FORWARD DYNAMIC STABILITY TEST FOR MANUAL WHEELCHAIRS
Ronald P. Gaal, Peter F. Pfaelzer, Ralf D. Hotchkiss Wheeled Mobility Center, School of Engineering San Francisco State University San Francisco, CA
ABSTRACT
In the interests of wheelchair safety and design, we have developed a simple, repeatable test that can be used to determine the forward dynamic stability of a wheelchair. This is done by measuring the wheelchair's ability to handle a common riding situation: hitting a low obstacle while rolling down a sloped plane.
Test methods included creation of a variable-height wheelchair obstacle and means of guiding a rolling wheelchair and accurately determining its speed. An anthropomorphic test dummy was used.
Seven different wheelchairs and/or configurations were tested, and the results confirmed that the test is repeatable and can be used to quantify differences in forward dynamic stability between 1) different wheelchairs, 2) different rolling speeds upon impact with the obstacle, and 3) different types of caster wheels installed on the same wheelchair.
In the quantitative results there were no differences between the dummy being belted versus not belted into the wheelchair seat.
INTRODUCTION
Epidemiologists have acknowledged that adverse wheelchair riding incidents are a significant problem and that the majority of serious wheelchair rider injuries result from tipping or falling out of the chair (1-4). This highlights the limited stability of current indoor/outdoor wheelchairs. One avenue for addressing wheelchair safety has been wheelchair testing and the publishing of American National Standards Institute (ANSI)/RESNA and International Standards Organization (ISO) wheelchair testing standards (5). The standards do not yet include tests of dynamic stability for manual wheelchairs.
METHODS
The basic test
The wheelchair, loaded with a test dummy, was rolled down a 5ø sloped plane. At a specified speed of travel, its front wheels hit a low, rigid obstacle. A test observer noted whether the wheelchair overcame the obstacle or tipped forward, or if the dummy fell out of the seat. (For this test, a tip was defined as the wheelchair tipping far enough that one or both footrests touched the test plane, or the wheelchair tipped over.) If the wheelchair overcame the obstacle with no tip or dummy fall, then the test was repeated with an obstacle of incrementally greater height. This procedure was repeated until eventually tipping and/or falling were observed. The primary result for each wheelchair, at each tested speed, was the height of the largest obstacle that did not cause the wheelchair to tip or the dummy to fall out.
The obstacle
The variable-height obstacle was built up using 1/4 in. (6 mm) thick, 3 in. by 48 in. (80 mm by 1,200 mm) rectangular pieces_an aluminum top step and plywood understeps. These were bolted to the asphalt test plane, across the path of travel (Figure 1). The aluminum top step had one rounded edge where it first contacted the oncoming wheelchair (radius = 1/4 in. or 6 mm). The plywood understeps were slotted at the mounting bolt locations, so that they could be easily added or removed to adjust obstacle height incrementally from 1/4 in. up to 3 in. maximum.
Generating wheelchair rolling speed
Each wheelchair (or configuration thereof) was tested at two rolling speeds, 3 and 6 mi/hr (1.3 and 2.7 m/s). To generate and control wheelchair speed we released the wheelchair from a standstill-using only gravity to accelerate it down the slope-and measured the speed immediately before impact. The impact speed was adjusted as needed by moving the wheelchair's starting position up or down the slope. (This method was used in stability tests performed by Kirby et al (6).) We were able to set the test speed of 3 mi/hr to within ñ 2%, and 6 mi/hr ñ 1%.
Wheelchair direction control
To generate 6 mi/hr at impact with the obstacle, the wheelchairs needed to roll nearly 18 ft (5.5 m) down the test plane. A simple guide rail system was developed to insure that the wheelchairs' front wheels consistently struck the obstacle simultaneously.
The guide rail was a 7/8 in. diameter by 20 ft long (2.2 cm by 610 cm) steel tube, supported 11 in. (28 cm) above the test plane. A lightweight, low-friction guide arm assembly was attached to the wheelchair (Figure 2). This assembly was fabricated primarily of 3/16 in. (5 mm) diameter steel bar stock, a steel hinge, and nylon rollers (colored white). The hinge, at the wheelchair-end of the arm, allowed the roller-end of the arm to move vertically, so it could be lowered onto any part of the guide rail. A mounting cuff and stabilizing strut allowed easy, secure mounting on a variety of wheelchair frames, using common band clamps. The mounting location was usually on the wheelchair's footrest strut. Just before the wheelchair hit the obstacle, the roller assembly would run off the end of the guide rail, so the rail did not interfere with the collision.
Wheelchair speed measurement
Speed measurements were taken as close as possible to the time of impact with the obstacle, to minimize measurement error due to wheelchair acceleration. Our system measured wheelchair speed within the last 6.5 in. (170 mm) of travel before impact.
Returning to Figure 2, notice the secondary timer-trigger arm, mounted to the same hinge as the roller guide arm. It carried a 6.0 in. (150 mm) long black board that passed through a light gate. (The light gate was an infrared beam and optical sensor, appearing in the figure as a backwards-C-shaped item, mounted on a secondary rail). The timer (not shown) indicated wheelchair speed by measuring the time during which the board interrupted the light beam. An elastic cord attached to the timer-trigger arm minimized unwanted movement (play) in the hinge, to improve time measurement accuracy.
The use of 3/16 in. steel bar to fabricate the roller guide and timer-trigger arms allowed us to bend and align the assembly to fit a variety of wheelchairs, using only hand tools. The strength and stiffness of the arms were appropriate to the loads, and the entire assembly weighed just 11 oz (0.31 kg).
Test dummy
Test wheelchairs were loaded with a 190 lb (86 kg) anthropomorphic test dummy of the type used in automobile crash testing. Each wheelchair was tested with and without use of a lap belt to restrain the dummy.
Test wheelchairs
Test wheelchairs were configured in the manner specified in ANSI/RESNA standard WC/01-Determination of static stability, with the lowest part of the leg support/footrest 50 mm (2.0 in.) above the test plane. Rear wheels were adjusted to their farthest rearward position. Forward and backward static stability was measured.
The following seven wheelchairs/configurations were tested:
Hospital-Solid: Hospital-style folding wheelchair with solid tires, 8" caster wheels.
Breezy-Solid: Quickie Breezy folding wheelchair with 7.5" solid casters.
Breezy-Pneumatic: Quickie Breezy folding wheelchair with 8" pneumatic casters.
Quickie -Solid: Quickie 1 rigid wheelchair with 5" solid casters.
Quickie 1-Pneumatic: Quickie 1 rigid wheelchair with 6" pneumatic casters.
Whirlwind II: folding chair with 7" "Zimbabwe" T-section rubber caster wheels.
Omnidirectional: Prototype wheelchair using 8" omnidirectional front wheels.
Videotaping of test trials
Over 200 controlled test trials were run. The trials were videotaped to allow further analysis, including slow motion viewing. Figure 1 shows two refinements of the test setup: 1) a number flipchart for videotaping, that allows easy identification of test trials when scanning the videotape; and 2) visual markers on the dummy, wheelchair, and background wall, which would assist in making future kinematic analyses.
RESULTS
Test results are summarized in Table 1.
In the quantitative results (i.e. highest obstacle cleared) there were no differences between the dummy being belted versus not belted into the wheelchair seat, although we observed clear differences in how the wheelchair and dummy tipped and/or fell.
CONCLUSIONS
The test program was highly successful, producing useful relative measurements of forward dynamic stability. The test was repeatable, with consistent results. Tests of seven wheelchairs/configurations confirmed that the test can be used to reveal and quantify differences in forward dynamic stability between 1) different wheelchairs, 2) different rolling speeds upon impact with the obstacle, and 3) different types of caster wheels installed on the same wheelchair. Our test methods and specialized equipment could be re-created by other wheelchair test facilities at reasonable cost.
Since there was no difference in quantitative results between the dummy being belted versus not-belted into the wheelchair seat, the test might reasonably be simplified to employ an ANSI/RESNA (or ISO) standard-style dummy, secured to the wheelchair. This hypothesis should be tested.
In November 1995 we submitted draft language for a proposed new Part of ISO 7176-Wheelchairs, based upon this test, to the ISO Breakout Group on Wheelchairs-Part 2: Dynamic Stability.
REFERENCES
1. Calder CJ, Kirby RL. Fatal wheelchair-related accidents in the United States. Am J Phys Med Rehabil 1990:69:184-90.
2. Ummat S, Kirby RL. Nonfatal wheelchair-related accidents reported to the national electronic injury surveillance system. Am J Phys Med Rehabil 1994:73:163-7.
3. Kirby RL, Ackroyd-Stolarz SA, Brown MG, Kirkland SA, MacLeod DA. Wheelchair-related accidents due to tips and falls among noninstitutionalized users of manually propelled wheelchairs in Nova Scotia. Am J Phys Med Rehabil 1994:73:319-30.
4. Vlahov D, Myers AH, Al-Ibrahim MS. Epidemiology of falls among patients in a rehabilitation hospital. Arch Phys Med Rehabil 1990:71:8-12.
5. Horne J. U.S. committee adopts wheelchair standards. P T Bulletin, Feb 1991.
6. Kirby RL, McLean AD, Eastwood BJ. Influence of caster diameter on the static and dynamic forward stability of occupied wheelchairs. Arch Phys Med Rehabil 1992:73:73-7.
ACKNOWLEDGEMENTS
This project was supported in part by the San Francisco Injury Center for Research and Prevention, through grant #R49/CCR90397-06 from the US Centers for Disease Control and Prevention (CDC). Rehabilitation engineering graduate students Khaled Daoudi and Mohammad Abu Shalbak were instrumental in carrying out the test program, with technical support from the School of Engineering at SFSU. Matthew Peak helped with graphics.
Ronald (Ronny) P. Gaal, P.E. Wheeled Mobility Center San Francisco State University 2746 1/2 Fulton Street Berkeley, CA 94705-1032 (510) 849-3202 rgaal@sfsu.edu Forward Dynamic Stability Test