History of Crash Test Dummies

Through the hundred plus year history of the automobile, safety has always been a serious concern. In fact, the fatality rate of 15.6 per 100 million vehicle miles traveled in the 1930s was many times our current rate of 1.8, even though we have millions of more cars on the road today. This notable progress is due in part to manufacturers’ diligent efforts to design cars so that fewer injuries occur during accidents. Crash test dummies, like those developed and manufactured by Humanetics Innovative Solutions, play a major role in making cars safer.

The History

1930s Fatalities per 100 million vehicle miles reach 15.6. Auto designers begin to pay serious attention to safety.
 
1949 Alderson Research Labs (ARL) Sierra Engineering build "Sierra Sam," an engineering dummy.
 
Early 1950s Cornell Aeronautical Laboratories study vehicle accidents to determine how to make cars safer. "Gard Dummy," a research dummy, is produced by Grumman-Alderson.
 
1950 Hollaman Air Force Base conducts crash tests using the ARL VIP 50th Dummy and Sierra Sam.
 
Late 1950s First cars with significant safety features introduced.
 
1950 - 1970 Automotive crash test dummies are developed based on aerospace models. 50th and 95th percentile males and 5th percentile female dummies produced.
 
1971 The Hybrid I, a standardization of the ARL & Sierra 50th percentile male dummies, is introduced.
 
1972 The Hybrid II is developed, with improved shoulders, spine and knees. It also offered better documentation than the Hybrid I
 
1973 The standard Hybrid II 50th percentile dummy is introduced. National Highway Transportation Safety Administration (NHTSA) contracts with General Motors to produce improved heads, necks, joints, ribs, knees, human-like posture and a new spine design. (ATD 502) Highway Safety Research Institute (HSRI) receives contract from Motor Vehicle Manufacturers Association (MVMA) to develop a 50th percentile male dummy with a new head, neck, thorax, spine, lumbar, pelvis, legs and joints.
 
1976 Hybrid III is introduced. General Motors improves ATD 502 with a new neck, thorax and more transducers for more extensive data.
 
1979 - 1987 NHTSA contracts with the University of Michigan Transportation Institute (UMTRI) to produce a new side impact dummy (SID). It is a Hybrid II type dummy with a new thorax.
 
1988 - 1989 Humanetics and SAE develop Hybrid III type small female and large male scaled dummies from Hybrid III 50th dummy. General Motors and Society of Automotive Engineers (SAE) develop Biosid, a Hybrid III based biofidelic side-impact dummy. AATD (advanced dummy project) completed. University of Michigan and Wayne State University receive NHTSA contract to develop an advanced dummy. First Technology Safety Systems is a subcontractor.
 
1995 - 1996 First Technology Safety Systems and Occupant Safety Research Partnership jointly develop the SID IIs, a small adult/teenager side impact dummy for side air bag development.
 
1996 First Technology Safety Systems develops the FT-Arup™ FE-Model Series, a highly precise and detailed finite element crash test dummy computer model.

Setting The Standards

Early in the century, there were virtually no safety regulations in place. During the 1930s, the industry took notice of very high injury and fatality rates. After World War II, action was taken to begin making cars safer. A major emphasis was on reducing injuries incurred when in an accident. Crash test dummies, such as those developed by Humanetics Innovative Solutions, have contributed greatly to identifying ways of making cars safer. Regulations also have played a large part in improving automobile safety. Following are some key regulatory actions that have taken place:
 
1966 President Lyndon Johnson signs the National Traffic and Motor Vehicle Safety Act, requiring the government to set safety standards for new vehicles.
1969 Nixon administration proposed passive restraints to protect unbelted passengers.
1971 Ford builds an experimental air bag fleet.
1973 General Motors manufactures 1,000 Chevrolets with experimental air bags.
1973 The first passenger car with an air bag sold commercially is an Oldsmobile Toronado.
1977 Carter Transportation Secretary Brock Adams announces that all new cars sold in US must have front air bags or passive safety belts that fasten automatically by 1984.
1981 Reagan administration delays passive restraint rule by one year. It is later canceled altogether.
1983 Supreme Court rules against the Reagan Administration and orders NHTSA to review the case for air bags.
1986 NHTSA allows auto makers to meet the passive restraint requirement with a driver side only air bag through 1990.
1987 NHTSA extends the driver side only air bag requirement through 1994.
1988 Chrysler becomes first US auto maker to make driver side air bags standard equipment.
1991 President Bush signs a law requiring air bags in all cars as of 1998, and truck as of 1999.
1993 Auto makers required to begin phasing in of passenger air bags in cars and light trucks

Dummies: Past & Present by Joe Smrcka

Introduction

Naked man is a fragile creature who must continually avoid harsh environments. When clothed, he can function in less favorable conditions, but he remains extremely vulnerable to blows, impacts, and accelerations. In early historic times, protective systems such as soldiers shields were devised and later elaborated into full suits of medieval armor. The growing severity of environments rendered such protection primitive indeed, and even before the advent of the space age, the concept of human packaging within vehicles was introduced.

