Citation
Evaluation of laboratory compaction methods for pervious concrete

Material Information

Title:
Evaluation of laboratory compaction methods for pervious concrete
Creator:
Brown, Bojana Barovic
Publication Date:
Language:
English
Physical Description:
x, 98 leaves : illustrations ; 28 cm

Subjects

Subjects / Keywords:
Lightweight concrete ( lcsh )
Compacting ( lcsh )
Compacting ( fast )
Lightweight concrete ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 96-98).
Statement of Responsibility:
by Bojana Barovic Brown.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
656836972 ( OCLC )
ocn656836972
Classification:
LD1193.E53 2010m B76 ( lcc )

Downloads

This item has the following downloads:


Full Text
EVALUATION OF LABORATORY COMPACTION
METHODS FOR PERVIOUS CONCRETE
by
Bojana Barovic Brown
B.S.,University of Colorado at Boulder, 2003
A thesis submitted to the University of Colorado Denver In partial fulfillment Of the requirements for the degree of Masters of Science Civil Engineering
May 2010


(2010) by (Bojana Barovic Brown) All rights reserved.


This thesis for Masters of Science
Degree by
Bojana Barovic Brown Has been approved By
Cheng Li, Ph.D.
Date


Brown, Bojana Barovic (M.S. Civil Engineering)
Evaluation of Laboratory Compaction Methods for Pervious Concrete Thesis Directed by Assistant Professor Stephan A. Durham
ABSTRACT
As societys infrastructure continues to grow and develop, pervious open areas are being depleted and replaced by impervious surfaces such as rooftops, sidewalks, and parking lots. Storm water runoff is an increasing concern of many municipalities as city engineers must design for increased storm drain and sewer system capacities. To address these concerns, pervious concrete has become a popular product that can reduce water runoff by allowing water to permeate through the pavement surface and into the underlying soil.
Pervious concrete contains little to no fine aggregate, thus allowing voids of 15% to 35%. Currently, very few testing standards exist for pervious concrete resulting in decreased quality control and in some cases poor pervious concrete pavements. This thesis evaluates laboratory compaction methods for pervious concrete. Many of pervious concretes properties are related to the amount and type of compaction. Compressive strength, durability, unit weight, and porosity are all properties that can be directly related to compaction of a pervious concrete mixture. This thesis evaluated four methods of laboratory compaction. Results from this study demonstrate that the weight versus volume method produced pervious specimens with superior strength when compared to three other methods. In addition, a common industry method, the jigging method, produced acceptable results. The research herein provides recommendations for pervious concrete sample preparation for laboratory testing.
IV


This abstract accurately represents the content of the candidates thesis. I recommend its
publication.
Stephan A. Durham, Ph.D.
v


TABLE OF CONTENTS
Figures...............................................................viii
Tables...................................................................
1 INTRODUCTION.......................................................1
1.1 Overview........................................................1
1.2. Research Objective..............................................1
2.1 Description.....................................................3
2.2 Consistency.....................................................6
2.3 Unit Weight.....................................................8
2.4 Segregation....................................................11
2.5 Curing.........................................................13
2.6 Strength.......................................................15
2.7 Permeability...................................................17
2.8 Durability.....................................................18
2.9 Infiltration Rate..............................................19
2.10 Research Limitations and Scope.................................21
2.11 Pervious Concrete in Colorado..................................21
2.12 Compaction.....................................................22
3 PROBLEM STATEMENT.................................................24
4 RESEARCH PLAN.....................................................28
4.1 Concrete Consolidation and Effects on Other Properties.........28
4.2 Mix Design.....................................................31
4.3 Consolidation Method 1: Traditional Consolidation..............33
4.4 Consolidation Method 2: Jigging Method Described In ASTM C29...35
4.5 Method 3: Compaction As A Percentage Of Volume.................36
4.6 Method 4: Weight vs. Volume Method.............................39
4.7 Compressive Strength...........................................43
4.8 Neoprene Pads..................................................45
4.9 Sulfur Capping.................................................45
4.10 Porosity Testing...............................................47
5 EXPERIMENTAL RESULTS..............................................49
5.1 Experimental Procedures........................................49
5.2 Unit Weight/Density Results Batch 2..........................60
5.3 Porosity Results- Batch 1......................................65
5.4 Compressive Strength Results Batch 1.........................68
vi


5.5 Batch 1 Results............................................75
5.6 Lessons Learned and Applied to Batch 2.....................76
5.7 Batch 2 Result Data........................................80
6 CONCLUSION...................................................92
6.1 Experiment Conclusions.....................................92
6.2 Recommendations............................................94
BIBLIOGRAPHY.....................................................96
vii


FIGURES
Figure 2.1: Impervious Layer Zone (Hager, 2009).........................13
Figure 2.2: Drain Time Testing Apparatus (Hager, 2009)..................20
Figure 2.3: Strike off and Compaction of Pervious Concrete
with Steel Roller Screed (Hager, 2009)..................................23
Figure 3.1: Consistency by Visual Observation (Tennis, 2004)............25
Figure 4.1: Compaction Tool.............................................37
Figure 5.1: Failed Sulfur Capping Images................................50
Figure 5.2: Sulfur Poured On Top of Standing Specimen...................53
Figure 5.3: Failed Attempt at Pouring Sulfur............................54
Figure 5.4: Tops of Cylinders Not Level.................................55
Figure 5.5: Visible Variation of Aggregates.............................56
Figure 5.6: Tops of Cylinders Not Level After Saw Cutting...............57
Figure 5.7: Broken Core Samples.........................................58
Figure 5.8: Compression Testing of a Non-level Cylinder Surface.........59
Figure 5.9: Unit Weight of Cylinders for Batch 1 -
Methods 1,2,3 & 4.......................................................61
Figure 5.10: Unit Weight of Cores for Batch 1 Methods 1, 2,3 & 4......62
Figure 5.11: Unit Weight of Cylinders and Cores for Batch 1 Methods 1, 2,
3 & 4...................................................................63
Figure 5.12: Porosity of Cylinders for Batch 1
Methods 1,2,3 & 4.......................................................66
Figure 5.12: Porosity of Cylinders for Batch 1 Methods 1, 2, 3 & 4....67
Figure 5.14: Porosity of Cylinders and Cores for Batch 1 -
Methods 1, 2,3 &........................................................68
Figure 5.15: Compressive Strength of Cylinders Batch 1
Methods 1, 2,3 & 4......................................................72
Figure 5.16: Compressive Strength of Cores for Batch 1
Methods 1, 2,3 & 4.....................................................73
Figure 5.17: Compressive Strength of Cylinders and Cores for Batch 1
Methods 1,2,3 & 4.......................................................74
Figure 5.18: Results for Batch 1 Cylinders Methods 1, 2, 3 & 4........75
Figure 5.19: Results for Batch 1 Cores Methods 1,2,3 & 4..............76
Figure 5.20: Batch 2 Unit Weight........................................83
Figure 5.21: Batch 2 Porosity...........................................84
viii


Figure 5.22: Compressive Strength at 7 Days...............................85
Figure 5.23: Batch 2 Method 4 Compressive Test at 7 Days..................86
Figure 5.24: Porosity vs. Strength at 7 Days..............................89
Figure 5.25: Porosity vs. Strength at 28 Days.............................89
Figure 5.26: Unit Weight vs. Strength at 7 Days...........................90
Figure 5.27: Unit Weight vs. Strength at 28 Days..........................91
IX


