Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
APPENDIX E DESC~PlION OF S~OOTHNES~EASORINC EgOIP~ENT
- c) - <e - o on D -50 - ._ a, i_ ._ U) U. 11 Hi - 4) - o to ~ -10 JO J C Ho of ~ 00.00 Rougher Smoother _ 80 00 _ - 60.00 - _ 40.00 - _ 20.00- _ 1 1 1 ~ - e.0~ ~1 - 1 1 1 00 -40.00 -30.00 -20.00 -a 0.1 )0 10.00 20.00 30.00 40.00 50 _~ -20.00 -40.00 - Regression Equation: y = 1 .22 78x + 3 E -1 5 -60.00 -;0000T - ~v~vv _ _ _ % Change in Initial Smoothness (based on target=4.1) 00 |~BIuegrass Parkway WB Woodford Co. ~ Figure D-63. Sensitivity of initial serviceability for Kentucky AC/AC parkway pavements (roughness model approach; Trigger=2.751. - 1 00.00 Sm oother Rougher : ~80.00 ... 60.00 . Regression Equation: = = ~y . -0.3779x - 6E-15 I I I I ~ HI t ~I .00 -80.00 -60.00 -40.00 -20.0' onto -60.00 -80.00 1 00.00 % Change in Initial Roughness (based on target=55 in/mi) -a 1-96 EB lonia Co. .00 ::: 1-96 WB lonia Co. i' 1-94 EB Van Buren Co. ~ ~1-94 WB Van Buren Co. | Figure D-64. Sensitivity of initial roughness for Michigan JRC pavements (roughness model approach; Trigger=150 in/mi). D-40
APPENDIX E. DESCRIPTION OF SMOOTHNESS-MEASURING EQUIPMENT Introduction Over the last century, a vast array of equipment has been developed to evaluate the smoothness of pavements. The first of these devices was developed on early paving projects, while some have been developed In connection with the construction of test roads. Many of these employed some innovative approaches in the attempt to quantify and ensure pavement smoothness, and even the early models tended to be quite sophisticated. In recent years, the growing interest in pavement management systems has focused on pavement ride as the major factor In evaluating the serviceability of a road, Hereby creating the need for economically evaluating the extensive mileage of a highway system. Electronics and space age-technology have been applied, creating an ever-changing approach to the subject. This appendix describes the significant types of equipment that have been used for evaluating the smoothness of pavement surfaces. Those devices that are currently being used by major highway agencies, as well as those that are believed to provide a potential contribution in the immediate future, are covered in depth. Smoothness Measuring Equipment Used in Construction Currently, there are a variety of types of equipment in use for measuring the smoothness of newly constructed pavements. Table E-l shows the breakdown of usage among the 47 States and 3 federal agencies that responded to the agency questionnaire. Discussed in this appendix are the pavement smoothness measuring systems currently used for crucial construction smoothness quality control testing and listed In table E-~. Several other profile- and smoo~ness-measur~ng systems have been used for pavement management purposes, including numerous high-speed profile measuring systems. Certainly, some of these systems also have application in initial pavement smoothness measurement. New smoothness measurement equipment is being developed, and these systems are also described in the following sections. Straightedge / Stringline Perhaps the most basic method of locating surface longitudinal and transverse pavement irregularities is win me use of a straightedge or stringline. Straightedge devices are usually made of wood or metal and can be of any convenient length, typically in the range of ~ to 16 It (2.4 to 4.9 m). A wire or string stretched tightly from the ends of a bow-shaped form is also sometimes used. When the straightedge E-!
Table Eel. Agency use of smoothness-measur~ng equipment. 1 Equipment Type | No. of Agencies Using l l Californ~a-type Profilograph (generic) | 33 l Rainhart Profilograph 7 Mays Meter (Vehicle mounted) | 2 Mays Meter (Trailer mounted) | 5 l | GM Type Profilometer | 4 Rolling Straight Edge 1 | Straight Edge or String line | 33a l l a six agencies use this category as Me sole method of smoothness measurement, and me remaining use it to augment over smoothness testing. Or stringline is placed on the pavement surface, variations in the pavement surface from the straightedge can be readily seen and measured. Smoothness is controlled by allowing no more than a specified deviation from the straightedge when placed on the pavement surface. Use of this approach is very labor intensive, because We person doing the measuring must be very close to the pavement surface, generally in a kneeling position, to determine if any deviation exists. Continuous longitudinal pavement surface coverage using straightedges is usually not practical, but Me evaluation of localized areas can be addressed adequately. Application is limited to identification Of short wavelength roughness, since measurement accuracy diminishes for pavement surface wavelengths greater Man about one-half the straightedge length. Since Me straightedge looks at a pavement surface from a "snapshot" viewpoint, its output can vary greatly from that determined using a profilograph. For example, a New York DOT study in 1960 found that straightedge results from measured ramps noted 25 percent more displacement than a profilograph. However, on mainline highways the straightedge results contained 50 percent less displacement (Scofield 1993~. This is assumed to be a result of Me tremendous effort required to measure mainline pavement surfaces and the associated fatigue effects. Although use of more sophisticated smoothness measuring equipment is the norm for use on most mainline paving projects, there are some situations in which the straightedge may be the primary method of measuring smoothness. About two- ~irds of the State agencies polled in 1994 indicate that straightedges are used to augment other forms of smoothness control. E-2
Rolling Straightedge Rolling straightedge equipment, as shown in figure E-1, consists of a rigid straightedge beam with wheels at each end. At the beam midpoint is a third wheel, riding on the pavement surface, which is linked to an Indicator showing the deviation from the straightedge plane. One United States manufacturer, ELK International-Soiltest supplies a generic rolling straightedge under the name Hi-Low Detector. It is a precision-made rectangular aluminum beam from 10 to 16 It (3.0 to 4.9 m) long with riding wheels at the ends and a detector wheel at the middle. The system is pushed lengthwise at walking speed by hand from the rear where the operator can steer the detector and control the spray marking system. If the deviation from the straightedge plane is greater than is allowable, the operator can trigger an inverted spray canister from the steering handle to mark the location of high or low pavement surface deviations. Calibration can be done by hand on a level surface. The rolling straightedge is slow and incapable of measuring "true" pavement surface profile. No record of the equipment output is provided by this system except for the paint markings on the pavements surface, but it is useful for bridge decks and small projects. Costs for the 10- to 16-ft (3.0- to 4.9-m) systems in 1995 range from $3,300 to $4,300. Figure E-~. Rolling straightedge. (Source: ELK) E-3
Profilograph Equipment The term profilograph refers to a rigid frame device carried by a system of support wheels that provides a dawn from which pavement surface deviation measurements can be made. A profile wheel, which rests on the pavement surface at the center of the unit, is linked to a mechanical strip chart recorder or computer, and variations In vertical movement of the profile wheel with respect to the established datum are recorded. For the mecharucal models, Me strip chart is then evaluated manually or electronically. Computerized models are programmed to produce a strip chart and a tabulation of pertinent smoothness information automatically. All of the devices in this category are operated In the range of 2 to 3 mi/hr (3.2 to 4.S km/hr). Two basic types of profiIograph have evolved, differing In We support wheel configuration. Support wheels on Me California-type profilograph have varied in number from four to twelve, with systems in many States using twelve wheels. These wheels are attached to the ends of a 25 It (7.6 m) truss and mounted on a mulliple-axie carriage Cat includes four wheels 17 in (432 mm) from Me truss centerline and two wheels 17 In (432 mm) on the opposite side of the truss centerline. The support wheels are commonly spaced at 2.7 it (0.82 m) intervals and positioned near the ends of the tmss, resulting In an overall profilograph span of about 33 It (10.0 m). The twelve support wheels for Me Rainhart-type profilograph are evenly spaced along its 24.75 it (7.5 m) span at offsets up to 22 in (559 mm) such that no wheel follows the same path. On bow systems, the front support wheels can be steered by the operator. The following sections will provide a short history of the profilograph systems and discuss several of the different models within the profilograph group. These include the Cox, McCracken, Ames, SSI, Paveset, Soiltest, and Rainhart systems. California ProfiIograph Background The development of the California profilograph followed many avenues in Me period from the 1930s to Me 1950s. Units win a beam length ranging from 10 ft to 30 ft (3.] m to 9.3 m) were used and the model that ultimately became popular has a length of 25 It (7.6 m). This type has come to be known generically as Me California profilograph, developed under Me direction of F. N. Hveem, former Materials and Research Engineer with Me California Division of Highways, and is described In a classic report from Mat era (Hveem 1960~. The basic 25-ft (7.~m) long California profilograph, as developed In the late 1950s, is typically supported be a carriage of six wheels at each end, arranged In an articulated manner and provided win steering capability. (Current models in use by the State of California consist of a single axle and two wheels at each end of the 25 It [7.6 m] beam.) A profile wheel is located at Me beam midpoint and is linked to a recorder Mat provides a paper strip chart showing changes in Me distance between the pavement at Me point of the profile wheel and the datum established by the E-4
