjustment
of the T1 position and acceleration was by loosening or tightening the seat-belts, but these parameters
could only be determined following a test when considerable post-processing had been completed.
Tests were undertaken in frontal and oblique impact conditions. At the oblique impact condition, the
Hybrid-III met the T1 input requirements, but the THOR-Alpha did not. The biofidelity of the THOR-
Alpha neck could not, therefore, be determined, but this was a test specification issue not a dummy
issue. The Hybrid-III dummy neck almost met the biofidelity requirements in the oblique tests. In the
frontal test condition, the THOR-Alpha tests again did not meet the input requirements. The Hybrid-
III tests did, but the dummy did not meet the biofidelity requirements. Durability problems with the
THOR-Alpha neck were found in separate neck flexion pendulum tests (Figure 12, right).

As a result of the difficulties experienced in trying to meet the input requirements to the biofidelity
tests, TNO redefined the tests, still based on the NBDL data, in terms of the global and local dummy
response. New biofidelity requirements were also derived.

                              FID project Final Publishable Report

Figure 12:   Neck tests at TNO (sled and pendulum tests)
             Left: Hybrid III in NBDL rigid seat oblique condition (note five point harness)
             Middle: THOR-Alpha in same condition (note optical ball markers for 3D motion analysis)
             Right: Neck damage in pendulum tests with Hybrid III certification pulse

4.4.6 TRL
TRL undertook out-of-position (OOP) airbag tests, seat-belt interaction and submarining sled tests
and an evaluation of the load sensing face of the THOR-Alpha dummy. In the OOP tests, two series
of ten tests were undertaken. In the first, the dummy was placed at a set distance 10 to 200 mm from
an airbag module. Dummy head, neck and thorax instrumentation were found to be sensitive to the
proximity of the airbag. In the second series, dynamic sled tests were undertaken with the dummy
seated in a vehicle buck and restrained by a lap belt and airbag. The time of deployment of the airbag
was varied to create various levels of out-of-position at deployment. None of the dummy
instrumentation was found to be sensitive to changes in this parameter. Durability problems were
experience with the neck and lumbar flex joint, along with many problems with smaller dummy parts.
Reliability problems with the chest compression instrumentation (Crux units) were found, but it was
very difficult to detect these without completely dismantling the thorax of the dummy.

Problems with retention of the shoulder belt in oblique impacts had been identified in a previous test
programme with an earlier prototype of the THOR dummy. When the seat-belt slipped off the
shoulder, the belt became trapped in the arm-shoulder joint and the dummy was artificially restrained
in a way that would not happen with a human occupant. It was found that design changes to the
THOR-Alpha had not changed shoulder belt retention in oblique impacts, but had considerably
reduced the tendency for the belt to become trapped in the shoulder joint (Figure 13, Left and middle).
The ability of a new transducer to detect submarining, where the lap belt rides up over the pelvis bone
and loads the abdomen, was evaluated in a small series of sled tests. Submarining will greatly increase
abdomen loading and could be highly injurious, but it is not easy to tell if it has occurred in a full-
scale vehicle crash test. The view for cameras is typically obscured by the bodywork of the vehicle
and it is not possible to tell whether increased abdomen loading is from the lap belt or from the
steering wheel or airbag. The new sensor measures loading to the left and right front of the pelvis. It
was found that the presence or absence of this pelvis loading correlated to submarining in well-
controlled sled tests. However, more work would be required in order to derive a regulatory

The THOR-Alpha face consists of five force plates, each supported centrally by a uniaxial load cell.
The face load cells were tested in quasi-static loading conditions. It was found that the accuracy of
force measurement was dependent on the position of application of the force on the force plate, being
worse further from the load cell. The error was up to 14 % at the edge of the mandible (chin) load cell,
which was considered to be unacceptable for a test device. The performance of the face was then
evaluated in isolated head drop tests on to a rigid bar and on to a range of steering wheels of known
aggressivity. It was possible to determine the distribution of face loading, which suggests that the face
could be used with localised injury criteria if the accuracy of the force measurement was improved. It

                              FID project Final Publishable Report

was also found that the face load measurements (along with head accelerations) could be used to rank
correctly the aggressivity of the steering wheels (Figure 13, Right).

Figure 13:   TRL sled en component tests with THOR-Alpha
             Left: THOR-Alpha seat belt interaction sensitivity test
             Middle: Still from high-speed film of the seat-belt interaction tests
             Right: THOR-Alpha head to steering wheel impact test
The following publication has been based on the above mentioned (paragraph 4.4.1 to 4.4.6) scientific

[16] B. van Don, M. van Ratingen, F. Bermond, C. Masson, P. Vezin, D. Hynd, D. Kallieris and L.
     Martinez, Evaluation of the Performance of the THOR-Alpha Dummy

4.4.7 Activity 4.9: Anthropometry
The objective of this activity was to measure the anthropometry of the Hybrid-III and the THOR-
Alpha dummy. This was performed by UPM. With the use of a Faro arm the external surface of the
Hybrid-III and THOR-Alpha was obtained. During the measurement process a local reference system
(location and orientation) was defined using clear and well identifiable points along the dummy parts.
The articulation points of the dummy parts were obtained, to allow the assembly of the individual
dummy parts together using CAD software. The measured points of the surface were processed to
obtain 3D surfaces of the dummy. The CAD model developed has been assembled jointly with the
model of the 50% percentile male generated by Robbins known as the UMTRI model. This model is
the results of the research project “Development of Anthropometrically based Design Specification
for an Advanced Dummy”, and has also been used to develop the WorldSID dummy. Comparison of
the three models (Robbins, Hybrid-III and THOR-Alpha, see Figure 14) allows the evaluation in
differences between the dummies, with special attention to the dummy part lengths and articulation
positions. The difference in volume of all the three models has also been developed.

