Modelling Studies

A NHTSA Modelling Study to Predict the Effects of Depowering an Air Bag on the Normally Seated Driver and Passenger Occupant During High Severity Frontal Impacts

A modelling effort was performed to examine the effects depowering the air bag could potentially have on occupants at higher speeds where energy absorption by the air bag is more critical. Therefore, the objective of this study is to provide an initial assessment of depowered air bag performance on the normally seated driver and right front seat passenger at higher severity impacts.

An analytical modelling study was conducted with the MADYMO rigid body analysis software package. Driver and passenger frontal impact simulations were conducted over a range of impact speeds with both baseline, and depowered air bag systems, and with 5th percentile females and 50th and 95th percentile male simulations.

Baseline Models

A MADYMO5.1.1 model of a 1991 Ford Taurus was chosen as the baseline vehicle model for the driver simulations in this study. This model was developed, and validated using experimental NCAP results by Lawrence F. Simeone at VNTSC⁽¹⁾.

The baseline passenger model used in the study was a generic MADYMO5.2ß vehicle model. A comparison was made of the chest and head acceleration responses of the model to a comparable FMVSS 208 vehicle compliance test. The results are relatively close in magnitude.

Test Matrix

Using the baseline driver and passenger MADYMO models, 108 simulations were conducted to evaluate the effects of depowering the air bag over a variety of high severity impact speeds. Both restrained and unrestrained occupants were evaluated.

Occupant size: The 50th percentile male Hybrid III anthropomorphic dummy was chosen as the baseline occupant in the study. The 5th percentile female and 95th percentile male MADYMO simulations were also conducted for this study. TNO's standard 50th percentile male Hybrid III occupant database⁽²⁾ was used to define the geometry, inertia, tree structure, and joint characteristics in MADYMO.

Impact speeds: The frontal impact simulations were conducted at three different impact speed levels: 40.2 kph (25 mph), 56.3 kph (35 mph), and 64.4 kph (45 mph). These impact speeds were chosen based on the availability of real-world crash data (i.e. the 40.2 kph simulations represented crashes from 32.2-48.3 kph, the 56.3 kph simulations represented crashes from 48.3-64.4 kph, and the 64.4 kph simulations represented crashes with impact speeds greater than 64.4 kph). Air bag initiation was also modified to be impact speed dependent (i.e. the change in velocity of the baseline models were used to approximate the corresponding air bag initiation times in the remainder of the models).

Belt Restraints: Driver and passenger occupants were modeled both with and without 3-point belt restraints. (The air bag was deployed in all simulations). No force-limiting or pretensioning devices were modeled in this study. Belts added to the baseline passenger model were constructed by John Guglielmi at VNTSC.

Air Bag Inflators: Three sets of air bag inflators were modeled for both the driver and passenger air bag systems in this study. These included: a baseline inflator, and two levels of depowering (20% and 40%) to represent the range of depowering being supplied by the industry, and those tested experimentally. The baseline driver inflator was a 1991 Ford Taurus 350x22 inflator. The inflator mass flow rate was generated from the MADYMO tank test analysis program. Depowering the air bag was modeled by reducing the gas supplied by the baseline air bag inflator by 20% and 40% (i.e. scaling the air bag inflator mass flow rate). This resulted in a corresponding reduction in peak pressure of the air bag inflator. A similar depowering procedure was conducted for the passenger simulations with the original air bag inflator of the model considered to be the baseline. No other modifications to the air bag configuration were made in this study (i.e. vent holes and air bag geometry were not changed).

Results

The HIC36 and 3ms chest acceleration results of the normally seated adult unbelted, and belted driver occupant simulations, and unbelted and belted passenger occupant MADYMO simulations for the baseline and depowered air bag systems are listed in Tables III-13 to Table III-16. The results are broken down by the three impact speed levels (40.2 kph, 56.3 kph, and 64.4 kph). Table III-17 provides a summary including 5th, 50th, and 95th percentile simulations.

The HIC and 3ms clip chest acceleration results for the belted driver and passenger generally remained relatively constant, or decreased slightly through the depowering of the air bag; therefore increases in fatality risk through depowering the air bag were not as significant as in the unbelted simulations.

Ford provided in their September 20, 1996 docket comments an analysis of their math modelling for drivers. Ford's model compared depowered current air bag inflators (350x22) to depowered inflators (300x12) at different speeds. HIC increased slightly at all speeds, chest g's go up by 2-3 g's, while chest deflection showed mixed results (some went up, some down) for the depowered systems. (See Table III-18).

