National Academies Press: OpenBook

Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates (2014)

Chapter: Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates

Page 1
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 1
Page 2
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 2
Page 3
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 3
Page 4
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 4
Page 5
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 5
Page 6
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 6
Page 7
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 7
Page 8
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 8
Page 9
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 9
Page 10
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 10
Page 11
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 11
Page 12
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 12
Page 13
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 13
Page 14
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 14
Page 15
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 15
Page 16
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 16
Page 17
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 17
Page 18
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 18
Page 19
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 19
Page 20
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 20
Page 21
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 21
Page 22
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 22
Page 23
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 23
Page 24
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 24
Page 25
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 25
Page 26
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 26
Page 27
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 27
Page 28
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 28
Page 29
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 29
Page 30
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 30
Page 31
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 31
Page 32
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 32
Page 33
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 33
Page 34
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 34
Page 35
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 35
Page 36
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 36
Page 37
Suggested Citation:"Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates ." National Academies of Sciences, Engineering, and Medicine. 2014. Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates. Washington, DC: The National Academies Press. doi: 10.17226/22374.
×
Page 37

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.

NATIONAL COOPERATIVE HIGHWAY RESEARCH PROGRAM Responsible Senior Program Officer: Edward T. Harrigan May 2014 C O N T E N T S Chapter 1—Introduction and Research Approach, 1 1.1 Background, 1 1.2 Problem Statement, 2 1.3 Research Objectives, 2 1.4 Scope of Study, 2 Chapter 2—Design and Conduct of the Study, 3 2.1 Testing Programs, 3 2.2 Test Methods, 3 2.3 Test Data, 3 2.4 Round Robin Study Instructions, 6 Chapter 3—Test Results and Analysis, 6 3.1 Precision Estimates of AASHTO T 96, T 304, and T 11, 6 3.1.1 AASHTO T 96, 6 3.1.2 AASHTO T 304, 8 3.1.3 AASHTO T 11, 10 3.2 Evaluation of the Effect of Sample Size on AASHTO T 11 Test Results, 13 3.2.1 Results of the Analysis, 13 Chapter 4—Evaluation of the Method of Washing, 14 4.1 Evaluation of Sieve Analysis Results, 14 4.1.1 Analysis of AGC Data, 15 4.1.2 Analysis of AGF Data, 17 4.1.3 Analysis of HMAIO Data, 20 4.1.4 Analysis of HMASE Data, 22 4.2 Evaluation of Degradation from Mechanical Washing, 24 4.3 Effect of Mechanical Washing Duration on Degradation, 29 Chapter 5—Conclusions and Proposed Changes to Standard Test Methods, 31 5.1 Summary and Conclusions, 31 5.1.1 Precision Estimates of AASHTO T 96, 31 5.1.2 Precision Estimates of AASHTO T 304, 31 5.1.3 Precision Estimates of AASHTO T 11, 31 5.1.4 Evaluation of the Effect of Sample Size on T 11 Test Results, 32 5.1.5 Comparison of the Method of Washing, 32 5.1.6 Evaluation of Degradation from Mechanical Washing, 32 5.1.7 Effect of Duration of Mechanical Washing on Degradation, 32 5.2 Proposed Changes to AASHTO Standard Test Methods T 96, T 304, and T 11, 32 References, 33 Unpublished Appendixes, 34 Appendix E, 34 PRECISION ESTIMATES Of AASHTO T 304, AASHTO T 96, AND AASHTO T 11 AND INvESTIgATION Of THE EffECT Of MANUAl AND MECHANICAl METHODS Of WASHINg ON SIEvE ANAlySIS Of AggREgATES This digest summarizes key findings of research conducted in Task Order #2 of NCHRP Project 10-87, “Precision Statements for AASHTO Standard Methods of Test,” by the AASHTO Materials Reference Laboratory under the direction of the principal investigator, Dr. Haleh Azari. Research Results Digest 389 CHAPTER 1—INTRODUCTION AND RESEARCH APPROACH 1.1 Background Under NCHRP Project 10-87, the AASHTO Materials Reference Labora- tory (AMRL) is conducting a multi-phase research project to improve estimates of precision in AASHTO test methods for a wide range of construction materials. AMRL has an extensive database of test results for the broad range of construc- tion materials collected through its Pro- ficiency Sample Program (PSP) that are used for developing precision estimates (1). Laboratories participating in the AMRL PSP receive annual or biannual shipments of paired proficiency samples, which are tested according to specified AASHTO test methods. The results of the testing are returned to AMRL for analysis, summari- zation, and reporting back to the labora- tories. The number of participants in the AMRL PSP program is sufficiently large enough to ensure a statistically sound basis for determination of estimates of precision for standard test methods. The technique developed by AMRL in NCHRP Project 9-26 is used for analyzing proficiency sample data (2). This four-step statistical method removes out lying results and analyzes the core data of a paired data set. The results of the analysis can then be used to obtain reliable single-operator and multilaboratory estimates of precision. This report includes the results from Task Order #2 of NCHRP Project 10-87 where PSP data from four different PSP testing programs of fine aggregate (AGF) and coarse aggregate (AGC) were used to update precision estimates for AASHTO Standard Test Methods T 96, Resistance to Degradation of Small-Size Coarse Ag- gregate by Abrasion and Impact in the Los Angeles Testing Machine; T 304, Un- compacted Void Content of Fine Aggre- gate; and T 11, Materials Finer Than 75-µm (No. 200) Sieve in Mineral Aggregates by Washing (3–5). In the 1996 revisions of AASHTO T 11, the required mass of fine aggregate was changed from 500-g to 300-g. The effect of this change on the variability of the results has not been investigated. As part of updating the precision estimates for AASHTO T 11, the effect of minimum sample size on the results of the test was also investigated.

2Moreover, AASHTO T 11 allows use of me- chanical apparatus for the washing operation, pro- vided the results are consistent with those obtained from the manual washing method. The consistency of the results of the two washing methods has not been evaluated. In this study the effect of manual and mechanical methods of washing on the sieve analysis of fine and coarse aggregates was also inves- tigated. For a comprehensive evaluation of the effect of washing method on sieve analysis results, all PSP data pertinent to the manual and mechanical wash- ing methods were analyzed. These are the aggregate gradation data from the Hot Mix Asphalt Extraction Ignition Oven (HMAIO) and Hot Mix Asphalt Sol- vent Extraction (HMASE), as well as the data from AGF and AGC testing programs. The aggregate gra- dation data from HMAIO and HMASE have been collected according to AASHTO T 30, Mechanical Analysis of Extracted Aggregate, and the aggregate gradation data from AGF and AGC have been col- lected according to AASHTO T 27, Sieve Analysis of Fine and Coarse Aggregates (6–7). 1.2 Problem Statement AASHTO Standard Test Methods applicable to highway materials require periodic studies to update estimates of precision to account for improvements in the test methods or inclusion of a wider range of materials. For AASHTO T 96, the collection of new data from testing a wide range of materials requires that the precision estimates of the test method be updated. For AASHTO T 304, the current preci- sion statement pertains to void contents determined on “graded standard sand” as described in ASTM C 778 (8), which is considered rounded; is graded from 600 µm (No. 30) to 150 µm (No. 100); and may not be typical of other fine aggregates. Additional precision data are needed for void content deter- mination of fine aggregates having different levels of angularity and texture. For AASHTO T 11, the results of the recent modification to the test method on the requirement of minimum sample size of 300-g have not yet compared with the results of the test based on the previous requirement of minimum sample size of 500-g. Moreover, the use of mechani- cal washing has been allowed as an alternative to manual washing for sieve analysis of aggregates in AASHTO T 11; however, the consistency of the results from mechanical washing with those from manual washing has not been evaluated. The use of mechanical washing has been also allowed in AASHTO T 27 and T 30. Therefore, for a compre- hensive comparison of the results from mechanical and manual washing, the PSP sieve analysis data collected according to AASHTO T 27 and T 30 will be included in the analysis. 1.3 Research Objectives The first objective of this work was to update precision estimates of AASHTO T 96, T 304, and T 11. The second objective was to examine the sig- nificance of the difference between variability of per- cent passing No. 200 sieve of 300-g and 500-g fine aggregate samples measured according to AASHTO T 11. The third objective was to evaluate the effect of manual versus mechanical washing, by comparing the results of sieve analysis of PSP samples, washed manually or mechanically prior to being tested according to AASHTO T 11, T 27, or T 30. 1.4 Scope of Study The scope of the work involved the following major activities: 1. Update precision estimates of AASHTO T 96, T 304, and T 11. a. Organize the most recent sets of PSP data collected according to each test method. b. Analyze the data to determine single- operator and multilaboratory estimates of precision. 2. Examine the significance of the difference between the variability of AASHTO T 11 sieve analysis results obtained from 300-g and 500-g sample sizes. a. Identify and organize PSP sieve analysis data from 300-g samples and those from 500-g samples of fine aggregates. b. Statistically compare the average and pooled variability of the results from test- ing the 300-g and 500-g samples. 3. Evaluate the effect of manual and mechanical washing on the sieve analysis of PSP aggre- gates washed prior to being tested according to AASHTO T 11, T 27, or T 30. a. Identify and organize the PSP gradation data resulting from manual and mechani- cal washing. b. Perform separate analyses of the preci- sion of aggregate gradation data resulting from manual and mechanical washing.

3c. Statistically compare the average and standard deviation of the percent passing various sieve sizes resulting from manual and mechanical washing. 4. Develop conclusions about the precision esti- mates prepared in this study, the suitability of a 300-g minimum sample size for AASHTO T 11, and the appropriateness of mechanical washing method for AASHTO T 11, T 27, and T 30 test methods. 5. Prepare proposed precision statements for AASHTO T 96, T 304, and T 11. CHAPTER 2—DESIgN AND CONDUCT Of THE STUDy This chapter provides information on the design and conduct of various elements of the study. 2.1 Testing Programs The data used for the evaluations in this study were collected from laboratories participating in four different PSP testing programs: AGF, AGC, HMAIO, and HMASE. 2.2 Test Methods The data analyzed for this study are the results of testing aggregates according to five different AASHTO test methods. These are T 96, Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Test- ing Machine; T 304, Uncompacted Void Content of Fine Aggregate; T 11, Materials Finer than 75-µm (No. 200) Sieve in Mineral Aggregates by Wash- ing; T 27, Sieve Analysis of Fine and Coarse Aggre- gates; and T 30, Mechanical Analysis of Extracted Aggregate. 2.3 Test Data For the analyses in this study, only the most re- cent proficiency sample data (after 1998) were used to account for changes in precision estimates result- ing from recent changes in the test methods or from testing a wider range of materials. Another reason for using the data sets collected after 1998 is that the numbers of participating laboratories are consider- ably larger in comparison to before 1998. Therefore, pooled results from the more recent samples would provide more reliable estimates of precision. One-hundred and fifty-five sets of data, col- lected from the laboratories participating in AGC, AGF, HMAIO, and HMASE testing programs of PSP, were used for various evaluations in this study. Table 2-1 through Table 2-5 provide information on the data sets, including the sample pair identifica- tion numbers, the testing program, the AASHTO Test Method PSP Tesng Program PSP Sample No. AASHTO T 96 Coarse Aggregate (AGC) 177 178 (Nov. 2012) 173 174 (Nov. 2011) 169 170 (Nov.2010) 165 166 (Nov.2009) 161 162 (Nov.2008) 157 158 (Nov.2007) 153 154 (Nov.2006) 149 150 (Nov.2005) 145 146 (Nov.2004) 141 142 (Nov.2003) 137 138 (Nov.2002) 133 134 (Jan.2002) 129 130 (Dec. 2000) 125 126 (Jan. 2000) 121 122 (Dec. 1998) Table 2-1 PSP data sets used for determining precision estimates of AASHTO T 96.

Test Method PSP Tesng Program PSP Sample No. AASHTO T 304 Fine Aggregate (AGF) 175 176 (March 2012) 171 172 (March 2011) 167 168 (March 2010) 163 164 (March 2009) 159 160 (March 2008) 155 156 (March 2007) 151 152 (April 2006) 147 148 (May 2005) 143 144 (May 2004) 139 140 (May 2003) 135 136 (May 2002) 131 132 (May 2001) 127 128 (June 2000) 123 124 (June 1999) Table 2-2 PSP data sets used for precision estimates of AASHTO T 304. Test Method PSP Tesng Program PSP Sample No. AASHTO T 11 AGC 177 178 (Nov. 2012) 173 174 (Nov. 2011) 169 170 (Nov.2010) 165 166 (Nov.2009) 161 162 (Nov.2008) 157 158 (Nov.2007) 153 154 (Nov.2006) 149 150 (Nov.2005) 145 146 (Nov.2004) 141 142 (Nov.2003) 137 138 (Nov.2002) 133 134 (Jan.2002) 129 130 (Dec. 2000) 121 122 (Dec. 1998) AGF 175 176 (March 2012) 171 172 (March 2011) 167 168 (March 2010) 163 164 (March 2009) 159 160 (March 2008) 155 156 (March 2007) 151 152 (April 2006) 147 148 (May 2005) 143 144 (May 2004) 139 140 (May 2003) 135 136 (May 2002) 131 132 (May 2001) 127 128 (June 2000) 123 124 (June 1999) Table 2-3 PSP data sets of fine aggregate (AGF) and coarse aggregate (AGC) of PSP data used for determining precision estimates of AASHTO T 11.

