U.S. Department of Transportation
Federal Highway Administration

TECHNICAL ADVISORY

T 5040.17
December 23, 1980

Par.

1. Purpose

2. Background

3. Skid Accident Reduction Program

4. Pavement Design, Construction, and Maintenance

5. Wet Weather Accident Location Studies

6. Pavement Skid Resistance Testing Program

1. PURPOSE. To provide guidance for State and local highway agencies in conducting skid accident reduction programs.

2. BACKGROUND

   1. This Technical Advisory provides a general overview of factors that should be considered as elements of any Skid Accident Reduction Program. This Technical Advisory supports current Federal Highway Administration (FHWA) policy and will be revised as appropriate to reflect changes in policy as they occur.

   2. The existing requirement for skid resistance pavementsare contained in several documents including Highway Safety Program Standard No. 12, Highway Design Construction and Maintenance (23 CFR 1204.4), Federal Highway Program Manual (FHPM) 6-2-4-7, Skid Measurement Guidelines for the Skid Accident Reduction Program. Other sources of technical advice are cited in the appropriate sections of this Technical Advisory.

   3. Highway Safety Program Standard 12 (HSPS No. 12) states that every State shall have a program of highway design, construction, and maintenance to improve highway safety. This program shall provide that "there are standards for pavement design and construction with specific provisions for high skid resistance qualities." The HSPS No. 12 also requires that each State have a "program for resurfacing or other surface treatment with emphasis on correction of locations or sections of streets and highways with low skid resistance and high or potentially high accident rates susceptible to reduction by providing improved surfaces." In discharging the responsibilities of FHWA, the Division Administrator should determine the acceptability of specification requirements and construction practices for placing, consolidating, and finishing both asphalt concrete and portland cement concrete pavements. Such determinations will rely on the highway agency to research, evaluate, and document the performance of the various aggregates, mix designs, and construction practices used.

   4. Even though the use of studded tires is beyond the control of most highway agencies, their use can cause significant wear on the pavement surface texture. For example, grooves sawed in concrete pavements have worn completely down in as short a time as 2 years. States are encouraged to ban or restrict the use of studded tires.

   5. Legislative actions in recent years support a general duty of any highway agency to "...maintain the roadway in a reasonable safe condition. This would involve, in essence inspection, anticipation of defects, and conformity with generally accepted standards and practices."* (Engineering and Government Liability, David C. Oliver, FHWA, an unpublished paper presented to the American Road and Transportation Builders Association Local Officials Meeting, St. Louis, Missouri, August 23, 1978.)

   The practical result is that highway agencies should have anorganized system to identify and correct hazardous locations in a cost-effective manner, as well as a comprehensive pavement management program to design, construct, and maintain highways in conformance with reasonable standards. Such a systematic process is the best way to execute the highway agency's duty to maintain a reasonable safe roadway.

3. SKID ACCIDENT REDUCTION PROGRAM. Each highway agency is encouraged to develop and mange a skid accident reduction program to reflect the individual needs and conditions within the State. the purpose of a skid accident reduction program is to minimize wet weather skidding accidents through: identifying and correcting sections of roadway with high or potentially high skid accident incidence; ensuring that new surfaces have adequate, durable skid resistance properties; and utilizing resources available for accident reduction in a cost-effective manner. A program comprised of at least the following three basic activities, if faithfully implemented, should enable the highway agency to comply with HSPS No. 12.

   1. The evaluation of pavement design, construction, and maintenance practices through its pavement management program to ensure that only pavements with good skid resistance characteristics are used.

   2. The detection and correction of locations with a high incidence of wet weather accidents utilizing (1) the State and local accident record systems, and (2) countermeasures for locations with high wet weather incidences, to ensure that existing highways are maintained in a safe condition.

   3. The analysis of skid resistance characteristics of selected roadway sections to:

      (1) ensure that the pavements being constructed are providing adequate skid resistance,

      (2) develop an overview of the skid resistance properties of highway systems,

      (3) provide up-to-date information of the pavement management process, and

      (4) provide data for use in developing safety improvement projects and the implementation of cost-effective treatments at appropriate locations.

4. PAVEMENT DESIGN, CONSTRUCTION, AND MAINTENANCE

   1. Pavement Design

      (1) Current pavement design practices should be evaluated to ensure that skid resistance properties are durable and suitable for the needs of traffic. Consideration of skid resistance levels, texture, aggregate availability, traffic volume, traffic speed, type of facility, rainfall, construction and maintenance practices, and accident experience are basic elements in such evaluations. Evaluations should document the compliance with the requirement for skid resistant surfaces and provide basic data for use in choosing corrective actions for locations with high wet weather accident rates.

      (2) One principal result of the evaluations is the development of a performance history for each particular pavement used by each highway agency. The performance of the existing pavement designs should be monitored and new designs should be evaluated to ensure that only skid resistant pavement surfaces are used. Information should be gathered as to the durability of a mix and the loss of skid resistance under traffic.