Today the packaging of the human to minimize or avoid injury is a major undertaking. Automotive and cockpit interiors are designed in conjunction with restraint systems to help safeguard the occupant. Ejection seats and capsules are in regular use to counter the numerous hazards of escape from high-performance aircraft. Highly sophisticated energy-absorbing couches are used to attenuate impact forces which astronauts may encounter.

In spite of the huge effort devoted to the development and improvement of protective systems, one essential link has been weak in the protective chain. All of these protective efforts are centered about a single object, the human body; but no human body is really available to test concepts and evaluate systems, except at low energies because of the unacceptable risks to subjects. Most needed, a real human subject would perish.

In the attempt to make valid tests of protective systems, engineers and scientists have turned to many alternatives. Cadavers have been employed as test subjects, in spite of the many problems created by their use, because they are the closest available substitutes for the living human. Cadavers have served certain functions in test work, but cadaver data has fallen short due to their scarcity. Chimpanzees, hogs and other animals have served as automotive passengers and cockpit occupants, but their dissimilarity to the human is far too pronounced and their capability for internal instrumentation far too limited to produce meaningful data.

Anthropomorphic test devices, commonly referred to as "dummies", have been the most widely used subjects for testing protective systems. The early, conventional dummies had only limited utility. While approximating human kinematics and providing acceleration and other data applicable to corresponding humans, such testing has not generally measured the stresses imposed upon the human in the test situation and could only roughly determine the potentials for severe or fatal injury associated with such stresses. These deficiencies led to the development of dummies with improved biofidelity (the degree to which pertinent human physical characteristics are incorporated in the dummy design) and greater measurement capacity. While some efforts have been made to develop an omni directional dummy, most of the effort has been directed toward developing separate dummies for frontal and side collision auto testing. These dummies are classified as "frontal impact dummies", "Side Impact Dummies" and, in addition there are "Aerospace Dummies." The dummies are also classified according to their physical size. The mid-size adult male dummy, the most utilized in automotive restraint testing, approximates the median height and weight of the U.S. adult male population. The small female and large male dummies approximate the height and weight of the fifth-percentile female and 95th percentile adult male. Heights and weights of child dummies approximate median heights and weights of children of the specified age grouped, without regard to sex.

The following chapters provide description of past and present dummies, their intended application, design specifications, physical characteristics, biofidelity levels and measurement capabilities.

Chapter 1

Past Dummies: 1949 - 1960
 
1949 - Sierra Sam This 95th percentile adult male dummy was developed by Sierra Engineering Co. under a contract with the United States Air Force, to be used for evaluation of aircraft ejection seats on rocket sled tests. It was subsequently used as a lap shoulder harness test device. The main features of the dummy were durability, serviceability, but poor repeatability. Ifs biofidelity was limited to its human-like exterior shape, body weight and the ranges of motion of its articulated limb joints. Its mechanical type lumbar spine and neck design had little resemblance to their human counterparts. The dummy was built to anthropometric data based on "Anthropometry of USAF Personnel." Its response measurements were limited. Only orthogonal linear head acceleration components were measured However the dummy represented the state-of the-art technology of that early period.
 
Mark 1-1952- This general purpose 95th percentile dummy was developed by Alderson Research Laboratories for use by U.S. and European Air Force. A full plaster cast was made of a life subject. Cuts through estimated joint centers (palpation method on the life subject), sagittal through shoulders, oblique through hips and transverse at vertebrae 07 and waist determined the basic segmentation. The head was a two-piece cast aluminum skull with cranial cavity to house accelerometers and pressure transducers with integral vinyl skin/foam covering. The neck was a series of precision investment cast semi-spherical ball-and-socket joints held together by a tensioned steel cable. A full ball-and-socket joint between head and neck provided most of the flexion and extension. The dummy featured a lumbar/thoracic spine with ball-and-socket joints separated by phenolic spacers and rubber elements to prevent metal to metal contact and dampen the range of motion. Ribs were constructed of round steel tubing and attached to the individual thoracic vertebrae. The entire assembly was held together by a tensioned steel cable.
 
Precision torque adjustable joints were used on all major limb joints with access ducts molded through the vinyl skin/foam flesh. The one-piece limb design resulted in elbow, wrist, knee and ankle articulations that were too stiff. The dummy represented ifs human counterpart in shape, size and total weight. Only a few prototypes were made.
 