5Test Method PSP Aggregate Type Sample Size PSP Sample No. AASHTO T 11 Materials Finer Than 75 µm (No. 200) Sieve in Mineral Aggregates by Washing Fine Aggregate 300 g 159 160 (March 2008) 155 156 (March 2007) 151 152 (April 2006) 147 148 (May 2005) 500 g 175 176 (March 2012) 171 172 (March 2011) 167 168 (March 2010) 163 164 (March 2009) 143 144 (May 2004) 139 140 (May 2003) 135 136 (May 2002) 131 132 (May 2001) 127 128 (June 2000) 123 124 (June 1999) 119 120 (May 1998) Table 2-4 PSP data sets used for comparison of percent passing 75-µm (No. 200) sieve by washing of 300-g and 500-g fine aggregate sample sizes following AASHTO T 11. Test Method and Descripon Aggregate Type Sample No. AASHTO T 11: Materials Finer Than 75 µm (No. 200) Sieve in Mineral Aggregates by Washing and AASHTO T 27: Sieve Analysis of Fine and Coarse Aggregates Fine Aggregate (AGF) 179 180 (March 2013) 175 176 (March 2012) 171 172 (March 2011) Coarse Aggregate (AGC) 177 178 (Nov. 2012) 173 174 (Nov. 2011) 169 170 (Nov. 2010) T 30: Mechanical Analysis of Extracted Aggregate Hot Mix Asphalt Ignion Oven (HMAIO) 25 26 (Feb. 2013) 23 24 (Feb. 2012) 21 22 (Feb. 2011) 19 20 (Feb. 2010) Hot Mix Asphalt Solvent Extracon (HMASE) 77 78 (Jan. 2013) 75 76 (Jan. 2012) 73 74 (Jan. 2011) Table 2-5 PSP data used for comparison of sieve analysis results from manual and mechanical washing.

6test methods according to which the materials were tested, and the dates the data were collected. Table 2-1 shows the sample round numbers of the AGC testing program used for preparing pre- cision estimates of AASHTO T 96. The data col- lected include percent loss of aggregate using the Los Angeles Testing Machine. Fifteen sets of data were available for the precision estimate evaluation of AASHTO T 96. Table 2-2 shows the sample round numbers of the AGF testing program used for preparing preci- sion estimates of AASHTO T 304. The data col- lected include the uncompacted void content of fine aggregate. Fourteen sets of data were available for the precision estimate evaluation of AASHTO T 304. Table 2-3 shows the sample round numbers of the AGF and AGC testing programs used for pre- paring precision estimates of AASHTO T 11. The data collected according to T 11 include the percent of the material finer than 75-µm (No. 200) sieve by washing. Fourteen sets of data each were available from AGF and AGC for the precision estimate eval- uation of AASHTO T 11. Table 2-4 shows the sample round numbers of the AGF testing program used for evaluating the dif- ference between the precision estimates of results from testing 300-g and 500-g samples according to AASHTO T 11. The data collected include percent of materials finer than 75-µm (No. 200) sieve by washing. Four sets of data were available from test- ing 300-g samples and 11 sets of data from testing 500-g samples. Table 2-5 provides the PSP sample round num- bers of AGF, AGC, HMAIO, and HMASE used for evaluation of the effect of method of washing (mechanical versus manual) on the sieve analy- sis results of AASHTO T 11, T 27, and T 30. The information on the method of washing has been col- lected from the PSP participating laboratories since 2010. There were three sets of data each from test- ing AGC and AGF samples according to both T 27 and T 11 test methods. There were four sets of data from HMAIO and three sets of data from HMASE samples tested according to T 30. 2.4 Round Robin Study Instructions Sample instructions and data sheets sent to the laboratories for each of the sample programs are pro- vided in Appendix A, which is not published herein but is available on the TRB website (http://www. trb.org) by searching for NCHRP Project 10-87. The question regarding the method of washing (manual or mechanical) from the participating laboratories have been included in the data sheets of the AGC, AGF, HMAIO, and HMASE testing programs since 2010. CHAPTER 3—TEST RESUlTS AND ANAlySIS This chapter includes statistical summaries of the data used in this study. The resulting precision estimates for AASHTO T 96, T 304, and T 11 and the outcome of evaluating the effect of minimum sample size on AASHTO T 11 test results are also presented in this chapter. An individual graph for each of the 155 proficiency sample pairs analyzed in this study can be found in Appendixes B through D, which are not published herein but can be found online by searching for NCHRP Project 10-87 on the TRB website (http://www.trb.org). 3.1 Precision Estimates of AASHTO T 96, T 304, and T 11 Using the most recent PSP data sets from testing the AGF and AGC samples (see Tables 2-1 through 2-3), the precision estimates of AASHTO T 96, T 304, and T 11 were updated. For each of the test meth- ods, a summary of statistics of individual data sets as well as the pooled statistics used for the update of repeatability and reproducibility estimates are pro- vided in the following sections. Proposed precision statements that include the precision estimates devel- oped in this chapter are provided in Appendix E. 3.1.1 AASHTO T 96 AASHTO T 96 is identical to ASTM C 131 (9) except for several provisions described in AASHTO T 96. The test method covers a procedure for test- ing coarse aggregates smaller than 37.5 mm (1½ in.) for resistance to degradation using the Los Angeles Testing Machine. A summary of statistics of percent loss of the 15 most recent pairs of PSP coarse aggre- gate samples tested according to AASHTO T 96 is provided in Table 3-1. The plots of the individual data sets are found in Appendix B. To decide whether to base the precision esti- mates on the standard deviation or the coefficient of variation, the relationship between the averages and the variability parameters was examined. Figure 3-1 shows the plots of the averages versus standard de- viations and averages versus coefficients of variation.

7PSP Sample No. No. of Labs Average Results Repeatability Reproducibility Reproducibility X Y 1s X Samples CV% Y Samples CV% X Samples Y Samples 1s CV% 1s CV% 177 178 513 24.08 23.90 0.517 2.1 2.2 1.112 4.6 1.075 4.5 173 174 480 13.64 13.77 0.328 2.4 2.4 0.752 5.5 0.735 5.3 169 170 492 26.46 27.34 0.804 3.0 2.9 2.079 7.9 2.098 7.7 165 166 476 21.80 21.75 0.414 1.9 1.9 1.226 5.6 1.242 5.7 161 162 456 56.82 56.73 1.204 2.1 2.1 1.941 3.4 2.007 3.5 157 158 417 13.98 14.18 0.299 2.1 2.1 0.734 5.2 0.740 5.2 153 154 444 36.98 37.20 0.969 2.6 2.6 2.141 5.8 2.102 5.7 149 150 438 42.95 40.22 1.597 3.7 4.0 2.800 6.5 2.400 6.0 145 146 412 27.29 27.48 0.863 3.2 3.1 1.600 5.8 1.500 5.4 141 142 398 13.58 13.72 0.569 4.2 4.1 1.100 7.9 1.100 7.9 137 138 394 16.22 16.02 0.577 3.6 3.6 1.100 6.5 1.100 6.8 133 134 363 21.52 21.35 0.631 2.9 3.0 1.500 6.8 1.500 7.1 129 130 335 20.27 20.30 0.768 3.8 3.8 1.400 7.1 1.500 7.3 125 126 310 41.85 41.83 1.185 2.8 2.8 2.600 6.1 2.600 6.1 121 122 290 13.32 13.19 0.612 4.6 4.6 1.100 8.5 1.200 8.9 Table 3-1 Summary of statistics of percent loss of coarse aggregate in Los Angeles Abrasion Testing Machine. y = 0.0236x + 0.1415 R² = 0.7308 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0 10 20 30 40 50 60 Re pe at ab ili ty Average Repeatability Standard Deviation vs. Average (AASHTO T 96) y = 0.0183x + 3.4767 R² = 0.0823 0 1 2 3 4 5 0 10 20 30 40 50 60 Re pe at ab ili ty Average Repeatability Coefficient of Variation vs. Average (AASHTO T 96) y = 0.0398x + 0.5078 R² = 0.6685 0 0.5 1 1.5 2 2.5 3 0 10 20 30 40 50 60 Re pr od uc ib ili ty Average Reproducibility Standard Deviation vs. Average (AASHTO T 96) y = 0.0509x + 7.5385 R² = 0.2477 0 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 60 Re pr od uc ib ili ty Average Reproducibility Coefficient of Variation vs. Average (AASHTO T 96) figure 3-1 Relationship between average and standard deviation and between average and coefficient of variation of the AASHTO T 96 test results.

8As indicated from the R2 values in the plots, there is a strong relationship between averages and both repeatability and reproducibility standard deviations (R2~ 0.7), while the correlation between averages and both repeatability and reproducibility coefficients of variation is very small (R2~ 0.07 and 0.27, respec- tively). Therefore, according to ASTM C 670 (10), the form of the precision estimates should be based on the sample coefficient of variation. The repeat- ability and reproducibility estimates of precision were then computed by averaging the coefficient of variation of the individual samples in Table 3-1. The resulting precision estimates for AASHTO T 96 are provided in Table 3-2. The numbers in the parentheses show the existing precision estimates of ASTM C 131. 3.1.1.1 Comparison of the New and Existing Preci­ sion Estimates of AASHTO T 96. The new and exist- ing precision estimates were compared to examine whether or not the precision estimates have changed as a result of inclusion of a wider range of materials. The existing precision estimates are based on aver- age percent loss in a range of 10% to 45%, while the new set of precision estimates is based on aver- age percent loss in a range of 13% to 57%. The comparison of the new precision estimates with the existing precision estimates of ASTM C 131 (Table 3-2) indicates that the 1s% coefficient of vari- ation of 3% computed in this study is 50% higher than the 1s% single-operator coefficient of varia- tion of 2% in ASTM C 131. Similarly, the new 1s% multi laboratory coefficient of variation of 6.2% is 38% higher than the existing 1s% multilaboratory coefficient of variation of 4.5%. The increase in both single-operator and multilaboratory coefficients of variation might be due to inclusion of data from a greater variety of coarse aggregates with a wider range of degradation resistance than those used for developing the precision estimates of ASTM C 131. 3.1.2 AASHTO T 304 AASHTO T 304 describes the determination of the loose uncompacted void content of a sample of fine aggregate. A summary of statistics of uncom- pacted void content of the 14 most recent pairs of PSP fine aggregate samples tested according to T 304 test method is provided in Table 3-3. The plots of the individual data sets are found in Appendix C. The plot of the averages versus standard devia- tions and the averages versus coefficients of varia- tion of the uncompacted void content are shown in Figure 3-2. As indicated from the figure, there are no relationships between the averages and either the standard deviations or the coefficients of variation. Therefore, as specified in ASTM C 670, the form of the precision estimates should be based on the sample standard deviation. The repeatability and reproducibility standard deviations in Table 3-4 were computed by pooling the sample standard deviations shown in Table 3-3 using the following equation (11): s n s n s n s n n n kp k k k ( ) ( ) ( ) = − + − + + − + + + − 1 1 . . . 1 . . . (Equation 1) 1 1 2 2 2 2 2 1 2 Where: sp = pooled standard deviation sk = kth standard deviation nk = number of laboratories analyzed resulting in kth standard deviation 3.1.2.1 Comparison of the New and Existing Precision Estimates of T 304. The current preci- sion estimates for AASHTO T 304 are shown in parentheses in Table 3-4. The comparison of the new and existing precision estimates of AASHTO T 304 would indicate if the precision estimates of the test based on actual fine aggregates are different Statistics 1s (%) d2s (%) Repeatability 3.0 (2.0) 8.5 (5.7) Reproducibility 6.2 (4.5) 17.6 (12.7) Note: The numbers in parentheses represent the existing precision estimates of ASTM C 131. Table 3-2 Pooled repeatability and reproducibility precisions of T 96 based on the samples’ coefficients of variation.

