IV. TEST DATA AND ANALYSIS OF TEST DATA

This chapter presents test data available to the agency on the various static and dynamic test procedures proposed or considered for advanced air bags. The test data, and analysis of the test data, are presented in the following format.

1. Static Tests (Out-of-Position)

1. Driver - 5th Percentile Female Dummy (MY99, MY98 and Pre-MY98 data available)
1. Position 1 (chin on module)
2. Position 2 (chest on module)
2. Passenger-side OOP
1. RFCSS 12-Month-Old Infant (CRABI) dummy
2. 6- Year-Old Child Dummy (MY99, MY98 and Pre-MY98)
(1) Position 1 (chest on module)
(2) Position 2 (head on module)
2. Vehicle Tests (In-Position)

1. Belted Tests
1. 56 kmph (35 mph), 0 degree, Belted Barrier Test
(1) 50^th Male
2. 48 kmph (30 mph), 0 degree, Belted Barrier Test
(1) 50^th Male.
(2) 5^th Female
3. 48 kmph (30 Mph) +/- 30 Degree (L or R) Oblique Belted Barrier Test
(1) 50^th Male
(2) 5^th Female (shown for comparison purposes only)
4. 40 kmph (25 mph), Offset Deformable Barrier (ODB), 40% Overlap, Belted Test
(1) 5^th Female
5. 60 kmph (37.5 mph), ODB, 40% Overlap, Belted Test (with Tibia Index data)
(1) 50th Male
(2) 5^th Female
2. Unbelted Tests
1. 40 kmph (25 mph), 0 degree, Unbelted Barrier Test
(1) 50^th Male.
(2) 5^th Female
2. 48 kmph (30 mph), 0 degree, Unbelted Barrier Test
(1) 50^th Male.
(2) 5^th Female
3. 48 kmph (30 mph) +/- 30 Degree (L or R) Oblique Unbelted Barrier Test
(1) 50^th Male
4. 56 kmph (35mph) ODB, 40% Overlap, Unbelted Test (without Tibia Index data)
(1) 50th Male
(2) 5^th Female
3. Summary of Pass Rates by Proposed Test Procedure
4. Test Procedure Stringency

1. Out-of-Position Static Test Procedures
1. 5^th Percentile Female Dummy
2. 6-Year-Old Child Dummy
2. In-Position Dynamic Test Procedures
1. 48 kmph (30 mph), 30 deg oblique unbelted FRB vs 40 kmph (25 mph) unbelted FRB vs 56 kmph (35 mph) unbelted ODB vs 48 kmph (30 mph) unbelted FRB
2. Left vs Right oblique
3. Left vs Right 35 mph ODB
4. 50^th Percentile Male vs 5^th Percentile Female Dummy Dynamic and Compliance Equivalency

A. Static Tests, Out-of- Position (OOP)

A.1.a. & b. Driver 5^th Percentile Female Dummy

Static deployment Positions 1 and 2 tests were conducted using the 5^th percentile female dummy representing the driver-side. These positions were the same as those proposed in the NPRM where in Position 1 the dummy's chin is placed on the inflator module and Position 2 the dummy's chest is placed on the inflator module. These same positions are proposed in the SNPRM. Table IV-1 shows that HIC₁₅, chest g's and chest deflection are not problem areas for all tests, whereas Nij is a problem area for both Position 1 and Position 2 testing based on MY98 and pre-MY98 vehicles. Compared to pre-MY98 vehicles, the magnitude of Nij has decreased in MY98 and MY99 and the Nij Pass Rates have subsequently improved. In MY99, the Nij Pass Rate improved to the 67 percent level for Position 1 and to the 100 percent level for Position 2.

Tables IV-2 and IV-3 show the individual Nij values by make/model/year used for the averages shown in Table IV-1.

Table IV-4 shows the individual chest deflection, chest g's and Nij values by make/model for 1996 versus 1998. Table IV-5 shows the same responses for individual MY99 make/models.

Passenger-Side OOP

Convertible Child Safety Seat and 12-Month-Old Infant (CRABI) Dummy

The new 12-month-old infant (CRABI) test dummy was sled tested per FMVSS No. 213, Child Restraint Systems, in a convertible child safety seat (can be used either forward facing or rearward facing) with, and without, an air bag present. Test results are organized by common test condition in Table IV-6a. The 213 sled test crash pulse was employed. The SNPRM proposes a Low Risk Deployment Test such as proposed in the NPRM or suppression. The sled tests with air bag deployments used a 1997 Ford Taurus or a 1998

Table IV-1
Summary of OOP Results by Model Year
Driver-5th Percentile Female Dummy
Numbers in (   ) indicate Pass Rate in Percent

Table IV-2
Static Out-of Position Driver Tests, Position 1, 5^th Percentile Female Dummy
Max Nij

Table IV-3
Static Out-of Position Driver Tests, Position 1, 5^th Percentile Female Dummy
Max Nij

Table IV-4
Static Out-of Position Driver Tests, Position 2
5^th Percentile Female Dummy
Chest and Neck Measurements

Table IV-5
Static OOP Driver Test Position 2
5^th Percentile Female Dummy

Ford Explorer air bag module. The HICs measured were an order of magnitude lower using the 12-month-old CRABI dummy compared to previous VRTC tests (shown in the August 1998 PEA) using a 9-month-old dummy. Overall Pass Rates are zero percent for each test condition which are summarized in Table IV-21. In the SNPRM, the manufacturers have a suppression option for meeting this static OOP test.

Tables IV-10, 11 and 12 in the Preliminary Economic Assessment (PEA) August, 1998 contained baseline 213 sled tests with a 9 month-old dummy as well as tests with air bags (mid-mount and top mount). The maximum HIC₃₆ using mid-mounted air bags ranged from 2,000-3,000 and the top mounted air bag HIC₃₆ was as high as 3,015. The new VRTC test series involving the 12-month-old CRABI dummy shows that HICs have been reduced by about an order of magnitude despite being computed over 15 ms rather than 36 ms. The test conditions in this test series were identical to those referenced in the PEA.⁽¹⁾

Tables IV-6a
FMVSS No.213 Sled Tests
12-Month-Old Infant (CRABI) Responses

R = child seat rearward facing. F = child seat forward facing.
1./ 1997 Ford Taurus and 1998 Ford Explorer air bag modules.
2./ For Test Conditions, see VRTC Report on the CRABI, Docket No. 99-5156-6.
3./ #3 and #4 are mid-mount air bags (M). #7 and #8 are top mount air bags (T).

