Dale E. Dawkins, Director
Vehicle Compliance and Safety Affairs
Chrysler Corporation
CIMS 414-01-22
12000 Chrysler Drive
Highland Park, MI 48288-0857

Dear Mr. Dawkins:

This responds to your September 4, 1992 letter, in which you notified the agency of your intention to modify the Hybrid III dummies to be used in your company's certification testing for Standard No. 208. You stated your opinion that the Hybrid III test dummy currently specified in 49 CFR Part 572 Subpart E is unacceptable for determining whether vehicles equipped with two-point automatic belt systems comply with the injury criteria in Standard No. 208. This opinion is based on the chest deflection measurements you have obtained, which you characterize as "erroneous, erratic, and nonsensical."

You indicated your belief that the inaccurate measurements arise because the ball on the end of the slider rod of the chest deflection transducer "frequently" will pop out of its guide track. According to Chrysler, this occurs because the rubber bump stops on the dummy's sternum act as a fulcrum to pry the ball on the end of the slider rod out of its guide track during crash tests, during which the two-point torso belt deflects the dummy chest in both the fore-aft and lateral directions.

Chrysler believes that this problem can be solved simply by modifying the thorax of the Hybrid III dummy by moving the rubber bump stops from the specified location on the sternum to the spine box of the dummy. Your letter indicates your belief that such a modification will not affect the calibration or measurement accuracy of the chest deflection transducer or the biofidelity of the Hybrid III thorax. You indicated that Chrysler intends to make this modification to the Hybrid III dummy thorax and use the modified dummy for Chrysler's certification testing of a 1994 model year vehicle for purposes of Standard No. 208.

NHTSA's position on the issue of what steps manufacturers must take before certifying that their vehicles or equipment comply with an applicable safety standard has been often stated and applies with equal force in this situation. The compliance test procedures set forth in Standard No. 208 must be followed by this agency during our compliance testing. In this instance, that means that NHTSA's compliance testing will be conducted using the Hybrid III test dummy specified in Part 572, Subpart E, with the rubber bump stops in the location specified therein.

Manufacturers certifying compliance with the safety standards are not required to follow exactly the compliance test procedures set forth in the applicable standard. In fact, manufacturers are not even required to conduct any actual testing before certifying that their products comply with applicable safety standards. However, to avoid liability for civil penalties in connection with any noncompliance that may be determined to exist, the National Traffic and Motor Vehicle Safety Act (the Safety Act) requires that the manufacturer exercise "due care" to assure compliance and in making its certification. It may be simplest for a manufacturer to establish that it exercised "due care" if the manufacturer has conducted testing that strictly followed the compliance test procedures set forth in the standard. However, "due care" might also be shown using modified test procedures if the manufacturer could demonstrate that the modifications were not likely to have had a significant impact on the test results. In addition, "due care" might be shown using engineering analyses, computer simulations, and the like.

In this case, then, the relevant issue is not whether Chrysler uses a modified version of the Hybrid III test dummy for its certification testing. Instead, the issue is whether Chrysler can show that it would exercise "due care" despite using a modified Hybrid III test dummy for its certification testing. This agency does not make any determinations as to whether "due care" has been exercised except in the context of an enforcement proceeding. Hence, Chrysler will have to make the initial decision as to whether a certification based on the modified Hybrid III dummy could be made in the exercise of "due care." Your letter states that Chrysler's modifications to the test dummy "will not affect the calibration or measurement accuracy of the chest deflection transducer or biofidelity of the Hybrid III thorax." If Chrysler can demonstrate that these statements are correct, Chrysler may be able to show that it exercised "due care" in connection with the use of the modified Hybrid III dummy.

Of course, a manufacturer that can show it exercised due care would still be subject to the statutory obligation to notify owners and remedy any vehicles that are determined not to conform to Standard No. 208. However, this same obligation would apply even to manufacturers that had conducted testing using an unmodified Hybrid III test dummy.

Turning from Chrysler's particular question to the standard in general, NHTSA notes that it has undertaken research examining the issue of chest deflection measurements by the Hybrid III test dummy. An extensive discussion of the agency research and analysis of this issue may be found in the interim final rule postponing the use of the Hybrid III dummy in vehicles that do not use either seat belts or air bags (55 FR 39280; September 26, 1990; copy enclosed). In that notice, NHTSA indicated that it had received significant data from sources outside the agency, including General Motors, Mercedes-Benz, Toyota, INRETS (a French government research and development group), and the Motor Industry Research Association (a British group).

The agency would be very interested in examining any data or test results that Chrysler may have on this subject, especially the assertion that the ball on the end of the chest deflection rod comes out of its track during crash test conditions. Please send all such information to: Barry Felrice, NHTSA Associate Administrator for Rulemaking, Room 5401, 400 Seventh Street, S.W., Washington, D.C. 20590.

I hope this information is helpful. Please let me know if you need any further information or have some further questions on this subject.


Paul Jackson Rice Chief Counsel


ref:208#572#VSA d:10/2/92