9PSP Sample No. No. of Labs Average Results Repeatability Reproducibility Reproducibility X Y 1s X Samples CV% Y Samples CV% X Samples Y Samples 1s CV% 1s CV% 175 176 535 44.376 44.433 0.268 0.6 0.6 0.570 1.3 0.570 1.3 171 172 484 42.657 42.609 0.266 0.6 0.6 0.504 1.2 0.524 1.2 167 168 468 43.321 43.018 0.342 0.8 0.8 0.586 1.4 0.568 1.3 163 164 466 43.733 43.842 0.295 0.7 0.7 0.760 1.7 0.763 1.7 159 160 443 41.505 41.555 0.272 0.7 0.7 0.871 2.1 0.863 2.1 155 156 396 43.104 43.125 0.262 0.6 0.6 0.573 1.3 0.560 1.3 151 152 410 43.055 42.940 0.348 0.8 0.8 0.720 1.7 0.670 1.6 147 148 387 42.623 42.631 0.343 0.8 0.8 0.770 1.8 0.780 1.8 143 144 367 43.053 43.085 0.403 0.9 0.9 0.790 1.8 0.790 1.8 139 140 345 43.268 43.393 0.432 1.0 1.0 1.100 2.6 1.200 2.7 135 136 287 42.695 42.749 0.383 0.9 0.9 1.100 2.6 1.100 2.6 131 132 242 43.267 43.183 0.378 0.9 0.9 0.920 2.1 0.860 2.0 127 128 211 42.663 42.682 0.353 0.8 0.8 1.300 3.0 1.300 3.2 123 124 183 42.794 42.806 0.417 1.0 1.0 1.100 2.6 1.100 2.7 Table 3-3 Summary of statistics of uncompacted void content of 14 sets of AGF sample pairs. y = 0.0015x + 0.4067 R² = 0.0003 0 0.1 0.2 0.3 0.4 0.5 41 41.5 42 42.5 43 43.5 44 44.5 45 Re pe at ab ili ty Average Repeatability Standard Deviation vs. Average (AASHTO T 304) y = 0.0251x + 1.8726 R² = 0.0137 0 0.2 0.4 0.6 0.8 1 1.2 41 41.5 42 42.5 43 43.5 44 44.5 45 Re pe at ab ili ty Average Repeatability Coefficient of Variation vs. Average (AASHTO T 304) y = 0.1195x + 5.9713 R² = 0.1007 0 0.2 0.4 0.6 0.8 1 1.2 1.4 41 41.5 42 42.5 43 43.5 44 44.5 45 Re pr od uc ib ili ty Average Reproducibility Standard Deviation vs. Average (AASHTO T 304) y = 0.3173x + 15.589 R² = 0.1274 0 0.5 1 1.5 2 2.5 3 3.5 41 41.5 42 42.5 43 43.5 44 44.5 45 Re pr od uc ib ili ty Average Reproducibility Coefficient of Variation vs. Average (AASHTO T 304) figure 3-2 Relationship between average and standard deviation and between average and coefficient of variation of the AASHTO T 304 data.

10 from the existing precision estimates that were developed based on the graded standard silica sand as described in ASTM C 778. For the new precision estimates, based on the PSP fine aggre- gates with uncompacted void content in a range of 42% to 45%, the 1s single-operator stan- dard deviation is 0.33% and the 1s multilabora- tory standard deviation is 0.81%. These values are significantly larger than the existing single- operator standard deviation of 0.13% and multi- laboratory standard deviation of 0.33%, respectively. The larger standard deviations of the PSP data in comparison to the existing standard deviations are expected since the basis of the existing preci- sion estimates is the uncompacted voids of the graded standard silica sand with round particles in a range of 600 µm (No. 30) to 150 µm (No. 100), as described in T 304, which may not be typical of other fine aggregates. 3.1.3 AASHTO T 11 AASHTO T 11 covers determination of the amount of materials finer than a 75-µm (No. 200) sieve by washing. Clay particles and other aggre- gate particles that are dispersed by the wash water, as well as water-soluble materials, will be removed from the aggregate during the test. A summary of statistics of percent finer than a 75-µm (No. 200) sieve of the 14 most recent pairs each of the AGF and AGC samples of PSP tested according to T 11 test method is provided in Table 3-5 and Table 3-6. The sample size of AGF samples used for this analysis is 500-g, which is the same as the sample size used Statistics 1s d2s Repeatability 0.33 (0.13) 0.95 (0.37) Reproducibility 0.81 (0.33) 2.29 (0.93) Note: The numbers in parentheses represent the existing precision estimates from ASTM C 670. Table 3-4 Pooled repeatability and reproducibility precisions of T 304 based on the samples’ standard deviations. PSP Sample No. No. of Labs Average Results Repeatability Reproducibility Reproducibility X Y 1s X Samples CV% Y Samples CV% X Samples Y Samples 1s CV% 1s CV% 177 178 1380 0.188 0.189 0.022 12 11.9 0.06 30.106 0.06 29.6 173 174 1371 0.263 0.268 0.041 15.5 15.1 0.12 44.2 0.12 43.8 169 170 1326 1.266 1.121 0.167 13.2 14.9 0.39 30.9 0.38 33.5 165 166 1223 0.298 0.299 0.028 9.5 9.5 0.07 24.9 0.07 24.9 161 162 1240 1.263 1.206 0.14 11.1 11.6 0.38 30.1 0.39 32.1 157 158 1128 0.298 0.278 0.037 12.5 13.4 0.14 45.2 0.13 47.0 153 154 1065 0.891 0.909 0.102 11.4 11.2 0.22 24.5 0.22 24.0 149 150 1111 0.538 0.492 0.091 17 18.6 0.21 39.5 0.2 41.4 145 146 1039 0.382 0.32 0.096 25 29.9 0.16 41.3 0.13 40.9 141 142 964 0.241 0.243 0.133 55.2 54.8 0.14 57.5 0.14 57.9 137 138 876 0.211 0.219 0.06 28.2 27.2 0.11 54 0.12 55.9 133 134 800 0.53 0.518 0.18 34 34.7 0.17 31.6 0.16 30.6 129 130 714 0.276 0.285 0.062 22.4 21.7 0.13 48.2 0.14 48.7 121 122 552 0.453 0.388 0.083 18.3 21.3 0.22 48.7 0.2 51.3 Table 3-5 Summary of statistics of percent finer than a 75-µm (No. 200) sieve in coarse aggregates by washing of 14 sets of AGC sample pairs.

11 for developing the existing precision estimates of AASHTO T 11. Figures 3-3 and 3-4 show the relationships between average percent passing sieve No. 200 and the repeatability/reproducibility standard devia- tions and the coefficients of variation. A review of the graphs indicates that the correlations between average and standard deviations are stronger than the correlations between average and coefficients of variation. However, to be consistent with the existing precision estimates, the precision estimates should be determined based on standard deviation. The repeatability and reproducibility estimates of precision for coarse and fine aggregates are computed by pooling the standard deviations of the individual samples in Table 3-5 and Table 3-6, respectively, using Equation 1. The computed precision estimate values for coarse and fine aggregates are provided in Table 3-7, with the existing precision estimates of AASHTO T 11 shown in parentheses. 3.1.3.1 Comparison of the New and Existing Pre­ cision Estimates of AASHTO T 11. The new and existing precision estimates for AASHTO T 11 in Table 3-7 were compared to determine if the new sets of precision estimates are the same as the existing ones. For the coarse aggregates, with the percent materials finer than a 75-µm (No. 200) sieve by washing in a range of 0.19 % to 1.23%, the new 1s single-operator and multilaboratory standard deviations are 0.10% and 0.21%, respec- tively. These values are compared with the existing single- operator and multi laboratory standard devia- tion of 0.10% and 0.22%, respectively, which were prepared from coarse aggregates having less than 1.5% finer than the 75-µm (No. 200) sieve. There is no change in the single-operator standard devia- tion and a 4.55% decrease in the multilaboratory standard deviation. For the new precision estimates based on the PSP fine aggregates with materials finer than a 75-µm (No. 200) sieve by washing in a range of 0.31 % to 2.54%, the 1s single-operator and multilaboratory standard deviations are 0.14% and 0.32%, respec- tively. These values are compared with the existing single-operator and multilaboratory standard devia- tion of 0.15% and 0.29%, respectively, which were prepared from fine aggregates having 1.0% to 3.0% finer than the 75-µm (No. 200) sieve. There is a 6.7% decrease in the single-operator standard de- viation and a 10.3% increase in the multi laboratory standard deviation. PSP Sample No. No. of Labs Average Results Repeatability Reproducibility Reproducibility X Y 1s X samples CV% Y samples CV% X samples Y samples 1s CV% 1s CV% 175 176 1354 0.313 0.305 0.042 13.3 13.7 0.13 42.3 0.13 43.7 171 172 1330 1.57 1.409 0.097 6.2 6.9 0.21 13.3 0.26 18.2 167 168 1303 0.786 0.837 0.059 7.5 7.1 0.14 18.3 0.13 16.0 163 164 1287 1.348 1.343 0.075 5.6 5.6 0.18 13.5 0.19 14.0 159 160 1171 1.425 1.419 0.077 5.4 5.5 0.2 14.2 0.2 14.3 155 156 1080 1.825 1.835 0.085 4.7 4.7 0.21 11.5 0.21 11.5 151 152 1125 2.423 2.446 0.144 5.9 5.9 0.34 13.9 0.34 14.1 147 148 1021 1.99 2.028 0.194 9.8 9.6 0.5 25.1 0.5 24.9 143 144 1015 1.277 1.271 0.136 10.7 10.7 0.28 21.8 0.27 20.9 139 140 926 1.988 2.536 0.259 13 10.2 0.46 23.4 0.6 23.6 135 136 810 1.789 1.78 0.145 8.1 8.1 0.41 23.1 0.42 23.5 131 132 698 1.366 1.365 0.147 10.8 10.8 0.34 24.9 0.33 24.5 127 128 625 1.87 1.852 0.209 11.2 11.3 0.41 21.9 0.41 22.1 123 124 587 2.128 2.107 0.145 6.8 6.9 0.5 23.7 0.51 24.2 Table 3-6 Summary of statistics of percent finer than a 75-µm (No. 200) sieve in fine aggregates by washing of 14 sets of AGF sample pairs.

12 y = 0.0925x + 0.0418 R² = 0.4472 0 0.05 0.1 0.15 0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pe at ab ili ty Average Repeatability Standard Deviation vs. Average (AASHTO T 11/Coarse Aggregate) y = 11.506x + 26.212 R² = 0.1179 0 10 20 30 40 50 60 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pe at ab ili ty Average Repeatability Coefficient of Variation vs. Average (AASHTO T 11/Coarse Aggregate) y = 0.2558x + 0.0503 R² = 0.8978 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pr od uc ib ili ty Average Reproducibility Standard Deviation vs. Average (AASHTO T 11/Coarse Aggregate) y = 15.588x + 47.239 R² = 0.287 0 10 20 30 40 50 60 70 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pr od uc ib ili ty Average Reproducibility Coefficient of Variation vs. Average (AASHTO T 11/Coarse Aggregate) figure 3-3 Relationship between average and standard deviation and between average and coefficient of variation of percent material finer than a 75-µm (No. 200) sieve by washing of AGC. y = 0.0925x + 0.0418 R² = 0.4472 0 0.05 0.1 0.15 0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pe at ab ili ty Average Repeatability Standard Deviation vs. Average (AASHTO T 11/Fine Aggregate) y = 11.506x + 26.212 R² = 0.1179 0 10 20 30 40 50 60 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pe at ab ili ty Average Repeatability Coefficient of Variation vs. Average (AASHTO T 11/Fine Aggregate) y = 0.2558x + 0.0503 R² = 0.8978 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pr od uc ib ili ty Average Reproducibility Standard Deviation vs. Average (AASHTO T 11/Fine Aggregate) y = 15.588x + 47.239 R² = 0.287 0 10 20 30 40 50 60 70 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Re pr od uc ib ili ty Average Reproducibility Coefficient of Variation vs. Average (AASHTO T 11/Fine Aggregate) figure 3-4 Relationship between average and standard deviation and between average and coefficient of variation of percent material finer than a 75-µm (No. 200) sieve by washing of AGF.