Table III-13 Normally Seated, Unbelted, 50th Percentile Male Driver High Severity Impacts of 40.2 kph, 56.3 kph, and 64.4 kph
MADYMO Response Unbelted Driver 50th Percentile Male	HIC36			Chest Acceleration
	Baseline	20% Depowered	40% Depowered	Baseline	20% Depowered	40% Depowered
40.2 kph	163.6	240.3	422.4	27.3	37.1	37.4
56.3 kph	467.5	607.2	530.1	38.0	33.0	49.2
64.4 kph	611.4	644.1	698.7	41.4	50.8	77.8

Table III-14 Normally Seated, Belted, 50th Percentile Male Driver High Severity Impacts of 40.2 kph, 56.3 kph, and 64.4 kph
MADYMO Response Belted Driver 50th Percentile Male	HIC36			Chest Acceleration
	Baseline	20% Depowered	40% Depowered	Baseline	20% Depowered	40% Depowered
40.2 kph	400.5	407.4	390.2	31.7	32.4	28.7
56.3 kph	600.5	678.5	649.2	42.3	44.6	44.3
64.4 kph	800.0	846.5	779.8	51.4	49.9	50.1

Table III-15 Normally Seated, Unbelted, 50th Percentile Male Passenger High Severity Impacts of 40.2 kph, 56.3 kph, and 64.4 kph
MADYMO Response Unbelted Passenger 50th Percentile Male	HIC36			Chest Acceleration
	Baseline	20% Depowered	40% Depowered	Baseline	20% Depowered	40% Depowered
40.2 kph	88.0	52.2	107.4	28.6	34.6	50.3
56.3 kph	372.6	369.9	525.9	45.0	55.1	73.5
64.4 kph	658.1	708.9	915.7	56.0	64.8	80.7

Table III-16 Normally Seated, Belted, 50th Percentile Male Passenger High Severity Impacts of 40.2 kph, 56.3 kph, and 64.4 kph
MADYMO Response Belted Passenger 50th Percentile Male	HIC36			Chest Acceleration
	Baseline	20% Depowered	40% Depowered	Baseline	20% Depowered	40% Depowered
40.2 kph	199.3	143.1	119.6	43.9	44.0	41.9
56.3 kph	482.4	388.9	382.1	64.4	64.5	61.9
64.4 kph	722.2	631.4	644.6	74.7	75.6	73.4

The majority of the HIC and 3ms clip chest acceleration results for the unbelted driver and passenger generally increased with increasing depowering, and resulted in a corresponding increase risk in fatality. However, the HIC results did not exceed the 1000 HIC injury measure in any of the simulations. However, there were a few cases with chest acceleration exceeding 60 g's. For example, this occurred in the 40% depowering simulations at 64.4 kph for the driver, and at 56.3 kph and 64.4 kph for the passenger.

Table III-17
(missing)

Table III-18 Prepared by Ford Motor Co. Biomechanics Adv. Safety
Peak Occupant Responses - Unbelted Driver
	HIC (36ms)		Chest Gs (3ms Cumdur)		Chest Deflection (inch)
Speed (mph)	350x12	300x22	350x12	300x22	350x12	300x22
14	42.2	45.4	18.1	20.1	0.417	0.466
25	223.4	238.0	42.7	45.8	0.71	0.703
30	316.8	364.5	48.3	50.6	0.789	0.738
35	684.9	748.0	53.7	55.2	1.343	1.258
Peak Occupant Responses - Belted Driver
	HIC (36ms)		Chest Gs (3ms Cumdur)		Chest Deflection (inch)
Speed (mph)	350x12	300x22	350x12	300x22	350x12	300x22
14	54.0	64.1	19.7	22.2	0.786	0.808
25	437.2	488.0	36.9	39.3	1.246	1.24
30	601.3	584.7	45.0	47.6	1.317	1.347
35	692.3	758.6	44.4	47.6	1.362	1.355

A NHTSA Analytical Modelling Study to Predict the Effects of Depowering an Air Bag on the Normally Seated Driver and Passenger Occupant in an Offset Frontal Collision

Previous analytical studies and experimental testing have been conducted to predict the effects of depowering vehicle air bag systems for a variety of occupant sizes, impact speeds, and restraint conditions based on a full frontal crash environment. Since some concern has been expressed that the simulated full frontal crash environment is not fully representative of real-world crashes, the agency decided to further examine the predicted performance of depowered air bag systems in an offset frontal impact environment.

Approach

An analytical modelling study was conducted with the MADYMO rigid body analysis software package. Driver and passenger frontal offset collisions were conducted with baseline and depowered air bag systems.

Baseline Models

Two previously developed MADYMO5.1.1 models, representative of a driver and passenger seated in a Ford Taurus vehicle, were used as a starting point in this study. The models were developed and validated in a full frontal crash environment using experimental NCAP test results by Lawrence F. Simeone at VNTSC⁽³⁾.

