Australian New Car Assessment Program (ANCAP)

      Guidelines for Crashworthiness Rating System

      November 2000 - Subject to change without notice
      This is NOT an official ANCAP web site.
      During 1999 ANCAP changed to Euro-NCAP test and assessment procedures. The following guidelines are still used for rating structure, restraint systems and head restraint design.

      For IIHS rating guidelines see the PDF downloads at http://www.hwysafety.org/vehicle_ratings/ratings.htm

      Web page created by Vehicle Design and Research Pty Limited
      email: mpaine@tpgi.com.au

      CONTENTS

      INTRODUCTION
      BACKGROUND
      RATING CATEGORIES
      SAFETY FEATURES
      STRUCTURE
        ASSESSMENT OF DEFORMATION
        STEERING COLUMN
        MEASUREMENTS OF DEFORMATION
      OCCUPANT RESTRAINT
      INJURY MEASURES
        COMBINED INJURY RATING
        OVERALL RISK OF LIFE-THREATENING INJURY
      HEAD RESTRAINT POSITION
      OVERALL EVALUATION
      REFERENCES
      SAMPLE FROM ANCAP BROCHURE
      DASH AND DOORWAY DEFORMATION

      These Guidelines have been published by the Technical Committee of the Australian New Car Assessment Program (ANCAP) for the information of vehicle manufacturers, regulators and consumer organisations. A knowledge of automotive engineering is assumed. The details of rating system were originally developed by Vehicle Design and Research Pty Limited (VDR) under contract to ANCAP.
      ANCAP reserves the right to amend these Guidelines without notice.
      Contact information

        Chris Coxon, Chairman, ANCAP Technical Committee
        Office of Road Safety
        Transport South Australia
        33 Warwick Street
        Walkerville South Australia 5081
        Australia


      For a less technical guide to the rating system see "How the evaluations are performed" (NRMA web site)

      Introduction

      This document describes the rating system used in Australian New Car Assessment Program (ANCAP). The system has applied to ANCAP brochures from November 1996. and allocates a rating of  GOOD, ACCEPTABLE, MARGINAL OR POOR for vehicle structural performance, occupant restraint/kinematics and key injury measurements. Appendix A contains a sample from a brochure.
      The document is intended to provide guidance for the preparation of NCAP brochures by persons with experience in crash investigation and crashworthiness evaluations. Due to its technical complexity it is not intended for general publication.
      The philosophy behind the rating system is that the resulting brochure should present the typical car buyer with key information in a manner which is simple, useful and credible. The system should attempt to spread the ratings across the entire range from "good" to "poor". A rating of  "good" should correspond to current best practice, not a theoretical ideal. Wherever possible, objective criteria should be used in arriving at ratings. Also it should be possible for different assessors (with suitable qualifications) to arrive at the same ratings.

      Background

      In 1978 the U.S. National Highway and Transport Safety Administration (NHTSA) began full frontal crash testing (56km/h) of popular vehicle models in the United States. The results were published for the information of consumers. This led to greater interest in the safety of vehicles.
      Since the mid 1970s at least one European car manufacturer has been conducting offset crash testing as part of its vehicle design program. In the late 1980s the German Automobile Club and the magazine Auto Motor und Sport began offset crash testing of popular European models (55km/h, 50% offset). The results are published in the magazine. In 1995 the US the Insurance Institute for Highway Safety (IIHS) began offset crash testing of U.S. models (64km/h, 40% offset).
       
      Overhead view of an offset crash test at 64km/h
      Offset crash test at 64km/h (100K animated GIF)
      The Australian  New Car Assessment Program (ANCAP) started in 1992 with full frontal crash tests. The results were published in April 1993. Offset crash tests were started in 1993 (results published in 1994 - initially 60km/h 40% offset and then 64km/h from mid 1995). The program is supported by all Australian automobile clubs and the state government road and transport authorities in Queensland, NSW and South Australia.