13 3.2 Evaluation of the Effect of Sample Size on AASHTO T 11 Test Results The new and existing precision estimates of AASHTO T 11 in Table 3-7 for fine aggregates are based on a nominal sample size of 500 g. A 1996 revision of this test method permits the fine aggre- gate test sample size to be the minimum of 300 g. Section 12.1.1 of AASHTO T 11 provides the preci- sion estimates based on analysis of the results from testing 300-g and 500-g test samples from Aggre- gate Proficiency Test Samples 99 and 100, which were essentially identical. As stated in the precision statement of the test method, there are only minor differences between the precision estimates from the two sample sizes. Note 7 of the test method states that the existing precision estimates would be revised to reflect the precision of the test based on 300-g test sample size when a sufficient number of Aggregate Proficiency Tests have been conducted using a sample size of 300 g. Four sets of PSP percent passing No. 200 sieve data, based on 300-g sample size, have been collected in recent years (see Table 2-4). To examine the reli- ability of the test results using a 300-g sample size, the precision estimates of the test were determined using the 300-g samples and compared with the pre- cision estimates presented in Section 3.1.3 prepared using 500-g samples. 3.2.1 Results of the Analysis Table 3-8 provides a summary of statistics of AASHTO T 11 test results based on four 300-g AGF sample sets. The pooled standard deviations and co- efficients of variation of the four data sets are pro- vided in Table 3-9 along with the precision estimates determined using 500-g samples (Section 3.1.3). As indicated from Table 3-9, the repeatability standard deviation associated with the 300-g sample size is the same as that associated with the 500-g sample size. However, the reproducibility standard devia- tion based on the 300-g sample size is 0.01 larger than the reproducibility standard deviation based on the 500-g sample size. The comparison of the coefficient of variation of the percent passing No. 200 sieve using 300-g and 500-g sample sizes is also shown in Table 3-9. As in- dicated, both repeatability and reproducibility coef- ficients of variation corresponding to 300-g samples Aggregate Type Stascs 1s d2s Coarse Aggregate Repeatability 0.10 (0.1) 0.28 (0.28) Reproducibility 0.21 (0.22) 0.59 (0.62) Fine Aggregate Repeatability 0.14 (0.15) 0.39 (0.43) Reproducibility 0.32 (0.29) 0.90 (0.82) Note: The numbers in parentheses represent the existing precision estimates. Table 3-7 Pooled repeatability and reproducibility precisions of percent material finer than a 75-µm (No. 200) sieve measured according to AASHTO T 11 based on sample standard deviation. Sample Size PSP Sample No. No. of Labs Average Results Repeatability Reproducibility Reproducibility X Y 1s X Samples CV% Y Samples CV% X Samples Y Samples 1s CV% 1s CV% 300 g 159 160 1139 1.526 1.526 0.100 6.5 6.5 0.213 13.9 0.212 13.9 155 156 1052 1.962 1.969 0.104 5.3 5.3 0.204 10.4 0.209 10.6 151 152 1075 2.554 2.571 0.148 5.8 5.7 0.328 12.9 0.329 12.8 147 148 1020 2.307 2.349 0.200 8.7 8.5 0.470 20.3 0.460 19.7 Table 3-8 Summary of statistics of percent passing 75-µm sieve by washing for four sets of 300-g PSP fine aggregate sample pairs.

14 are noticeably smaller than those corresponding to 500-g samples. Table 3-10 provides the results of statistical F-test on variance. As shown in the table, the computed F value for reproducibility is slightly greater than the critical F value, meaning that the reproducibility standard deviation of the 300-g samples is statistically larger than that from 500-g samples. However, from a practical standpoint, this is not considered signifi- cant since a difference of 0.01% in standard deviation translates into 0.03% percent allowable difference between two results, which is considerably smaller than the multilaboratory d2s of the AASHTO T 11 test method (0.90 %). Based on the analysis of per- cent passing 75-µm sieve data of PSP samples, there is no need to change the minimum sample size of the fine graded materials (nominal maximum size of 4.75 mm) of 300 g in AASHTO T 11. CHAPTER 4—EvAlUATION Of THE METHOD Of WASHINg Several test methods, including AASHTO T 11, AASHTO T 27, and AASHTO T 30, allow the use of a mechanical apparatus to perform the washing operation providing mechanical washing does not degrade the aggregates. To evaluate the difference between the sieve analysis results from mechanical and manual washing, the PSP gradation data result- ing from testing according to these test methods were analyzed. This chapter explores the effect of manual and mechanical washing on sieve analysis results of AASHTO T 11, T 27, and T 30. 4.1 Evaluation of Sieve Analysis Results The PSP collects sieve analysis data as part of the AGF, AGC, HMAIO, and HMASE testing programs. The data on the method of washing for the HMAIO and AGC samples have been collected since 2010. For the HMASE and AGF samples, the data have been collected since 2011. Using the sieve analysis data collected from samples with a known method of washing, the variability of the data from manual and mechanical washing was separately determined and statistically compared. The following sections pro- vide the results of statistical analysis using the data sets with a known method of washing for the AGC, AGF, HMAIO, and HMASE testing programs. The “p-values” in the table of statistical results indicate the statistical significance. If the p-value is less than 0.05, the difference in average or variability of per- cent passing resulting from the manual and mechani- cal washing is significant with 95% probability. If the p-value is smaller than 0.01, the differences are significant with 99% probability. The analyses of individual data sets are provided in Appendixes F through I, which are not published herein, but are available on the TRB website (http://www.trb.org) by searching for NCHRP Project 10-87. Sample Size Repeatability Reproducibility 1s CV % 1s CV% 300 g 0.14 6.5 0.33 11.7 500 g 0.14 10.1 0.32 20.9 Table 3-9 1s repeatability and reproducibility standard deviations and coefficients of variation of percent passing 75-µm sieve by washing from 300-g and 500-g samples. Comparison Computed F P Value Crical F Repeatability Reproducibility Repeatability Reproducibility 300 g vs. 500 g 1.00 1.06 0.023 0.003 1.04 Note: Degrees of freedom for both categories are greater than 1,000. Table 3-10 Results of F-test on variances for comparison of standard deviations of percent passing 75-µm sieve by washing of 300-g and 500-g sample sizes.

15 4.1.1 Analysis of AGC Data Table 4-1 through Table 4-3 provide the re- sults of a statistical comparison of the averages and repeatability/reproducibility standard deviations of the percent passing various sieve sizes pooled separately from manual and mechanical washing of AGC169- 170, AGC 173-174, and AGC 177-178 samples. Table 4-1 provides the results of the statistical t-test on average percent passing. As shown in the table, the p-values corresponding to all, except the 25.0-mm sieve, are less than 0.01, indicating that with a probability of 99%, the percent passing all sieve sizes, except the 25.0-mm sieve, of mechani- cally washed aggregates is significantly larger than those of the manually washed. The significance of the difference increases as the sieve sizes get smaller. The smallest p-value corresponds to the 75-µm sieve size from washing, meaning that with the highest probability, significantly larger amounts of materials finer than 75-µm sieve are washed away from the mechanically washed aggregates than from the manually washed aggregates. The fact that a higher amount of materials passes through smaller sieve sizes when aggregates are mechanically washed could be the result of two phe- nomena: (1) dust and fillers attached to the aggregates would be washed away more thoroughly during the mechanical washing and, therefore, particles would become smaller or (2) the aggregates break down Sieve Size Percent Passing, Manual Percent Passing, Mechanical Deg. of Freedom Computed T P Value 25.0 mm 99.93 99.93 496 0 1 19.0 mm 85.87 85.93 536 2.1 0.036 12.5 mm 50.43 50.63 544 2.7 0.007 9.5 mm 15.20 15.50 558 7.45 3.625E 13 4.75 mm 1.11 1.31 535 9.67 1.678E 20 75 µm Washing 0.55 0.69 432 11 4.286E 25 Note: Critical t for 1% level of significance is 2.58 and for 5% level of significance is 1.96. Table 4-1 Results of statistical t-test for comparison of average percent passing various sieve sizes after mechanical and manual washing pooled from AGC 169-170, AGC 173-174, and AGC 177-178 samples. Sieve Size Repeatability,Manual Repeatability, Mechanical Degrees of Freedom Computed F Crical F ( =0.01) Crical F ( =0.05) P Value 25.0 mm 0.118 0.120 387 & 3342 1.04 1.19 1.13 0.31 19.0 mm 0.442 0.413 3518 & 412 1.14 1.19 1.13 0.038 12.5 mm 0.829 0.801 3656 & 430 1.07 1.19 1.13 0.179 9.5 mm 0.539 0.536 3693 & 424 1.01 1.19 1.13 0.463 4.75 mm 0.195 0.170 3684 & 413 1.32 1.19 1.13 2E 04 75 µm Washing 0.095 0.105 369 & 3642 1.24 1.19 1.13 0.002 Table 4-2 Results of statistical F-test for comparison of repeatability standard deviations of percent passing various sieve sizes after mechanical and manual washing pooled from AGC 169-170, AGC 173-174, and AGC 177-178 samples.

16 during the mechanical washing, resulting in an in- crease of smaller particles. The comparison of the variability of the gradation measurements from man- ual and mechanical washing might explain which phenomenon is more likely. The results of the F-test on variance for the statistical comparison of the repeatability standard deviations of percent passing various sieve sizes from manual and mechanical washing of AGC 169-170, AGC 173-174, and AGC 177-178 samples are provided in Table 4-2. It is shown in the table that the variability of percent passing the 19-mm through 4.75-mm sieve sizes from mechanical washing is smaller than the variability of percent passing from manual washing; the results of the 19.0-mm and 4.75-mm sieves are statistically significant. However, the variability of percent passing 75-µm sieve from mechanical washing is significantly larger than that from manual washing. The graphical comparison of the mechanical and manual repeatability standard deviations is shown in Figure 4-1. The results of the F-test on variance for the sta- tistical comparison of the reproducibility standard deviations of percent passing various sieve sizes from manual and mechanical washing are provided in Table 4-3. It is shown in the tables that for all except the 75-µm sieve, the variability of percent passing from mechanical washing is lower than the variability of percent passing from manual washing Sieve Size Reproducibility,Manual Reproducibility, Mechanical Degrees of Freedom Computed F Crical F ( =0.01) Crical F ( =0.05) P Value 25.0 mm 0.159 0.148 3342 & 387 1.15 1.2 1.14 0.04 19.0 mm 0.662 0.603 3518 & 412 1.21 1.19 1.13 0.007 12.5 mm 1.497 1.448 3656 & 430 1.07 1.19 1.13 0.185 9.5 mm 0.882 0.775 3693 & 424 1.3 1.19 1.13 3E 04 4.75 mm 0.432 0.388 3684 & 413 1.24 1.19 1.13 0.002 75 µm Washing 0.222 0.246 369 & 3642 1.23 1.19 1.13 0.003 Table 4-3 Results of statistical F-test for comparison of reproducibility standard deviations of percent passing various sieve sizes after mechanical and manual washing pooled from AGC169-170, AGC 173-174, and AGC 177-178 samples. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 25.0 mm 19.0 mm 12.5 mm 9.5 mm 4.75 mm 75 m Washing St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-1 Repeatability standard deviations of percent passing various sieve sizes after manual and mechanical washing pooled from AGC 169-170, AGC 173-174, and AGC 177-178 samples.

17 and the variability associated with 19.0-mm, 9.5-mm, and 4.75-mm sieve sizes is statistically significant. The variability of percent passing the 75-µm sieve for mechanical washing is significantly larger than that for manual washing at both the 1% and 5% levels of significance. The graphical comparison of the mechanical and manual reproducibility standard deviations is shown in Figure 4-2. From the above, it is observed that with the excep- tion of percent passing the 75-µm sieve, the mechani- cal washing method has resulted in lower repeatability and reproducibility standard deviations of the percent passing of various sieve sizes. Therefore, it might be concluded that mechanical washing would improve removal of the filler and dust from coarse aggregates over manual washing. The results of the statistical test of significance for individual sample pairs of AGC 169-170, AGC 173-174, and AGC 177-178 can be found in Appendix F. 4.1.2 Analysis of AGF Data Table 4-4 through Table 4-6 provide the results of statistical comparison of the pooled averages and repeatability/reproducibility standard deviations of percent passing various sieve sizes from manual and 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 25.0 mm 19.0 mm 12.5 mm 9.5 mm 4.75 mm 75 mWashing St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-2 Reproducibility standard deviations of percent passing various sieve sizes after manual and mechanical washing pooled from AGC 169-170, AGC 173-174, and AGC 177-178 samples. Sieve Size Percent Passing, Manual Percent Passing, Mechanical Degrees of Freedom Computed t P Value 4.75 mm 99.93 99.97 389 -9.08 5.7E-18 2.36 mm 85.77 85.97 377 -5.75 1.8E-08 1.18 mm 71.04 71.15 378 -4.64 4.9E-06 600 µm 52.17 52.28 379 -2.39 1.7E-02 300 µm 19.61 19.81 378 -5.03 7.7E-07 150 µm 3.59 3.69 391 -6.82 3.5E-11 75 µm 1.09 1.15 404 -5.53 5.8E-08 75 µmWashing 0.95 1.03 423 -8.35 9.8E-16 Note: Critical t for 1% level of significance is 2.59 and for 5% level of significance is 1.96. Table 4-4 Results of the statistical t-test for comparison of average percent passing various sieve sizes from mechanical and manual washing pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples.