The driver and passenger MADYMO models were modified to simulate NHTSA's vehicle crash test #2075. In Test #2075, two vehicles, both moving forward at 61.5 kmph were impacted at a 30 degree angle. Vehicle 1 was a 1994 Ford Taurus GL 4-door sedan and vehicle 2 was a 1992 Ford Taurus LX 4-door sedan (Figure 1a & 1b). The crash signals, intrusion measurements, air bag deployment time, etc. were all modified based on Test #2075 through film and signal analysis. Comparisons of the experimental peak injury measures to the analytical MADYMO models for the driver and passenger occupant of vehicle 1 are listed in Table III-19.

Figures 12a & 12b

Vehicle-to-Vehicle Frontal 30 Degree Angled, 61.2
kmph Impact of a 1994 Ford Taurus GL 4-Door Sedan
and a 1992 Ford Taurus LX 4-Door Sedan

Table III-19 Comparison of Driver and Passenger Peak Injury Measures between Test #2075 and the Analytical MADYMO Simulations
	Driver 50th Percentile Male		Passenger 50th Percentile Male
	Crash Test #2075	MADYMO Model	Crash Test #2075	MADYMO Model
Head Injury Criteria (HIC)	410.9	436.5	257.0	356.9
3ms Chest Accel. (G's)	51.0	47.2	48.4	46.4
Chest Deflection (mm)	24.6	29.8	50.4	37.4
Left Femur Force (N)	5199.4	8857.4	4964.6	6269.8
Right Femur Force (N)	5823.7	6895.8	1647.9	2637.0
Shoulder Belt Force (N)	3118.1	3145.1	Not Meas.	7111.6
Res. Upper Neck Force (N)	2750.0	2634.4	3001.0	2322.6
Res. Lower Neck Force (N)	7050.0 (4)	3057.4	Not Meas.	2955.0
Res. Upper Neck Moment (Nm)	84.0	78.9	Erroneous	53.7
Res. Lower Neck Moment (Nm)	482.8 2	210.5	Not Meas.	173.9

Note: The lower extremity injury measures were difficult to match without extensive film analysis of the compartment intrusion (i.e. driver door bending, toe pan intrusion, steering column motion, etc.) Current estimations are based on pre- and post- intrusion measurements, timing of the vehicle-occupant interaction, and some film analysis. Since lower extremity injury measures were not significantly affected by depowering in previous simulation studies, the driver and passenger MADYMO models were used as a preliminary evaluation. Further documentation and model improvements will be pursued in the future.

Test Matrix

Using the baseline driver and passenger MADYMO models, twelve MADYMO simulations were conducted to evaluate the effects of depowered air bag systems in a frontal offset collision.

Occupant sizes: Both the driver and passenger occupants were represented by a 50th percentile male anthropomorphic test device (ATD). TNO's standard 50th percentile male occupant database⁽⁵⁾ was used to define the geometry, inertia, tree structure, and joint characteristics in MADYMO.

Impact speed: Due to the crash severity of Test #2075, the crash signals previously used in the driver and passenger MADYMO validation simulations were scaled to a softer, 48.3 kmph equivalent crash severity.

Belt Restraints: Driver and passenger occupants were modeled both with and without three-point belt restraint systems. (The air bags were deployed in all simulations). No force-limiting or pretensioning devices were modeled in this study.

Air Bag Inflators: Three sets of air bag inflators were modeled for both the driver and passenger air bag systems in this study. These included: a baseline inflator, and two levels of depowering (20% and 40%) to represent the range of depowering being supplied by the industry, and those tested experimentally. Depowering was modeled by reducing the gas supplied by the baseline air bag inflator by 20% and 40% (i.e. scaling the air bag inflator mass flow rate). No other modifications to the air bag configuration were made in this study (i.e. air bag venting, geometry, deployment strategy, etc.).

Results

Figures 13 to 15 are bar chart results of the peak driver and passenger occupant responses from the MADYMO simulations. Each chart compares the occupant response with the baseline air bag to the occupant response with the depowered air bags. Bar charts in the left column represent the belted dummy results and the charts in the right column represent the unbelted dummy.

From the results, HIC generally increased with increasing depowering for the driver and passenger occupants. The most significant increases were found in the unbelted passenger occupant, where the HIC exceeded 1000 with the 40% depowered air bag. However, since the kinematics of the occupants in the simulations implies significant lateral head motion, and impact loading, the predictive quality of HIC in lateral head loading could be questioned. The chest acceleration results for the driver and passenger occupant did not significantly increase with depowering of the air bag and the magnitudes were below 40G's for all cases. Chest deflection measurements decreased slightly with depowering for both the belted and unbelted, driver and passenger occupants and the magnitudes were all below 40 mm.