      (3) The level of skid resistance needed for a particular roadway depends primarily on the traffic volume, traffic speed, type of facility, and climate with additional consideration warranted at special locations such as steep hills, curves, intersections, and other sites which experience high demands for pavement-tire friction. It isdesirable to have one or more "skid resistant mixes" which have durable and higher than usual frictional properties for use in these special areas.

      (4) A pavement surface may provide adequate skid resistance at low speeds, yet be inadequate for high speed conditions. Pavement surfaces, therefore, should be designed on the basis of properties at expected operating speeds.

      (5) The American Association of State Highway and Transportation Officials (AASHTO) Guidelines for Skid Resistant Pavement Design, 1976, provide detailed information on the design of surfaces for both flexible and rigid pavements. The major considerations follow:

      (a) Flexible Pavements

      1 The skid resistance evaluation of bituminous pavements should include a determination that the aggregate used in the top layer of future pavements is capable of providing adequate skid resistance properties when incorporated in the particular mix and that the mix should be capable of providing sufficient stability to ensure the durability of the skid resistance.

      2 A bituminous pavement surface should contain nonpolishing aggregates. It isessential for good skid resistance that a mix design be used which allows good exposure of the aggregates. This requires that the pavement surface mixture be designed to provide as much coarse aggregate at the tire-pavement interface as possible.

      3 The open graded asphalt friction course (OGAFC), with a large proportion of one size aggregate, provides excellent coarse texture and exposes a large area of coarse aggregate. Guidance for this mix can be obtained from FHWA Technical Advisory T 5040.13, Open-Graded Asphalt Friction Courses, January 11, 1980.

      (b) Rigid Pavements

      1 The evaluation of portland cement concrete (PCC) pavements should include a determination that the finishing procedures, mix design, and aggregates provide the initial texture and necessary surface durability to sustain adequate skid resistance.

      2 In PCC pavements, the initial and early life skid resistance properties depend primarily on the fine aggregates for microtexture and on the finishing operation for macrotexture. Specifications for texturing concrete pavement should be carefully selected and enforced to ensure a macrotexture pattern appropriate to the type of facility.

      3 Regardless of the finishing or texturing method used, adequate durable skid resistance characteristics cannot be attained unless the fine aggregate has suitable wear and polish resistance characteristics. Research by the Portland Cement Association indicates that the siliceous particle content of the fine aggregate should be greater than 25 percent.

      4 If pavement evaluation studies indicate that the coarse aggregates will be exposedby the surface wear and have a significant effect on skid resistance of pavement, it too should have a suitable polish resistance characteristic.

      5 Metal tines, preceded by burlap or another type of drag finish, are recommended as being the most practical and dependable method of providing texture in PCC surfaces. Additional guidance can be obtained from FHWA Technical Advisory T 5140.10, Texturing and Skid Resistance of Concrete Pavements and Bridge Decks, September 18, 1979.

   2. Pavement Construction

      (1) Highway agencies are encouraged to adopt a policy of "prequalifying" aggregates to be used in surface courses. Prequalifying is a method by which aggregates can be classified according to their friction, texture, wear, and polish characteristics. Classifications should reflect performance related to traffic volume, operating speed, percent trucks, climate, geometric design, and other appropriate factors. Design procedures should be established to ensure that aggregates can be selected for each project which are suitable to the needs of traffic.

      (2) Prequalification may be accomplished by one of the following, or a combination of both:

      (a) A systematic rating of all fixed sources of aggregates (e.g., a commercial quarry which obtains aggregate from the same location for many years). Ratings should be based on standardized laboratory tests such as the American Society for Testing and Materials (ASTM) D 3319, Recommended Practices for Accelerated Polishing of Aggregates Using the British Wheel, or ASTM D 3042 Test for Insoluble Residue in Carbonate Aggregates, combined with data obtained from skid resistance tests of pavements in service. Other tests may be added or substituted if shown to predict pavement performance.

      (b) An evaluation and in-service history of the geologic or petrographic types of aggregates commonly used. Thus, when a new aggregate source is proposed, it can be accepted with minimum testing if an in-service history has been established for that type of aggregate.

      (3) Based on prequalification of aggregates, construction plans and specifications should define the friction quality of aggregate which will be acceptable. The following steps should be followed to assure acceptability of the as-constructed pavement surface course:

      (a) After the contractor has identified the particular aggregates and asphalt to be used on a project, it is recommended that a mix design be performed with the actual ingredients being used. Aggregates should be checked to determine if they are from prequalified sources or are an acceptable petrographic type.

      (b) Macrotexture and void content are important considerations in asphalt mixes. Since asphalts are often blended from several sources of crude oil that vary in temperature-viscosity characteristics, the mixing temperature should be determined for each project after establishing the characteristics of the selected asphalt. Allowable tolerances for asphalt content, mixing temperatures, and gradation should be established for each asphalt mix.

      (c) Job control of asphalt mixes should be designedto ensure that desired skid resistance properties are obtained. It should be recognized that small changes in aggregate gradation or asphalt content may significantly affect the macrotexture of finished surfaces.