18 mechanical washing of AGF 171-172, AGF 175-176, and AGF 179-180 fine aggregates. Table 4-4 provides the results of the statistical t-test on average percent passing from manual and mechanical washing of AGF 171-172, AGF 175- 176, and AGF 179-180 samples. As shown from the p-values in the table, the percent passing all sieve sizes from mechanical washing are significantly larger than those from manual washing. The results of the statistical F-test on variance for comparison of the repeatability standard devia- tions of percent passing various sieve sizes from manual and mechanical washing, pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples, are provided in Table 4-5. From the p-values, it is indicated that the repeatability standard deviation of percent passing of three sieve sizes, 4.75-mm, 300-µm, and 75-µm, are significantly different; the mechanical washing resulted in significantly lower standard deviation of percent passing 300-µm and 75-µm sieves, while manual washing resulted in sig- nificantly lower standard deviation of percent pass- ing of 4.75-mm particles. The differences between all other sieve sizes are not significant. The graphical Sieve Size 1s Repeatability, Manual 1s Repeatability, Mechanical Degrees of Freedom Computed F Crical F ( =0.01) Crical F ( =0.05) P Value 4.75 mm 0.056 0.065 335 & 3898 1.35 1.2 1.14 5.09E-05 2.36 mm 0.371 0.380 313 & 3873 1.05 1.2 1.14 0.262 1.18 mm 0.311 0.293 3900 & 317 1.12 1.22 1.15 0.089 600 µm 0.310 0.295 3882 & 323 1.1 1.22 1.15 0.128 300 µm 0.230 0.211 3871 & 321 1.19 1.22 1.15 0.022 150 µm 0.117 0.117 318 & 3809 1.01 1.2 1.14 0.461 75 µm 0.089 0.089 317 & 3816 1.01 1.2 1.14 0.42 75 µm Washing 0.080 0.074 3998 & 326 1.18 1.22 1.15 0.026 Table 4-5 Results of the statistical F-test for comparison of repeatability standard deviations of percent passing various sieve sizes after mechanical and manual washing, pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples. Sieve Size 1s Reproducibility, Manual 1s Reproducibility, Mechanical Degrees of Freedom Computed F Crical F ( =0.01) Crical F ( =0.05) P Value 4.75 mm 0.061 0.065 335 & 3898 1.11 1.2 1.14 0.084 2.36 mm 0.667 0.607 3873 & 313 1.21 1.22 1.15 0.015 1.18 mm 0.432 0.405 3900 & 317 1.14 1.22 1.15 0.064 600 µm 0.793 0.794 323 & 3882 1 1.2 1.14 0.479 300 µm 0.706 0.697 3871 & 321 1.03 1.22 1.15 0.387 150 µm 0.294 0.257 3809 & 318 1.31 1.22 1.15 0.001 75 µm 0.215 0.172 3816 & 317 1.56 1.22 1.15 2.05E-07 75 µm Washing 0.221 0.169 3998 & 326 1.71 1.22 1.15 5.57E-10 Table 4-6 Results of statistical F-test for comparison of reproducibility standard deviations of percent passing various sieve sizes after mechanical and manual washing pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples.

19 comparison of the mechanical and manual repeat- ability standard deviations is shown in Figure 4-3. The results of the statistical F-test on variance for comparison of the reproducibility standard deviations of percent passing various sieve sizes from manual and mechanical washing, pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples, are pro- vided in Table 4-6. It is indicated from the table that the reproducibility standard deviation of the major- ity of sieve sizes is smaller from mechanical washing than from manual washing. The p-values in Table 4-6 show that the reproducibility standard deviation of percent passing of four sieve sizes including 75-µm by sieving and by mechanical washing are signifi- cantly lower than those from manual washing. The differences between all other sieve sizes are not sta- tistically significant. The graphical comparison of the mechanical and manual repeatability standard devia- tions is shown in Figure 4-4. The results of statis- tical test of significance for individual sample pairs of AGF 171-172, AGF 175-176, and AGF 179-180 samples can be found in Appendix G. From the above, it is observed that the mechanical washing method has resulted in lower repeatability 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 4.75 mm 2.36 mm 1.18 mm 600 µm 300 µm 150 µm 75 µm Total 75 µm Washing St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-3 Repeatability standard deviations of percent passing various sieve sizes after manual and mechanical washing pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 4.75 mm 2.36 mm 1.18 mm 600 µm 300 µm 150 µm 75 µm 75 µm Washing St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-4 Reproducibility standard deviations of percent passing various sieve sizes from manual and mechanical washing pooled from AGF 171-172, AGF 175-176, and AGF 179-180 samples.

20 and reproducibility standard deviations of the per- cent passing of a majority of sieve sizes. Therefore, it might be concluded that mechanical washing would improve removal of the filler and dust from fine aggre- gates, compared with the manual washing. 4.1.3 Analysis of HMAIO Data Table 4-7 through Table 4-9 provide the results of the statistical comparison of the pooled averages and repeatability/reproducibility standard devia- tions of the percent passing various sieve sizes from manual and mechanical washing of HMAIO 19-20, 21-22, 23-24 and 25-26 samples. Table 4-7 provides the results of the statistical t-test on average percent passing from manual and mechanical washing of HMAIO 19-20, 21-22, 23-24 and 25-26 samples. As shown in the table, with the exception of the percent passing the 12.5-mm sieve Sieve Size Average Percent Passing, Manual Average Percent Passing, Mechanical Deg. of Freedom Computed t P Value 12.5 mm 94.68 94.68 755 0 1 9.5 mm 86.23 86.28 770 -2.55 0.011 4.75 mm 60.98 61.08 712 -3.81 0.000 2.36 mm 39.28 39.53 752 -5.64 2.37E-08 1.18 mm 25.60 25.83 748 -8.82 7.967E-18 600 µm 17.30 17.58 730 -10.60 1.554E-24 300 µm 12.40 12.75 716 -12.83 4.717E-34 150 µm 9.79 10.18 693 -13.93 4.817E-39 75 µm, Total 8.31 8.66 708 -11.97 3.501E-30 Note: Critical t for 1% level of significance is 2.58 and for 5% level of significance is 1.97. Table 4-7 Results of the statistical t-test for comparison of average percent passing various sieve sizes for mechanical and manual washing pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 samples. Sieve Size 1s Repeatability, Manual 1s Repeatability, Mechanical Deg. of Freedom Computed F Crical F( =.01) Crical F( =.05) P Value 12.5 mm 0.387 0.394 506 & 2321 1.04 1.17 1.12 0.297 9.5 mm 0.329 0.329 2273 & 500 1 1.18 1.12 0.497 4.75 mm 0.392 0.445 493 & 2212 1.29 1.17 1.12 8E 05 2.36 mm 0.732 0.756 507 & 2275 1.07 1.17 1.12 0.175 1.18 mm 0.324 0.329 495 & 2208 1.03 1.17 1.12 0.349 600 µm 0.322 0.333 491 & 2199 1.07 1.17 1.12 0.164 300 µm 0.306 0.316 489 & 2218 1.07 1.17 1.12 0.164 150 µm 0.305 0.308 478 & 2219 1.02 1.18 1.12 0.388 75 µm, Total 0.287 0.3 487 & 2211 1.1 1.17 1.12 0.094 Table 4-8 Results of statistical F-test for comparison of repeatability standard deviations of percent passing various sieve sizes from mechanical and manual washing pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 samples.

21 size, the percent passing all sieve sizes from mechan- ical washing is significantly larger than the percent passing manual washing. The results of the statistical F-test on variance for comparison of the repeatability standard devia- tions of manual and mechanical washing pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 sam- ples are provided in Table 4-8. The table shows that the repeatability standard deviations of the percent passing all sieve sizes are larger from mechanical washing than those from manual washing; however, only the repeatability standard deviation of the per- cent passing the 4.75-mm sieve from mechanical washing is significantly larger than that from the manual washing (p-value of 8 × 10-5). The graphical comparison of the mechanical and manual repeat- ability standard deviations is shown in Figure 4-5. The results of the statistical F-test on variance for comparison of the reproducibility standard de- viations of the percent passing various sieve sizes Sieve Size Reproducibility,Manual Reproducibility, Mechanical Deg. of Freedom Computed F Crical F( =.01) Crical F( =.05) P Value 12.5 mm 0.482 0.471 2321 & 506 1.05 1.18 1.12 0.26 9.5 mm 0.418 0.393 2273 & 500 1.13 1.18 1.12 0.04 4.75 mm 0.512 0.531 493 & 2212 1.08 1.17 1.12 0.147 2.36 mm 0.905 0.902 2275 & 507 1.01 1.18 1.12 0.469 1.8 mm 0.525 0.511 2208 & 495 1.06 1.18 1.13 0.223 600 µm 0.523 0.519 2199 & 491 1.01 1.18 1.13 0.433 300 µm 0.542 0.548 489 & 2218 1.02 1.17 1.12 0.377 150 µm 0.557 0.564 478 & 2219 1.03 1.18 1.12 0.351 75 µm, Total 0.562 0.575 487 & 2211 1.04 1.17 1.12 0.265 Table 4-9 Results of the statistical F-test for comparison of reproducibility standard deviations of percent passing various sieve sizes after mechanical and manual washing pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 samples. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 12.5 mm 9.5 mm 4.75 mm 2.36 mm 1.18 mm 600 µm 300 µm 150 µm 75 µm, Total St an da rd D ev ia ti on s Sieve Size Manual Mechanical figure 4-5 Repeatability standard deviations of percent passing various sieve sizes from manual and mechanical washing pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 samples.

22 from manual and mechanical washing, pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 samples, are provided in Table 4-9. It is indicated from the table that mechanical washing provided improved reproducibility precision of percent passing of six out of nine sieve sizes. The p-values in the table in- dicate that the only case of statistical significance corresponds to the percent passing the 9.5-mm sieve, where mechanical washing has significantly improved the variability of measurements. The graphical comparison of the mechanical and manual reproducibility standard deviations is demonstrated in Figure 4-6. From the above, it is observed that the mechani- cal washing method has resulted in improved repro- ducibility standard deviations of the percent passing of the majority of sieve sizes. Therefore, it might be concluded that mechanical washing could help in more consistent removal of the filler and dust from the extracted aggregates over the manual washing. The results of the statistical tests of HMAIO 19-20, 21-22, 23-24, and 25-26 samples can be found in Appendix H. 4.1.4 Analysis of HMASE Data Table 4-10 through Table 4-12 provide the results of the statistical comparison of the pooled averages and repeatability/reproducibility standard devia- tions of the percent passing various sieve sizes from manual and mechanical washing of HMASE 73-74, 75-76, and 77-78 samples. Table 4-10 provides the results of the statistical t-test on average percent passing values correspond- ing to manual and mechanical washing of HMASE 73-74, 75-76, and 77-78 samples. As shown in the table, for a majority of sieve sizes, the mechanical washing has resulted in a larger percent passing than the manual washing. The p-values in Table 4-10 indi- cate that for 4.75-mm, 300-µm, 150-µm, and 75-µm sieve sizes, mechanical washing resulted in a signifi- cantly larger percent passing than manual washing. The results of the statistical F-test on variance for comparison of the repeatability standard devia- tions of manual and mechanical washing, pooled from HMASE 73-74, 75-76, and 77-78 samples, are provided in Table 4-11. As indicated from the table, mechanical washing resulted in a larger repeatability standard deviation than manual washing for eight out of nine sieve sizes. However, the p-values indicate that none of the differences in percent passing val- ues are statistically significant. The graphical com- parison of the mechanical and manual repeatability standard deviations is demonstrated in Figure 4-7. The statistical F-test on variance for compari- son of the reproducibility standard deviations of percent passing various sieve sizes from manual and mechanical washing, pooled from HMASE 73-74, 75-76, and 77-78 samples, is provided in Table 4-12. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 12.5 mm 9.5 mm 4.75 mm 2.36 mm 1.8 mm 600 µm 300 µm 150 µm 75 µm, Total St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-6 Reproducibility standard deviations of percent passing various sieve sizes from manual and mechanical washing pooled from HMAIO 19-20, 21-22, 23-24, and 25-26 samples.