Finally, neck injury measures, such as: neck tension, neck extension moment, and neck shear, were generally found to increase with increasing levels of depowering. The most significant increases in neck injury measures were found when depowering to 40%. The magnitudes of the neck shear forces for the unbelted passenger occupant were abnormally low due to the angle that the dummy head impacted the air bag. However, this also resulted in the unbelted passenger exhibiting the only significant increase in neck compression.

Figures 16 to 19 are side and overhead views of the driver and passenger occupant kinematics as a result of depowering the air bag at the time of maximum forward occupant excursion. The figures demonstrate the trends associated with the depowered air bags being pushed out of the occupant's path, indicating a higher threat of hard interior compartment surface contacts, and possibly higher probability of partial ejection of drivers.

Generic Sled Pulse

The August 23, 1996 American Automobile Manufacturers Association (AAMA) petition sought an immediate amendment to revise the unbelted dummy test. The AAMA indicated that the immediate elimination of the FMVSS No. 208 unrestrained dummy test would be the most direct action to allow manufacturers to quickly initiate air bag design changes that can further reduce the injury risks related to air bag inflation. If the agency would not eliminate the unrestrained test, the AAMA then recommended that the current 30 mph unrestrained dummy vehicle barrier crash test protocol be replaced with a standard 30 mph unrestrained dummy sled test protocol. Under the AAMA proposal, a generic sled pulse of 143 msec would become the compliance test for the unbelted test condition to set the lower limit of vehicle unbelted crash performance. The AAMA provided an envelope for the crash pulse along with a suggested sine pulse that could be used for the test with a change in velocity of 302 mph. The sine pulse suggested by the AAMA is described by the mathematical function:

In its review of the suggested sine pulse, NHTSA suggested to the AAMA that the pulse was both too long in duration as well as too low in amplitude when compared to crash pulses observed in its FMVSS No. 208 compliance testing program. This observation was based on a review of the average crash pulse for vehicles tested in the MY 1996 test program. Figure 20 demonstrates this observation.

In addition, the agency ran a sled test using the 143 millisecond (msec) sled pulse (15 g peak) with an unrestrained (no belts and no air bag) dummy on a sled representing a 1993 Ford Taurus and found that the dummy passed all current FMVSS 208 injury criteria (neck injury instrumentation was not installed for the test).

Then the AAMA informed the agency that the suggested sinusoidal pulse could be shortened to an 125 msec duration. This change would require a 17.2 g peak to maintain the 48 km/hr (30 mph) change in velocity. Figure 21 provides a comparison of these two pulses, AAMA #1 is the 143 msec pulse and AAMA #2 is the 125 msec pulse. Figure 22 shows that a typical large car pulse in a 30 mph FMVSS 208 compliance test is similar to the 125 msec pulse, but this pulse is milder than the typical small car or utility vehicle, which is more representative of light trucks.

The AAMA stated that the sinusoidal crash pulse was a better representation of the vehicle-to-vehicle crashes that occur more frequently in the real world crash experience. The AAMA also stated that the crash pulse of the FMVSS No. 208 compliance test procedure was representative of only a small percentage of real world crashes.

In response to these claims, the agency examined the crash pulses measured in the offset crash tests that were conducted in its research program for the upgrade of frontal crash protection. Figure 21 provides the average crash pulse measured in the agency's collinear frontal offset crash testing (labeled as Car-to-Car, Collinear). It should be noted, however, that the car-to-car tests were conducted at a higher closing velocity than that required by the FMVSS No. 208 test procedure. The average change in velocity of the car-to-car tests was 60 km/hr (37.5 mph). Hence, for comparative purposes, the car-to-car average crash pulse was scaled to provide the 48 km/hr (30 mph) change in velocity. This scaled pulse is annotated as Car-to-Car (Scaled) in Figure 21. In examining these pulses, it is observed that the AAMA 143 msec sinusoidal pulse is longer in duration than the car-to-car pulses and the peak acceleration is lower than that found in the car-to-car pulses. In comparing the 125 millisecond, 17.2 g sinusoidal pulse with the scaled car-to-car pulse, it is observed that the sinusoidal pulse now more closely resembles that of car-to-car 30 mph crash severity.

Figure 22 shows the 125 msec pulse compared to a typical crash pulse from a three different types of vehicles.

Figure 23 shows the 125 msec sled pulse with the maximum and minimum corridors to be used in testing. The letters help define the curve in the standard.

AVS Technologies (74-14-N108-126) estimated through modelling that the generic sled pulse was equivalent to a 22 to 25 mph barrier test based on a comparison of HIC and chest g's.