1956- Models F, B & P Alderson Research Laboratories designed a modular series of general purpose test dummies to be used for a variety of applications not requiring full range of kinematic and dynamic responses needed in automotive testing, but which had fuller motion capabilities than were needed in most ejection-seat testing. Dummies of this type, frequently with special modifications, were used widely in many kinds of programs. Some of the dummies had elaborate, pressure tight instrument cavities and space-age finishes for project Apollo Landing Testing or for tests of the F-111 Escape Capsule. Others were fitted with frogmen's suits for underwater-escape tests. And still others were fitted for tractor-safety programs. These modular dummies also served in many other special automotive and aircraft programs. The dummies were available in 8 sizes, ranging from 3rd to 98th percentiles (Air Force Anthropometry). Weight distribution was based upon the then Gene-Rally-Accepted Data, with 15 lbs. allowance for instrumentation on models with chest cavities. Additional motion ranges were frequently supplied on a custom basis to meet particular test requirements. All models utilized a series of ball-and-socket joints for motions of the lumbar spine, model "F" had construction similar to the Mark I Dummy.
 
A simplified pelvic structure (no biofidelity attributes) contained pivot blocks for upper-leg flexion, extension, abduction & adduction motion ranges. All three models had little capability for dynamic simulation; appendages were steel weldments with clevis-type joints and friction washers served as pre4est stabilizers only. The dummies chests had no compliance. A seated-form pelvic section was later added as optional extra to give abdominal compliance and more accurate response to lap belts. The use of these dummies was discontinued in the late 1960's.
 
1960-The Gard Dummy (Grumman - Alderson Research Dummy) The development of the Gard Dummy became a necessity when aircraft ejection - seats with rocket catapults reached the testing phase. Such seats were designed so that the resultant vector of rocket thrust passes through the center of gravity of the man-seat system. Misalignment would produce a rotational moment and tend to impart a spin to the system. An anthropomorphic test dummy with a misplaced center of gravity might thus cause rejection of an adequate ejection seat or permit qualification of a faulty seat. Alderson Research worked closely with the Grumman Aircraft Engineers to produce a dummy integrated with sensors and telemetry instrumentation so that the combined center of gravity and moment of inertia were correct. The reference standards for matching these two factors were closely-similar groups of subjects for each percentile, placed in a face-curtain-operated ejection seat in the firing position. The instrumented Gard Dummy has a CG which lies (in side view) within a 3116 inch diameter circle surrounding the mean location of the subjects CG's. The Gard dummy is designed to measure many critical parameters of ejection-seat performance, including rotational stability, acceleration histories, and man-seat interface stresses. Instrumentation for the standard Gard sled test subject consists of telemetry package and 12 transducers. In addition, special instrumentation such as microphones and thermocouples determine possible adverse offends on flight personnel during dual ejections, and cameras to provide information on parachute openings. Used in all Navy programs, yielding data which are accurate and reproducible among the various aerospace companies and government facilities concerned with such testing.
 
Eight sizes are available ranging from the 3rd to 98th percentile in the US Navy and USAF Anthropometry. All models have an identical 25 lb. weight allowance for instrumentation packages. There is a considerable simplification in the design of motion ranges, primarily in the neck, the lumbar region, and the shoulders. These motions are not essential to the tests of the man-seat system, since the significant ejection events occur when the subject is restrained within the seat. The ruggedness of the Gard dummy is unparalleled. Free fall, when a parachute fails to deploy, seldom results in significant damage. All limb joints are fitted with friction washers capable of being set to a 1O-G level. These joints serve mainly as pre4est stabilizers and are not intended for dynamic simulation. The Gard Dummy is still in use.

Chapter 2

Past Dummies: 1966 - 1967
 
1966 - Model VIP Series (Very Important People). Prior generation of test dummies were developed to meet the needs of the aircraft industry for testing pilot escape systems. It was increasingly apparent that such dummies failed to meet adequately the requirements of the automotive field. For instance, they had no pelvic structure and lacked sufficient spinal articulation. These deficiencies comprised their usefulness in the evaluation of automotive restraint systems where kinematic realism is so essential, particularly with respect to jackknifing and submarining of the body. A program (supported by Ford Motor Co. and General Motors Corp.) was initiated to develop more advanced test dummies which would resemble the human more closely in dynamic response to crash decelerations, and which would yield more valid data than have been available in the past. The full simulation of all human characteristics of interest in crash research required a long-range, evolutionary program based on two objectives, immediate and long term. The immediate objectives were to match the motions produced in human subjects in crash decelerations, establish reproducibility of motions responses and match the impact response of the human rib cage.
 
The long term objectives were to match impact response of the human head, provide measurement of the parameters of crash injuries and match frequency response of the complete dummy. Body size was based on the 50th percentile H.E.W. (Health, Education and Welfare) anthropometric data. Linkage dimensions were in accordance to S.A.E. test specifications for dynamic testing of automobile restraint systems. Overall center of gravity and movement of inertia of the dummy was matched to human subjects of the same size. Segment centers of gravity followed S.A.E. specifications.
 