23 Sieve Size Average Percent Passing, Manual Average Percent Passing, Mechanical Deg. of Freedom Computed t P Value 12.5 mm 94.90 94.90 211 0 1 9.5 mm 86.70 86.73 205 -0.87 0.383 4.75 mm 60.30 60.37 194 -2.03 0.044 2.36 mm 38.20 38.20 233 0 1 1.18 mm 24.67 24.73 228 -1.25 0.214 600 µm 17.17 17.23 231 -1.21 0.228 300 µm 12.83 13.07 218 -4.11 5.625E-05 150 µm 10.50 10.77 213 -4.56 8.765E-06 75 µm, Total 9.07 9.23 213 -2.9 0.004 Note: Critical t for 1% level of significance is 2.59 and for 5% level of significance is 1.97. Table 4-10 Results of the statistical t-test for comparison of average percent passing various sieve sizes for mechanical and manual washing pooled from HMASE 73-74, 75-76, and 77-78 samples. Sieve Size 1s Repeatability, Manual 1s Repeatability, Mechanical Deg. of Freedom Computed F Crical F( =.01) Crical F( =.05) P Value 12.5 mm 0.3 0.315 144 & 654 1.11 1.34 1.23 0.209 9.5 mm 0.292 0.305 146 & 645 1.09 1.34 1.23 0.241 4.75 mm 0.223 0.244 132 & 600 1.2 1.35 1.24 0.079 2.36 mm 0.777 0.807 151 & 693 1.08 1.33 1.22 0.265 1.18 mm 0.396 0.396 658 & 146 1 1.37 1.25 0.509 600 µm 0.399 0.401 146 & 664 1.01 1.33 1.23 0.446 300 µm 0.391 0.398 149 & 664 1.04 1.33 1.23 0.374 150 µm 0.378 0.368 672 & 146 1.05 1.37 1.25 0.357 75 µm, Total 0.36 0.371 143 & 659 1.06 1.34 1.23 0.314 Table 4-11 Results of the statistical F-test for comparison of repeatability standard deviations of percent passing various sieve sizes after mechanical and manual washing pooled from HMASE 73-74, 75-76, and 77-78 samples. This table shows that for five out of nine sieve sizes, mechanical washing resulted in larger reproducibility standard deviations than manual washing. However, the p-values indicate that the differences in reproduc- ibility standard deviations of percent passing from manual and mechanical washing are not statistically significant. The graphical comparison of the mechan- ical and manual reproducibility standard deviations is presented in Figure 4-8. From the above, it is observed that the mechanical washing of HMASE has resulted in variability that is not significantly different from that of manual wash- ing. This and the fact that mechanical washing re- sulted in significantly more materials passing through

24 the 300-µm sieve and smaller suggests that better washing of aggregates would occur from mechanical washing. The results of the statistical test of signifi- cance for individual sample pairs of HMASE 73-74, 75-76, and 77-78 samples can be found in Appendix I. 4.2 Evaluation of Degradation from Mechanical Washing In previous sections, the comparison of the av- erage percent material passing various sieve sizes from mechanical and manual washing indicated that with mechanical washing a significantly larger per- cent of material passes through various sieve sizes. This might indicate that in addition to separation of filler from coarser aggregates, some degradation of the coarser aggregates is taking place. To explore the possibility of degrading of the mechanically washed aggregates, the percent loss or gain was computed for each sieve size. Tables 4-13 through 4-16 show the calculations for determining the amount of deg- radation of the aggregates in each of the AGF, AGC, Sieve Size Reproducibility,Manual Reproducibility, Mechanical Deg. of Freedom Computed F Crical F( =.01) Crical F( =.05) P Value 12.5 mm 0.391 0.395 144 & 654 1.02 1.34 1.23 0.417 9.5 mm 0.388 0.424 146 & 645 1.2 1.34 1.23 0.076 4.75 mm 0.342 0.344 132 & 600 1.01 1.35 1.24 0.457 2.36 mm 0.994 0.927 693 & 151 1.15 1.36 1.24 0.146 1.18 mm 0.622 0.578 658 & 146 1.16 1.37 1.25 0.136 600 µm 0.653 0.594 664 & 146 1.21 1.37 1.25 0.08 300 µm 0.616 0.631 149 & 664 1.05 1.33 1.23 0.342 150 µm 0.636 0.644 146 & 672 1.03 1.33 1.23 0.409 75 µm, Total 0.635 0.623 659 & 143 1.04 1.37 1.25 0.395 Table 4-12 Results of the statistical F-test for comparison of reproducibility standard deviations of percent passing various sieve sizes after mechanical and manual washing pooled from HMASE 73-74, 75-76, and 77-78 samples. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 12.5 mm 9.5 mm 4.75 mm 2.36 mm 1.18 mm 600 µm 300 µm 150 µm 75 µm, Total St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-7 Repeatability standard deviations of percent passing various sieve sizes after manual and mechanical washing pooled from HMASE 73-74, 75-76, and 77-78 samples.

25 0 0.2 0.4 0.6 0.8 1 1.2 12.5 mm 9.5 mm 4.75 mm 2.36 mm 1.8 mm 600 µm 300 µm 150 µm 75 µm, Total St an da rd De vi ati on s Sieve Size Manual Mechanical figure 4-8 Reproducibility standard deviations of percent passing various sieve sizes after manual and mechanical washing, pooled from HMASE 73-74, 75-76, and 77-78 samples. Sample ID Sieve Sizes % Retained Manual % Retained Mechanical %Loss/ %Gain %Degradaon AGF 171 172 4.75 mm 0.2 0.1 0.1 0.30 2.36 mm 14.3 14.3 0 1.18 mm 14.3 14.3 0 600 µm 18.4 18.5 0.1 300 µm 34.4 34.2 0.2 150 µm 13.6 13.7 0.1 75 µm total 3.1 3.2 0.1 Pan 1.7 1.7 0 AGF 175 176 4.75 mm 14.1 13.8 0.3 0.40 2.36 mm 15.2 15.4 0.2 1.18 mm 18.6 18.6 0 600 µm 31.9 31.8 0.1 300 µm 18.62 18.75 0.13 150 µm 1.2 1.2 0 75 µm total 0.08 0.08 0 Pan 0.3 0.37 0.07 AGF 179 180 4.75 mm 14.1 13.88 0.22 0.32 2.36 mm 14.69 14.78 0.09 1.18 mm 19.6 19.5 0.1 600 µm 31.39 31.41 0.02 300 µm 15.84 15.91 0.07 150 µm 3.19 3.23 0.04 75 µm total 0.14 0.16 0.02 Pan 1.05 1.13 0.08 Table 4-13 Percent degradation of fine aggregates from mechanical washing.

26 Sample ID Sieve Sizes % Retained Manual % Retained Mechanical %Loss/ %Gain %Degradaon AGC 169 170 25.0 mm 0.2 0.2 0 0.6 19.0 mm 13.8 13.7 0.1 12.5 mm 33.6 33.5 0.1 9.5 mm 37.9 37.5 0.4 4.75 mm 12.1 12.3 0.2 75 µm washing 1.2 1.3 0.1 Pan 1.2 1.5 0.3 AGC 173 174 19.0 mm 13.6 13.5 0.1 0.3 12.5 mm 35 34.8 0.2 9.5 mm 34.4 34.5 0.1 4.75 mm 16.5 16.61 0.11 75 µm washing 0.24 0.25 0.01 Pan 0.26 0.34 0.08 AGC 177 178 19.0 mm 14.8 14.8 0 0.1 12.5 mm 37.7 37.6 0.1 9.5 mm 33.4 33.4 0 4.75 mm 13.66 13.66 0 75 µm washing 0.26 0.3 0.04 Pan 0.18 0.24 0.06 Table 4-14 Percent degradation of coarse aggregates from mechanical washing. HMAIO, and HMASE samples. The details of the calculation are explained as follows: 1. For each sieve size, the percent retained is com- puted using the percent passing corresponding to both manual and mechanical washing. 2. Assuming that no degradation is taking place from manual washing for each sieve, subtract- ing the percent retained corresponding to the manual washing from that of the mechanical washing would provide the percent change in a particular sieve, either from loss or gain. 3. For each sieve, a negative value indicates loss of aggregates and a positive value indicates gain of aggregates for that sieve size, which have resulted from loss of aggregates in upper sieve sizes. 4. Since the loss of aggregates in upper sieves would result in the gain of aggregates in the lower sieves, summation of all the positive values should always be equal to the summa- tion of all the negative values. The absolute value of the percent loss is the measure of degradation. The following observations are made from evaluation of the degradation values in Tables 4-13 through 4-16: 1. As indicated from the tables, the largest and smallest degradation corresponds to coarse aggregates (AGC169-170 and AGC 177-178) with the total degradation values of 0.6% and 0.1%, respectively. 2. On average, the degradation of fine aggregate and coarse aggregate is about 0.33% compared with degradation of aggregates from ignition oven or solvent extraction of about 0.45%. This difference suggests that extracted aggre- gates degrade more than the virgin aggregates. 3. The maximum percent loss that has been observed for any sieve size corresponding to AGF, AGC, HMAIO, and HMASE aggregates is 0.4%. This value is considerably smaller

27 Sample ID Sieve Sizes % Retained Manual % Retained Mechanical %Loss/ %Gain %Degradaon HMAIO 19 20 12.5 mm 5.8 5.8 0 0.4 9.5 mm 9.2 9.2 0 4.75 mm 22.3 22.2 0.1 2.36 mm 19.1 19 0.1 1.18 mm 16 16 0 600 m 11 10.9 0.1 300 m 6.2 6.1 0.1 150 m 2.75 2.77 0.02 75 m, Total 1 1.01 0.01 Pan 6.65 7.02 0.37 HMAIO 21 22 12.5 mm 4.4 4.5 0.1 0.5 9.5 mm 6.6 6.5 0.1 4.75 mm 27.4 27.3 0.1 2.36 mm 22 21.9 0.1 1.18 mm 12.6 12.6 0 600 m 7.6 7.6 0 300 m 4.9 4.8 0.1 150 m 2.9 2.8 0.1 75 m, Total 1.9 2 0.1 Pan 9.7 10 0.3 HMAIO 23 24 12.5 mm 5.6 5.6 0 0.5 9.5 mm 10.2 10.1 0.1 4.75 mm 24.5 24.5 0 2.36 mm 25 24.8 0.2 1.18 mm 13.2 13.2 0 600 m 6.2 6.3 0.1 300 m 3.7 3.6 0.1 150 m 2.1 2 0.1 75 m, Total 1.3 1.4 0.1 Pan 8.2 8.5 0.3 HMAIO 25 26 12.5 mm 5.5 5.4 0.1 0.5 9.5 mm 7.8 7.8 0 4.75 mm 26.8 26.8 0 2.36 mm 20.7 20.5 0.2 1.18 mm 12.9 13 0.1 600 m 8.4 8.2 0.2 300 m 4.8 4.8 0 150 m 2.7 2.7 0 75 m, Total 1.7 1.7 0 Pan 8.7 9.1 0.4 Table 4-15 Percent degradation of ignition oven aggregates from mechanical washing.

28 than the multilaboratory d2s for percent pass- ing various sieve sizes provided in AASHTO T 27 or AASHTO T 30. Therefore, from a practical point of view, the amount of degra- dation of these aggregates resulting from mechanical washing is not significant. Based on the findings above, use of mechani- cal washing can be allowed in place of the manual washing without any significant change in percent passing each sieve size. However, since the amount of degradation greatly depends on the type of aggre- gate used, it is recommended that a comparison of sieve analysis results from manual and mechanical washing of a laboratory-prepared aggregate blend of known gradation be made for each aggregate type. If the difference between percent passing each sieve size from manual and mechanical washing is smaller than the multilaboratory d2s values speci- fied in T 11, T 27, or T 30, then mechanical washing Sample ID Sieve Sizes % Retained Manual % Retained Mechanical %Loss/ %Gain %Degradaon HMASE 73 74 12.5 mm 4 4 0 0.4 9.5 mm 6.7 6.7 0 4.75 mm 27.4 27.3 0.1 2.36 mm 22.7 22.7 0 1.18 mm 13.5 13.5 0 600 m 8 7.9 0.1 300 m 4.5 4.4 0.1 150 m 2.4 2.3 0.1 75 m, Total 1.5 1.7 0.2 Pan 9.3 9.5 0.2 HMASE 75 76 12.5 mm 5.7 5.7 0 0.4 9.5 mm 10 10 0 4.75 mm 24.6 24.5 0.1 2.36 mm 23.9 24 0.1 1.18 mm 13.5 13.4 0.1 600 m 6.2 6.3 0.1 300 m 3.7 3.5 0.2 150 m 2.1 2.1 0 75 m, Total 1.3 1.3 0 Pan 9 9.2 0.2 HMASE 77 78 12.5 mm 5.6 5.6 0 0.4 9.5 mm 7.9 7.8 0.1 4.75 mm 27.2 27.3 0.1 2.36 mm 19.7 19.8 0.1 1.18 mm 13.6 13.5 0.1 600 m 8.3 8.3 0 300 m 4.8 4.6 0.2 150 m 2.5 2.5 0 75 m, Total 1.5 1.6 0.1 Pan 8.9 9 0.1 Table 4-16 Percent degradation of solvent extraction aggregates from mechanical washing.