NHTSA also conducted a 12-month-old infant (CRABI) low risk deployment test using new inflator technology, namely - the 1^st stage of an experimental (confidential make/model) dual stage inflator. The vehicle's passenger-side seat was placed full-forward with the seat back set at a 30 degree angle. The top center of a Century rear facing child safety seat (RFCSS) was aligned with the geometric center of the air bag. The center line of the RFCSS was aligned with the longitudinal center line of the test vehicle. The vehicle safety belts were cinched to secure the RFCSS. The CRABI dummy was belted in the Century RFCSS. The first stage of an experimental dual power experimental air bag was statically deployed.⁽²⁾ Table IV-6b shows that all the responses of the 12-month-old infant dummy were low and passed the proposed ICPLs.

Table IV-6b
Low Risk Deployment Test (n=1) with the 12-Month-Old Infant (CRABI) Dummy
1^st Stage of Experimental Dual Power Inflator

6-Year-Old Child Dummy (Passenger-side)

The static deployment OOP Positions 1 and 2 (0 mm clearance) were tested using the 6-year-old child dummy. Positions 1 and 2 were the same as those proposed in the NPRM such that in Position 1 the chest is placed on the module, whereas for Position 2, the head is placed on the module. These same positions are proposed in the SNPRM. As shown in Table IV-7, HIC₁₅ is not a problem in either Position 1 or Position 2 tests using MY99 vehicles or test bucks. Model year 1998 and pre-model year 1998 vehicles are matched. MY98 and MY99 vehicles are not matched by make/model. Nij, chest g's, and chest deflection, however, were problems using Position 1 and Position 2 position procedures based on MY98 and 99 vehicles or sled test bucks. Position 2 tests were not conducted in MY98 or pre-MY98. HIC₁₅, chest g's and chest deflection have improved with the new redesigned air bags. Similarly, Nij values have improved, but there is still a low Nij Pass Rate in MY98 and 99 vehicles.

Tables IV-8a and IV-8b contain the specific 6-year-old dummy responses by make/model for MY96, MY98 and MY99 for static OOP Position 1 with zero clearance. Tables IV-8b and IV-8c show a 100 percent Pass Rate for OOP Positions 1 and 2 for the Acura 3.5RL using the first stage of a two stage inflator.

Table IV-7
Summary of OOP Results by Model Year
Passenger 6-Year-Old Child Dummy
Number in (   ) indicates Pass Rate in Percent

Table IV-8a
Static Out-of-Position Tests with a 6-Year- Old Child Test Dummy
Position 1 @ 0 mm Clearance from the Air Bag Module

Bold Numbers exceed proposed ICPL values.

N.A. indicates data Not Available

Table IV-8b
Static Out-of-Position Tests With a 6-Year-Old Child Dummy
Position 1 @ 0 mm Clearance from the Air Bag
Module for MY 99

Bold Numbers exceed proposed ICPL values.

* Stage 1 fired and then Stage 2 fired with a 40 ms delay.

** Stage 1 fired only .

Table IV-8c
Static Out-of-Position Tests With a 6-Year-Old Child Dummy
Position 2 @ 0 mm Clearance from the Air Bag
Module for MY 99

Bold Numbers exceed proposed ICPL values.

* Stages 1 fired and then Stage 2 fired with a 40 ms delay.

** Stage 1 fired only.

3-Year-Old Child Dummy Static OOP Tests

The SNPRM proposes the same OOP Position 1 and 2 static deployment tests as proposed in the NPRM. The agency did not conduct new, post-August 1998 PEA, Position 1 and 2 tests using the 3- year-old child dummy because of time and resource constraints. The agency believes that if a 6-year-old dummy fails the OOP tests, it is likely that the 3-year-old dummy will also fail. However, if the 6- year-old dummy passes the OOP tests, there is no guarantee the 3-year-old will pass.

B. Full-Scale Vehicle Tests (In-Position)

Belted Test Procedures

56 kmph (35 mph), 0 Deg., Barrier Test, Belted, 50^th Percentile Male Dummy

For a limited set of matched pairs (n=14) as shown in Table IV-9a, MY99 and MY98 responses are not significantly different, except for Nij which was lower by 12-14 percent for the driver and passenger, respectively, for MY99. Given the similarity of the two model years, NHTSA has combined MY98 and MY99 vehicles into a "redesigned" air bag group for further analysis. For Pass Rates in Table IV-21, for example, MY98 and MY99 have been combined.

Table IV-9a
NCAP Test Results
Average NCAP Test Results for Matched Make/Models
Belted @ 56 kmph (35 mph), 50^th Percentile Male Dummy
MY 99 vs MY 98

All average responses rounded to the nearest whole number.

*MY 98 had two Ford Windstar Tests, whereas MY 99 had one Windstar.

Comparing MY99 make/models to matched pre-MY98 make/models, as shown in Table IV-9b, indicates that HIC₁₅ decreased 14 and 47 percent and chest g's increased slightly by about 3 g's for both the driver and passenger, respectively. Nij and chest deflection responses were mixed with no clear trend. Table IV-9b is based on matching MY98 and pre-MY98 make/models to MY99 make/models.

Table IV-9b
NCAP Test Results
Average NCAP Test Results for Matched Make/Models
Belted @ 56 kmph (35 mph), 50th Percentile Male Dummy
MY 99 vs Pre-MY 98

Table IV-9c is based on matching MY98 make/models and pre-MY98 make/models. Table IV-9c shows very little difference between responses for MY98 and pre-MY98 vehicles, on the average, for the belted 56 kmph (35 mph) fixed rigid barrier test condition.

Table IV-9c
NCAP Test Results
Average NCAP Test Results for MY98 vs Pre-MY98 (Matched Make/Models*)
Belted @ 56 kmph (35 mph), 50^th Percentile Male Dummy

Tables IV-9d and IV-9e compare NCAP results by model year for the driver and passenger, respectively, and show the same comparisons as above based on All Vehicles in the file, rather than matched make/models. Except for passenger Nij, there has been no substantive change year-to-year, on the average. Femur axial loads do not appear to be a problem. For the passenger-side, there has been a downward trend in Nij from 1996 to 1999. The values in parentheses are Pass Rates (%) to assess the effect of applying the 50^th percentile ICPL values proposed in the SNPRM to NCAP. For example, if the decision was made to extend the proposed ICPLs to the NCAP program, driver and passenger side failures would occur, but at the same rate as earlier years. The agency conducted several 56 kmph (35 mph) belted crash tests (n=2) using the 5^th percentile female dummy using a 1988

Table IV-9d
NCAP Test Results
56 kmph (35 mph) NCAP Average Responses & Pass Rates x Model Year
Belted, 50^th Percentile Male Dummy
DRIVER (All Vehicles)

Table IV-9e
NCAP Test Results
56 kmph (35 mph) NCAP Average Responses & Pass Rates X Model Years
Belted, 50^th Percentile Male Dummy
PASSENGER (All Vehicles)

and 1993 Ford Taurus. As shown in Tables IV-9f, there was a HIC₁₅ failure for the 1988 Ford Taurus and an Nij failure for the 1993 Ford Taurus.

