Smoothness Specifications for Pavements Final Report (1997) / Chapter Skim
Currently Skimming:
Chapter 5. Study of Smoothness-Measuring Equipment and Indices
Chapter 5. Study of Smoothness-Measuring Equipment and Indices
Pages 163-200
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 163... ...
Due to the integrated nature of a smoothness index and smoothness-measur~ng equipment In an overall test procedure, this project studied each side of the smoothness issu~quipment and indices to arrive at recommended specifications for equipment used to measure ~rutial pavement smoothness. This was accomplished through a review of current smoothness-measuring equipment and Trough an evaluation of current smoothness summary statistics.
|
From page 164... ...
A few agencies are using noncontact pavement profilers for Me measurement of initial pavement smoothness. These devices offer We advantage of providing a pavement profile from which a more accurate representation of Me smoothness characteristics of Me roadway can be made.
|
From page 165... ...
A testing procedure should have the support of managers and technicians. Although the review indicated several pieces of equipment Cat could be used In measurement of initial pavement smoothness, it is apparent that the capabilities and characteristics of the necessary smoothness-measuring equipment is largely driven by the smoothness index that it measures.
|
From page 166... ...
However, the response of highway users and perhaps the additional pavement damage from dynamic loads should be controlling factors In defining acceptable levels of pavement smoothness. Because of this strong desire in the highway community to correlate roughness indices with user response, He roughness-user response relationship has been explored in various studies for both response-type and profile-based numerics Janoff, Nick, Davit, and Hayhoe 1985; Sayers and Gillespie 1986; lanoff 1988; Al-Omari and Darter 1992~.
|
From page 167... ...
combined, MRN measures do not accurately predict rideability danoff 19881. A 1987 Kansas DOT research project using 108 AC, PCC, and composite test sections resulted In moderate panel rating R2 values of 0.67 for both AC and PCC pavements (Moore, Clark, and Plumb 1987~.
|
From page 168... ...
It is evident from this figure that the true pavement surface profile is not measured by RTRRM systems. Variability in RTRRM systems has been documented since 1980, when Gillespie et al.
|
From page 169... ...
. Profile Index (PA from a Profilograph The profilograph, commonly used for initial pavement smoothness control, applies a mechanical rolling straightedge filter to a pavement profile, measuring wavelengths within the range of 1 to 75 ft (0.3 to 23 m)
|
From page 170... ...
However, the statistic is not based on signal theory, and it applies a linear weighting to profile data wavelengths that is not based on analytical investigation of the relationship between user response and pavement wavelengths. However poorly the profilograph measures the true pavement surface profile, the PI, as derived from the profilograph output, has served me highway corrununity fairly well as an easily understood index of ~rutial pavement smoothness.
|
From page 171... ...
This application allows PI values to be more easily obtained. In addition, it provides a more accurate crucial pavement profile from which other pavement smoothness indices can be obtained.
|
From page 172... ...
Thus, indices based on measurement of true pavement profile are investigated for their ability to more completely relate with user response and provide a practical method of measuring nutial pavement smoothness. Profile-Based Indices- Mechanical System Simulation Since Me advent of high-speed profile measuring devices, analog and digital filtering have been used to remove unnecessary and unproper wavelengths from measured "true" surface profiles, that is, to amplify or attenuate selected wavebands.
|
From page 173... ...
applied to the pavement surface profile is the determining factor needed, In addition to the RMS or ARS reduction method, for defining a profile-based smoothness Index. RARV and RARS con Distal filters for pavement profiles sometimes attempt to simulate the response of mechanical filters or isolate the profile wavelengths that are well correlated with some factor, such as user response.
|
From page 174... ...
Computation of the I:RI from filtering and weighting the measured pavement profile attenuates and amplifies the profile wavelengths, as shown in figure 95. This model, used in ASTM E Il70-87, provides a band pass filter at frequencies of approximately 0.014 to 0.177 cycles/ft (0.046 to 0.558 cycles/m)
|
From page 175... ...
~ other locations the IR} decreased by 31 to 46 percent from summer to winter. Analysis of these changes led Michigan DOT researchers to conclude Cat composite pavements subject to frost action are rougher in winter than in summer, and composite pavements not subject to frost action are rougher in summer Man In winter (Novak and DeFrain 1992~.
|
From page 176... ...
Individual wheelpatn roughness tends to be canceled by the profile averaging process, so that for very short wavelengths, the roughness amplitude of He averaged profile is 70.7 percent of the amplitude of a single wheel track profile. For very long wavelengths, the averaging process does not vary between single and averaged wheel Profiles.
|
From page 177... ...
When Sayers reevaluated the data using IRl model, the PCC correlation increased to 0.61 and the AC correlation increased slightly to 0.90. The large increase in PCC correlation resulted from the HRI's attenuation of short wavelength surface deviations, more generally associated with PCC pavements.
|
From page 178... ...
1 10 A recent investigation of the measurement of Me ride quality of flexible pavements, prepared for the Soup Dakota DOT, introduced a smoothness measuring index based on computer-modeled straightedge analysis of Soup Dakota profiler surface maces measured at I-ft (0.305-m) intervals.
|
From page 179... ...
data spacing precludes accurately measuring surface profile wavelengths less than 2 it (0.61 m) , TRS does not respond to this small portion of Me short surface wavelengths believed to affect rider comfort.
|
From page 180... ...