carrying wheels. The strip chart produces the deviations on a true vertical scale and on a 1:300 (25 ft/in [0.3 m/mm]) horizontal scale. Overall pavement smoothness is expressed by this equipment in terms of its "profile index" (Pl) and a tolerance for individual bumps. Pi is expressed in inches per mile (mm per km) and represents the total accumulated excursion of the strip chart trace beyond a tolerance zone. That tolerance zone typically ranges from 0.2 in (0.5 mm) to less than 0.001 in (0.025 ~run). For individual bumps the tolerance limit is expressed as some height, usually 0.3 in (7.6 mm) over a 25 ft (7.6 m) base length as read on Me strip chart. Some agencies use 0.4 in (10.2 mm) or 0.5 in (12.7 mm) tolerance levels. When the specifications are exceeded, it is necessary to identify the individual bumps on the pavement surface so that grinding can be completed to remove the bumps. Methods for analysis if the strip chart profiIogram include manual and computerized techniques. Manual evaluation of the strip chart is done using two clear plastic templates. The profile index is determined for each 0.! mi (0.16 km) of pavement using a plastic template 21.12 In (536.45 mm) long (21.12 in x 25 ft/in = 528 It [161 mid. The template is scribed in 0.1 in (2.5 mm) vertical increments to allow the vertical excursion to be counted. A smoothness tolerance represented by a blanking band runs the length of Me template. The blanking band is typically 0.2-in ,_ ~ ~ .. . . ~ ~ ~ ~ . `~ _ ~ . ~ ~ ~ ~ to.l-mm) wine; however, some agencies nave used a U.1-ln t;~.~-mm) wine nancl and some have used a 0.0-in blanking band. The blanking band, when used, is placed on the profile to cover as much of Me trace as possible. The accumulated distances of the trace above and below Me blanking band are used to determine the profile index. Individual bumps are manually identified using a second clear plastic template with a ~ in (25 mm) scribed base length and a parallel I-in (25-mm) long slot separated by Me bump tolerance limit, typically 0.3 in (7.6 mm). Computerized recording and analysis of profiIogram traces is completed by the manufacturer's software, once the appropriate parameters have been defined. Prior to completing the measurement pass, the critical bump height and blanking band width defined by the applicable specification must be defined. In addition, the computer filter setting, which essentially defines the baseline trace, is selected. The Cox and McCracken manufacturers have developed data filters in an effort to match the electronic data profiIogram to Me trace Mat would be measured by a manual profiIograph. Cox produces a baseline trace and accompanying null band by filtering a copy of the profiIogram to remove short-wavelength chaNer and produce a smooth trace that matches the profile trend. The McCracken profiIograph software reportedly breaks Me section into shorter intervals and analyzes the short intervals by the standard manual methodology (Scofield 1993~. The 25-ft (7.6-m) long frame for the profiIograph is comprised of five segments for ease of transportation and storage. All of the frame segments and the wheel assemblies, steering mechanism, and recorder can be carried in a pickup or van and assembled at the job site. Propulsion can be manual or by the use of a small E-5
motorized unit that is attached alongside the profilograph. carries the operator as well as propelling the equipment. The propulsion unit California implemented a profiIograph specification for concrete pavements . beginning In IV58 All of the California ~01~ prohlographs have been constructed ~ the machine shop of the Materials and Research Department, and the fleet eventually grew to about 20 units. They were built with various materials over the years. Early frames were constructed using plywood boxes, and later devices were built with rectangular aluminum tubing. Different recorder designs have been used and the wheel assembly was redesigned in 1983 to consist of a total of four wheels rather than We twelve Cat were used: earlier. As other agencies adopted the California profiIograph, obtaining the device was somewhat difficult. Some States constructed them in Weir own shops or In a local machine shop using plans obtained from CALTRANS. As Me demand grew, Me units were made available commercially. At the present time, Me California profilograph is being marketed by five firms: Cox and Sons, McCracken Division of International Pipe Machinery Corp, Ames Corporation, ELK International-Soillest, and Surface Systems and Instruments-~LC. A dual-lane testing system was also recently Introduced by Paveset America. The accuracy, repeatability, and ease of use reportedly varies between these systems. Numerous State Agencies have developed detailed descriptions for operating profilographs and for evaluating the profilogram traces. The American Society for Testing and Materials has developed a standard method for profilograph testing- ASTM E 1274, "Standard Test Method for Measuring Pavement Roughness Using a Profilograph." In 1978, Me California DOT prepared their latest revision of California Test 526 "Operation of the California Profilograph and Evaluation of Profiles." The increased use of Me California-type profilograph over Me past 35 years suggests that it has filled a definite need in the area of construction smoothness control. However, there is growing concern about certain lunitations of this instrument, including the following (Scofield 1992; Scofield, Kalevela and Anderson 1992): · Speed The slow operating speed of 2 to 3 mi/hr (3.2 to 4.8 km/hr) is becoming less tolerable. Agencies need a faster way to obtain profile v information. Precision- Recent studies have focused on the precision of Me instrument and trace evaluation methods operating under Me current state of the art. It is questionable whether the profile index can be obtained win adequate precision to justify its use in specifications that have incentive/dis~ncentive payments based on ~ in/ml (0.02 m/k~rl) increments. For example, a 1992 study by the Arizona DOT found that Me average standard deviation was 1.9 n/mi for four trace reduction operators on new pavements win average PI ratings from 2.6 to 4.S ~n/mi (Scofield 1992~. On four rougher sections, with an average PI of 8.6 ~n/mi (136 mm/km), Me average standard deviation was E-6