As early as 1960's mathematical models of the human body were found to be reasonably good when the head, shoulders, thorax and pelvic regions and abdominal regions were considered as masses linked by springs with viscous damping. The immediate objective in this development was to match primary motions in crash decelerations and impact response of the rib cage. All motions of the spine were clevis type friction joints, capable of adjustment to 10 G's in the neck and 5 G's in the trunk. Adjustable springs in the neck provided for return or rebound of the head as in whiplash motions. Adjustable springs in the trunk provided limited degree of spring return to help establish initial motion response and align. the spinal column automatically. Axial spring in lumbar spine attempted to match human spinal elongation and compression. Rib cage motion consisted of chest compression, moving a one-piece sternum back toward the spine, a steel structure, bored to relieve three perpendicular hydraulic cylinders fitted with bypass bores and needle-valve controls for viscous damping characteristics. The cylinders had stops to limit motion at the top of sternum, mid-sternum and at the sub-sternal. The shoulder design linking the shoulder pivot with the sternum, an aluminum casting formed rigid quasi-clavicle and simulated the. upper margin of the scapula. The region between the casting and the shoulder skin was filled with vinyl to give human-like compressibility for shoulder harness.
 
Spring action for depression of the shoulder girdle was provided by a pneumatic cylinder with a piston dampened by rubber washers on each side. In the seated-posture pelvic assembly, the pelvic structure (still used today) is a one-piece aluminum casting. The spatial geometry of the pelvic structure is representative of the human pelvis with the exception of the anterior-superior Iliac spines, which are to some degree deficient in anatomical definition. Also deficient is the left-right symmetry of the femoral ball socket flange surfaces. Ball joints have individual and fully independent joint load adjustments.
 
All limb joints with the exception of ball joints at the hips were clevis or sleeve type joints. Clevis joints had phenolic washers between steel surfaces. The dummy flesh was molded in integral vinyl skin and vinyl foam. The principal objective in the flesh design was to minimize involvement of the flesh in skeletal motion, which would affect calibrations and limit reproducibility of dummy response. To improve anatomical fidelity in the abdominal region the simple butt joint used in previous designs was replaced by a separate flesh molding representing roughly the abdominal contents. This was shaped on the top surface as a body of revolution generated near the motion axes of the lower spine, permitting smooth mating between chest and abdominal flesh. Provisions for instrumentation was provided for femur load cells and three-axis accelerometers at the centers of gravity of the head and upper thoracic section. After a series of extensive tests based on comparison with a similarly upgraded Sierra Engineering Dummy "Sierra Stan" and air force volunteers at the Holloman Air Force Base in New Mexico the VIP5O dummy was modified and became VIP-50A, "The first standard automotive crash test dummies. The first production dummies were delivered to Ford Motor Co., General Motors, and the National Bureau of Standards in Washington D.C. in early 1968.
 
1967 - Sierra Stan An adult1 50th Percentile, male dummy, designed to H.E.W. (Health and Education Welfare) anthropometric data by Sierra Engineering Co. Like the VI P50, the dummy was specifically designed to meet the requirements of auto-motive industry testing. The dummy was modeled in a semi-seated posture which allows it to assume almost an upright position, but also a seated posture suited to automotive testing. The dummy's skeletal structure was enclosed by vinyl skin and polyurethane foam flesh and accessible by zipper closures. The chest design included a potentiometer and provided force deflection measurements. The shoulder structure manifested a unique linkage design, telescoping rods connecting shoulders to sternum by means of ball joint rod ends. A "ball-and-socket" design was used for neck and lumbar spine. The abdominal area included a viscera sac, to mimic the actions of the dummy's human counterpart when testing seat belts. Space was provided in the head, chest and thighs for instrumentation.

Chapter 3

Past Dummies: 1970 - 1973
 
1970- VIP95 & VIPF5 This adult large male and small female test dummies are basically similar within their respective scale factors to the VIP-50A Test Dummy. In their key anthropometry, and joint motion they comply with the same 50th percentile standards established by SAE J-963. In both, their joint load adjustability and instrumentation provisions conform to requirements as called out in NHTSA's standard 208. Their biofidelity is limited to their humanlike exterior shape. These dummies are still currently used for lap shoulder harness fitting by some auto manufactures.
 
1970 - Sierra Susie An Adult, 5th percentile female dummy. Her interior design was structurally the same as that of Sierra Stan.. The smallest of the adult dummies, Susie's weight was 104 lbs. and her seated height was 30.9 inches. Her petite torso area, however, incorporated the same structural design as that of Sierra Stan. Susie presented the maximum in aesthetic appearance for test dummies, she wore a realistic wig.
 
