Quasi-static methods to evaluate seat strength in rear impacts

Abstract Objective Various methods have been used in the past 50 years to apply Quasi-static load to a seat in the rear direction and measure seat performance in rear impacts. This study compared five of the most-common test procedures to evaluate seats. In addition, occupant mass and center of gravity are discussed as important characteristics of rear loading of seats. Method Data was collected and analyzed from five different seat pull tests, including FMVSS 207, modified FMVSS 207, QST, body block and FRED II. Test data included peak force, moment and angle at peak moment. Occupant loading height of was determined using body segment weights and position in the forward (x) and vertical (z) directions based on anthropometry data. Results Some of the inherent differences in the tests are shown by comparing data with the same seat structure. The QST and FRED II use a lower height of loading than FMVSS 207. The QST and FRED II peak moment and force did not coincide with the same seatback angle as in FMVSS 207 and body block testing. Center of gravity height varies depending on whether the whole body or only the upper torso is considered. For the 50th male, it is 171.5 mm (6.8”) with the whole body and 246.7 mm (9.7”) with the upper torso. Conclusion Results from different tests cannot be readily compared because of different loading conditions, including body shape and height of load about the H-point, which can cause the seat structure to respond differently.


Introduction
Some have advocated up to 11,298 Nm (100,000 inlb) seat strength in FMVSS 207 tests prompting a debate about seat strength and yielding in rear impacts . Some argued that stronger seats may prevent contact with 2 nd row seat occupants and/or structures, and some argued that stronger seats may enhance injury potential to front-seat occupants in more-common, less severe crashes, for out-ofposition occupants and for occupants with spine degeneration (Warner et al. 1991;Strother et al. 1994;Benson et al. 1996;Viano and Parenteau 2008;Viano 2011;. Seats have become stronger overtime, despite FMVSS 207 requirements that remained constant, in response to new head restraint requirements and other design changes (Saunders et al. 2003;Viano and White 2016). Over more than 50 years of research and study a common need has been the measurement of strength and stiffness characteristics of seat structures as it relates to occupant loading in rear impacts.
Seats and head restraints provide the main restraint early in rear crashes. They support the occupants, and yield and absorb energy to "ride down" the acceleration forces. Seat performance has been studied since the 1950s. The Society of Automotive Engineers (SAE 1963) approved recommended practice J879 for the testing and evaluation of seat strength. One procedure required a 480 Nm (4,250 inlb) static rearward moment on the seatback. Federal Motor Vehicle Safety Standards (FMVSS) 207 and 202 were issued in the late 1960s to address seat geometry and strength in FMVSS 207, 32 FR 2415(NHTSA 1967) and in FMVSS 202, 33 FR 2945(NHTSA 1968. FMVSS 207 was based on SAE J879. A portion of the standard required seats to withstand a rearward moment of 373 Nm (3300 inlb) with a rearward load applied at the upper cross member of the seatback frame. This is still in effect today. The moment was calculated about the seating reference point (SgRP). Direct loading of the seat frame creates a test that is simple to conduct with high repeatability. There are drawbacks to this loading method when relating the results to seat safety such as ignoring the seat to occupant interaction and an unrealistically high loading condition. The initial pull height varies, but for a modern seat can be 46 cm (18.1") or more above the SgRP.
FMVSS 207 requires that a seat support a rearward moment about the SgRP of 373 Nm (3,300 inlb) from a horizontal point load on a reinforcing bar at the upper cross member of the seatback frame. The point load is applied at the mid-point of the seatback. The height above the SgRP of this loading point varies across seat designs and can require some estimation on high back designs without a clear upper cross member. As a result, the maximum load required varies depending on the frame height and construction. While the FMVSS requirement is 373 Nm (3,300 inlb), the test is often conducted to an ultimate load. This loading at the top cross-member often damages the seatback frame in areas higher than would be expected with occupant loading. For purposes of comparing seat strengths across various designs, Ford has conducted a series of rearward pull tests on seats using a consistent 31.8 cm (12.5") pull height above the SgRP. This height better approximates the center of gravity of the 50 th percentile upper torso than the top crossmember as required in FMVSS 207.
Loading the seat frame directly ignores the interaction between the occupant and the seat. Various methods were developed to simulate occupant loading into the seat trim and frame. The quasi-static tests include pull tests using various rigid body blocks, and push tests such as the General Motors (GM) Quasi-static Seat Test (QST) which utilizes a Hybrid II Anthropometric Test Dummy (ATD). The body block test is a modification of the FMVSS 207 test procedure. It involves a quasi-static rearward pull on the seatback using a rigid body block in place of direct loading to the seat frame. Severy et al. (1968) conducted a series of rear crash tests and developed a body block method to load the seat rearward. Body block pull test techniques developed by  have been used by Strother and James (1987) and Warner et al. (1991) with modifications. The size and shape of the block has been described and the results of over 300 tests of seat strength have been measured since the mid-1960s. The bulk of the testing has been conducted by Collision Research & Analysis (CRA) and the test procedure has been described by Viano and White (2016). The rear pull load is applied at a point initially 35.6 cm (14") above the seat cushion.
GM developed a Quasi-static Seat Test (QST) to assess occupant interactions with the seat in rear loading (Viano 2002. The test involves a modified Hybrid II crash test dummy placed in the design seating position and subjected to rear displacement by a hydraulic ram loading through the flexible lumbar spine element. This distributes force through the buttocks and upper torso and simulates occupant loading of the seat in a rear impact. The hydraulic ram is supported on sliding rails so the dummy can freely move upward and sideways as the seat deforms. The load is applied at a point initially 15.5 cm (6.1") above the H-point of the crash dummy.
FRED II was developed using the SAE J826 H-pt manikin as a refinement to the body block testing technique . The FRED device is constructed utilizing J826 compliant back and butt forms mated together with a mechanism hinged at the h-point and ballasted to match the full weight and distribution of the J826 manikin. When adjusted on the seat according to the SAE procedure its H-point will match the SgRP, providing a simple calculation of SgRP moment. The pull height is adjustable to suite the purpose of the test, but has been evaluated in comparison testing at a location of 17.1 cm (6.75") above the SgRP.
The objective of this study was to compare and describe the most common quasi-static seat test procedures including the regulated and modified FMVSS 207, body blocks, QST and FRED II. The seated occupant mass and center of gravity was also discussed as an important influence as an occupant loads the seat in a rear impact.