      In 1996 the Japanese National Organisation for Automotive Safety and Victims' Aid (OSA) commenced publishing results for full frontal crash tests in Japan (55km/h). In 1997 Euro-NCAP published the results of its first series of crash tests (64km/h 40% offset, side impact and pedestrian impact tests).
      Rating systems for these crash tests have evolved in response to consumer demands for a simple, non-technical presentation of the results. NHTSA recently introduced the star rating system, based on the risk of the driver or front passenger receiving a serious or life-threatening injury in a full frontal crash at 56km/h. A shortcoming of this approach is that it does not take into account structural performance or occupant restraint performance. It is possible for a test to result in "good" head and chest injury measurements (and therefore a good star rating) even though the structure performs poorly. In these cases the risk to life in a slightly different crash configuration, or slightly higher speed, could result in a much higher risk of serious injury.
      IIHS introduced a new style of rating system for its publication "Crashworthiness evaluations - midsize 4-door cars". This provides a separate rating for structure, occupant restraint and injury measures, together with an overall rating. Ratings are also given for head restraint design and bumper performance. An overview of the IIHS rating system is given in a paper issued at the 1996 ESV Conference (O'Neill et al 1996). This is the rating system now used for Australian NCAP, but modified to include analysis of full frontal crashes and front passenger performance. More information about occupant restraint/kinematics is given in the paper by Estep & Lund (1996).

      Rating Categories

      Rating categories should be consistent for each evaluation. In accordance with IIHS practice,  there are four categories for ratings.
      Good, Acceptable, Marginal and Poor
      Injury measurements should be described  in terms of protection from serious injury. For example a HIC of less than 750 indicates that there is  "good"  protection from serious head injury. (consumer research indicated that terms such as  "high risk of head injury" were confusing).

      Safety Features

      This section of the brochure lists notable safety features which, in general, go beyond those normally fitted by manufacturers. Examples (in Australia) include air bags and ABS brakes.
      Possible design shortcomings could also be noted. An example is a hard protruding object which could be contacted by the head in a side impact or rollover crash.
      Safety features are not rated..

      Structure

      The main item of interest is the loss of occupant survival space.
      The performance of the vehicle structure over the entire crash event is evaluated. The integrity of the occupant space is assessed through analysis of measurements of deformation, technical reports, photographs and high-speed film. Key areas of interest are movement of the dash and steering column, reduction in the width of the doorway opening, deformation of the firewall, footwell and floorpan, movement of pedals, integrity of the doors, ease of  opening doors after the crash, the deformation of the roof and pillars  and integrity of the seats. In assessing deformation, consideration should be given to how it affects the remaining survival space. For example, a large deformation might be acceptable if the initial clearance between a hazardous surface and the occupant is large and, at the height of the crash, there is no significantly increased risk of the occupant contacting that surface. In practice, popular vehicle models rarely have the luxury of such large initial clearances.

      Assessment of deformation

      It is evident from analysis of high speed film that many vehicles experience substantial elastic deformation at the height of the crash (noticeable in the above animation). Components such as the dash then spring back so that the residual (plastic) deformation is much less than the peak. No measurement of the elastic deformation of these components is taken during the crash (in any case, such measurements are likely to be difficult to achieve with on-board equipment). In the circumstances, elastic deformation should not be used as a factor in the  rating of structure but it can help to explain some events, such as unusual head or knee contacts. An increase in the rate of deformation can also indicate that catastrophic failure is occurring and the structure is no longer absorbing significant amounts of energy.

      Steering column

      The steering column usually collapses when struck by the occupant (including when an air bag is deployed). Therefore residual movement of the steering column  is not an appropriate factor is assessment of the  loss of survival space. However, the direction in which the steering column moves can affect air bag performance. For example, in several offset tests (Australian and IIHS) the steering column moved substantially towards the centre of the vehicle. The driver therefore hit the air bag on its outboard side and this resulted in undesirable occupant kinematics. In one case the driver's head rotated through 180 degrees so that the back of the head hit the A-pillar.
      The motion of the steering column can also increase the severity of head or chest impacts. In some cases the steering column was observed to have a substantial upwards motion at the moment when it was struck by the driver's head. The head was swinging downwards by this time and the resulting impact was more severe.