1970- Sierra Sammy & Sierra Toddler These six and three year old child dummies were developed to anthropometric data provided by the Institute of Transportation and Traffic Engineering, University of California, Los Angeles. Although distinct in its appearance, neither dummy was built to the correct weight distribution. The toddler, featured an articulated neck and lumbar spine, a flexible rib cage and a head with movable mandible. Sammy and toddler round out the "Sierra Dummy Family" which included Sam, Stan, Saul, Susie and Sue.
 
1971 - VIP 3C & VIP 6C In late 1971 Alderson Research Laboratories undertook the fresh and totally new design of a similar pair of crash test dummies to simulate the 50th percentile three-year old and six-year old child. The available data upon which to base these designs were meager, consisting principally of external body anthropometry compilation which have been picked from former surveys, mainly "Anthropometry for Child Restraints" by H.W. Stoudt, which was supplied to SAE Crash-Test Dummy Subcommittee in March 1971. The overall body weights for the dummies were selected from the McConville and Churchill Data "Source Data for the Design of Simulated Children's Body Forms" dated 1964. This data coincided with the FMVSS 208 requirement. It was, of course, desirable to duplicate accurately the segmental properties of centers of gravity and weight distribution, but reliable data on these parameters were tacking. Mechanical joint motion ranges were generally targeted to be the same as those specified for the adult in SAE J963. Another design goal was the physiology to be duplicated by these child dummies. It was desirable to simulate the immaturity of the key aspects of skeletal structure thought to be critical in the evaluation of automotive restraint systems. Finally, it was felt necessary as a generality to endow the dummies with the suppleness characteristic of children. Another design goal was the consideration of the uses to which these dummies were to be put and the associated requirements of the dummy users. The attitude of the dummies came first to mind and it was considered essential to design into these models the capability of either sitting or standing, as opposed to the by then commonly accepted permanently seated posture of adult dummies. Instrumentation needs were considered too1 and it felt essential to incorporate provisions at the C.G. of the head mass and at the C.G. of the upper thorax for the mounting of accelerometers for three axes.
 
The VIP 3C (Part 572) and VIP 6C round out the Alderson VIP Automotive Family. Both models provide full range of joint mobility, including realistic sitting or standing capabilities. Heads are constructed of hard elastomers and have too low natural frequency response. Necks and lumbar spines are of rubber column construction. Clavicular motion is provided by ball joints. Rib cages are one piece hard elastomer construction. Because of their limited biofidelity and response measurement capabilities, the usefulness of these dummies as injury-predicting surrogates is limited, and as such must be classified as first approximations of their subject populations.
 
1972 - HYBRID II (Part 572 Dummy)1-4 This is a 50th percentile adult male dummy specified by Part 572 of the Code of Federal Regulations1 to be used for compliance testing of cars equipped with passive restraints. The basis for the Hybrid II was the VIP-SO dummy, developed by Alderson Research and subsequently modified by General Motors to become a more repeatable lap/shoulder harness test device. It is used for limited qualification testing of air cushion restraint systems.4 The main features of the dummy are its good repeatability, durability, and serviceability. It's biofidelity is limited to its humanlike exterior shape, body weight, and range of motion of some of its articulated joints. Its response measurements are quite limited. Only orthogonal linear head and chest acceleration components and axial femoral shaft loads are measured. Because of its limited biofidelity and response measurement capabilities, the usefulness of this dummy as an injury-predicting surrogate is limited. However, the dummy does provide a basis for judging whether or not the repeatability and reproducibility of responses of other dummies are acceptable. The Part 572 dummy represents the state-of-the-art of dummy technology in the early 1970s.
 
1972 - Supermorphic Dummy Developed by Alderson Research Laboratories and Vector an Aydin Company for testing of The Yankee Escape System for the Navy EA6B Program. These 3rd and 98th percentile dummies comprised the most realistic effort, so far, in the area of human simulation for hazardous environment test programs, they incorporated the most sophisticated instrumentation system ever utilized in human simulation. Full articulations of limbs, torso, neck and head allowed the dummy to be placed in virtually any position which may be assumed by his human counterpart. Fidelity of joint structures was achieved by utilizing friction clutch assemblies and potentiometers to monitor each joint motion independently. Axial compression data was gathered from the lumbar and thoracic region of the dummy by the use of quartz load cells which could detect minute incremental forces at any force level in its ran9e. The primary reference for the dummy's acceleration was determined by three accelerometers located at the center of gravity of the total dummy. External dynamic pressure was measured by a transducer mounted in the head. Three individual gyros were mounted in the dummy's chest cavity for roll, pitch and yaw rate measurements. The upper thigh areas housed battery power supply so that the dummy system in any test environment could be independent of external cabling for power. The supermorph was a highly instrumented design (36 measurements) but too fragile for ejection testing.
 