      (4) The frictional properties of pavement surface types should be randomly tested within 6 months after opening to traffic to verify that the anticipated characteristics are present. Evaluation tests should involve direct measures such as the skid tester (ASTM E 274), or an acceptable alternative, but may use surrogate measures such as those which evaluate texture (for example, ASTM E 303, Standard Method for Measuring Surface Frictional Properties Using the British Pendulum Tester; and sand patch tests as described in the American Concrete Paving Association Technical Bulletin No. 19, Guidelines for Texturing Portland Cement Concrete Highway Pavements, Measurement of Texture Depth by the Sand Patch Method).

      (5) In cases where the skid resistance properties of a pavement are found to be questionable or inadequate, appropriate warning signs should be placed immediately as an interim measure. A complete evaluation and any remedial action needed should be effected as soon as possible.

   3. Pavement Maintenance. The same procedures and quality standards used in construction should be used in the maintenance operations.

5. WET WEATHER ACCIDENT LOCATION STUDIES. The purpose of this type of study is to identify locations with high incidence of wet weather accidents, determine corrective measures, and take appropriate actions in a timely and systematic manner. This activity should be conducted as part of the highway agency's safety improvement program and should make effective use of the agency's accident data file. Items to be considered for retrieval from the accident and traffic records are total accidents (rate), wet weather accidents (rate), and the wet/dry ratio.

   1. Identification of Wet Weather Accident Sites

      (1) Accident records, which are developed in compliance with Highway Safety Program Standard No. 9, Identification and Surveillance of Accident Locations, should be searched at least annually to identify sites which have a high incidence of wet weather accidents. It is essential to have a standardized highway location reference system for correlating data from different sources. Accident rates at a site will be of greatest value if:

      (a) the traffic volume is relatively high (i.e., approximately 1,500 vehicles per day or greater),

      (b) the period of accident data is at least two years, and

      (c) rainfall data are available for the same period as the accident data.

      (2) Rainfall patterns for the years in which skid resistance and accident data were compiled should be acquired for each area in the highway agency's jurisdiction. A suggested method is presented in Appendix A.

      (3) There are several methods in use by highway agencies to evaluate wet weather accident locations. One such method is the Wet Safety Factor (WSF), which is presented in appendix A.

   2. Field Review. A list of all sites ranked in order of WSF or another appropriate measure should be prepared as the basic list of candidate sites for remedial treatments. The selected locations should then be skid tested and reviewed by a team representing various disciplines such as highway materials, design,construction, maintenance, traffic and safety. See Appendix B for skid testing procedures. The review team should determine probable reasons for the high incidence of accidents and recommend corrective actions. Once the review team has recommended appropriate corrective treatments, a priority list of projects can be prepared based on benefits and expected costs.

   3. Priority Program. An assessment should be made of the benefits relative to the cost of providing remedial treatments for high priority projects. A number of highway agencies have their own methods for conducting benefit cost analyses of alternative remedial treatments. Some of these remedial methods are tied into traffic engineering or pavement management programs. A specific program for evaluating the benefits and cost of alternative treatments is presented in reference 1, Appendix C.

   4. Evaluation

      (1) Evaluation of completed projects as required in Highway Safety Program Standard No. 9 and FHPM 8-2-3, Highway Safety Improvement Program, should be well documented and should include a representative sample of completed projects. A sampling plan should be established, using accepted statistical methods, to evaluate projects with a range of such variables as classes of roadways, traffic volumes, types of countermeasures, pavements used, and other pertinent factors. On hazard elimination projects, these data should be correlated with accidents and traffic exposure and other pertinent factors inbefore/after analysis. See reference 2 in Appendix C.

      (2) The evaluation of completed safety projects should be continuing process to ascertain the long-term performance of corrective actions such as skid resistant overlays. The evaluations should address at least:

      (a) the overall effectiveness of the program in reducing accident rates at the corrected sites,

      (b) the adequacy of the various materials, designs, or methods used, and

      (c) recommendations for changes in the program, practices, or needed research and development.

      (3) As a secondary benefit, the evaluation process should provide input to an overall pavement management process.

6. PAVEMENT SKID RESISTANCE TESTING PROGRAM

   1. General Description of Program. The actual testing of pavement friction provides basic data for use in the three activities introduced in paragraph 3. Figure 1 graphically presents the interrelation between these activities. The upper portion of Figure 1 provides an overview of data to be collected to serve the safety, construction, and maintenance functions of high-way organizations concerned with the skidding properties of pavement surfaces. The lower portion of Figure 1 indicates the various uses of the skid testing data, along with weather and accident data. Some of these data are evidence of the durability of particular surfaces, while other data provide a general overview of the skid resistance characteristics of the highway system.

      Figure 1
      Model Skid Accident Reduction Plan

      (1) Skid resistance testing should be organized to support the following activities:

      (a) Pavement evaluation studies in which measurements of the skid resistance of test sections are made to determine the skid characteristics of typical mix designs. Sufficient numbers of measurements should be made to determine the level of pavementfriction, wear rates, and speed gradient of the pavement under various traffic exposures. These test sections should include the new projects to be tested as described in paragraph 4b(4).