TABLES
Table 4.1: Compaction Factors for estimating in-situ density (Kevern, 2009).29
Table 4.2: Summary of Specimens to be Created for Methods 1,2 &3.......31
Table 4.3: Summary of Specimens to be Created for Method 4.............31
Table 4.4: Summary of Total Specimens Needed...........................32
Table 4.5: Material Properties.........................................32
Table 5.1: Unit Weight (lb/ft3) for Cylinders and Cores................60
Table 5.2: Porosity (%) for Cylinders and Cores........................65
Table 5.3: Compression Strength Results for Method 1...................70
Table 5.4: Compression Strength Results for Method 2...................70
Table 5.5: Compression Strength Results for Method 3...................71
Table 5.6: Compression Strength Results for Method 4...................72
Table 5.7: Batch 2 Weights (yd3).......................................81
Table 5.8: Unit Weights of Fresh and Hardened Concrete.................81
Table 5.9: Batch 2 Results.............................................82
Table 5.10: Prediction for 28 Day Strength Results.....................87
Table 5.11: Compressive Strength Increase between 7 and 28 days........88
x


1 INTRODUCTION
1.1 Overview
As our society continues to replace open space with parking lots, building roof tops and other impervious areas, storm water management has become more challenging. Government jurisdictions restrict the amount of rainwater a new development can add to the existing drainage channels. Stormwater retention ponds are costly and take up usable space on a property. A way to reduce or eliminate stormwater ponds is to incorporate pervious concrete as the pavement choice for parking lots. Captured stormwater can seep through the pervious concrete and directly flow into the ground replenishing groundwater for existing landscape. By having a porous concrete surface, stormwater runoff is reduced. In addition, pervious concrete was named a Best Management Practice and is recommended by the Environmental Protection Agency (EPA), making it a concrete worthy of further research and development.
1.2. Research Objective
The objective of this thesis is to develop and evaluate testing standards for pervious concrete with the main focus being compaction methods. Testing standards of conventional concrete are not appropriate for pervious concrete and therefore should not be adopted as standards. Currently, the American Concrete Institute (ACI) does not have any accepted testing standards for pervious concrete, but is in the process of accepting new American Standards for Testing and Materials (ASTM) specifications regarding pervious concrete. Recommendations for testing pervious concrete exist, but need to be evaluated for their effectiveness and consistency. Testing standards for conventional concrete cannot be used for pervious concrete and need to be rethought and modified in
1


order to serve as tests for pervious concrete. The level of compaction of pervious concrete is a factor that directly impacts most relative properties of pervious concrete such as: unit weight, compressive strength, and porosity. The more compacted a specimen is, the less porous it is. As porosity goes down, strength and unit weight are increased. Four different methods of compaction are examined and discussed with more detail as noted in the following paragraphs.
In method 1, the concrete is compacted by rodding in accordance to the ASTM C138 specification for traditional concrete. Compaction method 2 follows the new ASTM Cl688 specification for pervious concrete and uses the jigging procedure for compaction. Compaction method 3 is a new method that uses a compaction tool. Method 4 is the weight versus volume compaction method. Method 4 determines the weight of concrete by multiplying the desired density by the volume. This quantity of concrete is compacted to the designed volume independent of compaction strikes or individual.
This thesis focuses on the effects of four compaction methods and the impact they have on unit weight, compressive strength, and porosity. Compaction methods were compared not only on laboratory results, but also on consistency and repeatability by individuals (presumably in the field) with only a basic knowledge of the subject. In comparing these four compaction methods, recommendations were provided for the best suited compaction method or methods given a specific specimen type (ie. cylinder vs. block mold).
2


2 LITERATURE REVIEW
2.1 Description
Pervious concrete has been used in the United States since the 1970s, in Europe since the 1910s, and has recently gained popularity in the United States due to its acceptance as a sustainable building material (Offenberg, 2008). The most common and likely use of pervious concrete is pavement application. Used as a pavement, pervious concrete provides a durable surface that is porous and allows water to pass through into the ground. This water penetrating surface produces less surface runoff, reduces peak water flow into storm sewers, and decreases or eliminates the area required for site retention ponds (Paine, 1992). As defined by the American Concrete Institute (ACI), the term pervious concrete typically describes a zero-slump, open graded material consisting of Portland cement, coarse aggregate, little or no fine aggregate, admixtures, and water. The combination of these ingredients produces a hardened material with connected pores, ranging in size from 0.08 to 0.32 in. (0.20 to 0.81 cm), that allow water to pass through easily. The void content ranges from 15% to 35% with typical compressive strengths of 400 to 4000 psi (2.75 to 27.6 MPa). The drainage rate of pervious concrete pavement varies with aggregate size and density, but will generally range between fall in the range between 2 to 18 gal./min/ft2 (1.4 to 12.63 L/min/m2) (ACI, 2006).
In order for pervious concrete to gain acceptance in the industry, standardized test methods need to be recognized and adopted by ACI. This typically comes in the form of specifications and testing standards. ASTM testing standards have recently been developed, but not yet accepted. To date no adopted standardized tests (fresh or hardened) exist for pervious concrete. Testing standards are critical for pervious concrete to be widely accepted and utilized. Testing standards must be easily performed and
3


duplicated by any individual with little experience and basic knowledge of the subject. The consistency and behavior of pervious concrete is significantly different from conventional concrete. Normal testing procedures (for conventional concrete) based on slump and cylindrical strength are not applicable to pervious concrete (County of Fairfax, Virginia, 2007), therefore the same standards should not be used for both types of concrete. New or modified testing methods need to be developed in order to accurately determine pavement performance and quality (Paine, 1992). Recommendations from various sources exist on testing fresh and hardened properties of pervious concrete, but no standard has been adopted by the ACI.
ACI committee 522 addresses pervious concrete, but as stated previously, ACI 522 does not officially have any recognized tests for pervious concrete. ACI 522.1 currently recommends density testing in accordance with American Society for Testing Materials, ASTM C138 (Standard Method for Testing Density for Conventional Concrete), Standard Test Method for Density (Unit Weight), Yield, and Air Content of Concrete. These tests have not been widely accepted or successful for pervious concrete because the tests were developed and aimed at conventional concrete mixtures. Fresh concrete properties tests for conventional concrete cannot be performed for pervious concrete because pervious concrete has a different concrete structure and contains large voids. Pervious concrete also compacts differently than conventional concrete. Traditional concrete is rodded during consolidation; however, rodding does not provide adequate consolidation for pervious concrete. In recognition of pervious concrete, ASTM has adopted some newer testing standards. In October 2008, ASTM Subcommittee C09.49 released C 1688, Standard Test Method for Density and Void Content of Freshly Mixed Pervious Concrete (Palmer, 2009). The consolidation method used in ASTM 1688 is the
4