Table IV-9f
NCAP Test Results
Belted @ 56 kmph (35 mph), 5th Percentile Female Dummy
1988 & 1993 Ford Taurus (NHTSA data)

Bold Numbers indicate test values exceeded the proposed ICPL.

R indicates right femur had maximum axial load.

* 1988 Ford Taurus did not have air bags, whereas the 1993 Ford Taurus had driver and passenger air bags.

48 kmph (30 mph), 0 Deg., Barrier Test, Belted, 50^th Percentile Male Dummy

Based on the 1997-98 NHTSA/Transport Canada test program, the 48 kmph (30 mph) belted 50^th percentile male responses in the August, 1998 PEA consisted of 3 driver (MY98 + pre-MY98 vehicles) and 7 passenger (MY98 + pre-MY98 vehicles) test points with 100 percent Pass Rates. If adjusted to the HIC₁₅ and Nij, it is believed the Pass Rates for all responses would remain at 100 percent. Based on the 1998-99 NHTSA/Transport Canada test program using 18 - MY 1999 test vehicles, after applying the new injury criteria and ICPL values, there were no Nij or other response failures. Although confidential compliance data received from [name confidential] was incomplete relative to the proposed ICPLs in the SNPRM (e.g., contained driver and passenger HIC₃₆ and chest g's data only) there was a 100 percent Pass Rate for driver and passenger side for a large sample of pre-MY98 and a few MY98 [name confidential] make/models. These Pass Rates are reflected in Table IV-21, Summary of Pass Rates by Proposed Test Procedure.

48 kmph (30 mph), 0 Deg., Barrier Test, Belted 5^th Percentile Female Dummy

Table IV-10 shows that for the same test condition as above, the 5^th percentile female dummy would experience Nij failures on the driver-side and Nij and chest g's failures on the passenger side.

Table IV-10
48 kmph (30 mph) Belted Barrier, 5^th Percentile Female Dummy
MY 98 and 99 Combined, Average Responses

Note: Table IV-10 contains NHTSA/Transport Canada test data. For the passenger side, the air bag did not deploy for either the 1998 Honda Civic or the 1999 Hyundai Accent as these test vehicles were purchased without passenger-side air bags.

48 kmph (30 mph), +/-30 Degree Oblique (L or R), Belted Test

In the SNPRM, NHTSA is proposing an "up-to- 48 kmph (30 mph)," +/- 30 degree (L or R) oblique belted test for the 50^th percentile dummy. The tests shown in Table IV-11 are identical to the current FMVSS 208 procedure. Two 1999 Dodge Intrepids were tested in a left-side (driver's side) oblique test, one with 50^th percentile dummies and the other with 5^th percentile dummies. Although only the 50^th percentile dummy is being proposed for use in the belted oblique tests, the 5^th percentile female dummy data is shown for comparison purposes. This is the same test as currently required by FMVSS 208 for the 50^th percentile male dummy, except the SNPRM proposes to add HIC₁₅ and Nij. The responses were very low resulting in a 100 percent Pass Rate for both dummies. This is a very benign test because (1) an oblique impact angle crushes soft sheet metal resulting in a soft crash pulse, (2) the dummies were belted and do not normally move out-of-position, and (3) the air bag "fire time" was very timely, similar to a full frontal fixed rigid barrier test. The maximum femur load recorded was 6,376 N for the right femur of the 50^th percentile driver dummy, well below the ICPL maximum of 10,000 N. [Name confidential] provided 30 mph, +/-30 degree oblique (left and right) belted test data based on HIC₃₆ and chest g's responses for the driver and passenger 50^th percentile male dummy. The [name confidential] make/models in this confidential data set were predominantly pre-MY98 with a few MY98 make/models. Although the Pass Rate was 100 percent, a full set of responses equivalent to the SNPRM were not available for review.

Table IV-11
30 mph (48 kmph) , +/-30 Degree Oblique (Left shown), Belted
50^th %-ile Male and 5^th%-ile Female
1999 Dodge Intrepid (n=1 test per dummy size)

NHTSA/Transport Canada cooperative research data.

40 kmph (25 mph), ODB, 40 Percent Overlap, Belted, 5^th Percentile Female Test Dummy

NHTSA is proposing a new test procedure in the SNPRM which is the same as was proposed in the NPRM, namely a 40 kmph ( 25 mph), offset deformable barrier (ODB) test, with 40 percent overlap, using a belted 5^th percentile female test dummy. As shown in Table IV-12, HIC₁₅, chest g's and chest deflection did not exceed the proposed ICPL values on the driver's side. However, Nij did exceed the proposed ICPL value of 1.0 on the driver's side (3/5 or 60% for MY99 passed and 8/14 or 57% passed for MY98). On the passenger-side, there were no HIC₁₅ , chest g's or chest deflection failures. There was a 100 % pass rate for MY99 vehicles based on Nij and an 86% (12/14) pass rate for MY98 vehicles based on Nij. [There was one Right side 40 kmph (25 mph) ODB, 40% overlap, belted, test conducted by NHTSA/Transport Canada using a 1999 Ford Crown Victoria, but this test was excluded from this analysis.]

Table IV-12
40 kmph (25 mph) ODB, 40% Overlap, Belted, Left, 5^th Percentile Female Dummy
Average Responses for MY98 and MY99 Combined
(   ) indicates Pass Rate in Percent.

NHTSA/Transport Canada cooperative research data.

Note: n=19 vehicles. 14- MY1998 vehicles and 5 MY1999 vehicles. Table IV-21, Pass Rates, shows n=20 for the same crash test procedure as Left side (n=19) and right side (n=1 MY99 vehicle) impacts were combined. Also, note on the passenger-side, air bags did not deploy for the 1999 VW Beetle or the 1999 Toyota Camry.

60 kmph (37.5 mph), Offset Deformable Barrier, 40% Overlap, Belted Tests

NHTSA is proposing in the SNPRM an 35-56 kmph (22-35 mph), unbelted, offset deformable barrier (ODB) test procedure that may be conducted on the left or right front of the vehicle. Tests with both 50^th and 5^th percentile dummies would be required. The ODB test at 56 kmph (35 mph) produces lower toe pan intrusion and lower potential for debilitating lower leg injury than a test at 60 kmph (37.5 mph). The European lower leg injury threshold has a Tibia Index with an IARV of 1.3 for both dummies. The Tibia Index was discussed briefly in Chapter III, Injury Criteria. It is being used for analysis purposes and is not part of the SNPRM proposal. Table IV-13 shows belted results for the 50^th percentile male dummy (both driver and passenger sides), however, at a speed slightly higher than the SNPRM maximum (37.5 mph) and belted. The agency's rationale being -- if they fail the ODB test belted, they would most probably fail the same test unbelted. Table IV-13 shows that despite a

Table IV-13
60 kmph (37.5 mph), 40% Overlap, Left, Offset Deformable Barrier Test Comparison
Belted, 50^th Percentile Male Dummy including Tibia Index

Bold Numbers exceed proposed ICPL values.