      (b) Evaluation of friction characteristics at locations which have a high incidence of wet weather accidents.

      (c) System status for which measurements of the skid resistance of a representative sample of roads are made to develop the general levels of pavement friction on all roads in the highway agency's jurisdiction.

      (2) Accurate location of sites or road sections requires the use of a standardized reference system. Often each element of the State which collects highway data uses its own reference system. For example, police accident reports may locate accidents by distance to a landmark, pavement records may be kept by project number and geometric features may be identified by station. A unified reference system has many benefits, especially in pulling together technical data for identifying and analyzing locations with a high incidence of wet weather accidents.

      (3) Pavement evaluation study sites and wet weather accident sites should be identified by the element within the highway agency responsible for thoseprograms. The skid testing can then become a routine matter for the element charged with operation of the skid test equipment.

      (4) A total skid inventory of all roads and streets in a highway system has proven to be impractical and is not necessary to carry out an effective skid accident reduction program. Roads and streets which are used primarily by vehicles traveling at low speeds are not highly susceptible to skid accidents and accordingly can be eliminated from routine sampling of highway sites. For urban areas, this means that most city arterials would be sampled but residential streets and roadways with low speed limits would not. Nearly all rural highway sections could be sampled, since such roads are liable to high-speed use.

      (5) Another practical consideration in determining which roads should be sampled is traffic volume. In urban areas, most roads with high speeds have moderate to high traffic volumes whereas this is not the case for rural highways. Relatively few rural roads are used by more than 1,000 vehicles per day. On a cost-effectiveness basis, such roads can seldom justify resurfacing on the basis of safety considerations alone; therefore there is little benefit in routine sampling of low-volume rural roads.

      (6) Highway sections within the constraints of higher speeds and volumes need not be tested every year, since few roads vary substantially in skid resistance in any two or three-year period. Beyond this period, however, roads may lose significant skid resistance and may pose a serious danger to users. Using these criteria as part of a sampling plan will permit most if not all highway agencies to make maximum use of skid resistance data without increasing the amount of skid testing undertaken.

      (7) Skid resistance measurements should be made with a calibrated locked-wheel skid tester using the ASTM E 274 method and supplemental procedures described in Appendix B or an acceptable alternative method. Locations such as intersections and sharp curves which are not easily measured with the locked-wheel skid tester at the standard speed of 40 miles per hour should be tested at a lower speed. Such tests should be supplemented with texture measurements topermit extrapolation of available skid resistance to operating speeds. Alternative methods of measuring pavement friction properties may be used provided they correlate well with the locked-wheel skid tester.

      (8) In analyzing the skid numbers obtained, the time of year the measurements were taken has to be considered. Several States have published the results of their analyses and have developed methods for correcting skid number measurements taken during various periods and for different pavement surface types. See references 5 and 6 in Appendix C.

   2. Specific Data From Sample Sites. In conjunction with skid resistance measurements, pavement wet time and accident records are desirable for each roadway section in the sample. The highway location system should be used for correlating data from different sources. An example of specific data which is desirable at each sample site is given in Appendix D.

   3. Sites with Low Skid Resistance. When sites with low skid resistance are identified during the testing of system status, these sites should be analyzed for corrective action. This can be done through a pavement management program, a high hazard elimination program, or other efforts. If the high hazard elimination program is used, the analysis should be in accordance with FHPM 8-2-3.

signed
R.D. Morgan
Associate Administrator for
Engineering and Traffic Operations

signed
Lorenzo Casanova
Associate Administrator for Safety

Appendixes

EVALUATION OF WET PAVEMENT TIME AND ACCIDENT DATA

A.1 The quantity of rainfall (inches) recorded by weather stations may be used to calculate the percentage of pavement wet time. Wet pavement time (WPT) may be estimated from total annual rainfall in inches (AR) as follows:*

WPT = 3.45 ln (AR) - 5.07

*(This equation is based on a relationship developed by K.D. Hankins in "The Use of Rainfall Characteristics in Developing Methods for Reducing Wet Weather Accidents in Texas," Texas State Department of Highways and Public Transportation Study No. 135-4, July 1975.)

Dry pavement time may be estimated by subtracting the amount of wet time and ice and snow periods from the total time in the period analyzed. Data from rainfall stations maintained by the National Oceanic and Atmospheric Administration's Weather Service may be used for wet and dry pavement time estimates for various areas within a State.

Isohyetal maps may be used to develop site wet pavement times. If ice and snow cover pavements for a significant portion of the time, a map for dry time should be prepared as well. Figure A-1 provides an example of a wet time map drawn from isohyetal charts.

Figure A-1
Percent of Time Pavement Is Wet

A.2 Wet Safety Factor (WSF)

There are a number of ways to evaluate the relative safety of the subject location, one of which is the wet safety factor (WSF) approach.**

**(The WSF is a generalized form of an index referred to as the "skid trap ratio" and recommended for use in NCHRP Report 37, Tentative Skid-Resistance Requirements for Main Rural Highways, " by H.W. Kummer and W.E. Meyer, Highway Research Board, Washington, D.C., 1967.)