jigging procedure. This is the first step in a series of tests necessary for pervious concrete to become a more predictable quality controlled material. Widely accepted and easily reproduced testing standards provide the industry more confidence that pervious concrete is a quality controlled and desired product. To date, pervious concrete testing standards have not proven to be consistent and accurate. Additional research is needed to develop standardized tests that can be easily and consistently performed both in the laboratory and in the field environment. The best compaction method(s) used for these tests is to be developed.
To develop standard testing methods applicable to pervious concrete, all concrete properties in the fresh and hardened state are considered and the best curing and handling practices are noted. Applicable fresh properties include: consistency and unit weight. Applicable hardened properties include: strength, durability, and permeability (Kosmatka, 2002). Density, being the most measurable and applicable property for pervious concrete, is determined for both plastic and hardened concrete (Kevem, 2009).
The University of Colorado has tested the performance of pervious concrete pavement systems in Denver, Colorado. Recommendations for design and construction were made for pervious concrete pavements in Colorados environment of fluctuating temperatures. Frequent freeze thaw cycles and very low humidity (Hager, 2009). A previous concrete test pavement was constructed in a parking lot of the Auraria campus at University of Colorado Denver. An extensive laboratory and field examination were included in this study. The laboratory phase examined the effects of cementitious content, w/cm, and sand content on the structural and hydrological performance of pervious concrete mixtures. It was concluded that the optimum cementitious content, w/cm, and sand
5


content were 525 lb (238 kg), 0.30, and 7.5% by total weight of aggregate. In addition, it was determined that cement replacement up to 20% with Class C or Class F fly ash could be used and meet structural and hydrological requirements. Additional research demonstrated that air-entraining admixtures increased the freeze/thaw resistance of the pervious concrete mixtures.
The pervious concrete pavement test section contained 20% fly ash, crushed recycled concrete as underplaying coarse aggregate layer, and 10% replacement of sand with crushed g lass i n t he f ine a ggregate u nderlayment layer. F ield i nvestigations o f t he pervious concrete pavement were performed and included monitoring of heat island effects, water quality, deterioration, clogging and permeability. The study looked into the effects of deicing agents commonly found on Colorados roads. It was found that deicing agents strip the top bonds between aggregates and accelerate deterioration of pervious pavement (Hager, 2009).
2.2 Consistency
Consistency of conventional concrete is typically measured by a slump cone test. For a particular mixture, the slump should be consistent throughout the concrete placement. In the field, a change in slump between batches typically means an undesired change in the ratio of the concrete ingredients; therefore, raising a red flag before the concrete is placed. The slump test is not applicable for pervious concrete due to the very low water cement ratios of a typical mixture (Obla, 2007). Slump test results for pervious concrete would always measure close to zero (inches) due to the nature of the dry mix (Palmer, 2009), therefore a new method of testing is needed to measure pervious concrete consistency. The only existing method for determining the correct consistency of
6


pervious concrete is to roll the sample into a ball in the palm of ones hand and visually judge the consistency (Tennis, 2004). The individual is to look for an even distribution of aggregates and for the concrete ball to hold its shape. This testing method could be effective for a single batch, but very subjective and inconsistent for several batches of concrete. Results will undoubtedly reflect the opinion of the person performing the test and not necessarily reflect the desired outcome. Therefore this method cannot be qualified as the sole testing method for consistency of pervious concrete.
Another possible option which has been used for testing the slump of dry conventional concrete mixtures is the modified vebe apparatus. This method can be reproduced in the laboratory and the field. It is applicable for no or low slump concrete having aggregates smaller than 2 inches (5cm.) (U.S. Army Corps of Engineers, 2001). Further examination of this method is needed to determine if it is a viable test option for pervious concrete.
Perhaps the best current method of testing slump of pervious concrete is the inverted slump cone test. It is an altered slump cone method that is suggested for testing the consistency of pervious concrete (Concrete Promotional Group, 2009). With conventional concrete, a slump cone is the primary quality assurance test for determining the consistency of concrete, but because it is not applicable for pervious concrete, the Portland Cement Association (PCA) developed a new method to use the slump cone that can be applied to pervious concrete (Design of Pervious Concrete Mixtures, PCA-Kevem, 2009). This method attempts to reproduce the effect of pervious concrete flowing down a concrete truck chute. The test method determines workability and changes in consistency in a mix (Missouri/Kansas Chapter of the American Concrete Pavement Association, 2009). The procedure for applying the inverted slump cone test includes:
7


Fill the cone with freshly mixed pervious concrete without any compacting or rodding.
A full inverted slump cone is then lifted and given a mild shake in order to loosen the mix and initiate flow.
If the material begins to flow and drip out of the cone, this indicates that the material will flow from the truck and place correctly. If the mixture stays lodged in the inverted slump cone, this suggests that the pervious mixture will have a difficult time coming out of the truck and is not workable. The mixture not flowing through the inverted cone also suggests that the concrete will have a high porosity and low strength (Kevem, 2009), both undesirable results. Pervious concrete that will not flow through the inverted cone suggests an ineffective mixture that must be corrected prior to placement. It is suggested that one method to alter this mixture onsite is to increase the workability by Add 50% of original dosage of either the water reducer or the hydration stabilizer in addition to 1 to 2 gallons (4 to 8 liters) of water per cubic yard of pervious concrete. (Missouri/Kansas Chapter of the American Concrete Pavement Association, 2009). Additional testing of this method should be performed in order to gain a better understanding and develop control measures.
2.3 Unit Weight
Recently, acceptance of pervious concrete is based on calculated density or the unit weight of the in-place pavement (NRMCA, 2004). ASTM does have a testing standard for density of pervious concrete, ASTM Cl688 / C1688M 08: Standard Test Method for Density and Void Content of Freshly Mixed Pervious Concrete. This unit weight test method is the most useful test for pervious concrete. ASTM Cl688 provides a procedure
8


for determining the in-place density and void content of freshly mixed pervious concrete. The fresh concrete unit weight value is of importance because it has a direct relation to the hardened unit weight value. Unit weight of hardened concrete can be directly linked to porosity and compressive strength. The unit weight test is the only standardized test that can be performed on a fresh pervious concrete.
ASTM Cl688 states to take a sample of fresh pervious concrete and place it inside a standard measure (bucket of 0.25 cubic foot volume or 0.007079 cubic meters). Concrete is then consolidated using a proctor hammer. Proctor hammer is a mechanical weight of 5.51bs (2.5 kg) that is dropped on a specimen (mostly used in soil testing). The density and void are calculated based on measured consolidated concrete mass, volume, and total mass (ASTM Cl688, 2009). However, it is believed that ASTM Cl688 calculates the density of the sample being tested and varies from field conditions ((Palmer, 2009). In the field, the pervious concrete is not consolidated the same manner as the ASTM Cl688 procedure consolidates the test specimen (Palmer, 2009). Different equipment and procedures are followed to consolidate pervious concrete in the field. Contractors typically use rollers, truss or laser screeds, all resulting in different compaction levels (Johnston, 2009). The ASTM standard states that the fresh density and void content calculated from this test may differ from the in-place density and void content, and this test shall not be used to determine in-place yield (ASTM C1688, 2009). This test can be easily and consistently reproduced, but it is not comparable to field conditions (Palmer, 2009). Determining compaction method that can be used in a laboratory testing with similar compaction to field condition is the key to having testing standards.
9