      Measurements of deformation

      The following guidelines are based on residual (plastic) deformation. An analysis of numerous offset crash tests conducted by IIHS and Australian NCAP suggests that dash movement and the reduction in driver's doorway width are the best indicators of loss of survival space. The ratings for these measurements are:
      Ratings for dash and doorway
      Rating Measured change
      Poor 200mm or more
      Marginal 150-199mm
      Acceptable 80-149mm
      Good less than 80mm

      The worst of these two measurements should be used for an initial rating of structure. This initial structural rating can be modified by several other structural factors

      Examples of factors which modify structural rating
      Initial Rating
      Good Acceptable Marginal
      Item
      Modified Rating
      Substantial footwell intrusion (350mm or more) Marginal Poor Poor
      Moderate footwell intrusion (250 to 349mm) Acceptable Acceptable Marginal
      Foot trapped by pedals or deformed panels Marginal Poor Poor
      Tools required to open door Acceptable Marginal Poor
      Doors open during crash see "Occupant restraint"
      Pedal moves to within 100m of seat structure  Acceptable Marginal Poor
      Seat is detached (one or both sides of seat) see "Occupant restraint"
      Bonnet intrudes into survival space Marginal Poor Poor
      Fuel system leak - major (stream) * Acceptable Marginal Poor

       *revised wording

      Occupant Restraint

      The performance of the restraint system (seat belts, seats, air bag & head restraint), over the entire crash event is evaluated. The motion of the occupants is assessed through analysis of measurements (e.g. seat belt reel out), technical reports, photographs and high-speed film. Key areas of interest are head, chest and knee strikes; air bag deployment and occupant kinematics, including rebound and partial ejection.
      The table below sets out guidelines for assessment of occupant restraint and kinematics. The factors resulting in a "poor" rating are generally due to equipment failure or equipment not performing as intended.
      The worst of all four restraint ratings is used for the overall occupant restraint rating for a vehicle. However, greater weight is given to the driver  restraint, in recognition of the lower occupancy rate for the front passenger position. This is done by elevating the worst passenger result by one level, for the purpose of determining overall restraint rating. For example, if the results for a vehicle are:

       driver/full frontal: acceptable            driver/offset: good
       passenger/full frontal: marginal          passenger/offset: good

      then the "marginal" passenger rating is elevated to "acceptable" and the overall rating for occupant restraint  is "acceptable".
      Note that peak and 3ms clip decelerations are recorded for head impacts. These give an indication of the severity of the impact and the characteristics of the surface.

      Ratings for Occupant Restraint
      Rating Criteria
      Poor A door opens during the crash
      Poor A seat becomes fully or partially detached, slides forward on one or both sides (of the seat) or the seat back fails - this also applies to unoccupied seats.
      Poor An occupant slides off one side of an air bag and makes a hard contact (70g or more) (also applies where an air bag clearly bottoms out or fails to inflate)
      Poor Seat belt failure (webbing, buckle, anchorages, retractor)
      Poor An occupant's head is partially ejected (more than 25% outside vehicle). (Note that this factor does not apply if the partial ejection is through a side window and that window is normally fitted with impact resistant glazing)
      Poor Severe motion of steering column increases risk of injury*
      Marginal An occupant slides off one side of an air bag but does not make a hard contact (<70g)
      Marginal An occupied seat tilts or twists substantially (without failure or sliding)
      Marginal An air bag appears to deploy properly but a high deceleration is experienced (70g or more) and  HIC > 750
      Marginal The driver's head or chest strikes the steering wheel, resulting in hard contact (70g or more)
      Marginal The passenger travels forward sufficiently to strike the dash and make hard contact (70g or more)
      Marginal The passenger travels forward sufficiently to strike its knees
      Acceptable In the absence of an air bag, no head or chest strike occurs but a high deceleration is experienced (70g or more) (e.g a whipping effect and/or stiff restraint system)
      Acceptable An air bag  high deploys and a high head deceleration is experienced (70g or more) but  HIC is 750 or less
      Acceptable The driver's head or chest strikes the steering wheel but decelerations are low to moderate (<70g)
      Acceptable During rebound, the occupant misses the seat back or head restraint
      Acceptable During rebound, the occupant's head strikes a window frame, B-pillar or roof.
      Good No head/chest strikes (other than air bag), decelerations low to moderate (<70g) and rebound is well controlled, with little tranverse or vertical movement of the occupant.
       *revised wording