Figure 23
(missing)

In a review of the National Automotive Sampling System data in the research program for the upgrade of frontal crash protection, agency staff estimated the exposure population for all towaway frontal impacts for a number of proposed crash test configurations [Stucki, Sheldon L. and Hollowell, William T., "Development of New Test Procedures for Frontal Offset Crashes," Airbag 2000 International Symposium on Sophisticated Car Occupant Safety Systems, Karlsruhe, Germany, November 1996 (to be published)]. The proposed tests and the crash modes simulated are shown in Figure 24. As shown here, it is seen that the frontal barrier test condition of FMVSS No. 208 represented approximately 22 percent of the NASS frontal cases. A large majority of the cases (73.4 percent) were judged to be represented by car-to-car collisions.

The Stucki, Hollowell paper presented figures on drivers exposed to frontal crashes, which showed about 22% were in impacts assumed to be full barrier "type." Although the impact modes assumed to represent full barrier type impacts only represent about 22% of exposed drivers in frontal crashes, the percentages are much higher when considering serious injuries and fatalities: about 31 percent and 34 percent for serious injuries and fatalities, respectively (see Table III-20 and Figure 25).

The non-full barrier type frontal impacts include collinear, offset impacts, and oblique, offset fixed object and into other vehicles, and distributed, oblique impacts into other vehicles. The majority of these impacts are vehicle-to-vehicle, with lesser proportions into fixed objects and not-so-fixed objects. Also, based on the Stucki, Hollowell paper, the highest injury risk for offset impacts is when the overlap percentage is high, 2/3 and over. Thus, it seems that the impact which best represents these non-full barrier type impacts would be a vehicle-to-vehicle impact with substantial overlap. Figure 22 shows a comparison of the AAMA proposed pulse, the average pulse for cars in the 208 compliance test, and a 50% overlap, car-to-car test at about 70 mph closing velocity, scaled to get a velocity change of 30 mph (It should be noted that the collinear offset test with higher overlap would result in higher peak G pulse.) The car-to-car, offset test produces a crash pulse that is below the 208 profile but higher than the AAMA pulse and outside the corridors and with a peak that occurs earlier. It is the agency's judgement that this pulse would produce much different results on the unbelted occupants.

The NHTSA final rule for the 125 msec pulse includes the following factors and considerations:

1. Compliance testing for the belted condition would remain the same as the current vehicle test.

2. All unbelted testing could be done without vehicles, but on a sled. Using a sled would significantly reduce the time it takes to do certification testing, since many sled tests can be run in a day (up to 10) compared to usually only one vehicle barrier test per day. Using a sled would also significantly reduce testing costs, since a vehicle is not destroyed.

3. A vehicle platform would be tested unbelted utilizing the full vehicle on the sled, but could be tested using the same sled pulse.

4. The sled test is a direct frontal impact, with no 30 degree left or right impacts.

5. The agency is requiring dummy neck measurements be taken during the sled test and a set of neck criteria must be met during the test. These neck criteria are shown in Table II-4. Based on data supplied by AAMA (see Figure III-26), the neck criteria could not be met by an occupant unprotected by belts or air bags, but could be met by an unbelted occupant protected by an air bag or a depowered air bag. Figure III-26 is based on the 143 msec sled pulse. It is believed that the neck injury criterion could limit the potential amount of depowering.

Figure III-26

UNBELTED IN-POSITION OCCUPANTS - AAMA 143 msec GENERIC SLED PULSE

Baseline vs Depowered Inflator vs No Air Bag - 50th Male Hybrid III

NHTSA has conducted a series of tests on two vehicle platforms to evaluate the AAMA proposed test procedure that would replace the un-belted crash test requirement currently in FMVSS 208. The AAMA proposal would replace the crash test with a generic sled test that would allow air bags to be de-powered. Specifically the sled test consists of a half-sine pulse with a 125 ms duration, 30 mph change in velocity, and 17.2 g maximum acceleration.

Tables III-21 to 25 shows a matrix on the first vehicle platform of the tests conducted, and test results. Tables III-26 to 29 show results from the second vehicle platform. A vehicle body-in-white was equipped with the necessary parts to represent the vehicles interior and tests conducted on a HYGE sled. The H-III 50th percentile male dummy and the 5th percentile female dummy were instrumented to measure loads, deflections, and accelerations on the head, neck, chest, and femur.

Dummy Seating Procedure:

50th % Male - Seated using the 208 seating procedure on driver and passenger-side

5th % Female Driver - Seated in the full-forward seating position, unless interference with forward structure, in which case seat is locked one notch rearward from that position.