Prior to the publication of ASTM Cl688, unit weight/density of pervious concrete was tested using ASTM 138 (Method for Testing Density in Conventional Concrete). The method called for classic rodding of the concrete inside the sample in order to obtain the desired consolidation (Palmer, 2009). In normal concrete, the aggregates are evenly grated allowing the paste between the aggregates to flow. When conventional concrete is rodded, the mixture can consolidate properly. With pervious concrete, there are little to no fine aggregate in the mixture. A very thin layer of cement paste bonds the coarse aggregates. The result is a mixture with large voids that will not flow when rodded (Offenberg, 2008). Rodding the pervious concrete does not produce consistent results because of the absence of small aggregates and nature of the dry mixture. It is believed that rodding over consolidates the concrete causing the larger aggregates to settle on the bottom. Result became more consistent as water was added to the dry mixture. The concrete paste has more water available, and was able to flow through the pervious concrete voids. However, adding additional water to the mixture would create another problem. Additional water would make the pervious concrete less permeable (Palmer, 2009).
Instead of rodding the concrete, it was suggested to jig the concrete, consolidating the concrete differently inside the sample. Jigging is used as a method in a test for aggregate testing and is specified in ASTM C 29. Jigging can be used to consolidate concrete, when performing both unit weight tests and cylinder tests (Paine, 1992). Instead of rodding the concrete 25 times per layer of concrete sample, one would jig or rock the concrete sample back and forth 25 times in each direction (ASTM C29, 2009). The unit weight measure if filled in three layers with each layer equivalent to one-third of the total volume of the measure. Jigging has proved to be a more effective representation of
10


pervious concrete consolidation as it were to appear in the field. According to the County of Fairfax, Virginia, the suggested quality control test for pervious concrete is the density or unit weight test. Density is determined using ASTM 138, Test Method for Density (Unit Weight), Yield and Air Content (Gravimetric of Concrete) by following the consolidation procedures in ASTM C29 (County of Fairfax, Virginia, 2007). In other words, the County of Fairfax, Virginia specifically recommends the density test be performed using the jigging method. When determining the density of hardened concrete, core samples can be taken in accordance with ASTM C42, Test Method for Obtaining and Testing Drilled Cores and Sawed Beams of Concrete. More than one core shall be taken in order to represent the entire placement site in accordance with ASTM D 3665, Practice for Random Sampling of Construction Materials. After core samples have been obtained, and the thickness determined, the density shall be tested in accordance with ASTM C140 (County of Fairfax, Virginia, 2007). ASTM C140 states the procedure for testing the unit weight/density that can be applied to hardened pervious concrete. The trimmed core samples shall be emerged in water for 24 hours, allowed to drain for one minute, surface water would be removed with a cloth, and the weight measured immediately (ASTM C140, 2009).
2.4 Segregation
Segregation of the components in a mixture results in a non-uniform mixture. Segregation can occur during mixing, transportation, placement, or compaction. Typically in conventional concrete, segregation occurs due to poor handling. For example, coarse aggregates can sink to the bottom and separate from the concrete matrix while the paste rises to the top, often the result of over vibration. One can see that this is
11


even more critical with pervious concrete due to the minimal cement paste content and ability to bond large aggregates to one another. A method in which segregation can be remedied in the field is to add water to the mixture. Typically, one gallon (4 liters) of water is recommended per yard of concrete. Of course, too much water increases the w/cm, thereby decreasing the compressive strength of the pervious concrete mixture (Obla, 2007). Adding water to a segregated mixture should only be a last resort effort in order to save the mixture. If segregation of the aggregates occurs, the necessary paste to aggregate bond will not occur. Because the paste would rise to the surface, it might clog the pores at the top of the slab. If the pores are clogged, water cannot penetrate the pervious concrete slab. In this scenario, the entire slab would be inadequate and need to be replaced. Segregation is expected to be overcome solely by careful handling.
Research conducted by the University of Colorado Denver, found an impervious layer developed about 1/3 the distance from the bottom of the specimen. See Figure 2.1. This impervious layer was found in specimens produced in the laboratory setting was believed to be caused by improper compaction. The addition of a hydration stabilizer helped to prevent this layer from forming in future mixtures (Hager, 2009).
12


Figure 2.1: Impervious Layer Zone (Hager, 2009)
A test has been proposed to monitor segregation. The test involves lOOOg of concrete to be placed through a #30 standard sieve. This sample is then vibrated for 60 seconds. The concrete matter which falls into the bottom pan is then measured. The amount of concrete passing the #30 sieve is used to determine the allowable segregation level (Offenberg, 2009). This test, however, is only mean to be used in a laboratory setting because it is too time consuming. This test is not intended for practical purposes.
2.5 Curing
Proper curing practice is essential to producing quality pervious concrete pavement (Obla, 2007). Curing is an essential step that ensures adequate hydration of cement paste in order to provide sufficient concrete bond and strength (Kosmatka, 2002). Water is a critical variable in pervious concrete and often needs to be adjusted in the field. Too little water can lead to improper curing. The porous surface of pervious concrete exposes more
13


surface area to air than conventional concrete, thus making over-evaporation of the mixture water more likely to occur. If not properly cured, the desired strength will not be reached causing the concrete to ravel and break away (Obla, 2007). Most of the problems that have occurred in practice can be contributed to improper curing practices (Palmer, 2009).
The best scenario in which concrete is cured would be one that allows all the moisture to remain in the mixture, allowing the concrete to cure slowly. Admixtures have been developed that help pervious concrete cure properly such as a hydration stabilizer added to the water. Topically applied coatings exist that help hydrate the concrete during curing. Effects that admixtures have on pervious concrete are not being covered in the scope of this thesis. Currently, it is common curing practice to cover the pervious concrete slab with a polyethylene sheathing to retain internal moisture. A study is currently being performed at the University of New Orleans that focuses on a better curing method for pervious concrete. The study proposed using a spray-on application of soy bean oil between the surface of pervious concrete slab and the polyethylene cover (Offenberg, 2009). Results from this study have yet to be published and may contribute to better curing practices for pervious concrete. Curing is not a property that can be tested. Instead, curing is a step in placement of concrete. Best curing practices should be followed by contractors during construction in order to obtain a quality pervious concrete pavement.
Hagers findings in Sustainable Design of Pervious Concrete Pavements were that proper curing of pervious concrete was even more critical when the pavement was being placed in a dry climate such as Denver, CO. When box specimens were being fabricated for
14


laboratory testing, all were wrapped with 6mm (0.006 inches) thick plastic and placed in a humidity controlled environment to cure. During the placement of the pervious concrete parking lot test pavement, concrete was covered with 6 mm (0.006 inches) thick plastic sheathing. Various weights were placed on top of the sheathing in order to secure the cover from moving. In addition, water was sprayed on top of the concrete during the initial fourteenth day after placement (Hager, 2009). Also, in an effort to accommodate Denvers very dry climate, a hydration stabilizer was added to the pervious concrete mixture used in Hagers study.
2.6 Strength
For conventional concrete, the most recognized and required hardened property is compressive strength. However, in pervious concrete, strength is considered but is not the property by which pervious concrete is accepted (NRMCA, 2004). Traditional concrete is typically tested with cylinder samples and in accordance with the ASTM C39, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens. The cylinder strength test following ASTM C39 is not an accurate method of testing compressive strength of pervious concrete. Due to the voids in pervious concrete, it is difficult to obtain the proper compaction, and thus difficult to determine the true strength of pervious concrete. It is even more difficult to create a high strength pervious mixture by using common materials, and is typically not performed (Fourtes, 2008). Making a cylinder with pervious concrete does not follow the same techniques as those followed when the pervious concrete is being placed in the field. Pervious concrete cannot be consolidated the same way as conventional concrete. Determining a proper compaction procedure would make it possible to test the compressive strength of pervious concrete.
15