       Injury measures

       
      Measurement Good Acceptable Marginal Poor
      HIC <750 750-899  900-999  1000 or more
      Chest Compression  <50mm  50-59mm 60-74mm 75mm or more
      Chest Acceleration# <60g 60-74g  75-89g 90g or more
      Femur Axial Force* <7.3kN 7.3-9.0kN  9.1-10.8kN 10.9kN or more
      Lower Leg Index <0.8 0.8-0.9 1.0-1.1 1.2 or more

      Notes:
       # Worst of chest compression and chest deceleration used as chest rating
       * IIHS set lower femur force limits for longer duration readings

      These criteria vary from those used in previous ANCAP brochures. In particular, the "high" (now "poor") HIC reading was 1250, compared with 1000 for IIHS (and FMVSS 208) so that the new criteria  result the overall injury rating for eight previously-tested Australian vehicles going from "marginal" to "poor". Also the inclusion of chest acceleration criteria  results in the chest injury rating changing from "good" to "acceptable" for some vehicles. IIHS criteria for leg injury are generally more lenient than previous Australian ratings.
      A broken dummy leg, foot or ankle is cause for a "poor" leg rating.

      Overall rating for each body region

      The worst of all injury measures is used in the overall injury rating for each body region. However, greater weight is given to the driver injury results, in recognition of the lower occupancy rate for the front passenger position. This is done by elevating the worst passenger result by one level, for the purpose of determining overall injury rating for a body region. For example, if the ratings for the head are:
       
         driver/full frontal: acceptable        driver/offset: good
         passenger/full frontal: poor           passenger/offset: good
      then the "poor" passenger rating is elevated to "marginal" and, since this is the worst of the four ratings, the overall rating for head injury is "marginal".

      Combined injury rating

      For the purpose of determining an overall crashworthiness rating for the vehicle (see "Overall evaluation") a combined injury rating is derived from the worst of head and chest ratings. If  all head and chest ratings are "good" and one or more leg ratings is "poor" then the combined injury rating is downgraded to "acceptable". Otherwise the leg rating does not have an influence on the combined injury rating (which is weighted towards life-threatening injuries). Note that leg injury rating might have an influence on the overall ranking of vehicles.
      The following table gives an example of the derivation of overall injury ratings. In summary: the passenger/full-frontal rating is elevated by one level (e.g. "poor" becomes "marginal", "good" is unchanged); an overall rating is assigned for each body region, based on the worst of the ratings for driver and the  elevated passenger rating; a combined injury rating is derived from the worst of the overall head and chest ratings (in this case "marginal"). These ratings are not intended to be published.
      Example of derivation of overall injury rating
      Driver Driver Passenger
      Overall Full-frontal Offset Full front
      Head Marginal Acceptable Good Poor
      Chest Good Good Good Good
      Upper legs Marginal Good Marginal Good
      Lower legs Poor - Poor -

      Note that passenger offset injury ratings are not usually published because they are generally much better than the other ratings and this can cause confusion about overall ratings. In the case of legs, the worst of left and right readings is used.

      Overall risk of life-threatening injury

      A risk of life-threatening injury, expressed as a percentage, is included for each vehicle. This risk is derived from the head (HIC) and chest (acceleration) measurements in accordance with the technique described in the November 1994 Australian NCAP Technical Report (contact the NRMA for a copy). This technique combines the risk for full frontal and offset tests.
      A rating (good to poor) for this risk is not provided because this might get confused with the separate ratings for head and chest injuries.
      Note that the term "life threatening injury" is based on an interpretation of the AIS Coding Manual which might not have been intended by the authors of that manual: the AIS injury severity codes are:
        1-minor,
        2-Moderate,
        3-Serious,
        4-Severe,
        5-Critical,
        6-Maximum.
      The risk is calculated on the basis of probability of head or chest injury of AIS 4 or more. The phrase "AIS 4 or more" is clearly inappropriate for a consumer publication. Although the AIS Coding Manual does not formally categorise these more severe injuries the introduction to the manual does refer to "serious and life-threatening levels (AIS 3 or more)". This suggests AIS 4 or more is life-threatening (since AIS 3 is "serious"). For the purposes of ANCAP the term  "life-threatening injury" is used for injuries of AIS 4 or more.