1972- Dynamic Dan A result of a fresh approach to analogs for obtaining human body responses to dynamic loads imposed by tests of manned systems. Dynamic Dan was designed and built by Wyle Laboratories/Payne Division in collaboration with Aerospace Medical Research Laboratories/USAF specifically for testing aircraft ejection seat systems to simulate the response of the seated human body to vertical acceleration. Stiffness of spring forming the spinal column selected so that longitudinal natural frequency and damping characteristics correspond to that of a human body. Access was provided to allow spring of different frequencies to be used for comparative purposes. Bones were composed of a fiberglass material of essentially the same modulus of elasticity as that of human bones, but with greater breaking strength in order to avoid unnecessary damage. Shoulder and hip joints were represented by a ball-socket for universal movement. Other joints were of a clevis type design incorporating motion-limiting stops which when calibrated, gave a measure of the energy absorbed. Each joint was provided with adjustable friction damping to simulate resistance to motion from "relaxed' to "rigid." Dynamic Dan was suited for a number of applications including vibration tests, parachute opening shock test and etc.

Chapter 4

Past Dummies: 1972 - 1976
 
1972 - Supermorphic Dummy Developed by Alderson Research Laboratories and Vector an Aydin Company for testing of The Yankee Escape System for the Navy EA6B Program. These 3rd and 98th percentile dummies comprised the most realistic effort, so far, in the area of human simulation for hazardous environment test programs, they incorporated the most sophisticated instrumentation system ever utilized in human simulation. Full articulations of limbs, torso, neck and head allowed the dummy to be placed in virtually any position which may be assumed by his human counterpart. Fidelity of joint structures was achieved by utilizing friction clutch assemblies and potentiometers to monitor each joint motion independently. Axial compression data was gathered from the lumbar and thoracic region of the dummy by the use of quartz load cells which could detect minute incremental forces at any force level in its range. The primary reference for the dummy's acceleration was determined by three accelerometers located at the center of gravity of the total dummy. External dynamic pressure was measured by a transducer mounted in the head. Three individual gyros were mounted in the dummy's chest cavity for roll, pitch and yaw rate measurements. The upper thigh areas housed battery power supply so that the dummy system in any test environment could be independent of external cabling for power. The supermorph was a highly instrumented design (36 measurements) but too fragile for ejection testing.
 
1972- Dynamic Dan A result of a fresh approach to analogs for obtaining human body responses to dynamic loads imposed by tests of manned systems. Dynamic Dan was designed and built by Wyle Laboratories/Payne Division in collaboration with Aerospace Medical Research Laboratories/USAF specifically for testing aircraft ejection seat systems to simulate the response of the seated human body to vertical acceleration. stiffness of spring forming the spinal column selected so that longitudinal natural frequency and damping characteristics correspond to that of a human body. Access was provided to allow spring of different frequencies to be used for comparative purposes. Bones were composed of a fiberglass material of essentially the same modulus of elasticity as that of human bones, but with greater breaking strength in order to avoid unnecessary damage. Shoulder and hip joints were represented by a ball-socket for universal movement. Other joints were of a clevis type design incorporating motion-limiting stops which when calibrated, gave a measure of the energy absorbed. Each joint was provided with adjustable friction damping to simulate resistance to motion from "relaxed' to "rigid." Dynamic Dan was suited for a number of applications including vibration tests, parachute opening shock test and etc.
 
1972 - OPAT Dummy The OPAT (Occupant Protection Assessment Test) Dummy was developed by David Ogle Ltd. (A British dummy manufacturer and MIRA (The Motor lndustry Research Association of Britain) under a contract with the British Government's Department of the Envirnoment1 Transport and Road Research Laboratory (TRRL) in 1972. The dummy was to be representative of the 50th percentile adult male in size and weight and was to provide humanlike behavior when used to evaluate lap-shoulder belt systems. The dummy features a humanlike clavicle and floating scapula design, and it's rib cage mimics the shape of the human. These features make the OPAT Dummy particularly useful in evaluating lap-shoulder belt systems. The chest structure has humanlike impact response for blunt frontal impacts. The dummy is equipped to measure orthogonal linear acceleration components of its head and chest and the axial loads in its femurs. Its repeatability is comparable to that of the Hybrid II (Part 572) dummy. The dummy is commercially available.
 