                             FID project Final Publishable Report

Figure 14:   Comparison of anthropometry: Hybrid III (left) and THOR-Alpha (right) with UMTRI
             data (green left and light blue right)
The following publication will be been based on the above mentioned scientific results:

[17] Martinez L, Ferichola G, Guerra L.J. “Anthropometry study of the THOR and HYBRID-III
     frontal impact dummies”. FISITA World Automotive Congress, Barcelona, Spain. (2004)

4.5   WP5: Modification and Testing of Dummy Enhancements

4.5.1 Activity 5.1: Update of the THOR multi-body Model.
The activities for update of the THOR multi-body model were closely linked with the activities in
WP4. Tests performed in WP4 and measurements taken formed the principal basis for the data
necessary to update and validate the multi-body model. The initial model of the THOR developed
during the ADRIA project was made reflecting the responses of the THOR dummy prototype
available for the ADRIA Consortium. After the end of the ADRIA project, NHTSA/GESAC has
updated the dummy and a THOR-Alpha has been available for the FID Consortium. One of the major
changes is that the THOR-Alpha included advanced lower legs, the THOR-Lx legs. Information
concerning the tests performed during the evaluation period of the THOR-Lx leg was available and
collected to develop and validate the THOR-Lx model.

For the development of the THOR-Lx model a similar process as described in the ADRIA deliverable
report D15 was used. From the 2D CAD technical drawings available at the NHTSA web-site a 3D
CAD model was developed. See Figure 19. This model was used to obtain mass properties of each
model segment (centre of gravity and inertia tensors). The external geometry of the THOR-Lx
described in the CAD model with smooth surfaces is meshed to facet surface included in the model.
See Figure 15 (middle). The special components like the Achilles tendon has been incorporated in the
model. The joint properties of the ankle with regards to the limitation of the motions are implemented.
See Figure 15 (right). The response of the THOR lower leg, ankle and foot assemblies has been
extensively evaluated using the verification and certification procedures developed for THOR-Alpha
by NHTSA and GESAC [18] and [19]. Results obtained in dynamic certification tests and data
registered in TRL's sled tests have been used to validate the THOR-Lx model in order to obtain a
similar response in the model than that in the real dynamic tests. Figure 16 (left) shows the heel and
foot impact tests simulation, these tests allowed the definition of the parameters for the tibia
compliant bushing, the contact function of the heel and ball region and the functions for the ankle
joint. The sled tests performed by TRL in WP4 have been used to validate the general response of the
lower leg model. See Figure 16 (middle).

                              FID project Final Publishable Report

The complete THOR model is prepared using the multi body facet technology. This approach result in
realistic external shape, this means an advance in the accuracy of the contacts calculated by the
model. New stress based contact functions are introduced instead of the penetration based contact
function used in an ellipsoid multi-body model. The performance of THOR-model torso is validated
against tests performed to different torso regions by INSIA in the WP4 (thorax, upper abdomen and
lower abdomen). Figure 16 (right) shows the simulation set-up for the upper abdomen certification
test. Finally the full dummy model has bee evaluated against the data registered in sled test performed
by INRETS, Figure 17 show images of one sled test simulated with the new THOR model. From this
model development it can be concluded that new model is large step forward with respect to the
model created in the ADRIA project. It includes accurate external shape with facet surfaces, the new
THOR Lx legs validated against dynamic and quasi-static tests, and thorax, upper and lower abdomen
validated against dynamic tests. Last hardware upgrades incorporated in the THOR-FT version
developed in the FID project are not included in this version of the model.

Figure 15:   THOR Lx leg
             Left: Hardware and CAD model (flesh parts removed)
             Mid: MADYMO facet-multi-body model
             Left: Foot model with the extreme motions, flexion, rotation and inversion-eversion

Figure 16:   THOR Lx leg and full dummy model validation simulations
             Left: Heel and foot impact
             Middle:   TRL foot well intrusion sled test
             Left: Upper abdomen steering wheel impactor certification test

                             FID project Final Publishable Report

Figure 17:   INRETS full dummy sled test simulation (images at 18, 58 and 98ms).

4.5.2 Activity 5.2 Design Briefs
In January 2002 the THOR-Alpha dummy became available for testing in the consortium. As a first
action in WP5 a design review was held with the delivered THOR-Alpha dummy. The objectives of
this meeting were:
• Review the design of the THOR-Alpha dummy
• Obtain detailed insight in the dummy’s features and construction
• Assess changes of the THOR-Alpha with respect to the recommendation made in the ADRIA
• Preparation of the preliminary design change recommendation list
• Establishment of the design team

During the design review the dummy was disassembled and assembled back to identify handling
issues, to see what changes had been made since the dummy was tested in the ADRIA project and to
inspect the dummy if it was acceptable for testing in the FID consortium. It was noted that a large gap
of 3 mm appears between the T12 load cell and the upper thorax interface, which prevented the parts
to be aligned properly and introduced being of the upper bracket. This was felt not to be a proper
engineering solution, or even a design or production fault. GESAC was asked to improve this, but did
take action on this recommendation. It was further stated that the provided THOR-Alpha dummy was
ready for testing. No serious deficiencies were noted during the design review meeting. Also a wish
list of the dummy manufacturer was discussed. The findings were discussed and prioritised and all the
design recommendation have been documented and reported.