. . 1.6 in/ml (25.3 mm/km). For mechanical and computerized profilographs, the range of test results for the same pavement with the same operator and trace reducer was between 3.5 and 7.0 in/ml (55.3 and 110.6 mm/km) for a smooth section and between 7.0 and 11.0 (~10.6 and 173.S mm/km) for a rougher wheelpa~. This is clearly more variability than can be considered acceptable for Implementing ~ncentive/disincentive specifications. The combination of variability from trace reduction, test path location, equipment properties, and computer filter settings makes obtaining precise smoothness statistics with the profilograph difficult. Smoothness Index With the growing Interest in pavement management systems and the Increased sophistication that is being applied to this activity, there is interest in moving away from Me profile index standard and adopting some other sublunary statistic such as the International Roughness Index (IRI) or Root Mean Square Vertical Acceleration (RMSVA), neither of which can be obtained using a profiIograph. This would allow a given index to be used for new construction and throughout the life of the pavement. Relation to User Response Because Me profilogram is known to amplify and attenuate Me true pavement surface profile, there is some concern about how well the profilograph output relates to the wavelengths of a pavement profile that are felt by highway users. Several cases have been documented in which new pavements that received low PI ratings, and in some cases incentive payments, were found by the roadway users to be uncomfortably rough. This calls into question the suitability of using profilograph data for construction control and suggests the need for refinement in evaluation procedures (Parcells 1992; Scofield 1992~. California ProfiIograph-Cox Models James Cox and Sons, Inc. of Colfax, California, began manufacturing profilographs in 1973. The basic parameters of the California type were adhered to, but Cox developed numerous innovations in design and construction. In the mid-19SOs, Cox developed a computerized version, shown in figure E-2, that is currently marketed as Model CS 8200 at a 1995 price of approximately $24,000. The CS 8200 is a microcomputer-based profile measuring device designed to measure roadway profiles and to reduce profile data in accordance with test method Califorrua 526 (1978~. The CS 8200 reduces Me measured profile data and generates graphic reports containing the measured profile annotated win stationing, excursion information, and documentation points. Profilograms may be reduced to obtain profile index or high points or may do bow sunultaneously. Cox no longer markets a manual profilograph. California ProfiIograph-McCracken The McCracken Division of International Pipe Machinery Corp. of Sioux City, Iowa, has marketed mechanical profilographs since 1984 and a computerized mode] since 1990. They are priced (1995) at approximately $16,000 and $26,000, respectively. E-7
Figure E-2. James Cox & Sons, Inc. computerized profilograph. (Source: Cox & Sons) The computerized version, illustrated in figure E-3, is fully capable of automatically providing Me full analysis of profile information in a manner similar to that performed by the Cox machine. The manufacturer indicates that PI repeatability with one unit and one operator is about 7 percent (0.5 in/ml [7.9 mm/km] for 7 in/ml [~10.6 mm/km] pavement). California ProfiIograph Ames The Ames Unit is manufactured by Ames ProfiIograph of Ames, Iowa was developed In 1986. Bow a manual and a computerized version are available, priced at approximately $10,000 and $22,000, respectively (1995~. The manual unit is referred to simply as the Ames ProfiIograph and the computerized unit is known as Computerized ProfiIograph Mode! 4000. The Ames units are designed to produce the same profile trace as the other California type units that have been mentioned, but it is considerably different In basic design, as shown in figure E-4. The wheel assemblies are similar to those of over models in that an articulated system consisting of a total of six wheels is used to support each enct of the device. E-8
Figure E-3. IPMC McCracken computerized profilograph. (Source: IPMC) .~.... Figure E-4. Ames computerized profilograph. (Source: Ames Profilograph) E-9
However, rather than utilizing a truss type of framework, the 25 it (7.6 m) beam portion of the Ames device is a 2 in by 6 In (51 mrn by 152 rnrn) aluminum box beam. A profile wheel is located at its midpoint, and a hinged leverage system or noncontact ultrasonic transducer system transmits movement of the profile wheel to a recorder located at the rear end of the unit. The computerized version can provide strip chart output at the conclusion of a run that shows the profiIogram trace, PRT values for each 0.: mi (0.16 km) and for the total length, and bump and dip locations and amplitudes. Band widths, bump limits, reduction length, and filler settings can be selected by the operator prior to each run. California ProfiZograph SoiZlest The Soiltest Products Division, ELK International, of Lake Bluff, Illinois markets a mechanical unit referred to as Me CT-4000 Pavement Profilograph. Priced at approximately $21,500, this unit features aluminum frame construction and provides a paper strip chart for evaluation by manual or electronic means. California ProfiIograph Pavese! Designed by Paveset Australia and developed by Paveset America in 1995, the mode] ESP 2000 profilograph measures In two wheelpaths sunultaneously. The system operates using a laptop computer In conjunction with Microsoft Windows software. Measured profile data can be stored digitally and plowed on a conventional printer. A profile index statistic is computed by the software win selectable options for 0.2 in (5 mm), 0.! in (2.5 mm), and 0.0 in blanking bands. Blanking bands are automatically positioned by Me software. Must grind locations are also provided by the software, based on user-selectable maximum bump heights. The Paveset profilograph is constructed of folding aluminum beams mounted on a long wheel base vehicle-towed trailer win rubber mounted suspension. The current version includes four in-line wheels at the support end of each beam. Foldout assembly of the system reportedly requires about 2 minutes. Similar to the Ames profilograph, the measuring wheel deflection in the middle of Me device is measured electronically to quantify Me change in pavement elevation with respect to the profilograph datum beam. The program software provides an audible alarm when the selected maximum allowable speed is surpassed. Turning can be accomplished by circling the tow vehicle in a folded or unfolded position, or disconnecting from the vehicle and turning within Me trailer length. While not currently used by any highway departments, the Texas Highway Department recently evaluated its repeatability and accuracy with respect to a manually operated McCracken profilograph. The repeatability of the system reportedly met Texas DOT standards, but accuracy information remains unavailable. The filtering system used in data reduction is reportedly still being modified to achieve the best correlations with State DOT profilographs. Cost of the system, including Me laptop computer and software, is nearly $27~000. E-10
Rainhart Prof~lograph The Rainhart Profilograph, Catalog No. 860 Profilograph, was developed by the Rainhart Company of Austin, Texas, in conjunction with the Texas Highway Depar~nent In 1967. Their studies served to establish parameters under which the device was designed and constructed. The device is 24.75 It (7.5 m) long overall and is supported by 12 averaging wheels spaced at 27-in (68~mm) intervals along its length. The wheels operate in groups of three, with the front two groups and the rear two group supporting minor trusses, which In turn support a major truss. A profile wheel is located at the center of the unit and is linked to a strip chart recorder that provides a trace of its movement with respect to the datum established by the support wheels. The major difference between the Rainhart Profilograph and the California type is that with the Rainhart the datum is established over Me entire length of the unit and over a width of 44 in (1,118 mm), whereas the datum for the California type is established near the ends of the 25-ft (7.6-m) beam. The trace developed by the Rainhart unit is evaluated In a manner similar to that of We California type, except that a blanking band of 0.1 In (2.5 mm) is normally used. Rainhart profilographs are reportedly heavier anct less maneuverable Wan the more commonly-used California- type profilograph systems (Kulakowski and Wambold 1989~. Currently, only manual data reduction methods are currently available for We Rainhart system, although the ProScan computerized data reduction system (discussed In the following section) can be used to electronically reduce Rainhart profilogram trace data. A 1985 study of five repeat runs at one PCC site and two AC test sites indicated that the repeatability of the manually-reduced Rainhart system is fair with an average coefficient of variation of 6.1 percent (Kulakowski and Wambold 1989~. This relates to a standard deviation of 0.4 in/ml (6.3 mm/km) for a 7 in/ml (110.6 mm/km) newly constructed pavement. High correlation (R2 = 0.82) is reported between manual trace reduction of Rainhart systems using a 0.1 In (2.5 mm) blanking band and California profiIograph systems using a 0.2 in (5.1 mm) blanl<ing band (Scofield 1993~. Electronic ProfiIogram Evaluation ProScan In 1992 and 1993 Kansas State University, under a contract with He Kansas Depa~-~rent of Transportation, developed an automated system to reduce profiIograph traces generated by a manual profiIograph (Devore and Hossain 1994~. Accomplishments included locating and marking the bumps that would require grinding and determining He profile index of He new riding surface. The results of He research have led to the development of a device that is now being marketed under He name ProScan. The scanning equipment for this device is shown in figure E-5. E-~!