5th % Female Passenger - Seated with passenger seat in the mid-track seating position

Test Numbers and conditions of the sled pulse for the first vehicle platform tested:

• Test 2 - 123.8 ms pulse duration, 17.1 g pulse, 29.8 mph deltav

• Test 3 - 123.6 ms pulse duration, 17.1 g pulse, 29.8 mph deltav

• Test 4 - 125.0 ms pulse duration, 17.3 g pulse, 30.0 mph deltav

• Test 5 - 124.9 ms pulse duration, 17.4 g pulse, 30.0 mph deltav

• Test 6 - 125.1 ms pulse duration, 17.2 g pulse, 30.0 mph deltav

• Test 7 - 126.3 ms pulse duration, 17.5 g pulse, 30.0 mph deltav

Table III-21 Test Matrix
Vehicle Passengers	Air Bag Inflator	Test No.
5th Female Driver 50th Male Passenger	No Deploy	Taur-Gen-003
5th Female Passenger 50th Male Driver	No Deploy	Taur-Gen-002
5th Female Driver 50th Male Passenger	Baseline Deployment	Taur-Gen-004
5th Female Passenger 50th Male Driver	Baseline Deployment	Taur-Gen-006
5th Female Driver 50th Male Passenger	De-Powered Air Bag Deployment	Taur-Gen-007
5th Female Passenger 50th Male Driver	De-Powered Air Bag Deployment	Taur-Gen-005
Vehicle: 93-95 Taurus Platform

Table III-22 50th % Male Sled Test Results with no Air Bag and Baseline Air Bag Deployment
Dummy Measurements	Driver No Air Bag	Driver Base Air Bag	Passenger No Air Bag	Passenger Base Air Bag
Test #	Taur-Gen-002	Taur-Gen-006	Taur-Gen-003	Taur-Gen-004
HIC	691(36ms)	168(36ms)	577(36ms)	100(36ms)
Chest g's(3 ms clip)	43	32	58	22
Chest Deflection (mm)	40	29	22	12
Left Femur (N)	5123	4645	4194	4232
Right Femur (N)	6752	5865	5108	3910
Neck Shear (N)	1065	1169	896	852
Neck Tension (N)	1821	750	1504	924
Neck Compression (N)	1991	443	1400	700
Neck Moment (N-m) (Flexion)	13	50	21	42
Neck Moment (N-m) (Extension)	72	14	221	39

Table III-23 50th % Male Air Bag Sled Test Results with De-powered Air Bag Deployment
Dummy Measurements	Driver De-powered Air Bag	Passenger De-powered Air Bag
Test #	Taur-Gen-005	Taur-Gen-007
HIC	289(36ms)	122(36ms)
Chest g's(3 ms clip)	39	23
Chest Deflection (mm)	24	6
Left Femur (N)	4683	4583
Right Femur (N)	6156	3819
Neck Shear (N)	1210	927
Neck Tension (N)	761	528
Neck Compression (N)	1634	1386
Neck Moment (N-m) (Flexion)	67	66
Neck Moment (N-m) (Extension)	13	35

Table III-24 5th % Female Air Bag Sled Test Results with no Air Bag and Baseline Air Bag Deployment
Dummy Measurements	Driver No Air Bag	Driver Base Air Bag	Passenger No Air Bag	Passenger Base Air Bag
Test #	Taur-Gen-003	Taur-Gen-004	Taur-Gen-002	Taur-Gen-006
HIC - calculated over 15 and 36 ms	269(15ms) 332(36ms)	93(15ms) 118(36ms)	627(15ms) 855(36ms)	50(15ms) 76(36ms)
Chest g's(3 ms clip)	38	35	57	23
Chest Deflection (mm)	47	27	44	14
Left Femur (N)	4161	3645	5123	3600
Right Femur (N)	2019	3173	6752	3329
Neck Shear (N)	1198	852	1244	890
Neck Tension (N)	707	924	1177	778
Neck Compression (N)	279	700	2705	293
Neck Moment (N-m) (Flexion)	89	42	11	60
Neck Moment (N-m) (Extension)	16	39	100	14

Table III-25 5th Female Sled Test Results with De-Powered Air Bags
Dummy Measurements	Driver De-Powered Air Bag	Passenger De-Powered Air Bag
Test #	Taur-Gen-007	Taur-Gen-005
HIC - calculated over 15 and 36 ms	122(15ms) 174(36ms)	54(15ms) 80(36ms)
Chest g's (3 ms clip)	34	21
Chest Deflection (mm)	31	6
Left Femur (N)	3535	3519
Right Femur (N)	3117	3274
Neck Shear (N)	347	959
Neck Tension (N)	808	381
Neck Compression (N)	153	809
Neck Moment (N-m) (Flexion)	12	77
Neck Moment (N-m) (Extension)	11	10
Proposed 208 sled test: Neck measurements filtered at SAE J211 recommendation (Forces - Class 1000; Moments - Class 600)

Test Numbers and conditions of the sled pulse for the second vehicle platform tested:

• Test # Hon-Gen-001 - 123.5 ms pulse duration, 29.2 mph deltav, 16.9 g's

• Test # Hon-Gen-002 - 123.4 ms pulse duration, 29.4 mph deltav, 16.7 g's

• Test # Hon-Gen-003 - 125.1 ms pulse duration, 29.4 mph deltav, 16.8 g's

• Test # Hon-Gen-003A - 124.9 ms pulse duration, 29.7 mph deltav, 16.8 g's

• Test # Hon-Gen-004 - 125.0 ms pulse duration, 29.3 mph deltav, 17.0 g's

• Test # Hon-Gen-005 - 124.3 ms pulse duration, 29.2 mph deltav, 16.8 g's

• Test # Hon-Gen-006 - 124.5 ms pulse duration, 29.3 mph deltav, 16.7 g's

Table III-26 50th Percentile Male with no Air Bag Deployment
Dummy Measurements	Driver, No Bag	Passenger, No Air Bag
Test #	Hon-Gen-005	Hon-Gen-002
HIC (36 ms)	460	386
Chest g's(3 ms clip)	64	44
Chest Deflection (mm)	42	24
Left Femur (N)	4085	5426
Right Femur (N)	3938	3803
Neck Shear (N)	1404	996
Neck Tension (N)	536	1180
Neck Compression (N)	3359	2464
Neck Moment (N-m) (Flexion)	17	31
Neck Moment (N-m) (Extension)	176	173

Table III-27 50th Percentile Male base and de-powered sled test results
Dummy Measurements	Driver Base Bag	Driver De-Powered Bag	Passenger Base Bag	Passenger de-powered Bag
Test #	Hon-Gen-004		Hon-Gen-003	Hon-Gen-006
HIC (36 ms)	167	NA	124	256
Chest g's(3 ms clip)	34		28	32
Chest Deflection (mm)	39		9	4
Left Femur (N)	3816		5055	4217
Right Femur (N)	4082		3698	4069
Neck Shear (N)	832		1180	1297
Neck Tension (N)	821		642	301
Neck Compression (N)	1687		1089	2013
Neck Moment (N-m) (Flexion)	69		113	81
Neck Moment (N-m) (Extension)	12		21	19

Table III-28 5th Percentile Female with no air bag deployment
Dummy Measurements	Driver, No Bag	Passenger, No Air Bag
Test #	Hon-Gen-006	Hon-Gen-001
HIC	861(15ms) 864(36ms)	744(15ms) 1005(36ms)
Chest g's(3 ms clip)	55	78
Chest Deflection (mm)	47	16
Left Femur (N)	4471	4615
Right Femur (N)	2605	3893
Neck Shear (N)	576	1314
Neck Tension (N)	2093	950
Neck Compression (N)	234	2252
Neck Moment (N-m) (Flexion)	37	18
Neck Moment (N-m) (Extension)	29	141

Table III-29 5th Percentile Female Base and De-powered sled test results
Dummy Measurements	Driver Base Bag	Driver De-Powered Air Bag	Passenger Base Air Bag	Passenger De-Powered Air Bag
Test #	Hon-Gen-003A		Hon-Gen-004	Hon-Gen-005
		NA
HIC	67(15ms) 118(36ms)		134(15ms) 227(36ms)	383(36ms) 324(15ms)
Chest g's (3 ms clip)	31		30	34
Chest Deflection (mm)	52		10	4
Left Femur (N)	2390		5512	3086
Right Femur (N)	2978		3907	3599
Neck Shear (N)	405		1047	1009
Neck Tension (N)	857		346	1342
Neck Compression (N)	93		1228	1118
Neck Moment (N-m) (Flexion)	22		86	43
Neck Moment (N-m) (Extension)	13		15	8

Proximity Testing

The agency has presented results from tests in out-of-position conditions that represent some of the worst cases that could occur. It has been shown that in these positions, any air bag deployment has a significant probability of causing a serious to fatal injury. The level of de-powering required to reduce the probability of serious injury is expected to drop dramatically for occupants who are not centered on the air bag or are a few inches from the air bag module. These tests examine longitudinal proximity and lateral proximity changes of the dummy position from the air bag. On the driver-side (see Table III-31 and 32), a 5th Percentile Female dummy was used at different positions away from the air bag and with varying steering wheel tilt angles. At 4 inches back, neck extension was still above the standard of 57 Nm. On the passenger-side (see Table III-30), a 3-year old H-III dummy was used in varying positions around two different air bag systems. At 175 mm (7 inches) back, 3-year old dummy measurements reduced dramatically.

Proximity Testing with H-III 3YO. Base dummy positions against the air bag module were OOP test positions (Position 1 and Position 2 used here). Proximity movement was based on moving the dummy a particular distance away from the air bag, maintaining the same dummy attitude (spine angle, height, sitting/standing posture, etc..)