The work done in FID WP 4 (Testing on the THOR-Alpha between January and October 2002)
showed satisfactory biomechanical responses and sensitivity, however further deficiencies of the
dummy were brought to light. Durability issues were found in the neck and lumbar spine which
continuously failed, flesh parts pelvis, shoulder, thighs which easily damaged compared to other
dummies, sternum material (bib) ripping, sternum mass continuously dropping off, rib damping
material de-bonding and tibia flexible element failures. Durability of zippers was found to be a
general problem on all positions in the dummy. The problems were so large that sufficient
repeatability testing could not be performed. Regarding the instrumentation also problems were
found. Head acceleration noise, problems with the CRUX's and DGSP’s (thorax and abdomen
deflection sensors) were found, face load cells appeared to be sensitive to shear loading. Also some
handling and usability issues were identified. All of these issues combined with the complexity of the
dummy to disassembly and put back together largely impaired the user friendliness and handling.
Summarising the THOR-Alpha showed satisfactory biomechanical responses and sensitivity, but
unsatisfactory durability, handling and usability. The objective of developing the THOR-FT (FID
Technology) was to have a dummy with better handling, durability, instrumentation, etc., but with the
same biofidelity and injury measuring capability of the THOR-Alpha. All test results of WP4 where
summarized in a meeting held October 2002. At that point the final list of problems to be solved was
agreed by the consortium partners, providing input to the design and build activity of task 5.3. The
final FID design changes list was also discussed with NHTSA and compared to a similar list they

                             FID project Final Publishable Report

presented based on experience in the USA with the THOR-Alpha dummy. Some issues were taken
onboard in the FT design changes list. It was decided to work on the durability and handling
improvements on a global dummy level to bring the dummy closer to crash test application rather than
concentrate on the improvement of two or three body regions.

Figure 18:   THOR-Alpha in review workshop at INRETS January 2002
             Left   Canvas upper and lower abdomen bags and CRUX
             Middle Pelvis flesh foam (perpendicular seating posture)
             Right Complete dummy THOR- Alpha dummy

4.5.3 Activity 5.3 Prototype Design and Build
The design changes list agreed upon (see activity 5.2) served as guide line for the development of the
THOR-FT (FID Technology) dummy prototype. The complete dummy was modelled in 3D CAD
software from the 2D drawings of the THOR-Alpha. See for the model Figure 19.

Figure 19:   THOR-FT dummy CAD-model

This allowed accurate assessment of the dummy anthropometry in the design stage. Comparison with
the requirements brought some anthropometry issues to light that had not been found previously and
needed correction. The main issues found were too short femurs (28 mm), deviation of the pelvis
outer shape, H-point deviation and clavicles protruding outside the outer surface model. New pelvis
and thigh flesh parts and clavicles were designed to correct the deviations. The dummy was changed
to integrate more reliable IR-TRACC deflection sensors (Figure 20, left), replacing the CRUX's in the

                                  FID project Final Publishable Report

thorax and DGSP’s in the abdomen. The dummy was design with provisions to integrate on board
data acquisition systems up to 128 measuring channels (Figure 20, right). Many other issues were
solved, to improve handling, durability and user friendliness of the dummy. The THOR-FT is a fully
metric dummy, which further improves user friendliness. During the design stage two progress-
meetings and one teleconference were held with the dummy manufacturer to monitor the development
progress and to discuss design issues. The deviations in the pelvis and thigh anthropometry were
discovered late in the design stage (Figure 20, middle). The correction required the development and
manufacturing of three completely new and complex tools for the flesh parts. The prototype dummy
was delivered to the consortium for testing early August 2003.

CAD-model with Thorax IR-TRACC's          Anthopometry comparison (issues        Data acquisition unit locations two at
at 7 locations (4 thorax and 3 abdomen)   upper leg length and pelvis shape)     lumbar spine two in thigh flesh foam

IR-TRACC: Infrared sensor in telescopic New pelvis and lower abdomen foam        Thigh flesh foam part with replacement
housing with two hinge potentiometers   parts (pelvis: new shape an high mass)   data acquisition unit insert (RHS)

Figure 20:    THOR-FT Design and built highlights

4.5.4 Activity 5.4 Prototype Testing/Validation
The objective of the subtask 5.4 is to examine the THOR-FT prototype behaviour with respect to the
requirements set for THOR in general and more specify for THOR-FT. After delivery of the THOR-
FT prototype dummy TNO, INRETS and TRL have the task to perform dummy validation tests. TNO
with the objective to examine the dummy design, handling and certification procedures in order to
check the design improvements and to contribute to a comprehensive user manual. INRETS with the
objective to redo the tests performed with PMHS's and THOR-Alpha in a car environment with and
without airbags for comparison. TRL with the object to examine several kinds of sensitivity issues.

In preparation of the test session after delivery of the dummy at TNO, the certification procedures
specified for THOR-Alpha were reviewed. This extensive review was reported on June 20, 2003 in a
document sent to NHTSA with the request to comment upon. The report criticised some design details
of THOR and proposes a reduction and rationalisation of the certification effort required for the
dummy. The proposals comprise a reduction from 13 full body and 12 component tests to 5 full body
and 16 component tests.

The late delivery of the dummy (Figure 21), without the new pelvis and thigh flesh foam parts, in the
first week of August 2003 did not allow much time for validation testing at TNO, INRETS en TRL.

                             FID project Final Publishable Report

The second week of August 2003 was used to summaries the states of the delivered hardware in a
design review. The dummy supplier was not able to show compliance with the performance
requirements. Almost all body parts showed the need of small to considerable performance tuning.
The lag of performance compliance reduces the value of the tests performed with the dummy. It was
decide at TNO to proceed with an adopted test program to explore the certification procedure
alternatives and contribute to outstanding dummy performance-tuning issues. The test work
performed in the second and third week of august focused on the head and face impact certification
tests, the neck bending certification test and the upper thorax certification tests. Alternative
certification test-set-ups were explored to come-up with recommendations for a rationalised and
reduced certification effort. The full body tests with impact on the thorax were performed with the old
THOR-Alpha pelvis flesh foam part and skin foam lead granulate bandage around the femurs. Big
problems with the IR-TRACC (3 dimensional Infrared displacement transducers) were identified.
Seven of these instruments are installed in the dummy upper and lower thorax and upper and lower
abdomen. To overcome these problems the complete calibration, data acquisition and processing
procedure for the IR-TRACC's was evaluated. The results of the certification procedure review, the
alternative certification procedure exploration tests and the IR-TRACC procedure evaluation is
documented in a FID report.