U.T. = Upper Tibia Index (L and R shown) or ICPL =1.3

L.T. = Lower Tibia Index (L and R shown) or ICPL = 1.3

ND - No data (either data was lost or lacked instrumentation).

100 percent femur axial load Passing Rate (both driver and passenger) for the 50^th percentile male dummy, there was lower leg injury potential as evidenced by the Tibia Index values exceeding 1.3 in 50 percent of the test vehicles. The predominant left maximum femur load reflects a near-side impact. The above data shows, for the driver-side, an 80 percent Pass Rate for Nij alone and a 50 percent Pass Rate. As these are left side or driver side offset tests, the passenger-side had a 100 percent Pass Rate. The overall Pass Rates are summarized in Table IV-21, Summary of Pass Rates by Proposed Test Procedure. However, the 35 mph unbelted ODB test procedure proposed in the SNPRM would result in a somewhat lower Pass Rate. Table IV-14 below shows similar results for the 5^th percentile female dummy.

Table IV-14
60 kmph (37.5 mph), 40% Overlap, Left, Offset Deformable Barrier Test Comparison
Belted, 5^th Percentile Female Dummy incl. Tibia Index
NHTSA data

Bold Numbers exceed proposed ICPL values. U.T. = Upper Tibia Index. L.T.= Lower Tibia Index

As shown in Table IV-14, using the 5^th percentile dummy in the belted 60 kmph (37.5 mph) ODB (left) test procedure, both the driver and passenger had problems passing Nij. Driver and passenger Pass Rates based on Nij alone were 16.7 percent and 50 percent, respectively. The Tibia Index was also a problem on for the driver-side, or the seating position on the impacted side of the test vehicle. The Pass Rate based on the Tibia Index alone was 16.7 percent for the driver and 100 percent for the passenger-side, despite the fact that the driver and passenger femur axial loads had a Pass Rate of 100 percent. The maximum femur axial loads occurred on the impacted, left side of the test vehicle.

Table IV-14a compares the responses of the 50^th and 5^th percentile dummies for the same MMY test vehicles. Table IV-14b compares the Tibia Index values by dummy size for the same MMY test vehicles. Combining Table IV-14a and IV-14b responses, the passenger-side dummy (50^th or 5^th) passed almost all ICPL and IARV values, except for Nij in one case (1998 Chevy Venture, 5^th %-ile dummy). As expected, the left-side ODB impact affects the driver dummies as they had higher responses. Counting the driver and passenger responses for all three test vehicles and both dummies (n=48), the 5^th percentile female dummy exceeded ICPL values 2.6 times more often as the 50^th percentile dummy (27.1 % vs 10.42 %). Overall, the decision regarding compliance (P or F) using either of the two dummies was identical, with the 5^th percentile dummy exceeding the lower leg IARV values more frequently. This is reasonable considering the fact that the 5^th percentile female dummy's seated position is with the seat full-forward, whereas with the 50^th percentile male dummy the seat is positioned mid-track. Based on Tables IV-14a and IV-14b, both dummies appear to behave with dynamic equivalency (similar responses for the same test condition) with a few exceptions.

Table IV-14a
60Kmh (37.5 mph), ODB, 40% Overlap, Left, Belted by Dummy Type

MMY = Make/Model/Year

Table IV-14b
60 kmph (37.5 mph), ODB, 40% Overlap, Left, Belted
Tibia Index Value Comparisons by Dummy Type

European Tibia Index IARV is 1.3

MMY = Make/Model/Year

Unbelted Test Procedures

40 kmph (25 Mph) Unbelted Barrier Test, 50^th Percentile Male and 5^th Percentile Female Dummies

Tables IV-15a shows that for the 5 vehicles tested, the 50^th percentile male and 5^th percentile female dummies did not exceed any of the proposed ICPL values on the driver-side, whereas Table IV-15b

shows the only passenger-side ICPL exceeded for both dummies was the Nij. This occurred for the same test vehicles - namely the 1999 Toyota Tacoma.

Table IV-15a
40kmh (25 mph), 0 degree, Unbelted by Dummy Type
DRIVER

* Single stage inflator

** 1^st stage of dual stage inflator.

Table IV-15b
40kmh (25 mph), 0 degree, Unbelted X Dummy Type
PASSENGER

* Single stage inflator.

** 1^st stage of inflator used.

48 kmph (30 Mph) Unbelted Barrier Test (Matched Pairs MY99 vs Pre-MY98 and Matched Pairs MY98 vs Pre-MY98)

The SNPRM is proposing an "up-to-48 kmph (30 mph)," unbelted, fixed rigid barrier test for the 50^th and 5^th percentile dummies. Pre-MY1998 vehicles were required to meet this test using the 50^th percentile male dummy. In addition to the 5^th percentile female dummy being added to the test, HIC_15, Nij and revised chest deflection ICPLs are also be proposed in the SNPRM. Tables IV-16a and IV-16b show the average responses from the 30 mph test condition as required for the 50^th percentile male dummy using the SNPRM proposed injury criteria. For the redesigned air bags, there was little difference between the MY98 and MY99 responses. For purposes of analysis, MY98 and MY99 vehicles were not necessarily matched make/models.

Table IV-16a
Summary of 48 kmph (30 mph) Unbelted Barrier Tests, 50^th Percentile Dummy
Average Responses

* Sample sizes: MY99 - There were 6 - MY99 (VRTC). MY98 - There were 7 - MY98 (VRTC), one 1998 Ford Escort and one confidential make/model.

48 kmph (30 mph), Unbelted Barrier Test, 5^th Percentile Female Dummy

The SNPRM proposes an "up-to- 48 kmph (30 mph)" unbelted barrier test for the 5^th percentile female dummy. The test data in Table IV-16b represents the 48 kmph (30 mph) test condition for 1-MY1998, 3 - MY1999, 1 - confidential MM1999 and 3 confidential make/model/year. Nij and chest deflection were problems for the driver and Nij and chest g's were a problem for the passenger-side.

Table IV-16b
Summary of 48 kmph (30 mph) Unbelted Barrier Tests, 5^th Percentile Female Dummy
Average Response of Test Vehicles
(   ) indicate Pass Rate %

The driver sample (n=8) includes 1998 Ford Taurus from the August, 1998 PEA, 1999 Saturn SL1, 1999 Dodge Intrepid, 1999 Toyota Tacoma, and 1- confidential MY99 and 3 confidential MMY are included in this table. The passenger sample (n=7) employed only 2 of the above confidential MMY tests.