- ~ - - - - - - ~ - ~ i' Figure E-5. ProScan elec~Tomc trace reduction system. (Source: ICE Corporation) ProScan is an automated profiIogram scanning system manufactured by ICE Corporation of Manhattan, Kansas. It is presently being marketed at a price of approximately $6,000. This relatively new device is amply described as follows (Devore and Hossa~n 1994~: ProScan system is a software and hardware package that provides a computerized metho~for reducing (analyzing) traces produced by non- computerizedl proflographs. This eliminates any subjectivity in the reduction process, increases the accuracy of the profile roughness index (PRIJ, greatly reduces reduction lime (it scans and analyzes a {race at the rate 15 mi/hr [24 km/hr] versus I-2 mi/hr [~.6-3.2 km/hr] manualZyJ, and reduces the required expertise (or training {imeJ of the person responsible for the reductions.... The paper transport machine accepts pro.fiiZograms from all popular Rainhar! and California style profiIographs and the software program automatically regulates the advancement, stopping, and speed of the profilogram paper. The software is able to control such items as bZanking-band width, minimum scallop width, and grind[-template height, one can also change the standard reduction length (to a value other than 0.] mi [160.S my, scallop height measurement resolution, and several other parameters. The program is also metric-readly in that aZZ measurement and the profile roughness index (PRY) can be reported in metric units. E-12
A 486-SX33 computer or better is required for running the program. The computer must have an available expansion slot for the scanner interface card (thus notebook and many laptop computers are not usable), and must have an EGA or better (VGA) video display. Notebook computers are acceptable, provided that they have a gas-plasma or active-matr~x LCD display. The $6,000 price includes the software, the paper transport unit, and the scanner. The computer and printer are normally furnished by the user, but turnkey system (including those items with the scanner board anc! all software installed) are available upon request. Advantages to this system over computerized profilographs include: · Lower cost. · Existing profilograph equipment can be used without modification. · A single computerized reduction system can be used in conjunction with multiple profilographs when locations allow. · An original paper record of the profilogram exists (before being massaged by a program). · Reports averaging side-by side segments and complete tracks can be generated. · Data may be reanalyzed using different reduction parameters in a matter of seconds. The computer equipment is not subjected to extreme weaker conditions. Any failure of computer hardware does not halt Me collection of profilograph data. The computer can double as a general-purpose office computer. Response-Type Road Roughness Measuring Systems (RTRRMS) Response-type measuring equipment is used to evaluate road roughness by measuring the dynamic response of a mechanical device traveling over a pavement surface at a given speed. Both automobiles and standardized trailers have been used for this purpose, win measurements taken of Me vertical movement of the rear axle of Me automobile or of the axle of Me trailer win respect to the vehicle frame. Primarily, RTRRMS are used for data collection on highway networks for pavement management purposes; however, they have been used on ia limited basis for controlling smoothness of new pavement construction. Primary advantages of RTRRM systems are as follows (Woodstrom 1990~: · Initial and operating costs are low. Data are normally collected at a 50 mi/hr (80 km/hr) speed, ~us, a considerable length of pavement can be evaluated in a relatively short period of time. · Reasonably accurate and reproducible roughness data can be collected if the device is properly calibrated and maintained. E-13
Limitations of RTRRM systems are as follows (Woodstrom 1990~: The characteristics of the mechanical system and the speed of travel affect the resulting roughness measurements. Response-type road roughness measuring systems measure the dynamic effect of roughness but do not define true pavement profile features. Response-type devices must frequently be calibrated, Trough a range of operating speeds and against sections of known profile ranging from very smooth to rough, to provide accurate repeatable data. The cost of calibration can be quite high. The vehicles In which RTRRMS are installed contribute to many sources of potential variation, including rear suspension damping, tire variations, vehicle weight changes, and windage effects. Because of variations of the different mechanical systems, comparability of data among users is difficult. The best known of the RTRRMS devices Is the Mays Ride Meter, which is described in the following section. Mays Ride Meter The Mays Ride Meter, shown in ~. ~. . . ~ figure E-6, is a response-type instrument that logs the pavement surface irregularities by recording magnitude, direction, and summation of axle to body excursion of a test vehicle or a standardized trailer. In the front seat of the vehicle, a strip chart recorder produces a printout of Me data. Resolution of Me transmitter is one electric impulse for each 0.! in (2.5 mm) of vertical axle displacement. The chart feeds in increments of 1/64 In (0.016 mm) for each 0.! In (2.5 mm) of rear axle/body excursion. Thus, a perfectly smooth pavement will not drive the chart, and a rough pavement will produce a longer chart output. The chart produces a distance trace and allows a roughness summation to be calculated over any desired distance. ~ Testing can be conducted at speeds from 20 to 60 mi/hr (32 to 96 km/hr); however, testing speed affects the system output. Other factors that affect output are tire inflation, trailer wheel alignment, suspension system characteristics, and vehicle weight changes (Gillespie, Sayers, and Segal 1980; Kulakowski, Chapman, and Wambold 1987~. All of these factors are a function of Me host vehicle characteristics. As a result, when host vehicle properties vary or change (e.g., tire pressure, shock absorbers, wheel alignment) the Mays Ride Number can vary accordingly. This results in concern for the accuracy and repeatability of MRM systems when used in initial pavement smoothness testing and control. Gillespie, Sayers, and Segal concluded in 1980 that "in Me more critical functions of evaluating sections of individual roads (especially Me relatively smooth surfaces represented by new construction), Me remaining random error limits the usefulness of RTRRM Concluding Mays) systems." E-14
At: Figure E-6. Georgia DOT trailer-mounted Mays Ride Meter (Source: Stone 1988) Calibration and correlation of these systems must be conducted frequently to ensure accuracy. Standard methods for this have been prepared by the National Institute of Standards and Technology for the FHWA in 1988 (Vorburger, Robinson, Pick and Flynn 1989), and recent success has been achieved win correlations between MRN and output from inertial profile measurement system output (Woodstrom 1990~. The Ra~nhart Company of Austin, Texas, has manufactured Me Mays Ride Meter, Catalog No. 890, since 1970. It is currently (1995) priced at approximately $3,500, uninstalled. A standardized trailer, developed by the Texas State Depal-tr~lent of Highways and Public Transportation for the purpose of stabilizing the weight variable, permitting Me use of standard test fires, and standardizing the suspension system, is also available Trough Rainhart for about $9,500. International Cybernetics Corporation (ICC) of Largo, Florida, also supplies the Mays Meter. The company's literature lists two models, MDR 4010, Mays Roughness and Highway Features Inventory System, and MDR 2010, Mays Roughness and Highway Features Inventory System Using Lap-Top Computer. BPR Roughomeler The BPR Roughometer was first Introduced in 1925, and was recognized as being the best high-speed smoothness measuring device available at that time (Buchanan E-15