Figure 21:   THOR-FT dummy as delivered August 2003 (pelvis and thigh flesh foam parts missing)

Figure 22:   THOR-FT testing at TNO August 2003 - from left to right:
             head impact with 23.4 kg pendulum and face impact with 13 kg pendulum,
             neck bending on part 572 neck pendulum and upper thorax impact with 23.4 kg pendulum
The new pelvis and thigh flesh foam parts were delivered mid September 2003. At the end of
September 2003 the dummy was send to INRETS for testing. The same tests performed in WP2 with
PMHS and in WP4 with Hybrid III and THOR-Alpha dummy (see paragraph 4.2.1 “Activity 2.1:

                                                    FID project Final Publishable Report

Human Thorax/Shoulder behaviour during frontal impact" for details). Six tests with the new THOR-
FT were performed in one week (see Figure 23). Due to the short period allocated by the co-ordinator
to perform the tests, the study was focused on the channels that can allow the comparison with the
corridors defined in the WP2 for the Shoulder and the Thorax and the comparison with the same
channels mounted on the THOR-Alpha (WP4). These channels were manly tri-axis accelerometers
mounted on the spine (1st, 8th, 12th vertebra and sacrum), the sternum (upper and lower part), the 4th
and 6th ribs (left and right) and the shoulder including the arms (left and right). During these tests the
lumbar spine damaged, furthermore no major problems were encountered. The behaviour of the new
prototype of the dummy was correct. In terms of biofidelity, i.e. comparison with PMHS corridors,
the modification of the spine has improved the biofidelity (see Figure 24). On the other hand, the
modification on the sternum and on the shoulder did not improve the biofidelity, but the behaviour of
the new dummy was not too different from the THOR-Alpha behaviour (Figure 25).

Figure 23:                   INRETS sled tests with THOR-FT
                             Left: THOR-FT before impact
                             Middle: Test 50 km/h with airbag
                             Right: Test 30 km/h

                     0,00    50,00   100,00    150,00       200,00        250,00                                     0,00   50,00   100,00    150,00       200,00        250,00
                    40                                                             40,00                           40                                                         40,00
                                              PMHS 50km/h+AB-corridor                                                                        PMHS 30km/h-corridor
                                              THOR-FT 50km/h+AB-mean value                                                                   THOR-FT 30km/h-mean value
                    30                        THOR-alpha 50km/h+AB-mean            30,00                           30                                                         30,00
                                                                                                Acceleration (G)

                                                                                                                                             THOR-alpha 30km/h-mean value
 Acceleration (G)


                    20                                                             20,00                           20                                                         20,00

                    10                                                             10,00                           10                                                         10,00

                    0                                                              0,00                             0                                                         0,00
                         0    50      100        150          200            250                                        0    50      100        150          200            250

                                        Time (ms)                                                                                      Time (ms)

Figure 24:                   T12 resultant acceleration (12th vertebra on thoracic spine)
                             Left: Test 50 km/h with Airbag,
                             Right: Test 30 km/h

                                                        FID project Final Publishable Report

                        0,00    50,00     100,00     150,00        200,00         250,00                                  0,00       50,00   100,00     150,00      200,00        250,00
                       40                                                                40,00                          40                                                             40,00
                                                   PMHS 50km/h+AB-corridor                                                                            PMHS 30km/h-corridor
                                                   THOR-FT 50km/h+AB-mean value                                                                       THOR-FT 30km/h-mean value
                                                   THOR- alpha 50km/h+AB-mean value                                                                   THOR-alpha 30km/h-mean value
                       30                                                                30,00                          30                                                             30,00
                                                                                                                                                      PMHS 30km/h-mean value

                                                                                                     Acceleration (G)
                                                   PMHS 50km/h+AB-mean value
    Acceleration (G)

                       20                                                                20,00                          20                                                             20,00

                       10                                                                10,00                          10                                                             10,00

                       0                                                                 0,00                            0                                                             0,00
                            0    50        100         150           200           250                                       0        50      100        150          200            250

                                              Time (ms)                                                                                         Time (ms)

Figure 25:                      Upper Sternum resultant acceleration,
                                Left: Test 50 km/h with Airbag,
                                Right: Test 30 km/h

Mid October 2003 the dummy was send to TRL. This was too late to be effectively used in the
allocated test period. Problems with the THOR-FT dummy with the IR-TRACC sensors and the poor
compliance with certification requirement lead to the late availability of the dummy at TRL.
Alternatively the THOR-Alpha dummy was subjected to a reduced shoulder belt sensitivity test

5                       Management and Co-ordination Aspects

5.1 Consortium
This consortium brings together a unique co-operation of 6 experienced and multi-disciplinary
partners from 5 different EU countries This consortium brings together a unique co-operation of 6
experienced and multi-disciplinary partners from 5 different EU countries (UK, France, Germany,
Netherlands and Spain). Four of the partners are research institutes (TNO, BASt, INRETS and TRL) all
with a well-recognised experience in the field of passive vehicle safety. These partners have worked
closely together in the past 18 years in different Community funded projects on passive safety. These
projects have resulted in proposals for new vehicle safety regulations for instance in the field of
protection of car occupants in frontal and lateral collisions. Without the support of the European
Community these proposals would not have reached the current status. The other partners in this project
are universities with specific and complementary expertise in the field of passive safety needed to
complete the goals of this project. This project directly will contribute to the harmonisation of safety
regulations worldwide and in particular between Europe and the United States. In this project, a close co-
operation with the US-DOT/NHTSA will be established through EEVC and IHRA Biomechanics.