In addition to compaction difficulty, pervious concrete cylinders have large voids and are likely to fail prematurely. The axial load applied by the testing machine will cause failure along lines of larger voids within the concrete cylinder. This premature breaking is typical of dry concrete mixtures such as high strength concrete cylinders. Traditionally compacted and tested cylinders are not an accurate method of testing pervious concrete.
Flexural and compressive strength test results greatly depend on the degree of compaction the sample has received (Paine, 1992). A more accurate representation of the compaction found in the field is taking a core sample and testing it for compressive strength. Even though a core sample represents the concrete compaction found in the field, it can still produce unreliable results reflecting low strength. When the core sample is taken, the paste structure of the hardened concrete can be damaged during the coring process (Kevem, 2009). At the University of Colorado parking lot study, cylinders were cored from box specimens for the purpose of compressive strength testing (Hager, 2009).
Cylinder capping may produce strength results for pervious concrete similar to those in the field. Again, the compressive strength result greatly depends on the compaction technique used when the cylinders are made. But, when looking for a better method to test cylinders (with the best compaction method the user assumes), cylinder capping may be a better alternative to testing cylinders traditionally in accordance with ASTM C39. Cylinder capping has been shown to be the better method of cylinder testing with dry high strength concrete mixes from 8,000 to 12,000 psi (Torres, 2006). Cement capping is one of the most widely used methods to determine compressive strength in hollow concrete block masonry. The American Society of Testing Materials recognizes capping as an acceptable testing method for concrete cylinders in ASTM Cl231-09. The capped
16


pads deform in initial loading of the concrete cylinders. The pads are restrained from excessive lateral spreading by metal rings and provide a uniform distribution of loading from the bearing blocks of the testing machine (ASTM Cl231-09). Same principals can be applied to testing pervious concrete cylinders. In a study by Ozyldirim, compressive strength data were obtained from two capped cylinder methods; neoprene pads and sulfur-mortar caps. These two capping methods showed very little difference in compressive strength results (Ozyldirim, 1985). More research is needed to determine the compressive strength differences between neoprene pad capping and sulfur capping of pervious concrete cylinders.
2.7 Permeability
When examining permeability or porosity, there are no applicable test methods that can accurately determine the porosity of pervious concrete in its plastic state. Permeable porosity is a basic measurement of voids in concrete. This can be easily performed once the concrete has hardened as a procedure in the density test ASTM Cl688. Porosity is a percentage of volume of sample divided by a volume of sample with the voids filled with water. Porosity is very difficult to correlate in the field while the concrete is in the plastic state. In Hagers study at the University of Colorado Denver, porosity was tested for the pervious concrete cylinders by using the volume displacement method (Hager, 2009). This procedure was used for Batch lof this study and is described in detail in Section 4.10. A modified procedure was used for porosity testing for Batch 2 and is listed in Section 5.6.
Currently, ASTM is working on a standard test method that will field test permeability, titled ASTM WK17606 New Test Method for Field Permeability of Pervious Concrete
17


Pavements. Pervious concrete contains a porosity ranging between 15-25% (ACI 522, 2009). This method is a quality control test that can determine the level of maintenance that the pervious concrete requires. All pervious concrete is recommended to be cleaned and inspected on yearly bases. ASTM WK17606 monitors clogging and long term permeability of pervious pavements. (ASTM WK 17606).
2.8 Durability
ASTM is developing a standard to test durability. ASTM WK23367 is a new test method that evaluates the surface durability potential of a pervious concrete mixture. According to ASTM WK23367, the concrete producers need a tool to assess the impact of using different raw materials to make pervious concrete. Additionally, raw material suppliers need a way to access the impact of their raw materials in pervious concrete. This test is not intended to be used for acceptance. A factor of durability is with concrete resistance to freeze thaw cycles. ASTM C666 provides a method for testing resistance of conventional concrete to freezing and thawing cycles. The procedure calls for the concrete specimen to be submerged in water for a period of time and subjected to freeze and thaw cycles (ASTM C666, 2003). This testing standard, however, is not recommended for pervious concrete because it is not believed to be accurate. The repeated freezing and thawing stresses are to be resisted by the cement paste alone. Typically, pervious concrete does not contain air entrainment like conventional concrete to help resist the freeze thaw cycles. The paste thickness that surrounds the aggregate is very low, causing a rapid deterioration of the specimen. In fact, research conducted by Bass showed damage was found to be quite dramatic when specimens were subjected to ASTM C666 test (Bass, 2008). Pervious concrete is never likely to be submerged fully
18


with water. Rainwater would drain through the pervious concrete and the soil below. At the University of Colorado parking lot study, ASTM C666 was used to determine freeze thaw resistance. Due to porous nature of pervious concrete, the transfer frequency could not be measured and it was recommended that an alternate method was developed to determine the freeze thaw resistance of pervious concrete (Hager, 2009). In this study, Hager recorded the mass loss of the freeze/thaw beams after each 28 cycles. Thus, mass loss versus number of cycles was used to predict the pervious concretes resistance to freeze-thaw. Though the ASTM C666 test is an aggressive test for pervious concrete, the method does provide an indication of performance when compared to other pervious concrete mixtures. For example, if two mixtures are subjected to this test method and one mixture experiences less mass loss at an extended number of freeze/thaw cycles than another mixture, the overall performance can be examined. A modified version of ASTM C666 is recommended for determining the durability of pervious concrete. A modified version may include using Procedure A (freezing and thawing the pervious concrete when the sample is not submerged). Additional research is needed in this area.
2.9 Infiltration Rate
Since pervious concrete is expected to allow rainwater to pass through it, infiltration rate is another factor that is of importance when rating pervious concrete. ASTM C1701/C1701M 09: Standard Test Method for Infiltration Rate of In-Place Pervious Concrete tests the infiltration rate of hardened pervious concrete but does not address the flow of water through the plastic mixture. This test method does not provide a method to predict infiltration rate of fresh concrete. This test is designed to identify the need for maintenance of an existing slab. All pervious concrete needs to be cleaned and
19


monitored yearly and this test provides guidance for that. One of the limits to this test is that the infdtration rate obtained by this method is valid only for the area of the pavement being tested. If the entire pavement was to be tested, multiple test locations are necessary and the results in that situation are averaged. This test method does not, however, measure the influence on in-place infiltration rate due to clogging of voids near the bottom of the pervious concrete slab. It is recommended that visual inspection of concrete cores is the best approach for determining whether there sealing of voids is present (ASTM 1701).
At the University of Colorado Denver pervious concrete pavement test section, the drain time of a pavement was measured using the clogging test (Hager, 2009; Delatte et. al., 2007). The cogging test measured the length of time for a given volume of water to drain thorough the pervious concrete pavement. The test equipment is shown in Figure 2.2. In this procedure, a typical concrete compressive strength cylinder mold with a hole and stopper in the bottom was utilized. The mold was filled with water, the stopper was removed, and the time required to empty the mold was recorded.
Figure 2.2: Drain Time Testing Apparatus (Hager, 2009)
20