, and 25 to 50 It (7.6 m to 15.2 m) with relative weightings of 9, 6, and 3, respectively, hereby placing more emphasis on the shorter wavelengths that correlated more win user response in their field evaluations (see figure 98)
|
From page 181... ...
Profile measuring equipment capable of accurately measuring surface wavelengths from 2 to 50 ft (0.6 to 15.2 m) is required for this index.
|
From page 182... ...
,, (15) The Ptmoff has not reportedly been used by any highway agency for initial pavement smoothness control, but it is expected to be used by one highway department in 1997.
|
From page 183... ...
Table 44. Correlation of Janoff and Spangler/Kelly ride numbers with user response.
|
From page 184... ...
The effect of this equation is to compute the root-mean square of We spatial acceleration (ft/ft2 or m/m2) of the measured pavement profile (P)
|
From page 185... ...
(Sayers and Gillespie 1986~. QIr is not known to have been used as a summary statistic for Initial smoothness specifications.
|
From page 186... ...
Identification of these statistics then provides a basis for defining the equipment characteristics and properties needed to measure initial pavement smoothness. Review of Kev Selection Factors To select a smoothness index from the large number of available statistics described previously, it is necessary to first define the factors necessary for a good initial pavement smoothness statistic.
|
From page 187... ...
No information on correlation with user response was available for the SV and MO statistics, and estimates of the correlation were used for rating these pavement indices. Correlation With Other Smoothness Statistics SHAs currently use a variety of indices for defining initial pavement smoothness.
|
From page 189... ...
an an v) uO be o a; an cq o · o i Sl~:,~; ~ r: ran ~ ~ L - ~ ~ 1'~1 8-~8 , , 1 , 1 1 1 1~ ~ 1'¢ 1 z 1'~ 1~1 ~ ~zilotlz-1~1~1~11 189
|
From page 190... ...
The IRI is currently used as an index of initial pavement smoothness In Minnesota. For PCC surfaces, the reported correlation of PI with PST was moderate, and correlation for AC pavements is estimated to be slightly beUer, because Me PI responds to a long-wavelength range similar to the Ad.
|
From page 191... ...
Of these three, Me RQl is the only one currently used in smoothness specifications for new PCC and AC pavements. Indices Cat are currently used for initial pavement smoothness control, the PI and the MRN, fell short in correlation with user response.
|
From page 192... ...
This will provide measured pavement profiles that can be used to provide sufficient accuracy and breadth for use win all promising smoothness statistics. Profile Measuring Equipment Requirements Based on Selected Indices The primary Indices reconnnnended for potential use in initial s m oothness elm; ~ ~a +; ~ ~ PIT MAT Teal O~_l`~Ct~l=~, ~ A, 6~Janoff, $~saye=, RQI, and I~J have inherent characteristics that help to define He requirements of He equipment necessary to measure the pavement profiles needed for index computation.
|
From page 193... ...
Measured Profile Wavelengths Because each primary index responds to slightly different pavement surface profile wavelengths, it is critical Cat equipment for measuring construction pavement smoothness be capable of accurately measuring surface profile ~ these wavelengths. These critical wavelengths reported In the literature are listed In table 48 for the primary indices Janoff 1986; Paterson 1987a; Scofield 1992; Darlington 1995; Sayers and Karamihas 1996~.
|
From page 194... ...
or less, it is recommended that the sampling interval for the initial smoothness control equipment specification be I.0 in (25.4 mm) , unless analog antialiasing filters are used.
|
From page 195... ...
. Because Me static accuracy of a profile measuring device is not as critical as the dynamic accuracy, it is recommended that the static accuracy required In the crucial smoothness control equipment specification be better than the required dynamic accuracy)
|
From page 196... ...
690DNC systems were also tested at this site, and me average standard deviations for those systems were about the same as the North Central Region ProfiIometer@. The average absolute bias for these systems was larger, as shown In table 49.
|
From page 197... ...
Other Considerations Several other attributes of the Initial smoothness measuring equipment that may not be specifically recommended In the specification, but that will nonetheless be useful, include minimal weight, higher speeds, ease of use, must-grind location capability, horizontal location accuracy control, rapid availability of survey data results, and automated calibration, data collection, filtering, and reporting. An understanding by the construction contractor of the newly constructed pavement smoothness level widen days or hours instead of widen weeks can improve road surface quality and result in savings to both the contractor and the owner agency.
|
From page 198... ...
Many of the current smoothness-measuring practices are deeply Ingrained within Me highway agencies and paving industries that make it Impractical to quickly move toward the adoption of new equipment and smoothness indices. Highway agencies and paving contractors have invested significant resources In the current systems, which in many cases are providing satisfactory results.
|
From page 199... ...
if analog antialiasing used Sampling Interval l . Distance Accuracy Vertical Elevation Accuracy Static Precision/Bias Dynamic Precision/Bias Over Considerations 0.1 percent +0.005 in/+0.005 in (0.125 mm/0.125 mm)
|
From page 200... ...
Identification of these statistics then provides a basis for defining the equipment characteristics and properties needed to measure initial pavement smoothness. Review of Kev Selection Factors To select a smoothness index from the large number of available statistics described previously, it is necessary to first define the factors necessary for a good initial pavement smoothness statistic.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.