Table 4:                        Partner Information

Org. Name                         Count     Nr. Of           Business     Activity/Main                                          RTD Role in project
(abbr.)                           ry        Empl.            Mission/Area of Activity
TNO                               NL        4500             Research Organisation in                                            Project co-ordinator, co-ordinator of
                                                             field of biomechanics and                                           WP 3 and 5, definition of
                                                             dummies, manufacturer of                                            requirements,      dummy       design
                                                             crash test dummies                                                  concepts,     analysis    of     neck
                                                                                                                                 biomechanical data, test lab for
                                                                                                                                 component tests, sled test lab, injury

                             FID project Final Publishable Report

                                                            criteria research
BASt          GE      400      Research Organisation in     Co-ordinator of WP 1, supplier of
                               field of biomechanics and    accident data, test lab for component
                               dummies                      tests
INRETS        FR      411      Research Organisation in     Co-ordinator of WP 2, supplier of
                               field of biomechanics and    biomechanical data, analysis of
                               crash dummies                biomechanical data (shoulder, pelvis,
                                                            femur, knee, face), test lab for sled
                                                            and impactor tests, dummy design
                                                            concepts, injury criteria research
TRL           UK      500      Research Organisation in     Co-ordinator of WP 4, supplier of
                               field of biomechanics and    accident data and biomechanical data
                               dummies                      (leg), test lab for sled and impactor
                                                            tests, injury criteria research
UoH           GE      2000     University, Dept. of Legal   Supplier of biomechanical data and
                               Medicine                     dummy data
UoM           ES      7000     University,   Dept.     of   Sled and impactor tests, modelling
                               Mechanical Engineering       activities

5.2 Description of the Participants
Partner 1: TNO Crash-Safety Centre (TNO), Netherlands
The core business of the research organisation TNO is producing knowledge. The major activities are
the transfer of know-how and the application of advanced knowledge in products and processes. The
TNO Crash-Safety Centre is one of the largest crash-safety research centres in the world. Within the
TNO Crash-Safety Centre a unique combination of expertise in impact biomechanics, crash dummy
development, experimental crash facilities and computer simulation techniques are available. On a
regularly basis the Crash safety Centre advises the Dutch government experts with their regulatory
activities (vehicle safety regulations). TNO further contributes to the development of crash dummies
and mathematical dummies, accident reconstructions and human substitute tests. TNO acts as the
project co-ordinator and is co-ordinating WP3 and 5.

Partner 2: BASt
BASt is a technical and scientific institute responsible to the Federal Ministry of Transport, Building
and Housing (BMVBW). The scope of work is considerable, ranging from replying at short notice to
incoming enquiries to the co-ordination and carrying out of research projects over a period of several
years. A focal point of BASt's work results from the role it plays in the formulation of specifications
and standards applying to all fields in highway-related work.

Partner 3: INRETS
INRETS (The French national institute for transport and safety research) has a long experience in
conducting research in the field of impact biomechanics and crash protection. Two research units are
The LBA (Laboratory of Applied Biomechanics) is a common research unit between INRETS and
Medical School of the ‘Université de la Méditérrannée’ in Marseilles. This unit has performed several
studies in the field of leg injury assessment, especially for pedestrians. The involvement of the
medical school allows for the development multidisciplinary approaches combining medical and
engineering expertise.
The LBMC (Biomechanics and Impact Mechanics Laboratory) is developing activities in various
fields pertinent to achievement of Biomechanics project. The overall objective is to develop
simulation tools (dummies, numerical methods), to advise crash test procedures and to make
recommendations for standards regulations. The LBMC has 25 permanent staff members to which are
joined Ph.D. students, trainees and Foreign researchers. The team of researchers is multidisciplinary
(engineering, medicine, biomechanics, computer science, ergonomics). Apart from two crash test

                             FID project Final Publishable Report

facilities, the laboratory is equipped with large a data acquisition system, a family of anthropometric
dummies, computers and workstations, film and video cameras recorders completed by a film analysis
system. Researchers of the two units are involved in several activities such as EEVC, IRCOBI, etc.

Partner 4: Transport Research Laboratory (TRL)
TRL has been involved in the study of the biomechanics of humans in car impacts since the 1950’s. It
has been very much involved in the study of human tolerance to injury, principally by the use of
accident reconstruction using dummies or dummy body parts. It was one of the first organisations to
propose the expression of injury tolerance in terms of injury probability functions (1974 Stapp
Conference). In addition, TRL has been deeply involved in the design and development of crash
dummies; including the OPAT dummy and EUROSID, where TRL was responsible for the thorax and
shoulder design and also contributed to the pelvis design. TRL, in collaboration with TNO, developed
the protection criteria for the thorax of EUROSID. TRL has developed a novel means of assessing
facial injury and this now forms the basis of the assessment of the safety of steering wheels by the
international customers’ union. TRL was a partner in the EC 4th Framework Project ‘ADRIA’ and
acted as co-ordinator for one of the tasks.

Partner 5: University of Heidelberg (UoH)
The Institute for Legal Medicine of the University of Heidelberg is equipped with a deceleration sled
to simulate crash tests. The device was completed to simulate footwell intrusion for the ADRIA
project. With the existing data acquisition system 32 channels can be measured and evaluated; this
system can be extended to 48 channels. Two high-speed cameras are also available.