For each wet weather accident location, a WSF may be developed. This factor is expressed as follows:

WSF = (DA)(PWT)/(WA)(PDT)

Where:

DA = number of dry weather accidents

WA = number of wet weather accidents

PDT = percent of dry pavement time

PWT = percent of wet pavement time

This factor is the reciprocal of the risk of having a wet pavement accident relative to having a dry pavement accident. On a specific roadway section, each of these variables must be developed for the same time period; otherwise, traffic exposure must be taken into account. Criteria may be developed for further consideration of pavement sections. A WSF less than 0.67 suggests a wet weather problem. This criteria is based upon the conservative estimate of the overall likelihood of a wet weather accident being 1 1/2 as great as a dry pavement accident. This estimate assumes that wet weather accidents at the site or road section under consideration are attributable entirely to a skidding problem. A low WSF in most cases is due to poor skid resistance. However, traffic engineering evaluations may reveal deficiencies in sight distance, road markings, inadequate drainage, etc. Auxiliary information obtained during the test program should provide indications of the safety problems.

SKID MEASUREMENT SYSTEM DESCRIPTION AND OPERATING PROCEDURES

B.1 DESCRIPTIONS OF SKID MEASUREMENT SYSTEM

The requirements of American Society for Testing and materials (ASTM) E 274 states "The method utilizes a measurement representing the steady state friction force on a locked test wheel as it is dragged over a wetted pavement surface under constant load and at constant speed while its major plane is parallel to its direction of motion and perpendicular to the pavement."

Although this specification may be met by a system involving only one wheel attached to a towing vehicle and although a few such systems are in use, the vast majority of skid measurement systems in use and expected to be in use in the near future consist of a towing vehicle and two-wheel trailer. On many systems either wheel may be locked during testing, but most commonly, the left is used.

The ASTM considers testing the left wheel track to be "normal." However, a differential in friction levels between the left and right wheel track may exist. When testing a site where a differential may exist, especially a high wet weather accident site, all lanes and wheel tracks should be tested. If a two-wheel trailer system is used, it is desirable to have the capability of testing with either wheel.

A skid measurement system must have a transducer associated with each test wheel which senses a force equal or directly related to the force developed between the sliding wheel and the pavement during test, electronic signal conditioning equipment to receive the transducer output signal and modify it as required, and suitable analog and/or digital readout equipment to record either the magnitude of the developed force or the calculated value of the resulting skid number (SN).

The system must include a facility for the transport of a supply of water--usually 200 to 500 gallons--and the necessary apparatus to deliver a specified amount of water--4.0 gallons per minute per wetted inch of pavement at 40 miles per hour within specified limits in front of the test wheel.

Finally, the system must include provision for measuring (and preferably for recording) the speed at which the test is conducted.

B.2 FIELD OPERATING PROCEDURES

B.2.1 Field Force Verification

It is generally impractical to perform force plate calibrations at frequent intervals while the measurement system is in the field. Facilities should, however, be available to permit the operator to ascertain that significant changes have not occurred in the force measurement subsystem since the most recent force plate calibration.

If the measurement system uses a torque transducer and is adaptable to mounting a torque arm, the verification can be accomplished within a reasonable time and effort. This device, consisting of an arm capable of being bolted to the test wheel in a horizontal position and of supporting known weights located at specified distances from the center of the test wheel, may be used to test the torque transducer to predetermined values of torque. Typically, the test wheel of the inventory system is raised off the ground, the torque arm is attached to the test wheel and held in a horizontal position, the brake of test wheel locked, and a series of known weights are suspended on the torque arm. This procedure will induce a series of known strains on the transducer, resulting in a series of output signals through the signal conditioning equipment. The magnitude of these signals should then be compared to the magnitude of signals produced through use of the same technique immediately after the most recent force plate calibration. Adjustment of signal conditioning equipment gain setting may be made to offset small force measurement subsystem variations which could occur.

Verification should be repeated periodically.

B.2.2 Test Tire and Wheel Preparation, Control of Tire pressure

Tire Specification

Unless otherwise specified, all tests shall be performed with tires meeting the requirements of ASTM E 501, Standard Tire for Pavement Skid Resistance Tests, and all pertinent sections of that specification as well as ASTM E 274 should be observed in their use.

Tire Mounting and Break-in Procedure

The tire should be mounted on a Tire and Rim Association 6JJ rim. The rim should have been examined to determine that it has suffered no damage or misalignment in prior use. After mounting, and before break-in, the tire and wheel should be balanced. The tire should be subjected to a break-in of 200 miles use before being used for testing. This break-in may be accomplished by using the tire on the skid trailer wheel which is not used for testing. If the tire must be remounted before test use, it should be rebalanced after remounting.

Tire Warm-Up Procedure

The test tire should be inflated to 24 + 0.5 pounds per square inch measured at ambient temperature. After tire pressure measurement and adjustment, the tire should be subjected to a 5-mile warm up, traveling at conventional highway speeds, before tests are performed. The 5-mile warm-up should be repeated on any occasion when the measurement system is parked for a periodof 15 minutes or more.

