Click for next page ( 80

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 79
79 Fourier transform requirements in MatLab, texture profile had an R2 of 0.58 and indicated PW and texture direction as data from a larger interval (19 ft [5.8 m]) than the 6-ft (1.8-m) major factors, along with TR inversely related to SI. Both CT Meter test interval was used to generate the PSD curves. analyses indicated that pavementtire noise is affected by var- With a much smaller texture profile data population than the ious texture properties, some of which are embodied by tex- full-length profiles, the resulting spectral curves were more ture PSD function. The partially discrete SI-texture model variable and not as well defined. was likely weakened by the fact that the texture PSD param- Application of the partially discrete SI-texture model to eters were derived from two-dimensional texture profiles, the fully discrete data set did not produce good results. As instead of a three-dimensional profile from which SI noise shown in Figure 5-37, greater scatter and an over-prediction was measured. Other variables, such as pavement porosity of about 1.5 dB(A) occurred as a result of differences in the and stiffness and tire tread and inflation pressure, are consid- A1/A2 PSD parameter. On average, A1/A2 values computed ered to be factors in noise generation. They were not exam- from the discrete PSD curves (i.e., the 19-ft [5.8-m] profile ined in this study. intervals) were about 0.15 less than those computed from the full-length PSD curves; peak spectrums for the discrete PSD curves occurred at generally higher wavelengths than those Texture Construction Analyses for the full-length PSD curves. Nominal Versus Actual Texture To complete the second statistical analysis, the fully discrete data set was run through SAS to determine if the partially dis- An important consideration regarding PCC surface tex- crete SI-texture model could be improved or if a better model tures designed to satisfy friction, noise, and other require- could be developed. The SAS Proc CORR analysis identified ments is the assurance that the specified texture is actually A1/A2, L4/L63, PW, and TR as statistically significant. constructed. Often, various factors pertaining to the PCC While not strongly correlated, the correlation coefficients mixture (weather conditions; texturing equipment, and for the three PSD parameters (A1/A2, L4/L63, and PW) gener- operator) are encountered during construction that prevent ally indicated that reducing higher wavelength texture and a contractor from achieving the desired specified texture. In increasing lower wavelength texture (and subsequently shift- addition, the level of QC/QA of the process could influence ing the peak spectrum to a lower wavelength) results in reduced the produced texture. noise. In the case of TR, the correlation coefficient was not As part of the construction demonstration of test sections strong and contradicted the expectation that a lower TR will for this study on the Illinois Tollway (I-355 South Extension), result in lower noise. This contradiction is primarily attrib- measurements were made of the depths of tine grooves at dif- uted to the low noise and high TR values exhibited by the ferent times following construction and prior to traffic open- diamond-ground and heavy turf drag test sections. ing. A tire-tread depth gauge was used to measure the groove Modeling with these and other variables using the SAS depths near the right wheelpath of the outside lane every 30 ft proc REG procedure did not produce any particularly strong (9.2 m) or so, both immediately behind the texturing machine and meaningful SI-texture models. The best model obtained (while the PCC was still fresh) and a few weeks after the PCC 104 103 102 Predicted SI, dBA 101 100 99 98 Predicted SI based on Partially Discrete SI-Texture Model and 97 Fully-Discrete Texture Data 1-to-1 line 96 96 97 98 99 100 101 102 103 104 Actual SI, dBA Figure 5-37. Actual versus predicted SI, based on partially discrete SI-texture model applied to discrete location texture PSD parameter data.

OCR for page 79
80 Sect 2 (Long Tine, Std Depth) Sect 5a (Long Tine, Std Depth) Sect 6 (Long Tine, Shallow Depth) Sect 9 (Tran Tine, Std Depth) Sect 10 (Tran Tine, Std Depth) Sect 12 (Tran Skewed Tine, Std Depth) 3.5 3 Avg Groove Depth, mm 2.5 2 1.5 Design Depth Behind Tining Machine Weeks after Const Months after Const (Depth Gauge) (Depth Gauge) (CT Meter) Time of Measurement Figure 5-38. Tine depth measurements for the newly constructed longitudinal and transverse tine test sections. had hardened. A third set of tine depth measurements was finishing operations and the downward pressure applied to obtained from the CT Meter testing performed weeks later. the tining unit). Figure 5-38 shows the average values derived from these The considerable drop in average values (on the order of measurements. The first set of points represent the design/ 0.02 in. [0.5 mm]) between the two sets of measurements is specified tining depth. The measurements taken behind the believed to be due to the imprecision of measuring the depth tining machine were fairly close to the specified depths, but of grooves. The grooves are variable and not well defined there was a marked reduction in the depth values obtained a (both at the bottom and at the top, where mortar deposits lie) few weeks later. CT Meter tine depth readings are of the same and measuring them in a medium that is soft and malleable general magnitude, but are less consistent across textures. (i.e., plastic concrete) and the wearing away of the mortar Regarding the two sets of depth gauge measurements, it deposits that occurs as a result of construction traffic and any was noted that, despite averages being close to the specified post-construction power brooming or cleaning operations values, a significant amount of variation in the readings was affects the measured values. experienced for all textures, due in part to the weather condi- The progression of texture changes illustrates the impor- tions during construction that made it difficult for the tining tance of evaluating/testing the texture dimensions to ensure machine operator to continually adjust for the conditions compliance with specifications and also the need for accurate (both in terms of the distance maintained behind paving and and timely measurement methods/equipment.