Partner 6: Polytechnical University of Madrid (UoM)
The INSIA (Instituto Universitario de Investigación del Automóvil) of the Polytechnical University of
Madrid is a research institute in the field of road vehicles and provides R&D services and training
covering road vehicles and related systems. The main fields of experience of the Institute are:
• Multibody Simulation
• Vehicle Dynamics
• Structural Calculation
• Vehicles’ Passive Safety
• Accident research and reconstruction
• Component Vehicle Development
INSIA also includes a Testing and Homologation Department for checking compliance of national
and international standards in vehicle safety. It also includes a material and vehicle component testing

Figure 26:   THOR-FT with project team members after final project meeting (October 7, 2003)

                             FID project Final Publishable Report

5.3 Website
The Project Co-ordinator also developed a FID-logo (see on page header) and report cover and
developed and maintained a web site for FID (see Figure 10). The web site has been set up and is
hosted by TNO Automotive and linked to the European Passive Safety Network web site. The address
is Part of the site is open for the general public, the other part is solely
accessible for partners of the FID project.

Figure 10:   FID project report cover and web site

5.4 Harmonisation issues
The FID consortium has underlined the importance of harmonisation of safety regulations world-wide
and in particular between Europe and the United States. From the start, a close co-operation was
sought between US-DOT/NHTSA, in particular through EEVC, IHRA (=International Harmonised
Research Agenda) Biomechanics working group and industry forums.

Frequent and ongoing dissemination of research findings to the EEVC WG12 on Adult Crash
Dummies has been organised. In particular, the biofidelity response requirements and dummy updates
have been presented and agreed upon by the experts in EEVC. Likewise, various presentations
regarding FID have been given at ISO working groups TC22/SC12/WG5 (ATD) and WG6
(Biomechanics), related to dummy design, requirements and injury criteria for the lower extremity.
Further communication with regulatory bodies and industry was enhanced though various
presentations in the US (SAE government-industry meeting and 2002/2003 STAPP conference),
Europe (IRCOBI and Passive Safety Network conference 2003) and Korea (KATRI crash seminars in
2001 and 2003).

A presentation concerning the FID project has been given by J. Wismans at the meeting of the IHRA
Biomechanics Working Group held in May in Washington DC. M. van Ratingen has followed up this
presentation with others (e.g. at the IHRA Biomechanics meeting in Jacksonville, FL, 2002), updating
the group of the ongoing work. The co-ordinator however failed to establish a real work item on
frontal impact dummy harmonisation in IHRA, partly due to the group’s strong focus on side impact
(see also IHRA status reports included in the proceedings of the 17th and 18th ESV conference).

5.5 Dissemination
On the following conferences and seminars, publications generated in the FID project were presented:

1. International Research Council on the Biomechanics of Impact (IRCOBI) Conference, September
   18-20, 2002, Munich, Germany.
       References [1] and [7]

2. International Research Council on the Biomechanics of Impact (IRCOBI) Conference, September
   24-25, 2003, Lisbon, Portugal.
       Reference [3], [8] and [15]

                             FID project Final Publishable Report

3. STAPP Conference, November 11-13, 2002, Ponte Vedra Beach, Florida, USA.
      Reference [2]

4. STAPP Conference, October 27-29, 2003, San Diego, California, USA.
      Reference [16]

5. Congrès de l’Association des Morphologistes May 2002, Sousse, Tunisie.
      Reference [5]

6. Congrès de la Société de Biomécanique September 12-13, 2002, Valenciennes, France.
      Reference [4] and [6]

7. International Crashworthiness and Design symposium, December 2-4, 2003, Lille, France.
       Reference [9] and [13]

8. 17th International Technical Conference on the Enhanced Safety of Vehicles, June 4-7, 2001,
   Amsterdam, The Netherlands.
       Reference [11]

9. 18th International Technical Conference on the Enhanced Safety of Vehicles, May 19-22, 2003,
   Nagoya, Japan.
       Reference [10] and [12]

10. 28ème Congrès de la Société de Biomécanique, September 11-12, 2003, Poitiers, France.
       Reference [14]

11. Planned in future, anticipating on acceptance of the publication:
    FISITA, World Automotive Congress, May 23-27, 2004, Barcelona, Spain.
        Reference [17]

12.   FTSS-Japan 10 Years Anniversary Seminar, May 2000, Tokyo, Japan,
13.   KATRI Crash Seminar (with FTSS), July 19, Seoul, South Korea,
14.   SAE Government Industry Meeting, May 12-15, 2003, Washington, D.C., USA,
15.   KATRI Crash Seminar (with FTSS), September 30, 2003, Seoul, South Korea,
16.   4th EVPSN Annual Conference - November 6, 2003, Paris, France,
17.   Workshop “THOR-FT” at the opening of FTSS-Germany, November 11, 2003, Heidelberg,

Besides the conferences, congresses and symposiums referenced above, presentations about the FID
project research are given at the following meetings:

•     European Enhanced Vehicle safety Committee (including it working group WG12)
•     International Harmonised Research Agenda (IHRA) Biomechanics working group
•     International Standards Organisation working groups TC22/SC12/WG5 (Anthropomorphic Test
      Dummies) and WG6 (Biomechanics)

Five newsletters are submitted to disseminate the FID research effort on the occasions of conferences,
symposiums and seminars. The newsletters were sent by surface mail to interested parties all over the
world (The FID mailing list comprises about 100 addresses):
1. March 2001,          Newsletter 1: general introduction
2. September 2002, Newsletter 2: for the IRCOBI conference,
        `               September 18-20, 2002 in Munich, Germany

                           FID project Final Publishable Report

3. April 2003,         Newsletter 3: for the ESV conference,
                       May 19-22, 2003 in Nagoya, Japan
4. September 2003,     Newsletter 4: for the IRCOBI conference,
                       September 24-25, 2003 in Lisbon, Portugal
5. October 2003,       Exhibition leaflet for the Passive Safety Network conference,
                       November 6, 2003 in Paris, France