In the AAMA submission of January 29, 1997, two bar graphs were presented with a 6 year-old child dummy at 3.5 and 4 inches from the instrument panel. The graphs indicated that whether the dummy measurements exceeded Ford's injury criteria. The results were that all four neck injury criteria were exceeded at 3.5 inches from the instrument panel (upper neck shear, upper neck flexion, upper and lower neck compression), while at 4 inches away, only the upper neck tension injury criteria was exceeded.

Proximity Testing with H-III 5th Percentile Female Dummy. Base dummy position against the air bag was an OOP test position (Position 1 used here). Proximity movement was based on moving the dummy a particular distance away from the air bag, maintaining the same dummy attitude (spine angle, height, sitting posture, etc..)

Table III-30 Proximity Analysis of Passenger-side Air Bag Deployment with Static OOP Test Results of H-III 3YO Child Dummy
3YO, Vehicle I-96	Base Position 1	50 mm Back*	100 mm Back	175 mm Back	300 mm Back	100 mm Inboard 0 mm Back
HIC	254	384	541	194	26	86
Shear (N)	1309	1354	785	680	179	519
Tension (N)	2095	2262	2112	859	339	1194
Flexion (N-m)	0	1	14	25	8	4
Extension (N-m)	53	56	34	23	6	21
Ni,j	3.11	3.35	2.49	1.2	0.32	1.25
3YO, Vehicle B-94	Base B-94 Position 2	156 mm Inboard	156 mm Inboard 100 mm Back
HIC	3511	112	59
Shear (N)	1142	646	84
Tension (N)	2999	700	191
Flexion (N-m)	2	27	0
Extension (N-m)	111	4	5
Ni,j	4.77	0.78	0.3
* - Back refers to the direction towards the rear of the vehicle longitudinally

Table III-31 Proximity Analysis of Driver-side Air Bag Deployment with Static OOP Test Results of 5th Percentile Female Dummy
Dummy Measurements	Driver - Pos. 1, Base, Prox-1	Driver - Pos.1, 4" Back - Prox-2	Driver - Pos.1, 8" Back -- Prox-3	Driver - Pos.1, Tilt wheel Up, 9 -- Prox-4	Driver - Pos.1, Tilt wheel Down, 9 -- Prox-5
HIC	213(15ms) 235(36ms)	46(15ms) 51(36ms)	15(15ms) 15(36ms)	191(15ms) 213(36ms)	108(15ms) 109(36ms)
Chest g's(3 ms clip)	49.9 (Result) 26.3 (3ms)	16(Result) 10(3ms)	9(Result) 7(3ms)	53(Result) 22(3ms)	39(Result) 31(3ms)
Chest Deflection (mm)	35	20	2	31	35
Neck Shear (N)	2271	1849	373	3207	2217
Neck Tension (N)	2794	1951	547	2860	2075
Neck Compression	39	51	311	115	33
Neck Moment (N-m) (Flexion)	15	3	24	4	3
Neck Moment (N-m) (Extension)	96	89	9	151	109

Table III-32 Proximity Analysis of Driver-side Air Bag Deployment with Static OOP Test Results of 5th Percentile Female Dummy
Dummy Measurements	Driver - Pos.1, Tilt wheel Down, 9, 4" Back -- Prox-6	Driver - Pos. 1 Off-Center, 4" Left -- Prox-7
HIC	37(15ms) 39(36ms)	32(15ms) 51(36ms)
Chest g's(3 ms clip)	13(3ms clip) 13(result.)	14(3ms clip) 28(result.)
Chest Deflection (mm)	18	7
Neck Shear (N)	1398	775
Neck Tension (N)	1658	1558
Neck Compression	47	26
Neck Moment (N-m) (Flexion)	2	1
Neck Moment (N-m) (Extension)	77	37

¹ Technical Information Exchange: A MADYMO 5.1.1 Simulation of a 1991 Ford Taurus Driver in a 35mph Full Frontal Impact by Lawrence F. Simeone, dated March 13, 1996.

² MADYMO Databases Version 5.1, TNO Road-Vehicles Research Institute, September 1994.

³ Technical Information Exchange: A MADYMO5.1.1 Simulation of a 1991 Ford Taurus Driver in a 35mph Full Frontal Impact by Lawrence F. Simeone, dated March 13, 1996. Technical Information Exchange: A MADYMO5.1.1 Simulation of a 1993 Ford Taurus Passenger in a 35mph Full Frontal Impact Restrained by Seat Belts and an Air Bag by Lawrence F. Simeone, dated August 21, 1996.

⁴ Note: A spike occurs at 100 msec in the lower neck resultant force and the resultant lower neck moment signals (otherwise the resulting magnitudes would have been on the order of ~3000 N and ~280 Nm, respectively).

⁵ MADYMO Databases Version 5.1, TNO Road-Vehicles Research Institute, September 1994.