1973 - Repeatable Pete10 This 50th percentile adult male dummy was developed by The Highway Safety Research Institute (now called University of Michigan Transportation Research Institute) in 1973 under a contract with the Motor Vehicle Manufacturers Association. Repeatable Pete features head, neck and chest structures with humanlike impact response characteristics for a prescribed set of frontal impact conditions.1011 The head has been designed to have humanlike impact response for lateral impacts as well.12 The dummy has flexible thoracic and lumbar spine that allows it to be placed in a humanlike, automotive seating posture. Constant torque joints are used for major limb joints. The dummy is instrumented to measure orthogonal linear acceleration components of the head and chest, fore/aft chest compression, and axial femoral shaft loads. Based on the results of tests conducted with two prototype dummies, the repeatability and reproducibility were comparable to the Part 572 Dummy.3~5 This is not a commercially available dummy.
 
1976-HYBRID III (Part 571 Dummy)4-6 This is a 50th percentile adult male dummy specified by Part 572 of the Code of Federal Regulation to be used for compliance testing cars. The basis for the Hybrid ill is the ATD 502, an advanced test dummy developed by General Motors in 1973 under a contract with the National Highway Traffic Safety Administration. The ATD 502 featured a head with humanlike impact response characteristics for the hard surface forehead impacts.78 A curved lumbar spine was used to achieve a more humanlike automotive seating posture. Constant torque joints were incorporated in the knee, elbow, and shoulder joints to improve repeatability and minimize the time required to set joint torques. The shoulder structure was designed to improve belt-to-shoulder interfacing, which was a problem with the Hybrid II (Part 572) shoulder design The Hybrid III dummy retained these ATD 502 features, while design changes were made to improve the impact response biofidelity of its neck, chest, and knees. Transducers were incorporated into the Hybrid III design to measure the orthogonal linear acceleration components of the head and chest, the sagittal plane reactions (axial and shear forces and bending moment) between the head and the neck at the occipital condyles, the displacement of the sternum relative to the thoracic spine1 the axial femoral shaft loads.
 
Based on the testing of three prototype dummies, it was concluded that the Hybrid III repeatability was equivalent to that of the Part 572 dummy and that is appeared significantly more reproducible.
 
Since the publication of the paper by Foster et aI.4 describing the Hybrid III, its response measurement capacity has been significantly increased.6 Table 1 lists the measurement capacity of the fully instrumented Hybrid III. Note that the fully instrumented Hybrid III provides 44 response measurements for assessing occupant protection potential. The only biofidelity improvements made to the Hybrid III since the Foster et al. paper have been to the knee and ankle joints. The current knee joint design allows the leg to translate relative to the thigh in a humanlike fashion.9 The ankle joint allows lateral flexion.
 
The mean total body weight of the fully instrumented Hybrid III dummy exceeds the 50th percentile adult male median weight of 164lbs by 7.3 lbs. or 3.3 kg. The excess weight is due primarily to the provision of fully instrumented lower legs with ball joint ankles on the dummy. These modifications strengthen the lower legs and ankles to prevent them from breaking in crash tests resulting in extreme rearward displacements of the floor or firewall of the vehicle. The instrumentation allows the collection of data used to design against injuries to the lower leg such as breakage of the tibia, fibula or ankle joint. It is felt that the increased measurement capabilities more than make up for any shortcoming associated with the increased weight.
 
The Hybrid III dummy can be purchased commercially. It's use is Federally mandated for certifying the crashworthiness in frontal impact of all passenger cars sold in the United States and in some other countries. It is extensively used by General Motors and other vehicle manufacturers to assess the occupant protection potential of their new car designs.
 
TABLE 1. Measurement Capacity of Fully Instrumented Hybrid III
Measurement Data Channels
Head  
Triaxial acceleration 3
Angular acceleration 1
Facial Laceration (Chamois)
Neck  
Axial load 1
Shear load 1
Bending moment 1
Chest  
Triaxial acceleration 3
Sternum acceleration 2
Deflection 1
Pelvis  
Triaxial acceleration 3
Anterior/superior iliac spine load 6
Upper Extremities  
Lower arm bending moments 4
Lower extremities  
Femur load 2
Femur/tibia translation 2
Tibia bending moments 4
Tibia axial load 2
Medial/lateral tibia plateau load 4
Lateral or fore/aft ankle bending moment and shear load 4
Knee laceration (Chamois)
Total Data Channels 44

Chapter 5

Past Dummies: 1983 - 1987
 
1983 - Limb Restraint Evaluator The LRE Anthropomorphic Manikin was developed by Systems Research Laboratories, Inc., Dayton, Ohio with the support and assistance of Alderson Research Laboratories in Stamford CT., under a contract to U.S. Air Force. The objective of the LRE was to provide a capability that can be used to evaluate the effectiveness of various limb restraint devices in preventing limb flail injuries during emergency ejection from military aircraft. The design guidelines were provided by U.S. Air Force.
 