Tire Wear and Replacement Procedure

The standard pavement test tire has a series of visual wear guide sipes (small circular holes) cast into each of the outer ribs of the tire. The test tire should be withdrawn from testing use when wear has progressed to a point at which the wear guide sipes are no longer visible. During routine testing, test tires should be examined at least twice daily (and more frequently as tire nears unacceptable wear level) to determined that wear has not progressed beyond acceptable limits.

Additionally, after any series of tests on pavements having very high skid numbers (in excess of SN=70) or in the event of a deliberate or inadvertent dry skid, the test tire should be examined for the development of a flat spot. If a significant flat spot or spots develop on a test tire, it should be withdrawn from test use due to the tendency of the test wheel to seek out and return to such a flat spot in subsequent lockups.

B.2.3 Watering Subsystem Procedures

Daily Procedures

Prior to the beginning of each day's activity, the crew should perform at least the following functions with respect to the water subsystem:

1. Determine that the water nozzle (nozzles) when in the testing position assumes the proper angle with respect to the pavement (ASTM E 274 requires an angle of 25 + 5 degrees).

2. If the measurement system has provision for raising and lowering the nozzle between tests, determine that the mechanism is working properly and that the nozzle assumes a fully lowered position during the test sequence.

3. Determine that the nozzle, when in the test position, will discharge water directly in front of and centered on the test wheel.

4. Examine the nozzle outlet orifice to determine that it is free from damage or distortion.

The above inspections should be repeated during a day's testing in the event of operation on very rough highways (or in the event of any off-highway travel) which may have caused damageto the nozzle or adversely affected its orientation.

Water Trace Width Check

Periodically the crew should make a measurement of the water trace width as a gross measure of overall water subsystem performance. This may be accomplished by driving the measurement system over a pavement at a selected convenient speed (the same speed should be used on all occasions), initiating water flow without locking the test wheel brakes, and measuring the width of the resulting water trace on the pavement. The trace width measurement should be made as quickly as possible after passage of the inventory system (preferably within 30 seconds). This would require that one member of the crew drive and operate the measurement system while the other member is positioned off the side of the pavement at the location at which the measurement is to be made. Best results are achieved if this procedure is performed on a relatively smooth pavement surface (low macrotexture).

B.2.4 Instrumentation Calibration Verification

Provision should be made to allow for verification of the signal conditioning instrumentation calibration (to account for the effects of zero and gain drifts).

General Requirements for Calibration Signal

The minimum acceptable facility for verification of conditioning instrumentation is a calibration signal subsystem. The calibration signal should be provided from such a source and in such a manner that there is little likelihood of variation in the calibration signal itself. This assurance then permits theoperator to make adjustments in the measurement subsystem gain to offset the frequent small deviations which occur due to changes in ambient temperature and other operating parameters.

Force Measurement Calibration Signal

The most straightforward technique for providing a force measurement calibration signal is to make provisions for switching a high quality shunting resistor of known value in parallel with one arm of the force transducer strain gauge bridge. This induces an imbalance in the bridge equivalent to the application of a known force to the transducer. The resultant signal is sufficient to verify, or provide means of adjustment for, all elements of the force measurement system forward of the transducer itself.

Frequency of Use

Instrumentation calibration verification through use of calibration signals should be accomplished at the beginning of each day's operation after equipment warm up, at intervals of no more than 2 hours when the system is in continuous use, and upon the renewal of operation throughout the day after any period during which the signal conditioning equipment has been turned off or the unit has been allowed to stand without use for 30 minutes or more.

B.2.5 Check List

A check list should be available to the crew and should be used prior to the beginning of daily operations and on any occasion during the day when testing is suspended for 30 minutes or more or when instrumentation has been turned off. The check list varies from system to system due to differences between the systems, but should provide for at least the following checks:

1. all power subsystems on and providing proper levels of power

2. all signal conditioning subsystems on for adequate time to reach stable operation (typically 10 to 30 minutes)

3. all recording systems on and functioning properly

4. instrument calibration (described above) performed

5. tire pressure checked and adjusted if necessary

6. test tire checked for wear

7. water nozzles checked for position and condition

8. water tank adequately filled

9. fuel supply adequate

10. safety chains and all other connections between trailer and towing vehicle properly connected, positioned, and protected if necessary

11. trailer jacks (if available) in retracted position

12. all auxiliary equipment (air-compressors, lights, etc.) functioning properly

B.3 USE OF STATIC AND DYNAMIC CALIBRATION PROCEDURES

B.3.1 Purpose of Field Test Center

At the present time the highest order of calibration and evaluation available for a State skid measurement system is that provided through the Field Test Center established under contract by the Federal Highway Administration (FHWA). Arrangements to receive the services of the Field Test Center may be initiated by a State through submittal of a request for such services to the local FHWA division office.