anc} Catudal 19401. It consisted of a single-wheeled trailer that is towed by a car or a light truck at a speed of 20 mi/hr (32.2 km/hr) (Woodstrom l99o). The wheel is mounted on leaf springs supported by the trailer frame. Pavement surface contours cause the sensing wheel to oscillate vertically with respect to the frame. The vertical movement is accumulated by a numerical integrator, yielding a roughness statistic in terms of inches per mile. After some period of use, it was learned that the equipment was highly susceptible to changes in temperature and to the condition of its bearings and other mechanical components. In addition, it had a resonant frequency problem that, if excited, produced erroneous results. Vibrations were commonly noted at high roughness levels. As a result, its use has gradually clecI~ned. PCA Roadmeler The PCA Roadmeter was developed by the Portland Cement Association (PCA) in 1965 (Brokaw 1967~. This device measures the number and amplitude of vertical deviations between the body of a standard automobile and Me center of the rear-axIe housing. It was primarily used with pavement rating systems into the early 19SOs and, to a limited degree, as part of smoothness specifications. It provided the roughness level for a section in terms of counts per mile but did not have the capability of distinguishing where in a section the roughness was located. Results could be affected by the placement of weights (luggage and passengers) in the vehicle, tire pressures, cross-winds, and other factors. Variations in design features were incorporated into the original device and, in some cases, these new instruments attuned Weir own identity. An example of one such device is the Cox Roadmeler. Inertial Profilometer l An inertial profiIometer was developed In Me early 1960s by Me General Motors Corporation Research Laboratories (Spangler and Kelly 1966~. This development was made possible by Me availability of a high-quality force balance accelerometer used In the aerospace industry for inertial guidance, as well as by the availability of high- quality analog computer components, including the integrators used in profile computation. With this type of system, profiles of a pavement surface can be obtained at normal highway speeds. Several of the past and recent inertial devices are described in the following section. K.~. Law 690 DNC Profilometer General Motors Corporation made the inertial profiIometer technology available to the highway community Trough a licensing agreement with K.~. Law Engineers, Inc. The first commercial contact-type profiIometer was built in 1966 for the Texas Highway Department. New technology led to Me development of the Mode} 690 Surface Dynamics ProfiIometer in 1979 and later the Mode! 690 DNC (digital noncontact). E-16
This device measures and computes the longitudinal profile of the pavement through the creation of an inertial reference by using accelerometers mounted In the body of the measuring vehicle. Relative displacement between the accelerometers and the pavement surface was originally measured with tensioned contact wheels, but is currently measured with a noncontact light beam system mounted with the accelerometers in the vehicle. The pavement profile is computed in each wheelpath as a function of the distance traveled (Carmichael 1987~. The repeatability of these systems has been very good over the years with average measured TR! standard deviations of 4 in/ml (63 mm/km) for AC pavements and 2 in/m: (32 mm/km) for PCC pavements when seven 690 DNC systems were tested In a national study over eight sections at four sites. The ~ for these sites ranged from 55 ~n/mi (869 mm/km) to more Man 300 ~n/mi (4,740 mm/km) (Perera 1994~. In the same study, gooci agreement was found between the {RT values from the ProfiIometer and rod and level when Me {RI was greater than 100 in/ml (1,580 mm/km). Poor agreement was reported on sections where the IR] was less than 100 in/ml (1,580 mm/km). Wet and very dark pavements reportedly can affect the system's accuracy, and the problems of light saturation and lost lock inherent with optical height sensors have resulted in a trend in the industry away from white light optical distance measuring sensors. Accuracy of the distance measuring system for the 690 DNC is emanated at 0.05 percent. Numerous software programs are available for conducting quality control checks and for using the profile data to obtain s~rnulated roughness measurement indices. Roughness indices that can be computed include root mean square vertical acceleration (RMSVA), International Roughness Index (IRI), BPR Roughometer index, Mays Ride Number (MRN), PCA meter, Cox meter, and others. Although it has had limited use for construction smoothness control, this device is used for pavement management system surveys because of its accuracy and rapid rate of data collection. in addition, it is currently being used to collect profile measurements on me pavement test sections included in Me FHWA's Long Term Pavement Performance (LTPP) pavement monitoring program (SE]RP 1994~. The cost of the K.J. Law Mode} 690 DNC Inertial ProfiIometer was approximately $250,000 but this depended on what optional features and computer software programs were selected. In 1994, the 690 DNC Inertial ProfiIometer was replaced by the T6600 Mode! (described below). K.~. Law T6600 Inertial ProfiZomeler and T6500 Road Surveyor The T6600 replaces Me 690 DNC ProfiIometer and is a noncontact profiler system that measures and records pavement profile in each wheelpa~, as well as pavement rut depth. Shown in figure E-7, The basic system Includes two accelerometers and three infrared height sensors. The infrared light projection on Me surface is a rectangular spot 0.16 in by 1.46 in (4 mm by 37 mm) across the wheelpath. Up to 25 E-17
Figure E-7. K.J. Law Road Surveyor Profilometer. (Source: K.J. Law Engineers, Inc.) sensors are available for transverse profile measurement. Pavement profile elevations are taken at An (25.4 nun) intervals, averaged over a 12-~n (305 mm) interval, and recorded for every I, 2, or 6 In (25, 51 or 152 mm) of travel. Available profile utilization programs include bituminous fill smoothing, mill or grind correction, combined mill and bituminous fill optimization, slab faulting measurement and classification, strip chart profile plowing, sunulation of Me California profilograph, Mays Ride Meter, and variable length straightedge simulation. Output statistics include IRI, Ride Number UNCAP I-23), root mean square acceleration (RMSA), ASTM Ride Number (RN), and MRN. Section lengths and section breaks can also be defined In the software. The system is reportedly self-calibrating. Data reduction is fully automated, with output to disk or printer. Average profile bias reported by the manufacturer is 0.005 In (0.13 mm), and absolute bias reported by the manufacturer is 0.05 In (~.3 mm). Profile repeatability, as estimated by Me manufacturer, is 0.02 In (0.5 mm). The distance measuring system accuracy is reportedly 0.01 percent. The T6500 Road Surveyor is a modification of Me T6600 Cat is applicable for network measurement application. The cost of the T6600 and T6500 are approximately $190,000 and $160,000, respectively. E-~S
