Arc far Side Impact Collaborative Research Program – Task 5b: Test Procedures Crash Tests and Sled Tests for the Far-side Environment

5.3Results

5.3.1Evaluation of Countermeasures by Each Dummy Model

The 50 simulations were completed and animations analyzed for correctness. Any bugs in the simulations were addressed and removed. Each simulation outputted an animation accompanied with time history plots of predefined measurements for certain properties, such as forces and accelerations.

The following discussion is limited to modeling and simulation of the 5 anthropomorphic test devices (ATDs). The results of the human finite-element model and human faceted model are unique and require separate discussion.

5.3.1.1Baseline

For all 5 ATDs in the baseline configuration with 90 degree side-impact pulses, the restrained dummy moved towards the centerline of the vehicle. After translating, the hips of the dummy struck the center console while the upper body continued to move towards the far-side. The pelvis, restrained by the lap belt, held the dummy’s hips into the seat. Meanwhile the upper torso slid out of the shoulder belt and rotated over the center console. The five ATDs grossly performed similarly, with the head of the ATD extending towards the center of the passenger seat while and the upper torso rebounded back to the driver seat. Some slight differences were seen in the rotation motion of the dummy as a result of various spinal constructions.

Seen in the animations, the highest likelihood of injuries for this case occurred in the neck and head. As the upper body rotated around the center console, the head and neck behaved as a spring-mass system, where the head is a mass on top of a spring, the neck. The head pulled and whipped the neck as motion was displayed. The head failed to make contact with any hard surfaces.

Figure 5 charts each dummy’s prediction of neck injury. Three metrics were used – F_z, neck tension and compression; F_y, neck shear force; and M_x, the bending moment about x-axis. All these measurements were taken at the upper and lower bodies of the neck. The graphs display the peak forces and moments for each measurement.

The vertical axis represents the percent value of the injury assessment reference value (IARV). A chart of IARV’s used is shown below.

5.3.1.2Chest Bag

By using the chest airbag, the occupant initially translated toward the centerline of the vehicle. Once the pelvis, struck the center console, the upper torso began rotating until the chest airbag made contact with the ribs. Rib compression was seen and the shoulders rotated slightly around the chest airbag. Occupant excursion wasn’t nearly as far as baseline cases, however, whipping of the head and neck was still seen.

The SID2s stuck the chest bag with its shoulder, instead of rib cage. This is a result of the small stature of this dummy.

Neck measurement numbers while using the chest airbag are also shown in figure 5.

5.3.1.3Shoulder Bag

The shoulder bag restrained the upper torso of each dummy from rotating around the center console. However, each dummy reacted differently as a result of different shoulder construction. The SID2s struck the shoulder bag with its neck. Neck data for all dummies is in figure 5.

5.3.1.4Chest and Shoulder Bags

With results very similar to just the shoulder bag, the chest and shoulder bags together restrained the dummy from rotation around the center console. The result was an upper torso with very little motion and an unrestrained head whipping around. Data is in figure 6.

5.3.1.5Reverse Belts

The reverse 3-pt belt system restrained the dummy by the neck from rotated around the center console, preventing occupant excursion. However, resulting neck loads were significantly increased. The BioSID data was excluded due to instabilities in the model that could not be resolved.

5.3.2Human Finite Element Model

The human finite element was unable to run until completion due to mathematical instabilities, which were insurmountable. In the baseline setup, as with the other dummy models, the human translated toward the centerline of the vehicle. The hips of the dummy initially made contact with the center console and the pelvic area started to deform. At this point, the dummies’ upper bodies rotated around the center console. Unlike the dummies, the human continued to translate until the center console intruded far into the pelvic region. It was here that the model would go instable due to the large deformations and displacements.

In the cases of chest and shoulder bags, the airbags intruded the human in the ribs and shoulder respectively. Again causing large deformations and the model went unstable. When the human moved into the airbags very little rotation was observed.

5.3.3Human Faceted Model

The human faceted model, although experimental, showed the most similar results to the PMHS head strike observation. The human would translate into the center console and continue translating while rotating until the head reached the opposite side of the vehicle.

The chest and shoulder airbags effectively restrained the human into its seat. The results from this model were purely visual. The numerical outputs were more difficult to interpret and compare evenly with the ATDs. Therefore, the results for both human models are limited to visual observations.

5.4Discussion

For each of the simulations shown in the current base configuration without use of countermeasures, it was seen that occupant excursion existed towards the passenger seat, but did not reach the door. Although the dummies reacted similarly, differences still existed. For example, the BioSID, with a rigid spine was not able to bend easily around the center console, transmitting more load into the neck. Conversely, the Hybrid III with the exact same neck contained lower forces. The flexible spine of the Hybrid III allowed it to bend and thus decrease neck forces.

For all baseline simulations with the five dummies, injury prediction levels were below accepted injury threshold. Despite the amount of motion observed, there was little to cause injury.

Contrary to these results, it is known that far-side vehicle impacts cause a high number of fatal head injuries annually. Fildes, Sparke, et al. showed that a post mortem human subject’s (PMHS) head can strike the B-pillar in far-side impacts. In none of the simulations, did the simulated dummy make contact with any surface inside the vehicle. This is similarly confirmed by Fildes, Sparke, et al. by testing a BioSID, EuroSID 1, and WorldSID in comparable circumstances. None of the dummies tested struck the B-pillar. The MADYMO simulations included BioSID and EuroSID1 in addition to three other dummies with similar results.

Further, to assess the dummies’ suitability for far-side testing, countermeasures were introduced to prevent occupant excursion and test the dummy reaction. They were chest and shoulder airbags and a 3-point seatbelt in reverse direction. Each limited occupant excursion, however, exposed more unique possibilities of injuries.

A chest airbag prevented some excursion, but not nearly as much as the shoulder bag or reverse belt. However, it restrained each dummy in the rib region, which introduces the potential for rib injuries.

The shoulder airbag restrained each dummy at the top of the torso. This was effective at preventing occupant excursion, however, left the head unrestrained. By restraining the upper body and leaving the head unrestrained put additional load on the neck, which the Hybrid III demonstrated was close to injury value, but still below the injury threshold.

The chest and shoulder bags together acted similarly to the shoulder bag alone. This situation looked much like a near-side impact and contact with a door. Again, the head is unrestrained and put additional stress on the neck.

The SID2s was the only odd sized dummy used. The EuroSID I and II, BioSID, and Hybrid III are all based on a 50^th percentile male. The SID2s, based on an adolescent, exposed vulnerability of the chest and shoulder bags to odd sized occupants. The SID2s struck the chest bag with its shoulder and the shoulder bag with its head/neck. Positioning of these bags can be difficult for oversize and undersized occupants.

Additionally, the chest and shoulder bags showed possible weaknesses of the countermeasures for elderly occupants with weak bones. The chest bag put nearly all forces into the rib areas, and the shoulder bag onto bones of the shoulder.

The reverse 3-point belt is a very simple and cheap method for an automaker to restrain an occupant for far-side collisions. The simulations demonstrated the effectiveness of such a countermeasure. All ATDs were prevented from moving over the center console, by grabbing the neck. Although effective at restraining motion using the neck, this sensitive area of the body can lead to further injuries. MADYMO was able to show significantly higher neck load forces for this situation, many beyond injury thresholds.

For an assessment of countermeasure effectiveness, the individual results of the five dummies were averaged together for every configuration. Figure 7 shows the neck loads of the dummies combined. The near injury threshold of the reverse belts can be seen. For all other airbag countermeasures, all neck loads fall far below injury thresholds and difficult to distinguish among one another.