48 kmph (30 Mph), +/-30 Degree (L or R) Oblique, Unbelted Test, 50^th %-ile Male & 5^th Percentile Female Dummies

The SNPRM proposes a test speed of "up-to-48 kmph (30 mph)," for a +/- 30 degree (L or R) oblique unbelted test procedure for the 50^th percentile dummy. This test at 48 kmph (30 mph) is already required by FMVSS 208 using the 50^th percentile male dummy, except that HIC₁₅ and Nij are being proposed in the subject SNPRM. Oblique tests with the 5^th percentile female dummy are not proposed in the SNPRM, but response data is shown in Tables IV-17 and IV-18 for comparison purposes. As shown in Tables IV-17 and IV-18, driver and passenger responses were benign for the 50^th percentile male dummy with a 100 Pass Rate for the 30 mph, 30 degree oblique unbelted test. [Similarly, driver and passenger responses were benign for the 5^th percentile female dummy, except for the driver-side Nij for the 1999 Dodge Intrepid resulting in a 75 percent Pass Rate (considering all 4 test vehicles) for the 30 mph, 30 degree oblique, unbelted test. The maximum femur axial load occurred predominantly on the impacted or near-side and were not a problem in these tests. It is difficult to detect any significant differences in stringency between left or right side impacts for either dummy size. For the same MMY test vehicle, the impact-side dummy appears to have the higher responses, but responses were mixed.

Table IV-17
48 kmph (30 Mph), +/-30 Degree Oblique (L or R), Unbelted Barrier Test
50^th Percentile Male & 5^th Percentile Female Dummies
DRIVER

* Single stage inflator.

** Only the 1^st stage of a dual stage inflator fired.

Table IV-18
48 kmph (30 mph), +/-30 Degree Oblique, Unbelted 50^th Percentile Male & 5^th Percentile
Female Dummies
PASSENGER

* Single stage inflator.

** Only the 1^st stage of a dual stage inflator was fired.

Given the range of dummy response variability observed in other crash tests, these test vehicles would still be expected to pass if typical variations occurred as in other tests. This appears to be a very benign test because (1) an oblique impact angle involves a lot of crushed soft sheet metal and a soft crash pulse and (2) the "fire time" was very timely, similar to a full frontal fixed rigid barrier. [Name confidential] provided 48 kmph (30 mph), +/-30 degree oblique (L+R) unbelted 50^th percentile male dummy compliance data (confidential), but this was an incomplete data set (e.g., included driver and passenger HIC 36 and chest g's only). The [name confidential] data was predominantly pre-MY98 and a few MY98 make/models.

56 kmph (35 mph), ODB Test Procedure, Unbelted 50^th %-ile Male and 5^th Percentile Female Option 2, Full-Scale Test Procedures, proposes an unbelted, 35-56 kmph (22-35 mph), offset deformable barrier ( ODB) test procedure (L or R) using the unbelted 50^th percentile male and 5^th percentile female dummies. The following data is representative of the expected responses, but Tibia response data was not available for these tests.

Table IV-19
56 kmph (35 mph), ODB, Unbelted, 50^th Percentile Male & 5^th Percentile Female
Dummy Comparisons
DRIVER

Bold Numbers exceed proposed ICPL values.

No Tibia response data available for these tests.

* Single stage inflator.

** Stages 1 + 2 of dual stage inflator were fired.

Table IV-20
56 kmph (35 mph), ODB, Unbelted 50^th Percentile Male & 5^th Percentile Female
Dummy Comparisons
PASSENGER

Bold Numbers exceed proposed ICPLs.

No Tibia response data available for these tests.

* Single stage inflator. ** Stages 1 + 2 of dual stage inflator were fired.

Based on Tables IV-19 and 20, for the 1999 Dodge Intrepid driver-side using either the 50^th percentile male or 5^th percentile female dummy, both failed Nij. For the 1999 Toyota Tacoma driver-side using either dummy, both passed. For the 1999 Dodge Intrepid passenger-side, with left side impact, the 50^th percentile male passed and the 5^th percentile female dummy failed Nij. For the 1999 Toyota

Tacoma passenger-side using either dummy, both passed. For the 1999 Saturn SL1, the 5^th percentile driver failed chest deflection.

Left vs Right ODB Test - A comparison of left-side versus right-side impact responses for the 5^th percentile female dummy was conducted. The left-side impact was more stringent based on the driver-side for the 1999 Dodge Intrepid, where as L or R were almost identical based on the passenger-side of the 1999 Toyota Tacoma. Based on passenger-side responses for the 1999 Dodge Intrepid, the left-side impact was more stringent, whereas the right-side impact was more stringent for the 1999 Toyota Tacoma based on the passenger-side responses.

50^th vs 5^th driver and passenger - For the same MMY (a 1999 Toyota Tacoma) tested with both size dummies, there did not appear to be much difference in responses in the 50^th vs 5^th percentile dummies for either the driver or passenger sides of the test vehicle.

Near-side vs Far-side Impacts - For the driver-side, 8/10 or 80 percent of the near-side femurs had the highest maximum axial loads, whereas for the passenger-side, 9/10 or 90 percent of the near-side femurs had the highest axial loads.

C. Summary of Pass Rates by Proposed Test Procedure

Table IV-21 summarizes the Pass Rates for each of the full-scale vehicle test procedures and the static OOP test procedures proposed for Options 1 & 2. Static OOP test Pass Rates are also shown for the 5^th percentile female dummy, 6-year-old child dummy and the 12-month-old infant (CRABI) dummy. The agency combined MY98 and MY99 vehicles together, combined left and right oblique tests, or combined left and right ODB test data together, in order to calculate Pass Rates. Because of the very limited number of test vehicles available for the unbelted ODB test condition at 56 kmph (35 mph), two confidential MMY's were combined with three MY1999 vehicles NHTSA had tested to create a sample of 5. For the 25 mph unbelted test Pass Rates, MY1999 (n=3) test results and 1 confidential MMY were combined. Similarly, for the 30 mph unbelted barrier test Pass Rate, using the 5^th percentile female dummy, 1 - MY98 was combined with 4 - MY99 (one of which was confidential) and 3 confidential MMY for a total sample of 8. Also, for the 30 mph, 30 degree, unbelted oblique crash test Pass Rate, several confidential MMYs were combined with NHTSA's tests. Some of the pre-MY98 and MY98+99 Pass Rates shown in Table IV-21 were determined using confidential 208 compliance data. The Pass Rates are used in Chapters VI, Benefits, and VII, Cost and Leadtime, to calculated costs and benefits.

Table IV-21
Summary of PASS RATES by Proposed Test Procedure

Table IV-21 Footnotes are as follows: N.D. - No data available at this time. *CRABI dummy cannot measure deflection. **Suppression or low risk test proposed in the SNPRM. *** 1997 Ford Taurus and 1998 Ford Explorer inflator technology, **** 1^st stage of experimental inflator for static OOP test and ***** Upgraded structural changes may be required.