Lightweight Inertial Surface Analyzer (LISA) The LISA was developed by the Materials and Technology Division of the Michigan Department of Transportation (Darlington 1992~. The LISA measures profile by the GM concept at a far lower cost than the van-based unit and, because of its light weight, permits analysis of concrete pavements within hours after paving. An accelerometer and a vertical distance sensor are used to measure changes In pavement elevation on any pavement type. The accelerometer, having a resolution of 0.0001 g, is used In conjunction with a laser distance sensor to compute the pavement profile. The tow vehicle is a John Deere towhee! "GATOR" weighing about 800 Ib (360 kg), and operating at a speed of ~ to 12 mi/hr (12.9 to 19.3 km/hr). Pavement profile elevation values are recorded every 3 in (76 mm) of vehicle travel, providing a profile that is filtered for accurate presentation of pavement wavelength features from 1.S to 80 it (0.5 to 24.4 m). The system, shown in figure E-S, is capable of self- calibration testing and correction. An on-board computer collects the profile data and can be used to view the "true, measured profile or Me simulates! profiIograph profile. It also can calculate smoothness statistics, such as the Michigan Ride Qualitr Index (RQ1), the California profile index, the International Roughness Index, and others. Any blanking band can be selected for PRT calculation, and must grind locations, using any selected bump height, can be identified and listed on output tables and on profiIograph plots. The must grind analysis can also be completed on the "true" profile also. (Darlington 1992). Figure Em. Lightweight Inertial Surface Analyzer (Source: Transology Associates) E-19
The LISA is being marketed by Transology Associates of East Lansing, Michigan. About 60 minutes are required for operator training. The complete device, including profiling system, vehicle, and transporting trailer is approximately $35,000 to $40,000 In 1995. K.~. Law 8300 Sonic Profiler The K.~. Law Mode! 8300A Pavement Roughness Surveyor is a non contact pavement roughness measuring system using an ultrasonic probe and an accelerometer to measure road roughness. The output is RMSA (roof mean square acceleration) roughness and correlation equations are used to provide several response-type roughness Indices, such as the Mays Ride Number, Me PCA Roadmeter number, and the BPR Roughometer index. Measured profile data are not retained by this system, improving computation time and reducing data storage expense. However, this made it impossible to compute supplemental smoothness indices from original profile data. The cost for this device, when manufactured, was approximately $50,000, exclusive of Me vehicle. South Dakota Profling System A profiling system was developed by the Soup Dakota Department of Transportation in 1981 (Huff 1984~. It is typically mounted in a small to mid-si7ed van and measures pavement profile and rut depth, operating on an inertial reference profiling concept. Mounted on Me front of the ~rutial vehicles are an accelerometer and ultrasonic (acoustic) sensor for profile measurement in one wheelpa~ and Tree ultrasonic sensors for the measurement of rut depth. Profile elevation measurements are reported at ~ It (0.305 m) intervals and rut depth elevations are measured and reported at 2 ft (0.61 m) intervals. Testing speeds can range up to 65 m~/hr (104.5 km/hr). Roughness output has been reported by the Soup Dakota profiling system by a PST value computed from Me measured profile data. Profile measurements are processed nearly instantaneously by Me system software using correlations between measured profile values and rating panel values from surveys conducted in South Dakota. The South Dakota DOT has also developed the capability to generate International Roughness Index (IRI) and Mays Output (MO) values from Me measured profile data. Cost of the South Dakota system has been estimated at $25,000 to $30,000, excluding the cost of the vehicle on which the device is mounted. Operator training for this type of system generally requires one or more days, with additional support as needed. Approximately 30 States were using this type of device in their 1994 system survey work. E-20
ICC MDR Profiling System The South Dakota profiling system is now manufactured commercially by International Cybernetics Corporation (ICC) of Clearwater, Florida. This company has a family of road profiler equipment in its "Mobile Data Recorder" (MDR) product line, each mounted in a small- to mid-sized cargo van. These include the following: MDR 4084 Profile/Roughness System featuring Profile/IR] in one wheelpath; average rut depth in two wheelpaths to center of lane; one inertial reference accelerometer; three height sensors; one linear distance sensor, and a m~rumum recording interval of 3 in (76 mm). MDR 4085 Enhanced Profile/Rut depth/Roughness System featuring profile/IRI in two wheelpaths; average or half-car simulation {RI; average rut depth in two wheelpaths to center of lane; two inertial reference accelerometers; three height sensors; one linear distance sensor, and a 3 in (76 mm) minimum recording interval. MDR 4087 Enhanced Profile/Rut depth/Roughness System, shown in figure E- 9, featuring profile/IRI in two wheelpaths; average or half-car simulation IRI; average rut depth In each wheelpath; two inertial reference accelerometers; five or more height sensors one linear distance sensor, and a minimum recording interval of 3 in (76 mm). ~ _ _ ~ __ Figure E-9. ICC MDR 4087 profiler. (Source: ICC)
International Cybernetics Corporation offers ultrasonic or laser sensors on all current MAR systems. The lasers are more expensive than the ultrasonic sensors but offer several advantages. First, the lasers can be used when roads are wet as long as there is no standing water to alter the readings. Second, more data points can be sampled with the laser so that a better average height can be obtained, allowing better repeatability. Also, the laser is less sensitive to temperature changes, which may occur quickly in hilly or changing terrain. Finally, the accuracy of ultrasonic sensors is affected by macrotexture on asphalt pavements, according to a national study (Perera 1994~. Reported accuracy of the distance measuring system is 0.02 percent. The standard deviation of two laser-based {CC MDR 4087-L systems tested in 1993 on eight sections at four sites was 4 for AC pavements and 2 for PCC pavements. IRI at the sites ranged from 55 in/ml (869 mm/km) to more than 300 in/ml (4,740 mm/km) (Perera 1994~. Setup effort, including equipment warm up, calibration, and system initialization, for this type of system is minimal, requiring 15 to 30 minutes. Data can be collected at highway speeds, and is stored and processed by computer. Costs of these systems range from $80,000 to $110,000 in 1995 dollars. A single wheelpath portable laser profiler is also available for $45,000. Dynatest Profiling System In 1995, Dynatest introduced a fully-automated roughness measurement test system called the Model 5051 RSP (Road Surface Profiler) to meet the FHWA Long Term Pavement Performance monitoring specifications. The base system, mounted on Me front of a mid-s~zed van, generates response-type output using a single accelerometer. The system is shown in figure E-IO. Up to eleven lasers can be added to the response-type system, using the in-place accelerometer, to provide longitudinal and transverse pavement surface profile and rutting information ~ real tune at speeds up to 100 km/hr (62 mi/hr). The lasers and accelerometers can be mounted in single boxes, S.5-ft (2.~m) rut bars, and In fold-out extensions to measure up to Il.5 ft (3.5 m) of road width. The software is capable of automated data recording and post-process~ng, or real- t~me data processing of longitudinal or transverse profile and the TR] statistic. Longitudinal profile recording Intervals down to ~ in (25.4 mm) are available. Transverse profile measurement can be measured on the same frequency. With a single accelerometer, the software is capable of providing a response-type roughness index developed at the University of Texas (Walker 1991~. Using a combination of accelerometers and laser sensors, the Dynatest 5051 RSP system can provide real-time ~ values for each wheelpath and maximum "stringl~ne memos" rut values using the roadway cross section. Precise, high quality distance measurement Instrumentation is also provided. E-22