Physical Characteristics
  • LRE must respond to impact accelerations in the Z and X directions; and if possible, in all directions.
  • Parts from existing dummies should be considered for use where possible (ie.Hybrid II or Hybrid III).
  • Shape and inertia properties of the LRE must duplicate those of the human extremities.
Joint Design
  • All joints will be simplified representations of their corresponding human joints.
  • Maximum displacement of each joint shall be retained with normal limits.
  • Joint stiffness should be approximated.
  • Angular displacement measurements of the shoulder, hip, elbow and knee joints shall be made.
Other
  • Torsion moments shall be measured after the joint has reached its limit of motion.
  • Seat or restraint system shall not be altered or affected by the LRE.
  • The tension compression forces and bending moments in the bones of the upper and lower arms and upper and lower legs shall be measured.
This manikin was replaced by ADAM (Advanced Dynamic Anthropomorphic Manikin) and is no longer available. 1986 - ADAM The small and large Advanced Dynamic Anthropometric Manikin was developed by Systems Research Laboratories, Inc. (SRL), under contract to the Armstrong Aerospace Medical Research Laboratory (AAMRL) at Wright Patterson Air Force Base) Ohio. ADAM is a fully instrumented, high fidelity manikin used as sophisticated test device to support acceleration and ejection system tests. ADAM was developed to be the crew member and instrumentation package for testing of the U.S. Air Force Advanced Development Ejection Seat Program, Crew Escape Technologies (CREST)1 ADAM has direct applications in other important areas of replicating a human body's dynamic response during potentially dangerous conditions1 i.e. experimental parachute test, helicopter seat crashworthiness test, etc.
 
Biodynamic Fidelity
ADAM body segments approximate human surface contours, weights, moments of inertia, centers of gravity and joint center locations. Ranges of motion of 39 revolute joints replicate human articulations. "Soft stops" which yield human-like, non-linear torque deflection variances for each articulation, as well as providing absorption of impact loads. Independently adjustable friction mechanisms designed into each joint (excluding wrist and sternoclavicular joints) to provide for passive muscle resistive forces. Biodynamic responses for both low amplitude vibration and high impact environments.
 
Limb Construction
Limbs are constructed from stainless steel; the torso is constructed from aluminum alloy. A heat cured vinyl plastisol provides the proper outside flesh covered body contours which also represents the characteristics of human flesh. Vinyl plastisol foam in between outer and inner layers of vinyl plastisol is compliant and represents the shape of soft tissue covering All of the vinyl plastisol components for the small ADAM were developed and supplied by Alderson Research Laboratories, Stamford, CT., for the large ADAM by Humanetics Inc., Carson City, CA.
 
Dynamic Spine
The ADAM spinal system was designed to replicate the human spine's elasticity in the vertical direction. The Z-axis human spine response qualities are incorporated into ADAM by using a mechanical spring/damper system in the spine. This mechanical spine allows for human-like deformation in the z-direction and is equipped with non-linear stops at the end of the vertical travel. By providing a realistic spinal system, The ADAM's seated whole body center of gravity varies with the force applied to it and provides a realistic response to loadings in pitch, roll and yaw directions.
 
The ADAM was produced in two different sizes:
  Small ADAM Large ADAM
Weight 64.55 kg 98.43 kg
Sitting Height 87.63 cm 95.25 cm
Standing Height 168.28 cm 178.72 cm
 
Both manikins are available.
 
1987 - Small Female and Large Male Hybrid III - Type Dummies While the Hybrid III dummy provides excellent assessments of the effectiveness of automotive restraint systems for the midsize adult male occupant, it provides no information concerning restraint effectiveness for large- or small-size adult occupants. To fill this void, the Center for Disease Control (CDC) awarded a grant in 1987 to Ohio State University (OSU) to develop a multisized Hybrid III - based dummy family. To support the OSU effort the Mechanical Human Simulation Subcommittee of the Human Biomechanics and Simulation Standards Committee of the Society of Automotive Engineers (SAE) formed a Task Force of biomechanics, test dummy, transducer, and restraint-system experts. They defined the specifications for an adult-size small female dummy and an adult-size large male dummy having the same level of biofidelity and measurement capacity as the Hybrid III dummy.48 key body-segment lengths and weights were selected for each dummy based on the anthropometry data for the extremes of the U.S. adult population.49 Geometric- and mass-scale factors were developed to assure that each body segment had the same mass density as the corresponding Hybrid III body segment. Other pertinent dimensions were scaled from their corresponding Hybrid III dimensions using the corresponding geometric-scale factors.
 
The Hybrid III biomechanical response requirements for the head, neck, chest, and knee were scaled using the appropriate scale factors giving corresponding biofidelity response requirements for each size dummy.
 
The Hybrid III design drawings were scaled using these geometric-scale factors to produce design drawings for each dummy. This procedure assured that each-size dummy made according to its scaled drawings would meet its scaled biofidelity requirements. The dummies were instrumented identically to the Hybrid III dummy. Both are commercially available.