D. Test Procedure Stringency

Static Test Procedures, Out-of-Position

a. Static OOP Test (Driver-Side) - Position 1 vs Position 2 Stringency based the 5^th

Percentile Female Dummy

Considering that the dummy injury responses are equally weighted, and the limited number of data points available, OOP Position 1 would appear to be more stringent than OOP Position 2 based on Nij and HIC₁₅ , whereas Position 2 would appear to be more stringent than Position 1 based on chest g's and chest deflection. The data supports the idea that these tests are complementary, namely - OOP Position 1 is more of a "worst case" head/neck impact condition, while OOP Position 2 is more of a "worst case" chest impact condition. The agency has a limited number of data points because of resource and manpower constraints.

Static OOP Test (Passenger-Side) - Position 1 vs Position 2 Stringency based the 6-Year-Old Child Dummy

Recognizing the limited number of data points, OOP Position 2 would appear to be more stringent than Position 1 based on the magnitude of HIC₁₅, Nij and chest g's responses.

Static OOP Test (Passenger-Side) - Position 1 vs Position 2 Stringency based the 3-Year-Old Child Dummy

This was discussed earlier. Due to limited time and resources, the agency did not conduct any 3-year-old child dummy static OOP tests.

Dynamic Test Procedures, In-Position

a. Test Procedure Stringency Comparison 40 kmph (25 mph) Unbelted Barrier vs 48 kmph (30 mph) Unbelted barrier vs 56 kmph (35 mph) Unbelted, ODB (Left) ( 50^th & 5^th Percentile Dummies ) based on a 1999 Toyota Tacoma and a 1999 Dodge Intrepid

Methodology

Part I. Two test vehicles were used to assess test procedure stringency, particularly the stringency of the 35 mph ODB unbelted test procedure compared to a 25 mph unbelted FRB and a 30 mph unbelted FRB test. The two vehicles employed by the agency were a 1999 Dodge Intrepid and a 1999 Toyota Tacoma. Both 50^th and 5^th percentile dummies were used in the assessment. Tables IV-22a & 22b show the 50^th percentile dummy driver and passenger responses for the 1999 Dodge Intrepid by test procedure type. Tables IV-22c & 22d show the 5^th percentile dummy driver and passenger responses for the 1999 Dodge Intrepid by test procedure type. Tables IV-23a- 23b show the 50^th percentile dummy driver and passenger responses, respectively, for the 1999 Toyota Tacoma by test procedure type. Tables IV-23c & 23d show the 5^th percentile dummy driver and passenger responses, respectively, for the 1999 Toyota Tacoma by test procedure type. The percentage values shown in parentheses ( ) represents the percent difference between the subject test procedure and the 48 kmph (30 mph) unbelted test procedure. Tables IV-24a & IV-24b are further comparisons for the [confidential MMY] and Tables IV-25a & IV-25b are for the [confidential MMY], using the 50^th percentile male dummy, but were not used to rank the test procedures.

Table IV-22a
50^th Driver - 1999 Dodge Intrepid

All percentages calculated with respect to the 48 kmph (30 mph) unbelted barrier test condition.

Bold Numbers exceed proposed ICPLs.

Table IV-22b
50^th Passenger - 1999 Dodge Intrepid

All percentages calculated with respect to the 48 kmph (30 mph) unbelted barrier test condition.

Table IV-22c
5^th Driver - 1999 Dodge Intrepid

All percentages calculated with respect to the 48 kmph (30 mph) unbelted barrier test condition. Bold numbers exceed proposed ICPLs.

Table IV-22d
5^th Passenger - 1999 Dodge Intrepid

All percentages calculated with respect to the 48 kmph (30 mph) unbelted barrier test condition. Bold numbers exceed proposed ICPLs.

Table IV-23a
50^th Driver - 1999 Toyota Tacoma

All percentages calculated with respect to the 48 kmph (30 mph) unbelted barrier test condition. Bold numbers exceed proposed ICPLs.

Table IV-23b
50^th Passenger -1999 Toyota Tacoma

All percentages calculated with respect to the 48 kmph (30 mph) unbelted barrier test condition. Bold numbers exceed proposed ICPLs.

Table IV-23c
5^th Driver - 1999 Toyota Tacoma

All percentages calculated with respect to 48 kmph (30 mph) unbelted barrier test.

Table IV-23d
5^th Passenger -1999 Toyota Tacoma

All percentages calculated with respect to 48 kmph (30 mph) unbelted barrier test.

Bold numbers exceed proposed ICPLs.

Table IV-24a
CONFIDENTIAL

Table IV-24b
CONFIDENTIAL

Table IV-25a
CONFIDENTIAL

Table IV-25b
CONFIDENTIAL

Tables IV-25c shows 30 mph unbelted barrier test data for the [confidential MMY] using the 5^th percentile female dummy, while Table IV-25d shows 30 mph, 30 degree oblique unbelted test data for the [confidential MMY] using the 5^th percentile female dummy. Table IV-25e shows 25 mph unbelted barrier test data for the [confidential MMY] and [confidential MMY] using the 5^th percentile female dummy.

Table IV-25c
CONFIDENTIAL

Table IV-25d
CONFIDENTIAL

Table IV-25e
CONFIDENTIAL

Data Analysis: Test Procedure Stringency Comparison for 40 kmph (25 mph) Unbelted Barrier, 48 kmph (30 mph) Unbelted barrier and 56 kmph (35 mph) Unbelted ODB (Left) (50^th and 5^th %-ile Dummy)

Based on the above tables (excluding the confidential data), Tables 26a & 26b compare the stringency of each test condition by dummy size. The following questions were asked: (1) Combining driver and passenger responses, what percentage of the responses at 30 mph unbelted are more stringent (greater than) than the 25 mph unbelted responses? (2) Combining driver and passenger responses, what percentage of the responses at 35 mph ODB unbelted are more stringent (greater than) than 30 mph unbelted responses? (3) Combining driver and passenger responses, what percentage of responses at 35 mph ODB unbelted are more stringent (greater than) than 25 mph unbelted responses? Table 26a shows the answers to these questions based on "any measurable difference," whereas Table 26b answers the same questions based on "at least a 20 percent difference."

Table IV-26a
40, 48, & 56 kmph (25, 30 and 35 mph) ODB Test Procedure Stringency Comparison
by Dummy Size (Any Difference at All)

Source: Tables IV-22a to IV-22d and Tables IV-23a to IV-23d.

Given a sample of two test vehicles, with responses from all three test procedures (50^th and 5^th percentile dummies), the 35 mph unbelted ODB is less stringent than the 30 mph unbelted barrier test using either dummy as fewer than half (38% and 44% for the 50^th percentile male and 5^th percentile female dummies, respectively) of the responses at 35 mph ODB exceeded those at 30 mph.

On the other hand, 88% and 88% (50^th percentile male and 5^th percentile female dummies, respectively) of the 30 mph responses exceeded the 25 mph responses and 88% and 69% (50^th percentile male and 5^th percentile female dummies, respectively) of the 35 mph ODB responses exceeded the 25 mph responses.