Figure E-10. Dynatest 5051 Road Surface Profiler (Source: Dynatest Consulting, Inc.) FHWA PRORUT The PRORUT device was developed through an FHWA contract with the University of Michigan Transportation Research Institute (UMTRI) in 1983 (Gillespie and Sayers 1987~. PRORUT is an inertial profiling system that can be used to measure and record various roadway characteristics, including longitudinal pavement profiles, rutting, and roughness levels of the bow wheel tracks. The system is mounted In a van and uses commercially available components. Laser sensors and accelerometers are used to obtain the profile measurement in each wheel track. A centrally mounted laser sensor is also included for measuring the crown between Me wheel tracks and determining an average rut depth. The electronics and software system provide for two additional sensors so that Me rut depth in each wheel track can be measured, if desired. An IBM-compatible personal computer controls system operation and processes the data. The menu-driven software leads the operator through system calibration, component checks, and data collection. Data can Men be analyzed and presented graphically and numerically. The graphic records are useful in detailing the road profiles, and Me tabular output is preferable for listing Me International Roughness Index (IRI) and the mean rut depth averaged over selectable lengths of the test sections (Brown 1990~. E-23
IMS Laser RST Propter Swedish The IMS Laser RST Profiler is a road surveying system developed by the Swedish National Road Adm~rustration. It can be used to measure longitudinal pavement profile, rut depth, megatexture, and macrotexture at a recording interval of ~ in (25.4 mm) or more. An accelerometer is used to establish an inertial reference plane, and roughness is measured continuously at speeds up to 55 mi/hr (90 km/hr). The roughness measurement output is IRI, although over roughness indices can be computed. A modification of this system has been used for ~ree-dimensional profile mapping. Figure E-~1 shows a version of the Laser RST system. Danish Road Institute ProfiIograph The Danish Road Institute (DRI) ProfiIograph was developed in 1990 and is quite similar to the Swedish device mentioned above, except Cat it is mounted in a car rather than in a small van. As shown in figure E-12, it has a vertical accelerometer, three aclditional accelerometers, two gyroscopes and one odometer, so that all six degrees (three translatoric and three rotoric) are evaluated. Road surface elevation is measured by up to 17 laser sensors mounted on a beam attached to the front of the vehicle. Lasers are also mounted at an angle on We extendable support bar to measure surface profile for the entire travel lane width. Because of the large number of accurate sensors and positioning instruments, this system can provide a large amount of Information. This includes lon~tud~nal and transverse surface profile, ruding, side slope, surface macrotexture (influencing fire noise), surface microtexture (influencing friction characteristics), and many profile- and response-related pavement smoothness indices. In 1994 the Danish Road Institute ProfiIograph Incorporated GPS technology using We GPS satellite navigation system to identify the vehicle's position within a few centimeters. clinometer-Based Profiling Systems Three manually propelled profiling systems have been developed using highly sensitive inclinometer sensors. The Face Dipstick was developed for smoothness measurement on ultra-smooth warehouse floors and later has been used for calibration of high-speed road profile systems. On the over hand, We CSC Profilair was intended as a replacement for rod-and-level profile measurement of airfield runways. The ARRB Walking Profiler was developed and manufactured by We Australian Road Research Board and is marketed In the United States by Trigg Industries International, Inc. for pavements, building slabs, and game courts as a World Bank Class ~ profile measuring device. These systems are discussed below. E-24
Figure E-ll. Infrastructure Management Services Laser RST profiler. (Source: IMS) Figure E-12. DRI Profilograph system. (Source: Darrish Road Institute) E-25
E.W. Face Dipstick The Dipstick (DIP = Digital Incremental Profiler) is a precision smoothness measuring device, operated by hand. Its shape is somewhat similar to a walking cane with a I-ft (0.3 m) long base, as shown In figure E-13. In the base is a vertical displacement measuring device that provides slope information to a small attached computer. By rotating the base end to end, an operator can move along a highway at 3-, 6-, 9-, or 12-in (7~, 152-, 229-, or 305-mm) Increments, determining Me surface profile, with a resolution of 0.001 in (0.025 mm). About 900 measurements can be made In an hour at 1-ft (0.3-m) increments. The computer and available software can produce a record of the profile measurements and also calculate an TRT, RMSVA, Mays Output, PSI, PRI, instantaneous vertical and angular acceleration, F-numbers (floor smoothness ratings), and cumulative deviation from a selected straightedge. Surface contours can be plotted, overlay thicknesses can be determined, and cumulative and running IR! can be computed and displayed. cot Begs It hands MEOW .11 1 L - d Gem ~· Ale Sod, feel a: filet Cat -Cat LEO · , / ~ S - ah Kk#sor~d OIL ~3 Ad> Exam* ~ Figure E-13. Dipstick Auto-Read Profiler. (Source: Face Construction Technologies) E-26
A 1991 study by the Center for Transportation concluded that "the manual-read version of the dipstick is an effective Class ~ profiling instrument, as long as operational techniques, including loop closing and repeat runs are followed. It is many times faster than the rod-and-level surveying method, and has a resolution 12 times better, making it an extremely accurate device" (Bertrand, Harrison, and Hudson 1991~. Class ~ profiling instruments are assumed to provide absolute elevation relative to true horizontal profile with less than I.5 percent bias. In highway applications, the Face Dipstick is used primarily to calibrate other types of smoothness measuring equipment and is also used to evaluate the smoothness of warehouse floors when smoothness is a critical factor In automated storage/retrieval operations. Because of its accuracy, the Dipstick has been used recently by the FHWA to provide a baseline profile for verification of high-speed profile system accuracy. CSC Profilair Developed in 1992 by Civil Structural Consultants (CSC) of North Vancouver, B.C. and commercially available in 1994, He CSC Profilair was originally designed to replace rod-and-level profile measurement of airport runways. The system, shown in figure E-14, is a gravity referenced device using an inclinometer for measuring the vertical angular deviation between adjacent elevation points. The elevation difference is then computed and added algebraically to He value of the previous elevation. Sensitivity of the angular measurement is based on a + 14.5 degree range digitized into + 4,096 units of angle, providing a static resolution of 0.001 in (0.025 mm). The 3-wheeled device is comprised of a metal structure that moves by rolling on solid balloon-type natural rubber tires. It can be divided into three 70-Ib (32 kg), or less, sections for airline or vehicle transportation. Immediately behind He steering wheel is a fourth wheel for recording He surface elevation. Data from the inclinometer attacher! to Cat wheel is stored on a 3.5 in (89 mm) floppy disk. The interval for testing is ~ ft (0.3 m), and the recommended rate of data collection is 5,000 ff/hr (0.42 m/s). Power for He inclinometer data collection system is supplied by a 12-voll bakery for up to ~ hours of operation, and power for forward movement is provided by the operator. At the conclusion of data collection, the disk can be removed and installed in a portable computer for computation of He profile and roughness Indices IRI and RMSVA. Bump criteria for selected baselength can be input before processing with the result that locations of out-of-tolerance bumps are marked on the profile and listed In a separate table. The software makes corrections for temperature and state of motion variations. To unprove accuracy, a reverse run is generally necessary, and the software is designed to adjust for these closed-loop profiles. E-27