2.10 Research Limitations and Scope
This thesis focuses on compaction methods in laboratory testing for pervious concrete. Current ASTM pervious concrete standards can be successful in providing valid data if a compaction method is determined that is similar to field conditions. All pervious concrete has a site specific need and function depending on the area of the country, its purpose, and soil characteristics. This research does not address the pervious concrete system as a whole, but the pervious concrete material itself. For example, a clay soil will not allow the water to penetrate as quickly as a sand soil, and therefore, the pervious concrete being placed over clay soil may require additional sub-grade preparation. In addition, the sandy soil may be very dry and would require proper measurements to ensure the pervious concrete does not lose moisture in the bottom layers of the pavement. Different results will occur in the laboratory then the field due to varying field conditions. Adjustment should be made to the conclusions of this paper before applying it to any field condition.
2.11 Pervious Concrete in Colorado
Pervious concrete has been used as a pavement material in Colorado, and partial failures have occurred at these specific locations: Safeway Grocery store parking lot located at SE comer of 13th Ave. & Kremeria St., Denver, constructed in 2005: Walmart Super Center parking lot located at NW comer of Tower Rd. & 1-70, Denver, constructed in 2006; and Vitamin Cottage parking lot, located NW comer of Colorado Blvd. & Evans Ave., Denver, constructed 2007 (Hager, 2008). In June of 2008, the Urban Drainage and Floor District of Denver (UFFCD) set in place a temporary moratorium on pervious concrete placement. The UDFCD stated that the moratorium was set in place due to partial failures of pervious concrete pavements around Denver metropolitan area such as
21


the Safeway, Walmart, and Vitamin Cottage previously named. Forensic investigation was launched by UDFCD to determine the cause of these failures. Speculation to the cause of failures was chemical reaction of magnesium chloride (deicing agent common in Denver), abrasions due to traffic, freeze/thaw conditions, improper mix design and placement, and/or curing techniques. This moratorium was removed later in 2009 shortly after the publication of Specifiers Guide for Pervious Concrete Pavement Design (Bush et. al, 2009).
2.12 Compaction
Pervious concrete is inherently difficult to test and verify in-situ properties from field-placed samples or cores (Kevem, 2009). The existing unit weight testing standard ASTM 1688 are planned to be adopted by the ACI. Even the new ASTM 1688 standard only addressed one compaction method. The ASTM prescribed jigging will be evaluated as compaction method 2 (procedure explained in Section 4.4). The compaction method used governs the level of consolidation in a fabricated cylinder or other concrete mold. Field consolidation and laboratory consolidation need to be as close as possible in order to obtain valid data. Performance of pervious concrete can be predicted by using unit weight of the fresh mix (Kevem, 2009). However, in order to achieve the desired unit weight, the best and most appropriate compaction method should be identified.
Figure 2.3 shows the compaction of the pervious concrete pavement at the University of Colorado Denver. In this parking lot test pavement, the compaction method that was chosen included the use of a steel roller screed. During fabrication of cylinders for laboratory testing, the compaction method used was the rodding method (Hager, 2009).
22


The rodding method is the compaction method used for traditional concrete and is
referred to as method 1 in this study.
Figure 2.3: Strike off and Compaction of Pervious Concrete with Steel Roller Screed (Hager, 2009)
Compaction of pervious concrete during fabrication is directly related to its properties such as porosity, unit weight, and compressive strength. If the specimen receives a high level of compaction, it is expected to have lower porosity and higher unit weight and compressive strength. If a design unit weight is reached while concrete is still in its fresh state, it will govern the unit weight of the pervious concrete in its hardened state. If one can predict hardened unit weight, he also can predict the porosity and compressive strength. Using a correct compaction method controls the outcome of the final pervious concrete pavement. Fabricated samples or cores can be used to determine hardened concrete properties. Different compaction methods shall be examined to determine unit weight, porosity, and compressive strength.
23


3 PROBLEM STATEMENT
One way to reduce or eliminate storm water ponds is to incorporate pervious concrete as the pavement choice for parking lots. Captured storm water can seep through the pervious concrete and directly into the ground replenishing groundwater for existing landscape. Pervious concrete is a material that some owners or contractors shy away from because of the lack of testing standards needed for quality control.
Even though pervious concrete has been used sparingly in the United States since the 1970s, it has recently gained popularity due to its acceptance as a sustainable building material. Pervious concrete can be found as an environmentally responsible material in projects ranging from small sidewalks and driveways to large parking lot in commercial complexes. Economically pervious concrete can be a more affordable choice than conventional concrete, particularly if it were to replace for example, a collection pond. Though the installation of pervious concrete typically requires more soil preparation, good placement techniques, and proper curing, the overall labor involved in the installation process is similar or lower that of conventional concrete because it does not require the finishing steps necessary of conventional concrete.
Pervious concrete has a concrete matrix with little or no fine aggregates. Primarily, pervious concrete contains coarse aggregates, cement and water. Minimal sand up to 10% by weight of total aggregate can be added to improve strength. Typically, pervious concrete contains a void ratio between 15%-25%. This void structure can allow three to five gallons of water per minute for every square foot (1.0 1.75 Liters/min/m2) of pavement. Unlike conventional concrete, pervious concrete is not specified or accepted
24


based on strength. Currently, there are no standardized test methods to provide the tools
for quality control and assurance.
The best means to ensure good performance of pervious concrete is through the use of a well qualified and experienced contractor. Testing standards are not in place to ensure consistency and accuracy in the field. The only recommendations in place for quality pervious concrete placement are those judged by visual observation of an individual. Figure 3.1 shows the variance in consistency that a contractor would typically use to visually judge pervious concrete mix. The criterion for just right concrete varies with the individual and is subjective, therefore an easily reproduced standardized test needs to be developed in order to ensure quality placement.
Figure 3.1: Consistency by Visual Observation (Tennis, 2004)
Many different test methods have been proposed for testing pervious concrete in the fresh and hardened state. However, standardized tests need to be able to easily measure all the properties that are essential for a quality pavement such as porosity, density, and
25


compressive strength. These tests need to be easily reproduced and consistent across the board regardless of testing technician.
Because pervious concrete is a low slump concrete with water to cement ratios (w/cm) between 0.25 -0.35, the slump test does not apply as a valid test method. A slump test is typically a field quality control item that tests consistency during placement. Compressive strength of concrete is tested using standard molded cylindrical specimens. When pervious concrete cylinders are tested, the result is an inaccurate representation of the strength of concrete in the field. Pervious concrete aggregate bonds greatly depend on the confinement within the slab and this bond cannot be found along the perimeter of a cylindrical specimen. The highly porous cylinder shear prematurely along one of its voids and would not resemble that of a core sample. Cylindrical compression tests are not recommended because of the dependency of the result on the compaction of the concrete. It is believed that the compaction in cylindrical specimen does not resemble the same compaction achieved in the field with heavy machinery. A compressive test should exist that resembles the field conditions as closely as possible. Field tests are not in place for verifying the density of pervious concrete. Unit weight is a test performed only as part of post construction inspection. The most important criterion for pervious concrete is the void ratio. Currently, there are no methods to test the void ration until the mixture has already hardened.
Evaluation of existing and development of new testing standards is necessary for pervious concrete. The compaction method is directly linked to concrete properties such as unit weight, compressive strength, and porosity. The more compaction a specimen receives, the higher the unit weight, higher the compressive strength, and lower the
26


porosity. Unit weight is the only property that can be tested while pervious concrete is still in the fresh state. Being able to test concrete before placement allows an opportunity for field monitoring and correction. Unit weight of pervious concrete greatly varies depending on the compaction method used. Unit weight of fresh concrete is linked to the unit weight of hardened concrete, compressive strength and porosity. A compaction method should be developed that is best suited for pervious concrete. Determining a compaction method for pervious concrete will improve the accuracy of the standardized unit weight test. An accurate unit weight value can be used to monitor the concrete during placement and predict hardened concrete properties.
27