      Head restraint position

      The rating is based on IIHS criteria and manikin measurements: "height" is vertical distance from top of head restraint to top of manikin's head. "backset" is the horizontal distance from the front of the head restraint to the back of the manikin's head. The vertical reading (V) is reduced by 5mm and the horizontal reading (H) is reduced by 10mm in order to allow for different seat back angles (this has the effect of improving the evaluation). For adjustable head restraints which lock in position*   two sets of measurements are taken: one in the fully down, fully back position and the other in the optimum lockable position (usually fully up and fully forward). If the lowest position is in the same zone as the optimum position then the rating is based on this zone. Otherwise the zone for the optimum position is  used but is downgraded one level (e.g. good becomes acceptable) because so few motorists adjust their head restraints correctly

      *"lock in position" means the position of adjustment has a release device, such as a button,  and can withstand a 100N downward force  (proposed new wording) .


      The resulting zones are rated as follows:
       

      Zone for lockable optimum position
      Zone for lowest pos. Fixed or not  Lockable 1 2 3 4
      1 Good Good - - -
      2 Acceptable Acceptable Acceptable - -
      3 Marginal Acceptable Marginal Marginal -
      4 Poor Acceptable Marginal Poor Poor

      Overall evaluation

      A provisional overall rating is  based on the "average" of structure, occupant restraint systems and injury, according to the following  guidelines:
      Derivation of Overall Evaluation Rating
      Good Acceptable Marginal  Poor
      GGG AAA MMM PPP
      GGA GGM GGP APP
      - GAM GAP MPP
      - AAM GMM GPP
      - GAA GMP -
      - - AAP -
      - -  MMP -
      - - AMM -
      - - AMP -

      In this table the order is not important: GGA means two "good" and one "acceptable". However, the order might affect overall ranking of a vehicle, where priority is given to structure, then injury and finally restraints.
      Finally the ANCAP Technical Committee reviews all of the information and compares each vehicle with the other vehicles in the same group to determine whether the provisional overall rating is appropriate for each vehicle. This review process is more likely to result in an upgrading for borderline cases than any downgrading. However, it is possible that a review might pick up an undesirable crashworthiness feature that was not adequately covered by these guidelines and therefore deserves consideration of a downrating.
      This rating system for overall evaluation was used for the November 1996 ANCAP brochure and subsequent ANCAP brochures.

      References

      Australian NCAP (1996) 'Buyers guide to crash tests: Small Car Update',  November 1996.
      Estep C.R. & Lund A.K.(1996) 'Dummy kinematics in offset frontal crash tests', Proceedings of 15th International ESV Conference, Melbourne.
      IIHS (1996) 'Midsize 4-door cars - Crashworthiness Evaluations', Insurance Institute for Highway Safety.
      Hackney J.R., Kahane C.J & Chan R. (1996) 'Activities of the New Car Assessment Program in the United States', Proceedings of 15th International ESV Conference, Melbourne.
      O'Neill B., Lund A.K., Zuby D.S. & Estep C.R. (1996) 'New vehicle crashworthiiness evaluations by the Insurance Institute for Highway Safety', Proceedings of 15th International ESV Conference, Melbourne.

      World Wide Web sites concerning NCAP:

      FIA (Euro-NCAP)  http://www.fia.com/tourisme/safety/safint.htm
      IIHS  http://www.hwysafety.org/
      NHTSA   http://www.nhtsa.dot.gov/cars/testing/ncap/stars.html
      NRMA  http://www.nrma.com.au/crashtests
      OSA  http://www.osa.go.jp/esiken2.html
      VDR  http://www4.tpg.com.au/users/mpaine/roadsafe.html
       

      SAMPLE FROM ANCAP BROCHURE (December 1997)


       

      Dash and doorway deformation

      Road safety index