29 could be used for evaluation of other samples of the same aggregate type. 4.3 Effect of Mechanical Washing Duration on Degradation The degradation values computed in the previ- ous section were the average values resulting from various mechanical washing durations. The labo- ratories have been conducting mechanical washing for different durations, which could have signifi- cant effect on the degradation of aggregate. In fact, part of the between-laboratory variability could be caused by the different durations of the mechanical washing practiced by different laboratories. AMRL, in addition to the data on the method of washing, has been collecting data on the duration of the mechanical washing, which can be used for evalu- ating the effect of washing duration on aggregate degradation. For this evaluation, the information on the duration of mechanical washing was used to group the gradation data. Depending on the information received, the data were grouped in three or four categories. For the coarse aggregate, the sieve analysis data were grouped into three categories of 1-5 min, 6-10 min, and more than 11 min. For the AGF, HMAIO, and HMASE aggregates, data were grouped into four categories of 1-5 min, 6-10 min, 11-15 min and more than 16 min. Using the proce- dure explained in the previous section, the amount of degradation was computed for each washing duration group. Figures 4-9 through 4-12 show the effect of me- chanical washing duration on the amount of aggre- gate degradation. As indicated from the figures, for each of the aggregate types (AGF, AGC, HMAIO, HMASE), there is an increase in the amount of degradation with the increase in washing time. The deviation between gradations from the manual and mechanical seems to be considerable after 10 min- utes of agitation. To determine an appropriate washing duration range, the percent degradation for each washing period is compared with the acceptable range of two results (multilaboratory d2s) of the total per- centage passing different sieve sizes specified in AASHTO T 27 and AASHTO T 30. For both origi- nal and extracted aggregates, the comparisons of the degradation levels in Figures 4-9 through 4-12 with the multilaboratory d2s values in AASHTO T 27 and AASHTO T 30 indicate that washing duration should be limited to 10 minutes to ensure that percent deg- radation is smaller than the largest acceptable differ- ence between results of two laboratories, which is 0.6%. This limitation would improve the between- laboratory variability from mechanical washing as well as reducing the degradation of aggregates. The washing process can be ended earlier if water be- comes clear before the end of the 10-minute period. On the other hand, lack of clarity in water at the end of the 10-minute period would indicate that mechani- cal washing produces considerable amount of filler by degrading the aggregates and, therefore, a manual wash should be used. 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 5 6 10 11 15 16+ Pe rc en tE ro sio n (% ) Washing Duration Range (minutes) AGFWashing Duration vs Percent Erosion figure 4-9 Percent increase in degradation of fine aggregates with increase in mechanical washing duration.

30 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 1 5 6 10 11+ Pe rc en tE ro sio n (% ) Washing Duration Range (minutes) AGCWashing Durationvs Percent Erosion figure 4-10 Percent increase in degradation of coarse aggregates with increase in mechanical washing duration. 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1 5 6 10 11 15 16+ Pe rc en tE ro sio n (% ) Washing Duration Range (minutes) HMAIOWashingDurationvs Percent Erosion figure 4-11 Percent increase in degradation of ignition oven aggregate with increase in mechanical washing duration. 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 1 5 6 10 11 15 16+ Pe rc en tE ro sio n (% ) Washing Duration Range (minutes) HMASEWashingDurationvs Percent Erosion figure 4-12 Percent increase in degradation of solvent extracted aggregates with increase in mechanical washing duration.

31 CHAPTER 5—CONClUSIONS AND PROPOSED CHANgES TO STANDARD TEST METHODS 5.1 Summary and Conclusions This digest was prepared for Task Order #2 of NCHRP Project 10-87 to update precision estimates of three test methods pertaining to aggregate materi- als: T 96, Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Testing Machine; T 304, Uncompacted Void Content of Fine Aggregate; and T 11, Materials Finer Than 75-µm (No. 200) Sieve in Mineral Aggre- gates by Washing. As part of updating the precision estimates of AASHTO T 11, the effect of minimum sample size on the repeatability and reproducibility of the test results was investigated. The data from Fine Aggregate (AGF) and Coarse Aggregate (AGC) testing programs of the AMRL Proficiency Sample Program (PSP) were used for the evaluation of preci- sion estimates. In addition to updating the precision estimates, the effect of manual and mechanical methods of washing on sieve analysis of the aggregates was in- vestigated. For this analysis, the aggregate grada- tion data from the Hot Mix Asphalt Ignition Oven (HMAIO) and Hot Mix Asphalt Solvent Extraction (HMASE), as well as AGF and AGC testing pro- grams of PSP, were analyzed. The aggregate grada- tion data from HMAIO and HMASE were obtained according to AASHTO T 30, Mechanical Analysis of Extracted Aggregate, while the aggregate gradation data from AGF and AGC were collected according to AASHTO T 11 and T 27, Sieve Analysis of Fine and Coarse Aggregates. The following conclusions are drawn from the results of the study. 5.1.1 Precision Estimates of AASHTO T 96 For AASHTO T 96, the existing precision es- timates are based on an average percent loss in the range of 10% to 45%, while the new set of precision estimates is based on an average percent loss in the range of 13% to 57%. The comparison of the new and existing sets of precision estimates indicates that the 1s% single-operator coefficient of variation of 3% computed in this study is larger than the existing 1s% single-operator coefficient of variation of 2% in ASTM C 131. Similarly, the new 1s% multilaboratory coefficient of variation of 6.2% is larger than the ex- isting 1s% multilaboratory coefficient of variation of 4.5% in ASTM C 131. The increase in both the single- operator and multilaboratory coefficients of variation is likely due to inclusion of data from a greater variety of coarse aggregates with a wider range of degrada- tion resistance than those used for developing the pre- cision estimates of ASTM C 131. 5.1.2 Precision Estimates of AASHTO T 304 For AASHTO T 304, the comparison of the new and existing sets of precision estimates indicates that the new standard deviations are larger than the existing standard deviations. The new precision es- timates, based on testing of the PSP fine aggregates with uncompacted void content in a range of 42% to 45%, include the 1s single-operator standard deviation of 0.33% and the 1s multilaboratory standard devia- tion of 0.81%, which are significantly larger than the existing single-operator standard deviation of 0.13% and multilaboratory standard deviation of 0.33%, respectively. The larger standard deviations based on the PSP data compared with the existing standard deviations are expected since the basis of the existing precision estimates are the uncompacted voids of the “graded standard silica sand,” as described in ASTM C 778, with round particles in a range of 600 µm (No. 30) to 150 µm (No. 100), which is not typical of other fine aggregates. 5.1.3 Precision Estimates of AASHTO T 11 For the precision estimates of AASHTO T 11, the new 1s single-operator standard deviation is 0.10% and the 1s multilaboratory standard deviation is 0.21%, based on coarse aggregates with percent materials finer than a 75-µm (No. 200) sieve by wash- ing in a range of 0.19 % to 1.23%. These values are similar to the existing AASHTO T 11 single-operator standard deviation of 0.10% and multilaboratory precision of 0.22%, respectively. The existing set of precision estimates for coarse aggregate is based on aggregates having a nominal maximum size of 19.0 mm (¼ in.) with less than 1.5% finer than the 75-µm (No. 200) sieve. Likewise, the new 1s single-operator standard de- viation is 0.14% and the 1s multilaboratory standard deviation is 0.32%, based on PSP fine aggregates with materials finer than a 75-µm (No. 200) sieve by washing in a range of 0.31% to 2.54%. These values are also comparable with the existing single-operator standard deviation of 0.15% and multilaboratory pre- cision of 0.29%, respectively. The existing precision

32 estimates are based on fine aggregates having 1.0% to 3.0% finer than the 75-µm (No. 200) sieve. 5.1.4 Evaluation of the Effect of Sample Size on T 11 Test Results The comparison of the repeatability and repro- ducibility standard deviations of the percent passing No. 200 sieve using 300-g and 500-g sample sizes indicated that from a statistical point of view, signif- icantly smaller multilaboratory standard deviation can be achieved using 500-g sample sizes. However, a difference of 0.01% in the multilaboratory stan- dard deviations of 300-g and 500-g samples trans- lates to an allowable difference of 0.03 % in percent passing No. 200 sieve from two laboratories, which is not considered practically significant. 5.1.5 Comparison of the Method of Washing The comparison of the average percent passing various sieve sizes from manual and mechanical wash- ing of AGC, AGF, HMAIO, and HMASE indicated that mechanical washing would result in a statisti- cally significant larger percent passing for a major- ity of sieve sizes, as compared with manual washing. This could indicate better removal of dust and filler, more degradation of aggregates using mechanical washing than using manual washing, or both. The results of the F-test on variance for the comparison of the repeatability and reproducibility standard deviations of percent passing various sieve sizes from manual and mechanical washing indi- cated that for the AGC and AGF samples, the vari- ability of the percent passing of a majority of sieve sizes is improved when samples are mechanically washed. However, for the HMAIO and HMASE aggregates, although not statistically significant, re- peatability and reproducibility standard deviations are predominantly larger when samples are washed mechanically. This might be due to vulnerability of the aggregates to breakage after exposure to heat and chemical solvents during the removal of the asphalt binder. 5.1.6 Evaluation of Degradation from Mechanical Washing The fact that a significantly larger percentage of materials would pass through various sieve sizes from mechanical washing might indicate that in ad- dition to removal of filler, some degradation of ag- gregates is taking place. To evaluate the amount of aggregate degradation during mechanical washing, the percent loss or gain of aggregates for each sieve size was computed. The summation of losses from larger sieve sizes, which is equal to the summation of gains in smaller sieve sizes, was used as the mea- sure of degradation. It was discovered that, on aver- age, the aggregates from ignition oven and solvent extraction have an overall degradation of 0.45% and the virgin fine or coarse aggregates have an overall degradation of 0.33%. For both virgin and extracted aggregates, the overall degradation values are never- theless considerably smaller than the multilaboratory acceptable range of two results (d2s) for total percent- age of material passing specified in AASHTO T 27 and AASHTO T 30. Therefore, it may be concluded that use of mechanical washing does not significantly degrade the aggregates. 5.1.7 Effect of Duration of Mechanical Washing on Degradation The gradation data were organized into mechan- ical washing duration groups of 1-5 min, 6-10 min, 11-15 min, and more than 16 min, and the amount of degradation was computed for each duration group. The results showed that there is a significant increase in the amount of degradation when the washing time exceeds 10 minutes. Moreover, the amount of deg- radation for washing durations of less than 10 min is considerably smaller than the multilaboratory d2s values in AASHTO T 27 and AASHTO T 30. Use of various washing durations by the participating labo- ratories may contribute to the multilaboratory vari- ability of the sieve analysis results with mechanical washing. Therefore, limiting the washing duration to 10 minutes should improve the multilaboratory variability of sieve analysis results. 5.2 Proposed Changes to AASHTO Standard Test Methods T 96, T 304, and T 11 From the analysis of AMRL Proficiency Sample data in this research, the following changes to the three standard test methods are proposed. 1. Adopt the new precision statement for AASHTO T 96 developed as part of this study and presented in Appendix E. Al- though the new precision estimates indicate increases in both single-operator and multi- laboratory coefficients of variation, the new