- ~- ~ ~- - l ~ ~ - ~ - 111 1 ~: ~ ~- ~ ~ s.- ~ ~ or ~ ~_ ~ 1 _ f ~ ~- ~ ~ - AL. _ ~ _ .:. - :' - - a :>x : 4:.~ - ~w I: x ~ ~ i: ~I -- ~ i: 1 ,: ~ :: :: ~: ::: ~ ~ :::: ::::: i: :::: : : ~ ::: : : ::: :: : :: :::::: : : : ::: :::: :: :: ~::::~: ::::: :: ~: ~ :~ :: : ~ : : ::: ::: ~ ~ ~ ~ ~ : : :, ::~ id: ~:: ::::::::::: ::~-: . ~a : ~::: ~- :: - ~ ::: :: : ::: ID_ : : :: : -~ ~ :: :::: :: ~ - ~ - .,,._: :: ::: : ::: :~. ~ Figure E-14. CSC Profilair. (Source: CSC Profilair) ARRB TR Lld. Walking Profiler The Walking Profiler, developed and manufactured by the Australian Road Research Board, is a multi-wheeled incI~nometer-based device that is pushed by the Operator at walking speed (1 km/hr [3,280 ft/hr]~. It continuously records the relative height of successive points at intervals of 241 mm (9.5 in), storing the vertical and horizontal deviation from the starting position in an onboard laptop computer. Software is available to compute and graph surface profiles and determine IRI. Accuracy reported by the manufacturer is +2 mm (0.0787 in) per 100 m (328 It) of horizontal distance. Other Profile and Smoothness Measuring Systems In addition to the above described pavement smoothness measuring devices, several other devices have been used for smoothness evaluations. These Include the ChIoe slope variance measuring system, Me Siometer accelerometer-based system, the Transportation Road Research Laboratory (TRRL) ProfiIometer, and the multifunctional ARAN pavement evaluation system. CHLOE Profits Developed at the AASHO Road Test during the late 1950s by Carey, Huckins, Leathers, and other engineers, Me ChIoe Profiler provides a "slope variance" (SV) statistic as a measure of pavement roughness, which in turn was intended to be used as an element of serviceability (Carey et al. 19621. Me Chioe Profiler is a relatively simple electrical-mechanical device. It is a trailered unit 25.5 it (7.8 m) long that measures the difference in angles between the 20-ft (6.! m) long trailer tongue and a small beam with two wheels, 9 in (229 mm) apart, recording the measured pavement slope at 6-~n (305-mm) increments. When in operation, the device is limited to a E-28
speed of less than 5 mi/hr (8.0 km/hr) (Gillespie, Sayers, and Segal 1980; Carey et al. 1962). The gain from the CHLOE system is near 1.0 for wave numbers between 0.02 and 30 cycles/ft (0.006 and 98 cycles/m), and is less than 1.0 for higher wave numbers. A gain of 1.0 indicates prefect representation of the "true" pavement profile. This trueness to the actual slope profile is a positive advantage of the CHLOE system; however, the SV statistic is derived from an band of profile wave numbers much broader than is significant to an automotive vehicle, and as a result introduces a random error that degrades the agreement between RTRRM system output and SV (Gillespie, Sayers, and Segal 1980~. Walker Siomeler The Siometer accelerometer response device was developed at We University of Texas in the 1980s (Walker and Phung 1987~. Within the Texas SDHPT, the primary application of the Siometer is for evaluation of ride quality. The device became known as Me Siometer because its primary OUtDUt iS Me sewiceabili~ index (SI) for a particular pavement section. Currently, Me system consist of a sensor, a main control module, and a laptop computer for storing results. Included In the sensor system are an accelerometer, housed in a small case mounted vertically inside the trunk of a vehicle, and a distance measuring signal. The sensor unit and Me main control module cost $20,000 (1995~. -A - c, In 1995, the Siometer technology was incorporated into the Dvnatest Mode] 5051 RSP, described in an earlier section. This system provides a small van, a sensing accelerometer, hardware, and software for determining Me vehicle response-type roughness statistic. Including the test van, the 5051 RSP-R0.1 version is currently sold for about $41,000. - 1 - ~ . · . .. - TRRL ProfiIometer Developed by the United Kingdom TRRL, the TRRL ProfiIometer originally consisted of a 2-wheeled trailer coupled to a lowing vehicle Jordan and Porter 1983~. The current system consists of a single vehicle into which is incorporated a beam encapsulated in foam and 15.4 ft (5 m) long, upon which four noncontact laser displacement transducers are mounted. Each transducer measures the vertical distance of pavement surface below the beam. The foam encapsulation inhibits the development of vertical temperature gradients in the beam, which would otherwise cause the beam to bend. Through a series of aigoriduns, He pavement surface profile is measured without He use of accelerometers or gyroscopes. · . ~ ~ ~ ~ ~ ~ . ·~ ~ ~ . ~ Reported capabilities of the equipment include measurement of longitudinal profile in either wheelpath, measurement of average rut depth, surface macrotexture measurement, and measurement of gradient, cross-fall, and radius of curvature. Windshield-mounted video cameras are also available. As reporter} by the manufacturer, the system is capable of accurately measuring pavement surface E-29
wavelengths between 8 In (200 mm) and 493 it (150 m). accuracy is 0.5 percent. Distance measuring The equipment has been used to measure longitudinal profile on asphalt and concrete roads and airfields In bow the OK and continental Europe. It can operate at speeds from 3 to 62 mi/hr (5 to 100 km/hr). ARAN Profiler The Automatic Road Analyzer (ARAN) is a multi-function road quality surveying instrument. It is equipped with the following subsystems: (a) pavement surface roughness measurement, ebb rut clepth and transverse profile measurement, (c) gyro, (~) right-of-way video logging, (e) pavement condition video logging, and (f) pavement rating. System hardware Includes axle and body accelerometers, analog signal amplifiers, analog low-pass fillers, and a 12-bit analog to digital (A/D) converter. The software includes digital band-pass filters Passing wavelengths of 1 It ~_ ~ ~- ~ e e ~e ~bed ~ 1 ~ (A tO ;5tJU ~ (Ue;, m tO Yle4 m), alglta1 nlgn-pass nlters passing wavelengths of 2 It (Oe61 m) or less, and statistical models generating the reported roughness statistics RMSVA (roof mean square vertical acceleration), MAS (mean absolute sIope) and TEXTURE (surface texture) (Lu et al. 1994~. The current ARAN Laser SDP mode] provides surface profile, IRI, Serviceability Index (SI), and Ride Number output. The software is also capable of identifying must-grind locations, using algorithms developed by States. Accelerometers and laser-based distance measuring sensors are included In this self-calibrat~ng system mounted in a full-sized van. Smoothness Measuring Equipment of the Future What will be in Me future for pavement smoothness measuring equipment? Given Me rapid advances In technology, it would be hazardous to predict Me future. However, Improvements In accelerometers and gyroscopes advanced by Me space industry should make it possible to more accurately measure long wavelength (greater than 300 It [91.4 m]) surface deviations. Faster data collection systems, unproved data storage hardware, higher-speed computing capabilities, and better height sensing mechanisms should make it more possible to measure and evaluate megatexture (noise source) and microtexture (friction source) using noncontact profiling systems. Global positioning may be used to pinpoint areas of pavement roughness on maps that are linked to databases contouring construction, as well as maintenance and performance data for an entire States pavement inventory. As the capabilities of new equipment to measure pavement profiles Improve, the highway community will be unpolled to define the question, "How accurate is accurate enough?", especially in the area of new construction smoothness control. If drivers and vehicles do not respond negatively to very long or very short pavement surface wavelengths, it is not necessary for Crucial construction equipment to E-30
accurately measure these wavelengths. Of course, other uses can be made of the new information made available by technological advances, including friction and noise evaluations, and highway agencies will use these advances to maintain high quality pavements In an era of shrinking budgets. E-31