Table IV-26b
40, 48, & 56 kmph (25, 30 and 35 mph) ODB Test Procedure Stringency Comparison
by Dummy Size ( At least +20% Difference )

Source: Tables IV-22a to IV-22d and Tables IV-23a to IV-23d.

Given a sample of two test vehicles, with responses from all three test procedures (50^th percentile and 5^th percentile dummies), the 35 mph unbelted ODB appears to be less stringent than the 30 mph unbelted barrier test using either dummy as fewer than half (19% and 38% for the 50^th and 5^th %-ile dummies, respectively) of the responses at 35 mph ODB exceeded those at 30 mph by at least 20%.

On the other hand, 63 % and 63 % (50^th percentile male and 5^th percentile female dummies, respectively) of the 30 mph responses exceeded the 25 mph responses by at least 20 % and 69 % and 56 % (50^th percentile male and 5^th percentile female dummies, respectively) of the 35 mph ODB responses exceeded the 25 mph responses by at least 20 %.

Conclusion - Based on the above data sets (Tables IV-26a & 26b), and all dummy responses, the agency concluded that the 35 mph ODB 40 % overlap, unbelted test is less stringent than the 30 mph unbelted barrier test but more stringent than a 25 mph unbelted barrier test.

NHTSA also compared crash pulses of a 56 kmph (35 mph) offset deformable barrier (ODB) test with 40 kmph (25 mph) and 48 kmph(30 mph) rigid barrier tests using a 1999 Dodge Intrepid and a 1999 Toyota Tacoma. There is a long duration, low magnitude acceleration that is associated with the crushing of the honeycomb barrier facing during the ODB test. After the crushing of the deformable face takes place, the remaining segment of the crash pulse is similar to that of a 40 or 56 kmph (25 or 35 mph) fixed rigid barrier test and this portion of the acceleration profile generally would fall between the two rigid barrier pulses, if adjusted with a time shift. Therefore, the agency has concluded that the 56 kmph (35 mph) ODB equivalent velocity is approximately 20 percent lower than a 56 kmph (35 mph) full frontal barrier speed. That is, a 56 kmph (35 mph) offset test is approximately equal to a 45 kmph (28 mph) rigid barrier test. The crash pulse analysis is consistent with, and supports, the above dummy response analysis.

Part 2. 48 kmph (30 mph), +/- 30 degree (L or R) Oblique, Unbelted Test Procedure Stringency Comparison

A further comparison of the above test procedures was made for the 48 kmph (30 mph), 30 degree oblique unbelted test procedure using the 1999 Dodge Intrepid and response data from the 50^th and 5^th percentile dummies. Tables IV-27a and 27b employ the same 1999 Dodge Intrepid data as above, except 48 kmph (30 mph), 30 degree oblique (L+R) test data has been added for comparison. From this data set, the agency concluded that the 48 kmph (30 mph), 30 degree oblique test was only marginally less stringent than a 25 mph unbelted barrier test.

Table IV-27a
Stringency Comparison by Test Procedure by Dummy Size
1999 Dodge Intrepid
Driver 50^th vs 5^th Percentile Dummies

Bold Numbers exceed proposed ICPL values.

Table IV-27b
Stringency Comparison by Test Procedure by Dummy Size
1999 Dodge Intrepid
Passenger 50^th and 5^th Percentile Dummies

Bold Numbers exceed proposed ICPL values.

Conclusions: Based on both 50^th and 5^th percentile dummy responses, the 35 mph ODB unbelted test is more stringent than the 25 mph unbelted barrier test, but less stringent than the 30 mph unbelted barrier test. The 30 mph 30 degree oblique was only marginally less stringent than the 25 mph unbelted barrier test. Table IV-28 ranks the proposed test procedures from "least" stringent to the "most" stringent.

Table IV-28
Proposed Test Procedure Stringency Rank

The SNPRM proposes both left and right side 48 kmph (30 mph), 30 degree oblique belted impact tests using the 50^th percentile male dummy. This is the same as the current 208 standard except for HIC₁₅, Nij and chest deflection ICPLs. The following [name confidential] 208 compliance data compares the two test directions. Table IV-29a shows a few selected [name confidential] vehicles tested both in the left and right directions, whereas Table 29b compares All Vehicles in the [name confidential] file.

Table IV-29a
L vs R, 48 kmph (30 mph), 30 degree Oblique Impacts, Belted, 50^th Percentile Male Dummy Confidential

Although the number of tests are limited, and recognizing that the data available reflects pre-MY98 make/models, there does not appear to be any significant difference between a left or right 48 kmph (30 mph), 30 degree oblique belted impacts, on the average, when using the 50^th percentile male dummy.

Table IV-29b
L vs R, 48 kmph (30 mph), 30 degree Oblique Impacts, Belted, 50^th Percentile Male Dummy,
Confidential

Although the data available reflect pre-MY98 [name confidential] make/models and a few MY98 models, there does not appear to be any significant difference between a left or right 48 kmph (30 mph), 30 degree oblique belted impact when using the 50^th percentile male dummy based on All Vehicles. On the average, the left-side impact responses were higher numerically in all cases, except passenger-side chest g's. This suggest that the left-side oblique belted test condition might be slightly more stringent, on the average, compared to the right-side belted oblique test condition. Because the left impact and right impact samples of make/models were not necessarily matched, the higher numerical values could have been due to the differences in the make/models selected, rather than the side of the impact.

Left vs Right 48 kmph (30 mph), 30 degree Oblique, Unbelted Test Stringency

The SNPRM proposes both left and right side 48 kmph (30 mph), 30 degree oblique unbelted impact tests using the 50^th percentile male dummy. This is the same as the current 208 standard, except for the new proposed HIC₁₅, Nij and chest deflection ICPLs. The following [name confidential] 208 compliance data compares the two test directions. Table IV-30a shows a few selected [name confidential] vehicles tested both in the left and right directions, whereas Table 30b compares All Vehicles in the [name confidential] file.

Table IV-30a
L vs R, 48 kmph (30 mph), 30 degree Oblique Impacts, Unbelted, 50^th Percentile Male Dummy
Confidential

Although the number of tests are limited, and recognizing that the data available reflects pre-MY98 make/models, there does not appear to be any significant difference between a left (L) or right (R) 48 kmph (30 mph), 30 degree oblique unbelted impacts when using the 50^th percentile male dummy.

Table IV-30b
L vs R, 48 kmph (30 mph), 30 degree Oblique Impacts, Unbelted, 50^th Percentile Male Dummy
Confidential

Although the number of tests are limited, and the data available reflects pre-MY98 make/models, there does not appear to be any significant difference between a left (L) or right (R) 30 mph, 30 degree oblique unbelted impact, when using the 50^th percentile male dummy based on All Vehicles. On the average, all responses for the left-side impacts were numerically higher than the right-side oblique impacts. This suggests that the left-side unbelted oblique impact condition might be slightly more stringent, on the average, compared to the right-side oblique impact condition. However, because left-side and right-side impact make/models were not necessarily matched, the higher numerical values could be due to make/model differences.