The THOR FT (FID Technology) dummy was displayed at two occasions:
• Passive Safety Network (PSN) conference November 6, 2003, Paris, France.
• Opening seminar German office of FTSS, November 10 and 11, 2003, Heidelberg, German

5.6 Partners – Contact Persons
TNO Automotive                                                 Mr. Michiel van Ratingen
Schoemakerstraat 97                                            Tel : +31 15 269 6342
P.O. Box 6033                                                  Fax: +31 15 262 4321
NL-2600 JA, Delft                                    
The Netherlands
                                                               Mrs. Birgitte van Don
                                                               Tel:    +31 15 269 7374
                                                               Fax: +31 15 262 4321

                                                               Mr. Kees Waagmeester
                                                               Tel:   +31 15 269 6682
                                                               Fax: +31 15 262 4321

                                                               Mr. Bernard Been
                                                               Tel: +31 15 269 6681
                                                               Fax: +31 15 262 4321

INRETS-LMBC                                                    Mr. Fran?ois Bermond
Centre de Lyon-Bron                                            Tel:    +33 47 214 2378
25, Av. Francois Mitterand, Case 24                            Fax: +33 47 214 2360
F-69675 Bron Cedex                                   
                                                               Mr. Philippe Vezin
                                                               Tel:    +33 47 214 2379
                                                               Fax: +33 47 214 2360

INRETS-LBA                                                     Mrs. Catherine Masson
Centre Marseilles-Salon de Provence                            Tel:    +33 49 165 8015
Faculté de Médecine- Secteur Nord                              Fax: +33 49 165 8019
Boulevard Pierre Dramard                             
F-13916 Marseille cedex 20

                              FID project Final Publishable Report

BASt                                                               Mr. Stephan Knack
Section Passive Vehicle Safety, Biomechanics                       Tel:    +49 22 044 3657
Bruederstraße 53                                                   Fax: +49 22 044 3687
D-51427 Bergisch Galdbach                                
                                                                   Mr Roland Schaefer
                                                                   Tel.: +49 22 044 3655
                                                                   Fax: +49 22 044 3687

                                                                   Mr. Claus Pastor
                                                                   Tel:   +49 22 044 3656
                                                                   Fax: +49 22 044 3687

TRL (Transport Research Laboratory)                                Mr. David Hynd
Biomechanics and Injury Prevention Group                           Tel:   +44 1344 770310
Old Wokingham Road                                                 Fax: +44 1344 770149
RG45 6AU
United Kingdom

(Instituto Universitarion de Investigación del Automóvil)          Mr. Luis Martinez-Saez
Camino de la Arboleda s/n                                          Tel:    +34 91 336 5327
Universidad Politécnica de Madrid (Campus Sur)                     Fax: +34 91 336 5302
Ctra. de Valencia km. 7                                  
28031 Madrid

University of Heidelberg                                         Mr. Dimitrios Kallieris
Institute of Forensic Medicine                                   Tel:   +49 6221 56 8940
Voßstraße 2                                                      Fax: +49 6221 56 3508
D-96115 Heidelberg                      

European Commision                                                 Mr. W. Maes
Office: DM 28 1/10                                                 Tel:   +32 2 2963434
B-1049 Brussels                                                    Fax: +32 2 2965196

6    Results and Conclusions

The results identified of the FID project as defined in the Technical Implementation Plan are:
1. Identification of the principal occupant injuries in frontal impact car crashes and the injuries for
   which a legislative frontal impact dummy should have the capability to measure injury risk.
   Owners: BASt and TRL

                              FID project Final Publishable Report

2. New biomechanical data concerning the behaviour of the human thorax/shoulder,
    pelvis/femur/knee and the lower leg during frontal impacts.
    Owners: INRETS, UPM, UoH, TNO and TRL
3. Set of requirements for frontal impact dummies, consisting of biofidelity requirements for all
    important body parts, repeatability, durability and anthropometry requirements based on the most
    recent biomechanical data.
    Owners: TNO, BASt, INRETS, UPM, UoH and TRL
4. Review and revise (where appropriate) injury assessment values for a legislative frontal impact
    dummy, with special focus on the lower leg.
    Owners: TNO and TRL
5. A prototype instrumented frontal impact dummy (THOR-FT), based on the THOR-Alpha design,
    50th percentile male. Suitable for inclusion in the Frontal Directive as successor of the Hybrid-III
    Owners: BASt, UPM, TNO and TRL. With the IPR at TNO
6. A computer model of the THOR-Alpha
    Owners: UPM, TNO and TRL. With the IPR at UPM and TNO
All these results are of category "A" that means "results usable outside the consortium".

7    Acknowledgements

The FID consortium thanks the European Commission that sponsored the project, in particular the
project officers assigned to FID, Mr. Rene Bastiaans, Mr. Claude Morin and Mr. John Berry for their
support and advice. Also thanks to the European enhanced Vehicle Safety Committee for their
guidance of this research and NHTSA for their interest in the project. FID also likes to acknowledge
GESAC, responsible for developing and manufacturing the THOR-Alpha dummy, represented by Mr.
Rangarajan Mr. Platten, and Mr. Shams or their technical support and advice.

TNO Automotive thanks:
     First Technology Safety Systems (FTSS), Mrs.Youmei Zhao, Mr.Steve Moss and Mr. Arie
     Schmidt for their support during the development and testing of the THOR-FT dummy.