FMVSS 207
Figure 1 top image shows a FMVSS 207 setup with a point load applied rearward at the top cross-member of the seatback. A bar is attached to the top of the seatback and an eyelet at the centerline of the seat serves as a point-load attachment for a chain to the hydraulic ram. The load attachment to the seat is not specified and as a result lab to lab variability exists which can result in minor variation in the results. Pull bar height varies depending on seat design. Figure 1 shows a typical setup with a 41.1 cm (16.2") above the SgRP or H-pt. It also shows the setup of the modified test utilizing a pull bar attached 31.8 cm (12.5") above the SgRP. Because the moment arm is a measurement of the pull location above the SgRP the location of the SgRP on any particular seat design is required. This is obtained either through design information of the seat or a measurement on a fully trimmed seat using the SAE measurement device. Traditionally the moment is calculated using the initial pull height throughout the test. Alternately the adjusted moment can be calculated throughout the test using additional measurements to track the pull height. Care must be taken when comparing results of multiple tests that the methods match. Figure 1 shows setup with a horizontal and centered direction of rearward pull. The body block is positioned on the seat and loaded rearward by a cable with an initial height 35.6 cm (14") above the seat cushion surface. The seat is mounted to a portion of the vehicle floor pan so that the support for the seat tracks is equivalent to its installation in the vehicle. The seat is positioned mid-track with the seatback approximately 20 deg rearward of vertical. The body block simulates the force of an occupant's torso on the seatback in a rear crash. The block weighs 13.6 kg (30 lb). It is placed on the seat cushion and rests against the seatback. A hydraulic ram pulls the cable rearward. The rearward pull is horizontal to the fixture and is terminated when one of three conditions is met: (1) the seat angle reaches approximately 45 deg, (2) the seat structure cannot support an increase in load or (3) the body block migrates upwards more than a few inches. The moment on the seat is calculated using the initial 35.6 cm moment arm as a constant. Please see Viano and White (2016) for additional information on the test procedure and equipment.

QST (quasi-static seat test)
The QST setup for the seat loading is shown in Figure 1. It involves a Hybrid II ATD subjected to rearward displacement by a hydraulic ram loading through the flexible lumbar spine element, distributing force into the seat through the buttocks, lower torso and upper torso. The hydraulic ram is supported on sliding rails, so the dummy moves upward and sideways as the seat deforms. The system is counterbalanced so only the mass of the dummy vertically loads the seat. The kinematic sequence in the test mimics the dynamics of torso loading in a rear impact. More than fifteen channels of response data are collected in a QST. They include the trajectory of the occupant, occupant loads into the seat, forces supporting the front and rear attachments to the vehicle floor and rotation angle of the right and left side of the seatback.
Seat responses were calculated from the force and displacement data, including the H-point moment (M h ) and seatback pivot moment (M p with F x the horizontal ram load, z h the H-point offset, z p the z offset from the ram to the seatback point in the recliner and z the vertical travel of the ram. Twist of the seatback is computed as the difference in angle from the right and left sides of the seatback. The torsional load (T) on the seatback is T ¼ F x y with y the lateral displacement of the ram and dummy. Energy (E) transfer to the seat is computed by E ¼ SF x dx, the integration of the horizontal load from the dummy over the ram displacement. Summary data from different seats was compared using values at an average 60 deg seatback angle rearward of vertical.