B.3.2 Criteria for When to Use the Field Test Center

Each measurement system should be submitted for calibration and evaluation at the Center as soon as possible after its introduction into service. It should be resubmitted for calibration and evaluation whenever:

1. significant repair or modification has been accomplished by the owning agency which might reasonably be expected to affect test results, or

2. whenever it has experienced sufficient use such that normal wear in the various subsystems might be expected to have affected their operation.

   The second consideration suggests that each measurement system should be resubmitted at least every 2 years.

B.3.3 Calibration Services Provided by Field Test Center

The static and dynamic calibration services provided by the Field Test Center include the following:

1. Horizontal and Vertical Force Calibration. This provides for evaluation of the accuracy, linearity and hysteresis of the measurement system force transducers and signal conditioning equipment through use of an air bearing force plate maintained by the Center, and periodically calibrated by the National Bureau of Standards.

2. Flow Rate Evaluation and Adjustment if Required. This includes determination that the water delivery subsystem of the measurement system provides a quantity of water (dependent upon trace width) in front of the test tire which meets ASTM E 274 requirements at speeds between 20 and 60 miles per hour.

3. Static Evaluation of Water Distribution. This provides an evaluation of the uniformity with which the total water flow is distributed across the trace width and adjustment, if necessary, to assure that the water is in fact delivered uniformly and in line with the test tire.

4. Force Plate or Load Cells. The visitors force plate used for routine checks of the force measurement subsystem can be calibrated while at the Center.

5. Speed Calibration. The speed measurement (and recording if available) subsystem is evaluated, calibrated and, where necessary and possible, adjusted to produce accurate speed measurement values over the range of 20 to 60 miles per hour.

6. Tire Pressure Gauge Calibration. This provides assurance that tire pressures in the test wheels and in the speedmeasuring fifth wheel (if used) can be accurately measured and set.

7. Dynamic Correlation. Two such correlations are conducted: The first with the measurement system in the "as arrived" condition and the second after all of the foregoing evaluations have been conducted and indicated adjustments accomplished. The first correlation results in the development of mathematical relationships between the measurement system and the Area Reference Skid Measurement System that permit data collected by the measurement system, prior to its visit to the Center, to be adjusted to a common base provided by the use of the Area Reference System. The second correlation permits the development of similar relationships which may be used to relate the results of subsequent testing to the Area Reference System base. The data from the second correlation also provide an estimate of the system measurement variance.

B.4 MAINTAINING SYSTEM INTEGRITY BETWEEN FIELD TEST CENTER CALIBRATIONS

Two basic types of procedures are available for determining that significant changes have not occurred in the measurement system since its most recent evaluation and calibration at the Center. These involve techniques for evaluating important subsystem performance and techniques for evaluating performance of the total system.

B.4.1 Techniques to Evaluate Subsystem Performance

As a minimum, the owner of each measurement system should maintain and periodically make use of facilities for evaluating the force, water, and speed measurement subsystem of the inventory system.

Evaluation of Force Subsystem

The force subsystem should be evaluated through use of a force plate. An air-bearing force plate is recommended since its action is such as to essentially eliminate the effect of friction in the plate itself. If an air-bearing force plate is not available, any of several commercial mechanical force plates may be used. If a mechanical device is used, precautions should be taken to assure that all moving parts (particularly load application screws and spherical or roller bearings) are well lubricated and that the lubricant is periodically removed and replaced.

To conduct an evaluation, the test wheel of the measurement system should be centered on the force plate, the test wheel brake locked, and known frictional forces introduced to the tire-force plate interface through appropriate motion of the force plate. Frictional forces should be both increased and decreased in a stepwise manner to allow for detection of possible hysteresis effects. The indicated force readout values for the system should then be plotted against known force input values. The resulting plotted calibration line should be evaluated for nonlinearity and hysteresis characteristics. Also actual readout values for known force inputs should be compared with those readout values determined from tests conducted with the same equipment after the most recent Center evaluation.

Evaluation of Water Subsystem

The most effective evaluation of the water subsystem to discern variations in performance is that of flow. Flow rate may be evaluated by raising the rear wheels of the towing vehicle, running the vehicle at an indicated speed of 40 miles per hour (or any other desired speed), collecting the water pumped through the system and out the nozzle during a measured time period, and calculating the flow rate in gallons. This procedure should be repeated at two or more speeds to evaluate linearity of the water delivery subsystem with test speed.

The Pennsylvania State University has developed a water rate flow tank which is circular in cross section and of such size that it fits easily into a standard manhole. The tank has athreaded opening in the bottom for drainage and a stop-plug with a long handle which permits the plug to be removed and replaced from the top of the tank after it is hanging in the manhole. It also has a scale calibrated in gallons on the inside of the tank. This tank may be suspended in a standard manhole, the measurement system positioned so that the nozzle will discharge directly into the tank, the rear wheel of the towing vehicle raised, and total flow measured at any desired speed. The only additional equipment required is a stopwatch.

Evaluation of Speed Measurement Subsystem

The speed measurement subsystem should be evaluated by operating the measurement system at various test speeds over a measured mile course. If the basic speed measure is done through the use of the tow vehicle speedometer or through a tachometer-generator driven by the tow vehicle or by a fifth wheel, then the vehicle should be driven over the measured mile course at a selected speed and the time of transit measured with a stopwatch. The actual speed, calculated from the distance and the elapsed time, is then compared to the indicated speed.

If speed measurement is based upon a pulse generator driven by a fifth wheel, the accuracy of the speed measurement is directly dependent upon the accuracy of the fifth wheel for distance measurement. To evaluate this subsystem, the fifth wheel tire pressure is adjusted until the distance indicated agrees with the known distance traversed (the assumption being made here is that the electronic package which converts the pulses to velocity is functioning properly).

If tapeswitch event detectors, placed 200 feet apart, and an interval timer (+0.01 second resolution) are available to measurethe time required by the inventory system to travel 200 feet, a very accurate speed measurement is obtained to check against the indicated value.

Time Between Subsystem Evaluations

The force, water and speed measurement subsystems of the measurement system should be checked by the methods described above at intervals no greater than 3 months.

B.4.2 Techniques to Evaluate Total System Performance

Use of Measurement System Sample Variance as Performance Measure

A portion of the information furnished, as a result of an evaluation at the Center, is the pooled sample standard deviation of the measurement system for repeated test at three test speeds on five special test surfaces. If the sample standard deviation at the desired speed is squared, the resulting value, SD²_t is an estimate of the skid measurement system variance. Subsequent to the Center evaluation, the crew should periodically select a pavement location having a skid number of approximately 30 to 40 and run 20 repeat tests at the desired speed over the same location. From the results of these latter tests a new estimate, SD²_E, can be calculated. If the ratio SD²_E/SD²_t does not exceed 2.0, the chances are 19 in 20 that the system standard deviation has not doubled over that established during its visit to the Center. (If the system has not been to a Center to obtain an estimate of SD²_t, its crew should select a pavement location having a skid number of approximately 30 to 40, run repeat tests at each desired speed over the same location, and calculate the sample standard deviation at each such speed.)

As an alternative, the above procedure could be performed making only 10 repeat tests on the selected pavement. In this case, the ratio of SD²_E/SD²_t should not exceed 2.2. The chances are then four in five that the system standard deviation has not doubled over that previously established.

The above procedure should be performed at time intervals no greater than 3 months.

Short Term Checks of System Performance

The agency operating the measurement system should select several pavements located close to the site at which the system is normally garaged and perform repeated tests on the surfaces at quite frequent intervals, preferably weekly. Measured values of skid resistance on these surfaces will obviously change as thesurfaces change from traffic wear, environmental, and/or seasonal variations. However, these changes should occur in an orderly and predictable fashion and any abrupt change would be an indication of possible erratic performance of the measurement system. A continually updated record of the results of such tests should be maintained and examined after each updating for evidence of such erratic performance.

REFERENCES

The following is a selected list of references which may be helpful in implementing the program described in this Technical Advisory. This list is not intended to be a bibliography of all documents available in this field:

* Means that these studies are available through the National Technical Information Service, 5285 Port Royal Road, Springfield, Virginia 22161.

*1. Effectiveness of Alternative Skid Reduction Measures,Benefit Cost Model, Report No. FHWA-RD-79-12, Volume II, November 1978, Federal Highway Administration.

*2. Accident Research Manual, FHWA-RD-80-016, February 1980, Federal Highway Administration.

*3. Evaluation of Minor Improvements (Part 8), Grooved Pavement (Supplemental Report) CA-DOT-TR-2152-11-75-01, September 1975, R.N. Smith and L.E. Elliott, Office of Traffic, California Department of Transportation.

4. Evaluation of Minor Improvements (Part 9), Open Graded Asphalt Concrete Overlays, January 1972, James I. Karr, Transportation.

5. Variations in Skid Resistance Over Time, FHWA-VA-80-33, February 1980, S.N. Runkle, David C. Mahone, Virginia Highways and Transportation Research Council.

6. Seasonal Variations in the Skid Resistance pavements in Kentucky, Research Report 532, November 1979, James L. Burchett, Roland L. Rizenbergs, Kentucky Department of Transportation.

SPECIFIC DATA TO BE REPORTED FOR SAMPLE SITES

The following data should be collected in testing samples locations:

D.1 Skid numbers (SN) should be taken for major classes of roads stratified by traffic volume and geographical location.

D.2 Auxiliary data which should be included in order to establish distribution of skid numbers may include the following:

(a) Location of site or roadway section

(b) Responsible jurisdictional unit and route number or other designator

(c) Functional classification of road (e.g., two-lane, four-lane divided without full control of access, etc.)

(d) Surface type (e.g., bituminous, open-graded, concrete, tine finish, etc.)

(e) Average annual daily traffic (use traffic count data if available)

(f) Length of roadway section

(g) Lane where skid measurements are made

(h) Date of skid measurements

(i) Number of tests made in section

(j) Average SN

(k) Range of SN measurements

(l) Presence of atypical geometric or feature

(m) Evidence of skidding (e.g., skid marks, scarred posts, etc.)