      Ms. M. Ramet, MD for the experimentation on PHMS,
      Ms. K. Bruyere for the Thor’s face test in WP4 and for the literature survey in WP3,
      Ms S. Compigne for the literature survey in WP3,
      Mr. R. Bouquet and Y. Caire for the design of the seat for sled test in WP2 and WP4,
      Mr. A. Maupas, M. Maret, G. Goutelle, M. Callejon, A. Gilibert, J. Lardiere and P. Lapelerie,
      Ms. S. Serindat, Technical Staff of INRETS LBMC.
INRETS-LBA thanks the "Association pour la promotion des dons de corps de Marseille"
UPM/INSIA would like to thank to the Science and Technology Spanish Ministry for its partial
      funding in several of the task developed within this project.

University of Heidelberg thanks:
       Mr. Hans Riedl PhD for the co-operation
       Mr. Bernd v. Wiren for the experimentation support

                              FID project Final Publishable Report

8    References

1     Vezin, P. Bruyere, K., Bermond, F. (2002) Human response to a frontal sled deceleration. Proceedings of
      the International Research Council on the Biomechanics of Impact (IRCOBI) Conference.

2     Vezin, P., Bruyere, K., Bermond, F. and Verriest, J.P. (2002) Comparison of Hybrid-III, THOR-Alpha
      and PMHS Response in Frontal Sled Tests. STAPP Car Crash Journal, vol. 46, pp. 1-26, Paper 02S-06

3     Vezin, P., Verriest, J.P. (2003) Influence of the Boundary and Restraint Conditions on Human Surrogate
      Head Response to a Frontal Deceleration. Proceedings of the International Research Council on the
      Biomechanics of Impact (IRCOBI) Conference.

4     Masson, C., Cavallero C., Vinel, H., Brunet, C. (2002) Frontal impact on the human knee: experimental
      data. Archives of Physiology and Biochemistry, p. 31.

5     Masson C., Vinel, H., Portier, F., Ghannouchi S., Brunet, C. (2002). Presentation: Résponse de
      l’ensemble genou-femur-pelivs à un choc frontal à faible vitesse. 84ème Congrès de l’Association des
      Morphologistes. Sousse-Tunisie. Mai 2002.

6     Masson C., Cavallero C., Vinel H., Brunet C. Presentation of [4]: Réponse de l'ensemble genou-fémur à
      un choc frontal: approche expérimentale. XXVII Congrès de la Société de Biomécanique, 12-13
      septembre 2002, Valenciennes (France)

7     Masson C., Vinel H., Cavallero C., Brunet C. Response of the human pelvis-femur-knee complex during
      low speed frontal impact. IRCOBI September 18-20, 2002 - Munich (Germany)

8     C. Masson, C. Cavallero, Comparison between Hybrid III and cadaver knee response in frontal impact.
      IRCOBI 2003. Accepted for short communication.

9     J. Cardot, C. Masson, C. Brunet. Etude expérimentale et numérique du membre pelvien soumis à un
      impact frontal. International Crashworthiness and Design Symposium 2003.

10    Hynd D, Willis C, Roberts A, Lowne R, Hopcroft R, Manning P and Wallace W (2003). The
      development of an injury criterion for axial loading to the THOR-Lx based on PMHS testing. 18th
      International Technical Conference on the Enhanced Safety of Vehicles, Nagoya, Japan: US Department
      of Transportation, National Highway Traffic Safety Administration, Paper No. 078.

11    Kallieris D, Riedl H, Mattern R (2001) Response and Vulnerability of the Ankle Joint in Simulated
      Footwell Intrusion Experiments - A Study with Cadavers and Dummies, 17th International Technical
      Conference on the Enhanced Safety of Vehicles, Amsterdam, The Netherlands 2001, 9 Seiten (CD-

12    B. van Don and M. van Ratingen (TNO); F. Bermond, C. Masson and P.Vezin (INRETS); D. Hynd and
      C. Owen (TRL); L. Martinez (INSIA); S. Knack and R. Schaefer (BASt); on behalf of EEVC WG12;
      Biofidelity Impact Response Requirements for an Advanced Mid-Sized Male Crash Test Dummy; 18th
      International Technical Conference on the Enhanced Safety of Vehicles, Nagoya, Japan 2003; Paper No.
      76 (2003)

13    Vezin P., Verriest J.P. Behaviour of Head/Neck complex of two frontal crash test dummies submitted to
      a frontal deceleration as compared to that of a human surrogate, Proceedings of International
      Crashworthiness and Design Symposium Paper 03-OR-44, Lille, France 2003.

                             FID project Final Publishable Report

14   Bermond F., Vezin P., Bruyere-Garnier K., Verriest J.P. (2003) “Post mortem human subject and dummy
     response in frontal deceleration”. Proceedings of 28ème Congrès de la Société de Biomécanique,
     Archives of Physiology and Biochemistry, Swets & Zeitlinger Publishers, Vol. 111, supplementary
     volume September 2003, pp 88.

15   Martinez L, Ferichola G, Guerra LJ, van Ratingen M, Hynd D., Biofidelity and repeatability evaluation
     of the THOR dummy thorax, abdomen and femur, through a set of test, IRCOBI 2003.

16   B. van Don and M. van Ratingen (TNO); F. Bermond, C. Masson and P.Vezin (INRETS); D. Hynd
     (TRL); D. Kallieris (University of Heidelberg) L. Martinez (INSIA); Evaluation of the Performance of
     the THOR-Alpha Dummy; STAPP Car Crash Journal, vol. 47, Paper 03S-26; (October 2003)

17   (Future publication according planning, anticipating on acceptance)
     Martinez L, Ferichola G, Guerra L.J., Anthropometry study of the THOR and HYBRID-III frontal
     impact dummies, FISITA World Automotive Congress, Barcelona, Spain. 2004

18   GESAC (2001), Biomechanical response requirements of the THOR NHTSA advanced frontal dummy
     (Revision 2001.02). Report Nº: GESAC-01-04.

19   GESAC (2001), THOR Certification manual (Revision 2001.02). Report Nº: GESAC-01-05


Shared By: