音声ブラウザご使用の方向け: SKIP NAVI GOTO NAVI

Draft test method for the measurement of surface accessibility

Denise A. Chesney, Peter W. Axelson and Edward J. Hamilton* Beneficial Designs, Inc., Santa Cruz, CA *National Center on Accessibility, Martinsville, IN

ABSTRACT

ADA Accessibility Guidelines specify that ground and floor surfaces should be firm, stable, and slip-resistant. These specifications are subjective; objective methods for assessing surfaces are not given. Test methods for measuring surface firmness and stability are currently under development. One of these test procedures, the wheelchair rolling resistance test, utilizes the SMARTwheel to measure the work required to travel across different surfaces. This draft test procedure was evaluated on various surface types ranging from level rubber matting to indoor carpeting to pea gravel. Measurements were also taken on an adjustable ramp set to different grades, cross slopes and grade/cross slope combinations. The results demonstrated the feasibility of using this test procedure as an objective method to measure surface firmness. Comparisons were made between the work required to negotiate the level surfaces and the work required to propel up various ramp angles. Future work will include the determination of the specific ramp angle to be used as the performance criteria for accessible surfaces.

BACKGROUND

Current ADA Accessibility Guidelines for buildings and facilities contain subjective technical require-ments for ground and floor surfaces. According to ADAAG, accessible routes must be firm, stable and slip-resistant. Test methods exist for measuring slip resistance, but not for surface firmness or stability as it relates to access for people. Surface firmness primarily influences the effort required to roll/propel a wheelchair across the surface, while surface stability mainly affects wheelchair turning and maneuvering, and access by cane, crutch, and walker. In 1993, a Phase I research project was conducted to evaluate the feasibility of developing prototype devices to objectively measure the characteristics of a surface that affect its degree of accessibility [1]. Wheelchair rolling resistance was used as an indicator of the firmness of the surface and was determined by measuring the force required to pull a loaded wheel-chair across the surface. This method was found to be unreliable. An alternative wheelchair rolling resistance test method was also developed. This method utilized the SMARTwheel [2] to determine the amount of work required to travel across the surface. The work accomplished during this Phase I research project was the impetus for the development of a draft ASTM standard specification for the firmness of surface systems used under and around playground equipment.

RESEARCH OBJECTIVES

The measurement of surface firmness applies to all paths of travel, indoor and outdoor. Ideally, the same test method should be used to assess all types of surfaces ranging from indoor carpeting to outdoor recreation access routes and playground surface systems. The objectives of this research were to evaluate the wheelchair rolling resistance test procedure by measuring a wide range of surface types, and to compare the surface measurement results with measurements of the work required to roll up/across various grades and cross slopes.

METHOD

Test Equipment

The SMARTwheel is a specially instrumented main wheelchair wheel that is capable of measuring forces applied to the pushrim [2]. The SMARTwheel was used on a standard rehab wheelchair (Quickie2 by Sunrise Medical) with pneumatic rear tires and front casters, and a total weight of 15.5kg (34.2lbs.). During testing, the wheelchair was propelled by a 75.5kg (166lb.) able-bodied test subject. A 40-60% total weight distribution was used; 40±2% of the total weight of the wheelchair and test subject was distributed over the front casters; the rear wheels supported the remaining 60±2% of the total weight.

Test Surfaces

The test surfaces were divided into six different categories. The adjustable ramp was firm, stable and slip resistant (coefficient of friction greater than 0.8). The trail surface was hard-packed decomposed granite. The indoor accessible carpet used in testing had level cut pile 1.2cm (0.47in) in height and was not used with padding. The inaccessible carpet had a pile height of 2.1cm (0.83in) and was used with a 0.9cm (0.34in) composite foam pad.

1) Ramp grades: 2%, 5%, 6%, 7%, 8%, 10%, 14%, 20%, and 24% (1:50, 1:20, 1:16, 1:14, 1:12, 1:10, 1:7, 1:5, and 1:4, respectively)

2) Ramp cross slopes: 2%, 3%, 5%, 8%, 12%, 14%, 16%, and 20%

3) Ramp grade/cross slope combinations: 8%/3%, 8%/5%, 8%/8%, 14%/5%, and 14%/8%

4) Playground surfaces (all level): rubber, 2 engineered wood fiber surfaces, generic cedar chips, generic bark chips, #1 plaster sand, and 3/8"pea gravel

5) Trail surface: 2%, 8%, and 14% grade; 2%/5% and 8%/5% grade/cross slope combinations

6) Indoor inaccessible carpet: 2%, 8%, and 14% grade; 2%/5%, 8%/5%, and 14%/8% grade/cross slope combinations; 2% grade accessible carpet

Test Procedure

A computer data acquisition system recorded the forces measured by the SMARTwheel at a frequency of 50Hz. Only the forces tangential to the pushrim and parallel to the direction of travel were measured. Electronic triggers were placed 2m (6.56ft) apart and signaled the beginning and ending of each trial. The test subject propelled the wheelchair across the test surface a distance of 2m using four uniform pushes. Each trial was completed in 7.0 ±0.5sec. Trials were acceptable if the wheelchair came to a stop at the end of the 2m test run such that all the energy exerted by the test subject during the last propulsion stroke was used to move the wheelchair the specified distance. A minimum of seven trials were conducted on each test surface. Four of the surfacing materials (engineered wood fiber surfaces, generic cedar chips and bark chips) were tamped prior to each trial to remove any tracks created by the wheelchair during the previous trial. Pushrim force measurements were analyzed to determine the amount of work required to negotiate a each test surface.

Data Analysis

Torque versus time plots were generated for each trial and the impulse (area under the torque curve) was calculated using a trapezoidal approximation. The amount of work required to wheel across each test surface was determined by calculating the total impulse for all four propulsion cycles and dividing by the time to complete the trial. For each test surface, an average of five trials was calculated to determine the average work required to negotiate the test surface.

RESULTS

Work Required vs. Ramp Configuration

A total of 22 different ramp configurations were objectively measured. As expected, increasing the grade of the ramp increased the amount of work required to travel up the ramp. Grades of 14%, 20%, and 24% required 70%, 130%, and 180% more work, respectively, than the standard 8% grade ramp. Propelling the wheelchair such that it came to a stop at the end of the 2m test run was difficult to achieve on surfaces with cross slope only (no grade) and little rolling resistance. Therefore, a slight grade of 2% had to be used with all cross slope ramp tests in order to achieve accurate and repeatable results. Similar to the ramp grade results, the work required to travel across the ramp with cross slope increased linearly with the amount of cross slope. A cross slope of 20% required 70% more work than a 2% cross slope.

Multiple linear regression analysis was performed to determine the effects of grade and cross slope on work required. This regression explained all of the variation in the work required and was highly significant (p<0.01). As expected, both grade (p<0.01) and cross slope (p<0.01) were found to be significant factors that affect the work required to traverse the ramp. Based upon this analysis, an equation for predicting the amount of work required for any ramp configuration was determined:

Work Required (N-m) = [2.37 x Grade] + [0.70 x Cross Slope] + 3.36

The amount of work associated with ramp configurations that can not be measured according to the rolling resistance test procedure could then be estimated using this equation.

Comparison of all ramp results indicated that cross slopes of 16% or less require less work to negotiate than the standard 8% ramp with no cross slope. The work required to travel across the ramp with a 20% cross slope was not significantly different than the ramp with an 8% grade and 3% cross slope. These results also showed that grade/cross slope combinations of 8%/8% and 14%/8% required significantly less work than grades of 10% and 20%, respectively.

Work Required vs. Surface Type

All playground surfacing materials were significantly different (p<0.01) except the two engineered wood fiber surfaces. Compared to the engineered wood fiber surfaces (15.9N-m, 16.8 N-m), generic cedar chips (21.3N-m) required about 30% more work to negotiate, and generic bark chips (25.4N-m) required about 55% more work. Both engineered wood fiber surfaces required significantly less work than the ramp with a 6%grade (p<0.01) and an average of 25% less work than the standard ramp with an 8%grade. No significant difference was found between the cedar chips and the 8% grade ramp (p>0.20). The work required to negotiate bark chips exceeded that of the ramp with an 8% grade by 16.5%.

Testing could not be conducted on sand and pea gravel according to the draft test procedure. On both surfaces, the front casters imbedded into the surface such that the test subject was incapable of propelling the wheelchair.

The trail surface required 9-24% more work to negotiate compared to the corresponding grade/cross slope ramp configurations. The inaccessible carpet was 23-172% more difficult to negotiate than the ramp and this percentage decreased as the work required increased. Three surfaces (ramp, trail, and inaccessible carpet) were examined more closely to determine the effects of grade, cross slope and surface on the amount of work required. Grades of 2, 8, and 14% in combination with cross slopes of 0, 5 and 8% were compared. Within this small group of test surfaces, the amount of work required was not related to cross slope. Work required to negotiate the test surface was significantly related to grade (p<0.001) and surface (p<0.005). The effects of surface type on the amount of work required decreased as grade increased. In other words, surface type contributed less to the differences in the amount of work required at steeper grades compared to lower grades.

Comparisons (two tailed t-test) made between all test surfaces revealed that the following groups of test surfaces were not significantly different:

  • accessible carpet with 2% grade; one engineered wood fiber surface (p>0.20);
  • ramp with 12%cross slope, one engineered wood fiber surface (p>0.20);
  • ramp with 8% grade, inaccessible carpet with 2% grade, generic cedar chips (p>0.20);
  • ramp with 8% grade/5% cross slope, bark chips, inaccessible carpet with 5%cross slope (p>0.20);
  • ramp with 14%grade/5%cross slope, trail with 13% grade (p>0.20);
  • ramp with 14% grade/8% cross slope, trail with 13% grade (p>0.20);

DISCUSSION

The draft wheelchair rolling resistance test procedure was used on a wide range of surface types. Limitations occurred on surfaces that were clearly "inaccessible" (sand and pea gravel) and on surfaces that provided very little rolling resistance (rubber and ramps with cross slope and no grade). The repeatability and sensitivity of the test procedure were good. The test procedure was sensitive enough to differentiate between small changes in grade, cross slope, and surface type. Future testing will include the use of three different test subjects to further evaluate the repeatability of the test procedure. Funding for research to develop a portable device to quickly and easily measure firmness is pending.

The draft test procedure provided objective measurements of surface firmness that could be compared with the work required to negotiate various ramp angles. ADAAG specifies maximum slopes for accessible ramps (1:12 or 8.3%). Therefore, it is feasible that the work required to negotiate a specific ramp angle could be used as the ADAAG pass/fail performance criteria for accessible surfaces. Additional research must be conducted to determine the appropriate reference ramp angle to be used. For outdoor recreational environments, pass/fail performance criteria is not desired. Instead, a means for disclosing objective information about the firmness of an outdoor recreation access route or recreation trail is needed. This firmness measurement could be used to classify the surface into categories (e.g. paved, hard, firm, soft, very soft). This research focused on test methods for assessing surface firmness. Future research to investigate test methods and develop devices for measuring surface stability has been proposed and funding is pending.

References

1. Chesney, D.A., Axelson, P.W., Williams, J.R., "Surface Assessment Devices for Accessibility," Proceedings of the RESNA '94 Annual Conference, 17-22 June 1994, pp.275-277.

2. Asato, K.T., Cooper, R.A., Robertson, R.N., and Ster J.F., "SMARTwheel Development and Testing of a System for Measuring Manual Wheelchair Propulsion Dynamics," IEEE Transactions on Biomedical Engineering, Vol. 40, No. 12, 1993.

ACKNOWLEDGMENTS

This research was funded by the National Center on Accessibility in Martinsville, Indiana, a program of Indiana University in cooperation with the National Parks Service Office on Accessibility. Initial surface measurement research was funded by the National Center for Medical Rehabilitation Research at the National Institute on Child Health and Human Development in the National Institutes of Health through Phase I SBIR Grant No. 1 R43 HD30979-01.

Denise A. Chesney, M.E. Beneficial Designs, Inc. 5858 Empire Grade, Santa Cruz, CA 95060 tel: (408) 429-8447 fax: (408)423-8450 email: pax@netcom.com