FRED II
The FRED II device is still in a development phase and at this point there is not a large body of available tests. Based on limited test results, the FRED II testing method provides a good balance between the complex QST and the traditional body block method. Please see  for additional information on FRED II test set-up and comparison results to the traditional body block tests.

Seat tests and matches
A series of FMVSS 207 tests were conducted by MGA and Ford using a 31.8 cm (12.5") pull height above the SGRP on the trimmed seat. The seat was tested untrimmed. Thirtyfour included data on the force and angle derived from displacement. Table 1 lists the seats by make, model and model year range. The model year varied from 1993 to 2004. Seats were selected to represent single and dual recliner seats used with conventional restraint systems and seats with integrated restraints (SIR or ABTS). They were inspected to ensure no existing damage and the vehicle VIN was checked to exclude collision damage that might affect the seat. A series of body block tests were conducted by CRA using a body block with 35.6 cm (14") pull height on a fully trimmed seat. More-

Height of occupant loading
In many cases the purpose of a static strength test of the seatback is to provide insight on seat performance related to occupant loading. In this context it is helpful to understand the distribution of mass of the typical occupant. When an occupant loads the seat in a rear impact, the force can be resolved to a point load acting rearward at the center of pressure (cp). The cp is the center of gravity (cg) of the body with the assumption that the effective mass acts as a lumped mass. The cg of the 5 th , 50 th and 95 th seated occupant was determined using body segment weights and position in the forward (x) and vertical (z) directions based on US DOT data on anthropometry of motor vehicle occupants (Schneider et al. 1983). The coordinate system of the segment cg locations is centered at the seated occupant Hpoint. The overall occupant center of gravity was determined by the sum of each body segment mass times the position divided by the weight. Two calculations were made. One for the whole body and another for only the upper torso. Table 1 summarizes the peak force, moment and angle at peak moment for matched tests using the FMVSS 207 with 31.8 cm (12.5") pull height, body block with 35.6 cm (14") pull height and QST with varying pull height used in the calculation. There were 13 matches with the same or sister seat run with the three different methods. A 14th match includes only the body block and QST methods. In some cases, there were multiple tests with the same seat. The summary includes the average responses for the seat in a particular test. For the QST tests, the average height of the ram load was 17.5 cm ± 1.8 cm (6.9" ± 0.7") for the single and dual recliner seats and 25.1 cm ± 0.4 cm (9.9" ± 0.2") for the ABTS seats.

Results
There are inherent differences in the seat loading and response with different test methods used on the same seat structure. As the QST uses a lower loading height than the other test methods it is expected that a higher horizontal load would be required to get to the ultimate torsional strength of the seat. While this is reflected in most of the results it is not always the case. Additionally, the QST compared to the body block testing show the peak moment and force do not always occur at the same seatback angle. The results of the different test methods cannot easily be compared as differences in the test methods can cause the seat structure to respond differently. Table 2 summarizes the mass and position for the whole body and upper torso of the seated 5 th female, 50 th male and 95th male based on anthropometry of seated occupants. For the whole body, the height of the cg is 171/5 mm (6.8") for the 50 th male above the seated hip point. If only the upper torso is considered, the height of the cg is 246.7 mm (9.7") above the hip point. The individual body segment mass and position is provided in the Appendix (see online supplement) with a graphic position of each body segment with respect to the hip point. These results are reported relative to the hip point of the respective occupants which is comparable to the seating reference point of a fully trimmed seat. Table 2 also summarizes the typical heights of loading in seat tests. FMVSS 207 uses the top cross-member of the seat, which varies between seat designs but is about 457.2 mm (18") above the SgRP. The body block height is referenced to the seat cushion, which can be higher or lower than the seat SgRP height. The initial loading with the QST is 154 mm (6.1") and the dummy typically moves rearward and upward as the force increases. The FRED II has the ability to change the pull height and angle as needed.  used an initial height of 17.1 cm (6.75") above the SgRP.

Discussion
Various methods have been used to quasi-statically load a seat in the rearward direction to gain information related to seat performance in rearward impacts. Researchers have attempted to relate this information to a seat designs ability to reduce the risk of injury to occupants in rear impacts. There have been ongoing efforts to refine testing methods for rear loading of seats. This study compared different methods to evaluate rearward seatback strength.
The FMVSS 207 and the modified FMVSS 207 tests apply the load directly to the seatback frame. This results in a highly repeatable and simple test to conduct but does not include an occupant to seat interaction. The original name of the Federal Motor Vehicle Safety Standard was "207 Anchorage of Seats -Passenger Cars." The purpose was to "test seats, their attachment assemblies and their installation to minimize the possibility of failure by forces acting on the seat as a result of a vehicle impact" (NHTSA 1967). The test is useful in understanding how a seat structure manages load. It is useful in comparing seat structures and assessing changes. The direct loading of the seat frame and the height of the loading can produce local deformation patterns in the seatback structure that do not represent occupant loading in rear impacts. As a result, it is difficult to gain more than a basic understanding of a seats rearward capabilities from this type of testing.
There have been several methods developed to load the seat rearward with a body form of an occupant. This body block testing includes the occupant to seat interaction in a simple manner. The most common type of body block loading is presented here. The block loads the seatback trim and frame, rather than at the upper cross member of the seatback frame as in FMVSS 207. This testing can provide more information than a direct pull but has limitations in that it is only intended to simulate upper torso interaction. The height of the applied load and the shape of the block does not capture lower seatback interaction which can play an important role in seat to occupant interaction.
QST is the most complicated testing methods presented in this study. It captures lower torso interactions and applies the loading at a height representative of a human interaction. While the QST method is more representative of real-world occupant loading, it is too complex for routine seat testing. It requires a Hybrid II ATD as well as specialized fixturing and instrumentation to conduct. The QST has proven to be a useful research tool in understanding the interaction between occupant and seat in a rearward impact. The specialized equipment has prevented it from being a tool used in development of seats. Few labs can conduct this test today. FRED II may provide a balance between complexity and usefulness when studying occupant loading of a seat in rear impacts.
All the available quasi-static test methods are useful in understanding how a seat and occupant interact in a rearward impact, but care must be taken by the researcher in understanding the limitations of each. Some test methods are tailored toward understanding the seat frame response while others are more focused on the occupant to seat interaction. Determining the optimum seat characteristics for rear collisions remains a challenge due to the wide range of occupants and crash severities. Despite these challenges, rear collisions continue to be the safest of all crash types, including those rear collisions involving a high delta V (Parenteau and Viano 2021).
In our experience, quasi-static tests are useful during the development of a seat. It allows analysis of the seat as a component early in the design process before sled or crash testing is available. Each seat design has its own distinctive modes of frame deformation, and the same seat structure may have different modes whether it is a manual or power, or 2-way, 4-way, or 8-way design. An advantage of quasistatic testing is that good camera coverage can be made of specific areas of the seat in testing. In addition, since quasistatic testing occurs early in the development of a seat, the designer can modify the design to change the strength and areas of deformation

Limitations
Quasi-static testing is one way to determine characteristics of a seat in rear loading. It focuses on seat strength, which is a measure of occupant retention. The testing can be run until the seat no longer supports substantial load. This demonstrates the ultimate yield characteristics of the seat structure, which can be important in high-speed rear impacts or in rear impacts with very heavy occupants.
Quasi-static testing does not include head restraint interactions or seatbelt performance. Quasi-static testing can be done early in the development of a seat before sled or crash testing can be done with instrumented ATDs. The dynamic tests provide additional information about the seat and head restraint and can be tailored the crash pulse of a specific vehicle and different occupant sizes and seating position.
It would be interesting to compare the applied moment versus seatback angle in the quasi-static tests since this measurement would be useful. However, the QST is the only method that measures to changing moment arm of the loading and calculates the applied moment versus the average seatback angle (Viano 2002). It is thus not possible to compare the different methods using moment and angle. The various quasi-static tests show deformation modes of the seat if a sufficient load is applied in the test. The deformation varies by the seat type and design and whether it is a manual or a power seat and the degrees of adjustability of the cushion and seatback. There is some validation of the type of seat deformation during the development of the QST procedure, but none of the other methods have compare the seat deformation in the quasi-static test with sled tests. There is also minimal comparison of seat deformation between the various methods.
A series of identical seats were tested with the QST method and in sled tests to study the similarity of the deformation modes (Viano 2002). The matched testing on three different seat types showed the deformation areas in the QST match those seen in sled testing. The authors are unaware of other studies looking at the similarity of seat deformation areas in quasi-static and sled testing.
The seats used in this study had different designs including dual and single recliners. The all-belts-to-seats (ABTS) design was also assessed. The model year varied from 1993 to 2004. The collection of tests is a convenience sample from various sources.

Funding
The author(s) reported there is no funding associated with the work featured in this article.