4 RESEARCH PLAN
4.1 Concrete Consolidation and Effects on Other Properties
Consolidation is a process by which concrete is compacted and decreases in volume. Consolidation occurs when stress is applied to the concrete causing the paste to pack together tightly and shrink in volume. Proper consolidation is a step in measuring unit weight and fabricating cylinders for compressive strength tests. As noted previously, pervious concrete consolidates differently from conventional concrete. For this reason, traditional testing methods cannot be applied. Consolidation is directly related to the unit weight, strength, porosity, and overall performance of pervious concrete. Since compaction of the pervious concrete greatly influences concrete properties, an acceptable method of laboratory compaction must be developed. This research evaluated the following consolidation techniques:
Method 1: Traditional consolidation following ASTM Cl38 where cylinder is rodded 25 times
Method 2: Jigging method described in ASTM C29
Method 3: Compaction as a percentage of volume. Compaction percentage is determined based on the thickness of the slab. Compaction tool is used
Method 4: Weight vs. Volume Method
In order to create a consistent consolidation method for testing pervious concrete in the laboratory, it is necessary to understand how concrete is consolidated in the field. Pervious concrete is consolidated in the field by using compacting equipment different than that used for conventional concrete.
28


Two of the most common field consolidation methods are:
Pervious concrete slab is roughly finished slightly above desired elevation. A weighted roller is driven on top of the concrete until a desired final density is reached.
Pervious concrete slab is compacted using a roller screed.
Both of these common methods compact the concrete according to Table 4.1 (Kevem, 2009).
Table 4.1: Compaction Factors for estimating in-situ density (Kevern, 2009)
Slab thickness (in.) Compaction
4 25%
6 17%
8 13%
When pervious concrete is being compacted for any test (unit weight or cylinder strength), it should represent the field condition. A specific test does not exist that can correlate the cylinder compaction to the field compaction. In the field, the pavement percentage of compaction is listed in Figure 4.1. Compaction method 3 follows this percentage chart as a guide for compacting cylinders for laboratory testing. The compaction method used to fabricate concrete specimens for laboratory testing should simulate field condition. A recommendation for future research is to examine the difference in compaction between field condition and laboratory specimens.
To avoid unnecessary variance that could occur from one batch to the next, one large batch was made. Four different compaction methods were used to fill concrete cylinders and a concrete box. This thesis focuses on consolidation methods in order to test unit weight, compressive strength, and porosity. The concrete box was used to create additional specimens. Four drilled cores were obtained from each box. Cored samples
29


were compared to cylinders for each compaction method. The cores and cylinders were compared based on measured unit weight, porosity, and compressive strength.
To get consistent data, each consolidation method needed to fill 6 cylinders and 1 box. Unit weight was taken for each method. Unit weight for all methods was taken for all specimens by following ASTM Cl38 with some modifications. The difference of unit weight procedures was the method of compaction. Method 1 uses the traditional rodding compaction method and is explained in Section 4.3. Method 2 follows the ASTM 1688 jigging compaction procedure and is explained in Section 4.4. Method 3 uses a compaction tool and is explained in Section 4.5. Method 4 is the weight vs. volume compaction method and is explained in Section 4.6. After the concrete cured for 28 days, unit weight/density, porosity, and compressive strength were measured. Unit weight was calculated once more at 28 days for all specimens and compared to fresh unit weight. Unit weight of hardened concrete was calculated by dividing the weight of specimen by the measured volume. Porosity was calculated as a percentage of concrete voids and the procedures are described in Section 4.10.
Tables 4.2 and 4.3 lists the amount of cylinders and boxes needed for each consolidation method. Samples within each method were named A, B, C, etc. Each sample is listed under the test that was performed. It is to be noted that method 4 requires a different number of cylinders. Additional cylinders were fabricated for methods 1, 2, and 3 in order to saw cut the cylinder and examine density as a function of cylinder height. Since method 4 specimen are 4in. x 4in. (10.16 x 10.16 cm) cylinders (half size), the saw cut test was not performed on method 4. The wet saw used for cutting cylinders into 1/3s is not intended for cutting lin. (2.54 cm) strips.
30


4.2 Mix Design
Batch 1 was made large enough to have concrete for all the methods. Each method consisted of 6 cylinders, 1 box from which 4 core samples were drilled, and one unit weight. An additional 15% was mixed in order to accommodate waste. Table 4.4 provides the necessary volume for batch 1. Total volume of concrete batched was 3.41 ft3.
Table 4.2: Summary of Specimens to be Created for Consolidation Methods 1, 2
&3

Total Made Neoprene Pads Sulfur Capped Saw Cut 1/3 height UW Test Porosity Test Compressive Test
Cylinders A, B, C, D, E,F A, B C, D E,F A, B, C, D, E, F A, B, C, D, E, F A, B, C, D
#of Boxes 1 box made
Cores from each box A, B, C, D A, B, C, D A, B, C, D
Table 4.3: Summary of Specimens to be Created for Consolidation Method 4

Total Made Neoprene Pads Sulfur Capped Saw Cut 1/3 height UW Test Porosity Test Compressive Test
Cylinders A, B, C, D, A, B C, D A, B, C, D, A, B, C, D, A, B, C, D
# of Boxes 1 box made
Cores from each box A, B, C, D A, B, C, D A, B, C, D
31


Table 4.4: Summary of Total Specimens Needed
Testing Specimens Required # Volume (ft3)
Compressive cylinders 24 1.40
Boxes filled for cores 4 1.37
Total 3.02
x 1.15 3.47
The w/cm remained constant at 0.30. This value is within the range of typical pervious concrete. The design density was at 124 lb/ft3 (1986 kg/m3). Fine aggregate was not used for the mixture. From the above criteria, the mix proportions were calculated and are shown in Table 4.5
Table 4.5: Mixture Characteristics
w/cm 0.30
Unit Weight 124.6 lb/cf
Cement 550 lb/cy
Rock 2636 lb/cy
Water 179 lb/cy
Inverted slump cone test was used prior to fabricating specimens for each method. The inverted slump cone procedure follows:
Fill the cone with freshly mixed pervious concrete without any compacting or rodding.,
A full inverted slump cone is then lifted and given a mild shake in order to loosen the mix and initiate flow.
32


4.3 Consolidation Method 1: Traditional Consolidation
Consolidation Method 1
Method 1 follows the consolidation used for conventional concrete. The concrete was rodded.
Objective:
Objective was to record unit weight for method 1 of the pervious concrete mixture. Consolidation method used for method 1 was in accordance with ASTM Cl38. In each procedure step, a note may have been added on slight differences in procedure.
Prediction:
It was expected that the unit weight will be very low. The large voids in the specimens were expected to low unit weight, very high porosity, and very low strength. This prediction pertains to cylinders and drilled cores.
Terminology:
Yield volume of concrete produced per batch, cubic yard, or cubic meter Air content percentage of air voids by volume of concrete Apparatus:
Measure a cylindrical metal watertight measure, a yield bucket (0.25 ft3 or .007079 m3)
Tamping Rod (5/8in. or 1.58 cm diameter)
Mallet rubber, weighing approximately 1.25 lb (2.75 kg)
Flat plate Procedure:
1. Weigh the empty measure.
33