33 precision estimates are based on the proper- ties of a greater variety of coarse aggregates with a wider range of degradation resistance than those used for developing the precision estimates of ASTM C 131. 2. Adopt the new precision statement for AASHTO T 304 developed as part of this study and presented in Appendix E. The new precision estimates are significantly larger than the existing single-operator and multi- laboratory standard deviations; however, the new precision estimates are based on prop- erties of a wide selection of fine aggregates, while the existing precision estimates are based on the uncompacted voids of a “graded standard silica sand,” which is not typical of other fine aggregates. 3. Adopt the new precision estimates for AASHTO T 11 developed as part of this study and presented in Appendix E. Although the new and existing precision estimates are com- parable, there is a slight decrease in the multi- laboratory standard deviation of the percent finer than 75-µm sieve for both coarse and fine aggregates, which could reflect improvement in the sieve analysis process. 4. For the fine aggregate sample size of AASHTO T 11, the multilaboratory standard deviation of percent passing 75-µm sieve size using a 500-g sample was 0.01% smaller than that using a 300-g sample, which was statisti- cally significant. However, from a practical standpoint, this is not considered significant since a difference of 0.01% in the standard deviation translates into a 0.03% percent al- lowable difference between the two results, which is considerably smaller than the multi- laboratory d2s of 0.82% specified in AASHTO T 11. Therefore, the minimum sample size of fine aggregates in AASHTO T 11 should remain as 300 g. 5. Based on the sieve analysis of virgin and extracted aggregates used in the PSP, use of mechanical washing is acceptable despite the significantly larger percent of materials pass- ing all sieve sizes from mechanical washing than from manual washing. This is because the amount of degradation that could occur during mechanical washing was found not to be significant from a practical standpoint. 6. Since the amount of degradation greatly de- pends on the aggregate type, a comparison of sieve analysis results from manual and me- chanical washing of a laboratory-prepared aggregate blend of known gradation should be made for each aggregate type. If the dif- ference between the percent passing of each sieve size from manual and mechanical wash- ing is smaller than the multilaboratory d2s values specified in T 11, T 27, or T 30, then mechanical washing can be used for evalua- tion of other samples of the same aggregate. 7. The duration of the mechanical washing should be limited to 10 minutes. This limita- tion would improve the multilaboratory vari- ability from mechanical washing and would reduce the degradation of aggregates. The washing process can be ended earlier if water becomes clear before the end of the 10-minute period. On the other hand, lack of clarity in water at the end of the 10-minute period would indicate that a considerable amount of filler is being produced as a result of ag- gregate degradation and, therefore, manual washing should be used. REfERENCES 1. AASHTO Materials Reference Laboratory, 2013. http://www.amrl.net/. 2. Holsinger, R., Fisher, A., and P. Spellerberg. NCHRP Web­Only Document 71: Precision Estimates for AASHTO Test Method T 308 and the Test Methods for Performance­Graded Asphalt Binder in AASHTO Specification M 320. Transportation Research Board of the National Academies, Washington, D.C., 2012. 3. AASHTO, Designation T 11, “Materials Finer Than 75-µm (No. 200) Sieve in Mineral Aggregates by Washing,” Standard Specifications for Transporta­ tion Materials and Methods of Sampling and Testing, 32nd Edition, AASHTO, Washington, D.C., 2012, CD-ROM. 4. AASHTO, Designation T 96, “Resistance to Degra- dation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Machine,” Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 32nd Edition, AASHTO, Washington, D.C., 2012, CD-ROM. 5. AASHTO, Designation T 304, “Uncompacted Void Content of Fine Aggregate,” Standard Specifications for Transportation Materials and Methods of Sam­ pling and Testing, 32nd Edition, AASHTO, Wash- ington, D.C., 2012, CD-ROM.

34 6. AASHTO, Designation T 30, “Mechanical Analysis of Extracted Aggregate,” Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 32nd Edition, AASHTO, Washington, D.C., 2012, CD-ROM. 7. AASHTO, Designation T 27, “Sieve Analysis of Fine and Coarse Aggregates,” Standard Specifications for Transportation Materials and Methods of Sampling and Testing, 32nd Edition, AASHTO, Washington, D.C., 2012, CD-ROM. 8. ASTM, Designation C 778, “Standard Specification for Standard Sand,” Annual Book of ASTM Stan­ dards, Volume 4.01, ASTM, West Conshohocken, PA, 2012. 9. ASTM, Designation C 131, “Standard Test Method for Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Testing Machine,” Annual Book of ASTM Standards, Volume 4.02, ASTM, West Conshohocken, PA, 2007. 10. ASTM, Designation C 670, “Standard Practice for Preparing Precision and Bias Statements for Test Methods for Construction Materials,” Annual Book of ASTM Standards, Volume 4.02, ASTM, West Conshohocken, PA, 2013. 11. Ku, Harry H., “Statistical Concepts in Metrology,” NIST Special Publication 300, Volume 1, 1969: pp. 316-40. UNPUBlISHED APPENDIXES The following appendixes are not published herein, but can be found online at http://www.trb.org by searching for NCHRP Project 10-87. The appen- dixes are titled as follows: Appendix A: Proficiency Sample Data Sheets and Instructions Appendix B: T 96 Coarse Aggregate Graphs Appendix C: T 304 Fine Aggregate Graphs Appendix D: T 11 Coarse and Fine Aggregate Graphs Appendix F: Coarse Aggregate—Washing Method Tables and Graphs Appendix G: Fine Aggregate—Washing Method Tables and Graphs Appendix H: Hot Mix Asphalt Ignition Oven— Washing Method Tables and Graphs Appendix I: Hot Mix Asphalt Solvent Extraction—Washing Method Tables and Graphs APPENDIX E—PRECISION STATEMENTS fOR T 96, T 304, AND T 11 Precision Estimate for AASHTO T 96— Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the los Angeles Testing Machine E.1 Precision and Bias E.1.1 Precision. Criteria for judging the acceptabil- ity of resistance to degradation results obtained by this method are given in Table E-1. E.1.1.1 Single­Operator Precision (Repeat- ability). The figures in Column 2 of Table E-1 are the coefficients of variation that have been found to be appropriate for the conditions of test de- scribed in Column 1. Two results obtained in the same laboratory, by the same operator using the same equipment, in the shortest practical period of time, should not be considered suspect unless Statistics Coefficient of Variation 1s (%)a Acceptable Range of Two Test Results d2s (%)a Single-Operator Precision LA Abrasion Loss (%) 3.0 8.5 Multilaboratory Precision LA Abrasion Loss (%) 6.2 17.6 aThese values represent the 1s% and d2s% limits described in ASTM Practice C 670. Note – The precision estimates given in Table E-1 are based on the analysis of test results from 15 pairs of AMRL coarse aggregate proficiency samples. The data analyzed consisted of results from 290 to 513 laboratories for each of the 15 pairs of samples. The average percent LA Abrasion Loss ranged from 13% to 57%. The details of this analysis are presented in the main text of NCHRP Research Results Digest 389. Table E-1 Precision estimates.

35 the difference in the two results, expressed as a percent of their mean, exceeds the values given in Table E-1, Column 3. E.1.1.2 Multilaboratory Precision (Reproduc- ibility). The figures in Column 2 of Table E-1 are the coefficients of variation that have been found to be appropriate for the conditions of test described in Column 1. Two results submitted by two differ- ent operators testing the same material in different laboratories shall not be considered suspect unless the difference in the two results, expressed as a percent of their mean, exceeds the values given in Table E-1, Column 3. E.1.2 Bias. The bias of the procedure in this test method cannot be determined. Precision Estimate for AASHTO T 304— Uncompacted void Content of fine Aggregate E.2 Precision and Bias E.2.1 Precision. Criteria for judging the accept- ability of void content obtained by this method are given in Table E-2. E.2.1.1 Single­Operator Precision (Repeat- ability). The figures in Column 2 of Table E-2 are the standard deviations that have been found to be appropriate for the conditions of test described in Column 1. Two results obtained in the same labora- tory, by the same operator using the same equip- ment, in the shortest practical period of time, should not be considered suspect unless the difference in the two results exceeds the values given in Table E-2, Column 3. E.2.1.2 Multilaboratory Precision (Repro- ducibility). The figures in Column 2 of Table E-2 are the standard deviations that have been found to be appropriate for the conditions of test described in Column 1. Two results submitted by two different operators testing the same ma- terial in different laboratories shall not be con- sidered suspect unless the difference in the two results exceeds the values given in Table E-2, Column 3. E.2.2 Bias. The bias of the procedure in this test method cannot be determined. Precision Estimate for AASHTO T 11— Materials finer Than 75-µm (No. 200) Sieve in Mineral Aggregates by Washing E.3 Precision and Bias E.3.1 Precision. Criteria for judging the accept- ability of the percentage of materials finer than a 75-µm (No. 200) sieve by washing obtained by this method are given in Table E-3. E.3.1.1 Single­Operator Precision (Repeat- ability). The figures in Column 2 of Table E-3 are the standard deviations that have been found to be appropriate for the conditions of test described in Column 1. Two results obtained in the same labo- ratory, by the same operator using the same equip- ment, in the shortest practical period of time, should not be considered suspect unless the difference in Statistics Standard Deviations 1s a Acceptable Range of Two Test Results d2s a Single-Operator Precision Uncompacted Voids (%) 0.33 0.95 Multilaboratory Precision Uncompacted Voids (%) 0.81 2.29 aThese values represent the 1s (or 1s%) and d2s (or d2s%) limits described in ASTM Practice C 670. Note – The precision estimates given in Table E-2 are based on the analysis of test results from 14 pairs of AMRL fine aggregate proficiency samples. The data analyzed consisted of results from 183 to 535 laboratories for each of the 14 pairs of samples. The average percent uncompacted voids ranged from 42% to 45%. The details of this analysis are presented in the main text of NCHRP Research Results Digest 389. Table E-2 Precision estimates.

36 the two results, exceeds the values given in Table E-3, Column 3. E.3.1.2 Multilaboratory Precision (Reproduc- ibility). The figures in Column 2 of Table E-3 are the standard deviations that have been found to be appropriate for the conditions of test described in Column 1. Two results submitted by two differ- ent operators testing the same material in different laboratories shall not be considered suspect unless the difference in the two results exceeds the values given in Table E-3, Column 3. E.3.2 Bias. The bias of the procedure in this test method cannot be determined. Table E-3 Precision estimates. Condition of Test Standard Deviation 1s a Acceptable Range of Two Test Results d2s a Single-operator Precision Percent finer than75- µm sieve by washing (%) Coarse Aggregate 0.10 0.28 Fine Aggregate 0.14 0.39 Multilaboratory Precision Percent finer than 75- µm sieve by washing (%) Coarse Aggregate 0.21 0.59 Fine Aggregate 0.32 0.90 aThese values represent the 1s (or 1s%) and d2s (or d2s%) limits described in ASTM Practice C 670. Note – The precision estimates given in Table E-3 are based on the analysis of test results from 14 pairs of coarse aggregate and 14 pairs of fine aggregate of the AMRL Proficiency Sample Program. The data analyzed consisted of results from 552 to 1,380 laboratories for each of the 14 pairs of samples of both coarse and fine aggregates. The average percent finer than a 75- µm sieve was less than 1.5% for coarse aggregate and in a range of 1% to 3% for fine aggregate. The details of this analysis are presented in the main text of NCHRP Research Results Digest 389.

Transportation Research Board 500 Fifth Street, NW Washington, DC 20001 These digests are issued in order to increase awareness of research results emanating from projects in the Cooperative Research Programs (CRP). Persons wanting to pursue the project subject matter in greater depth should contact the CRP Staff, Transportation Research Board of the National Academies, 500 Fifth Street, NW, Washington, DC 20001. COPYRIGHT INFORMATION Authors herein are responsible for the authenticity of their materials and for obtaining written permissions from publishers or persons who own the copyright to any previously published or copyrighted material used herein. Cooperative Research Programs (CRP) grants permission to reproduce material in this publication for classroom and not-for-profit purposes. Permission is given with the understanding that none of the material will be used to imply TRB, AASHTO, FAA, FHWA, FMCSA, FTA, or Transit Development Corporation endorsement of a particular product, method, or practice. It is expected that those reproducing the material in this document for educational and not-for-profit uses will give appropriate acknowledgment of the source of any reprinted or reproduced material. For other uses of the material, request permission from CRP. ISBN 978-0-309-28400-4 9 780309 284004 9 0 0 0 0 Subscriber Categories: Materials

Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Research Results Digest 389: Precision Estimates of AASHTO T 304, AASHTO T 96, and AASHTO T 11 and Investigation of the Effect of Manual and Mechanical Methods of Washing on Sieve Analysis of Aggregates updates precision estimates of American Association of State Highway and Transportation Officials (AASHTO) T 96, T 304, and T 11.

The report also examines the significance of the difference between variability of percent passing No. 200 sieve of 300-g and 500-g fine aggregate samples measured according to AASHTO T 11.

In addition, NCHRP RRD 389 evaluates the effect of manual versus mechanical washing, by comparing the results of sieve analysis of PSP samples, washed manually or mechanically prior to being tested according to AASHTO T 11, T 27, or T 30.

The following appendixes are not included in the print or PDF version of the publication, but are available for download from the project page for NCHRP Project 10-87. The appendixes are titled as follows:

* Appendix A: Proficiency Sample Data Sheets and Instructions

* Appendix B: T 96 Coarse Aggregate Graphs

* Appendix C: T 304 Fine Aggregate Graphs

* Appendix D: T 11 Coarse and Fine Aggregate Graphs

* Appendix F: Coarse Aggregate—Washing Method Tables and Graphs

* Appendix G: Fine Aggregate—Washing Method Tables and Graphs

* Appendix H: Hot Mix Asphalt Ignition Oven—Washing Method Tables and Graphs

* Appendix I: Hot Mix Asphalt Solvent Extraction—Washing Method Tables and Graphs

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!