Click for next page ( 36


The National Academies | 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 35
35 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS The density that can be obtained under normal rolling con- of the sample, an error occurs. The allowable absorption ditions is clearly related to the t/NMAS. For improved com- level to use the displacement test method is specified as pactibility, it is recommended that the t/NMAS be at least 3 2 percent in AASHTO T166, but this level of absorption for fine-graded mixes and at least 4 for coarse-graded mixes. can create accuracy problems, as shown in this report. It The data for SMA indicate that the ratio should also be at least is recommended that the absorption limit for the displace- 4. Ratios less than these suggested numbers could be used, but ment test method be reduced to 1 percent. If the vacuum- more compactive effort would generally be required to obtain seal method is adopted on a project, the measured voids may the desired density. In most cases, a t/NMAS of 5 does not now be somewhat higher than with the water displacement result in the need for more compactive effort to obtain maxi- method. mum density. However, care must be exercised when the The water displacement method was accurate for all water thickness gets too large to ensure that adequate density is absorption levels encountered for mixes that were fine-graded obtained. (ARZ gradations). For mixes having gradations near the max- The results of the evaluation of the effect of mix tempera- imum density line (TRZ) or coarser (BRZ and SMA), the ture on the relationship between density and t/NMAS indi- level of water absorption at which AASHTO T166 began to cate that one of the reasons for low density at thinner sections lose accuracy was between 0.2 and 0.4 percent. (lower t/NMAS) is the more rapid cooling of the mixture. For mix design samples and other laboratory samples that Hence, for thinner layers it is even more important that rollers are compacted to relatively low voids, the displacement method stay very close to the paver so that rolling can be accomplished will provide reasonably accurate answers. However, for field prior to excessive cooling. For the conditions of this study, the samples where the void levels will typically be 6 percent or mixes placed at the NCAT test track at 25-mm thickness higher, it is important to evaluate absorption to determine if the cooled twice as fast as mixes placed at 37.5-mm thickness. For vacuum-seal method needs to be used. thicker sections (larger t/NMAS), the rate of cooling is typi- Care must be used when using the vacuum sealing method cally not a problem. to measure density. Many times the plastic bag develops a leak The in-place void content is the most significant factor during the test, leading to an error in the result. Weighing the impacting permeability of HMA mixtures. This is followed sample in air after measuring the submerged weight will indi- by coarse aggregate ratio and VMA. As the values of coarse cate if a leak has developed. If a leak is identified, the test must aggregate ratio increases, permeability increases. Permeability be repeated until an acceptable test is achieved. decreases as VMA increases for constant air voids. There appears to be a need for a correction factor for the The variability of permeability between various mixtures vacuum-sealing and water displacement methods to provide is very high. Some mixtures are permeable at the 8 to 10 per- equal measured air void contents even when the air void level cent void range and others do not seem to be permeable at is low. The correction factor for the mixtures evaluated in these higher voids. However, to ensure that permeability is this report was approximately 0.2 percent air voids. A better not a problem, the in-place air voids should be between 6 and determination of the correction factor can be made for specific 7 percent or lower. This appears to be true for a wide range dense graded mixes by compacting samples in the Superpave of mixtures regardless of NMAS and grading. gyratory compactor to approximately 4 percent air voids When laboratory prepared samples having low levels of (design air void content) and testing using the two test methods. water absorption were evaluated, the dimensional method The difference between these two tests will be the correction resulted in the highest air void contents followed by the gamma factor for the mix. ray method. The vacuum-sealing and water displacement The in-place air voids of the 20 field projects were high. (AASHTO T166) methods resulted in similar air void con- Fourteen of the 20 mixes tested had average in-place air voids tents when the water absorption level was low. The vacuum above 8 percent and seven of the mixes had average air voids seal method is an acceptable method to use for low and high over 10 percent (based on test results with the vacuum-seal void levels. method). This low density on a high percentage of random At low levels of water absorption, the water displace- projects is disturbing because this lower density will most ment method is an accurate measure for bulk specific grav- certainly lead to significant loss in pavement life. ity. The error develops when removing the sample from More emphasis must be placed on obtaining adequate water to determine the SSD weight. When water flows out density. Regardless of the method of density measurement

OCR for page 35
36 used, some cores have to be taken and tested for calibration. Even though there is a lot of scatter within and between The most reliable way to measure density is to take cores projects, most field results support the finding that higher for density testing. If the amount of absorption during den- t/NMAS ratios generally provide lower void levels. Coarse- sity measurement exceeds 1 percent, the T166 method will graded mixtures generally have higher permeability values likely provide a higher measured density than the true den- than the fine-graded mixtures for a given air void level. Air sity. The vacuum seal method is one approach to measure a voids were clearly shown to be a key determinant of perme- density more accurately when the water absorption exceeds ability. However, many times the air voids were reasonably 1 percent. low (5 to 7 percent) and the permeability was still high.