Left vs Right 56 kmph (35 mph), ODB, 40% Overlap, Unbelted Test Stringency

Table IV-31 shows a matched pair of 1999 Dodge Intrepid left (L) and right (R) side impacts, for the 5^th percentile dummy, based on the proposed 56 kmph (35 mph), ODB, 40% overlap, unbelted test procedure. There was no significant difference between L and R side impacts based on the driver-side responses Either crash direction would have resulted in a compliance test failure. However, on the passenger-side, responses were mixed with the passenger-side dummy failing Nij and chest g's in a left-side ODB test and passing Nij and chest g's on the right-side ODB test. Despite these differences, both left and right ODB tests would have resulted in compliance test failures based on the 5^th percentile dummy. However, left side impact responses for HIC₁₅, Nij, chest g's and chest deflection were higher for both the driver and passenger. Recognizing the limited number of test points, the left impact condition would appear to be somewhat more stringent having higher response values across the board.

Table IV-31
56 kmph (35 mph), ODB, L vs R, 40% Overlap, Unbelted Test,
5^th Percentile Female Dummy
1999 Dodge Intrepid (Left n=1/Right n=1)

Sled Tests

BMW provided a comparison of 50^th and 5^th percentile dummy responses based on the FMVSS 208 (S13) sled test. (See Docket No. 98-4405-59) Figure IV-1 shows a good comparison of the dynamic responses of both dummies under identical, low variability test conditions. This bar chart compares the dynamic equivalence of the two dummies. For example, the 5^th percentile dummy is more vulnerable in the areas of fore/aft neck shear and neck moments (extension and flexion) compared to the 50^th percentile dummy, whereas the 50^th percentile male dummy is more vulnerable in the chest g's area for the same test condition.

Figure IV-1 BMW 208 sled test (S13) bar chart comparing 50^th and 5^th percentile dummy responses. [See Docket No. NHTSA-1998-4405-59]

Full-Scale Vehicle Tests: Dummy Response Comparison

48 kmph (30 mph), 30 degree oblique (L) belted test condition

Table IV-11 shows that the 50^th and 5^th percentile dummies responded almost identically in this test. Both dummies would have resulted in the 1999 Dodge Intrepid passing the compliance test procedure.

60 kmph (37.5 mph), ODB, 40% overlap, belted test condition

For the 1998 Dodge Neon, driver and passenger responses were very similar. The 1998 Dodge Neon would have passed the compliance test using both the 50^th and 5^th percentile dummies. For the 1998 Ford Contour, driver and passenger responses were very similar, except the 5^th percentile dummy failed Nij on the driver's side. Therefore, based on the 50^th percentile dummy the Ford Contour would have passed the compliance test, whereas based on the 5^th percentile dummy the same vehicle would have failed. For the 1998 Chevy Venture, compliance test failure would have occurred using either dummy. The 50^th percentile dummy driver failed Nij and the 5^th percentile driver failed both Nij and HIC₁₅. The 5^th percentile passenger failed Nij. Therefore, for 2 out of the 3 MY1998 make/models studied, the use of both dummies would have resulted in the same compliance test decision, whereas in 1 out of 3 there would have been a conflict. (See Tables IV-14a and IV-14b)

For the 60 kmph (37.5 mph), ODB, 40% overlap, belted test condition with the Tibia Index, and full-forward seating position for the 5^th percentile female dummy, shows that overall test stringency can be enhanced by the incorporation of a lower leg injury criteria. Given the 48 test responses measured over 3 test vehicles, including the Tibia Index, the 5^th percentile dummy exceeded the ICPL or IARV values 27 percent of the times, whereas the 50^th exceeded the ICPL or IARV values only 10 percent of the time. (See Tables IV-14a and 14b.)

30 mph, 0 Degree Fixed Rigid Barrier, Unbelted Test Condition

Table IV-32 compares the responses of the 50^th percentile male and 5^th percentile female dummies for the 30 mph, 0 degree impact, unbelted, fixed rigid barrier test condition using a 1999 Saturn SL test vehicle. The driver and passenger responses for both dummies were very similar. Using either dummy and the new proposed injury criteria and associated ICPLs, the subject vehicle would have passed the compliance test.

48 kmph (30 mph), 30 degree oblique (L) unbelted test condition

As shown in Tables IV-17 and IV-18, for the 1999 Dodge Intrepid (L), the responses of the 50^th and 5^th percentile dummies were very similar, except for driver-side Nij failure for the 5^th percentile female dummy. The 50^th percentile male dummy Nij was 0.37, whereas the 5^th percentile female dummy Nij was 1.693. Passenger-side responses were very similar and did not exceed any of the applicable ICPLs. Therefore, for this test vehicle, introduction of the 5^th percentile female dummy and concomitant ICPLs results in compliance test failure, whereas passage would have occurred with the 50^th percentile male dummy alone. [NOTE: The 5^th percentile female dummy is included here for comparison purpose only and is not proposed for this test in the SNPRM.]

Table IV-32
50^th Percentile Male vs 5^th Percentile Female Dummies
48 kmph (30 mph), 0 Degree Fixed Rigid Barrier, Unbelted Test

56 kmph (35 mph), ODB, 40 % Overlap, Left, Unbelted Test Condition

For the 1999 Dodge Intrepid (L), driver responses were very similar and both failed Nij. The passenger-side responses were not as similar as the 5^th percentile dummy failed Nij and chest g's. Use of either dummy would have resulted in a compliance test failure. For the 1999 Toyota Tacoma (L), driver responses were very similar and none of the responses exceeded the applicable ICPLs. Similarly, the passenger-side responses were very similar and none of the responses exceeded the applicable ICPLs. Use of either dummy would have resulted in passing the compliance test. Therefore, these two specific test vehicles, the usage of either dummy led to the same compliance test decision. (See Tables IV-19 and IV-20.)

Conclusion: Overall, the two dummies appear to be dynamically equivalent as they have very similar responses for the same dynamic test conditions and appear to be equivalent from a compliance point of view. However, there were several cases where the dummies were not equivalent from a compliance point of view. In each of those cases, the 5^th percentile female dummy was more vulnerable to failure than the 50^th percentile male dummy.

^1. Preliminary Economic Assessment, FMVSS No. 208, Advanced Air Bags, August 1998, Office of Regulatory Analysis and Evaluation, Plans and Policy, NHTSA/DOT.

^2. "Evaluation of the CRABI 12-Month-Old Infant Dummy and Its Comparison with TNO P3/4, February 1999," Hagedorn, A.V. and Pritz, H.B., Vehicle Research and Test Center, East Liberty, OH, Docket No. NHTSA-99-5156-6.