2. Fill the measure with freshly mixed concrete in three layers of approximately
equal volume. Rod each layer 25 times with the tamping rod.
Note: This procedure step was altered for batch 2 and is listed in Section 5.6 under title Bettering Procedures for batch 2. It is to be noted that cylinders, the unit weight bucket, and the box were all rodded 25 times at each 1/3 for batch 1.
3. After each layer is rodded, tap the sides of the measure 10-15 times with the
mallet (this procedure is required to release any large trapped air bubbles).
After consolidation, the measure must not contain any excess of concrete
protruding above (approximately 1/8 inch) the top of the yield bucket.
Note: Striking off was difficult to do with pervious concrete, instead, the best level surface was created with a flat board. It is to be noted that it is not possible to strike off pervious concrete.
4. Strike off the top surface with a sawing motion of the flat trowel (using little vertical pressure).
5. Clean all excess concrete from the exterior of the measure (use a dampened towel if necessary, and then dry). Note: Pervious concrete needs significantly more cleaning than regular concrete. Wet towel was used.
6. Weigh the measure with concrete.
7. Calculate the unit weight of concrete as the ratio between weight of concrete and measure volume:
yconcrete = Wconcrete/ Vmeasure
where: yconcrete = unit weight of concrete (lb/ ft3)
Wconcrete = net weight of concrete (lb)
Vmeasure = volume of measure (ft3)
34


4.4 Consolidation Method 2: Jigging Method Described In ASTM C29
Consolidation Method 2
Consolidation method 2 was in accordance with the jigging procedure.
Objective:
The objective was to record unit weight for method 2 of pervious concrete mixture. Consolidation method used for method 2 was in accordance with ASTM C29 using the jigging method. In each procedure step, a note may have been added on slight differences in procedure.
Prediction:
It was expected that the unit weight will be more consistent and higher than unit weight in method 1. Consolidation method 2 was also expected to have lower porosity and higher strength than method 1.
Apparatus
Measure a cylindrical metal watertight measure (0.25 ft3 or 0.007079 m3) Procedure:
1. Fill measure in three approximately equal layers (by volume), compacting each layer by placing the measure on a firm base, such as a concrete floor, raising the opposite sides about 2 inches (5 cm), allowing the measure to drop in a such manner as to hit with a sharp, slapping blow.
2. Compact each layer by dropping the measure 50 times, 25 times on each side.
Note: The procedure was followed exactly when filling the unit weight bucket and when consolidating the box. This procedure was also followed exactly when filling the cylinders. However, it is to be stated that it was apparent that dropping a heavy 12x12x7 inch (30.48 x 30.48 x 17.78 cm) box 2 inches (5 cm) 25 times each side of each layer produced a much higher compaction than when following that method for cylinders.
35


3. Level the surface of concrete with fingers or a straight edge.
Note: It was difficult to level off pervious concrete. A straight edge was attempted, but ultimately, a flat square plate was used to level off the box and unit weight bucket. While plate leveling may produce additional compaction, this was performed consistently for all the methods.
Calculation
1. Calculate the unit weight, ybulk = (G T) / V where:
ybulk = unit weight of the aggregate, lb/ft3 G = mass of the mixture plus the measure, lb T = mass of the measure, lb V = volume of the measure, ft3
4.5 Method 3: Compaction As A Percentage Of Volume
Consolidation Method 3
Consolidation method 3 was similar to the traditional compaction method 1. The only difference in the procedures was that method 1 uses a rod and method 3 used a compaction tool. Picture of a compaction tool is in Figure 4.1.
Objective:
The objective was to record unit weight for compaction method 3. Consolidation method 3 used a compaction tool seen in Figure 4.1. Compaction method 3 follows ASTM C138, with the exception listed in the procedure. In each procedure step, a note may have been added on slight differences in procedure. Compaction tool was defined as solid piece of steel tubing (without the interior
36


void) that has a level surface on the striking side. Compaction tool is not weight dependant.
Compaction Tool
Figure 4.1: Compaction Tool (Offenberg, 2009)
Prediction:
It is expected that the unit weight will be higher than that for methods 1 and 2. Procedure:
Follow method 1. Instead of rodding the sample during compaction, use the compaction tool. Same procedure applied for cylinders compacted with the compaction tool. The only difference is the tool being used; compaction tool vs. rod. Figure 4.1 shows a typical compaction tool. This particular tool has a round plate with a diameter slightly smaller than that of a cylinder so it can easily be used for cylinder compaction.
1. Weigh the empty measure.
2. Fill the measure with freshly mixed concrete in three layers of approximately equal volume. Tamp each layer 25 times with the compaction tool.
37


Note: This procedure step was altered for batch 2 and is listed in Section 5.6 under title Bettering Procedures for batch 2.
3. After each layer is tamped, tap the sides of the measure 10-15 times with the mallet (this procedure is required to release any large trapped air bubbles). After consolidation, the measure must not contain any excess of concrete protruding above (approximately l/8in. or .317 cm) the top of the yield bucket.
4. Strike off the top surface with a sawing motion of the flat plate.
Note: Striking off was difficult to do with pervious concrete, instead, the best level surface was created with a flat board. It is to be noted that it is not possible to strike off pervious concrete.
5. Clean all excess concrete from the exterior of the measure (use a dampened towel if necessary, and then dry).
Note: Pervious concrete needs significantly more cleaning than regular concrete. Wet towel was used.
6. Weigh the measure with concrete.
7. Calculate the unit weight of concrete as the ratio between weight of concrete and measure volume:
yconcrete = Wconcrete/ Vmeasure
where: yconcrete = unit weight of concrete (lb/ ft3)
Wconcrete = net weight of concrete (lb)
Vmeasure = volume of measure (ft3)
38


4.6 Method 4: Weight vs. Volume Method
Consolidation Method 4
Consolidation method 4 is the weight vs. volume method. A weighted amount of concrete was placed into a predetermined volume. Concrete was compacted as much as necessary to be able to be placed inside the volume.
Objective:
The objective was to record unit weight for method 4 of pervious concrete mix. Four cylinders and one box were filled per compaction method 4. The basic concept of method 4 was to compact a predetermined weight of concrete into a volume by striking it as many times as necessary. The concrete weight was determined by taking the desired density value and multiplying it by the volume. Then, the weighted concrete is place into the volume and compacted with compaction tool as necessary.
Prediction:
It was expected that the unit weight for method 4 will be higher than method 1 and 2. It is uncertain how it will compare to method 3. The porosity for the method 4 was expected to be less than any of the other methods. In addition, since method 4 proposes 4 x 4in. (10.16 x 10.16 cm) cylinders instead of 4 x 8in (10.16 x 20.32 cm) cylinders, it was expected that the concrete tests higher in strength than any of the other methods.
Procedure:
This method was proposed by Matthew Offenberg, U.S. Technical Service Manager at W. R. Grace & Co. (Offenberg, 2009). According to his experience, 4 x 4 in. (10.16 x 10.16 cm) cylinders are a best representation of the field
39


conditions. Instead of filling six 4 x 8 in. (10.16 x 20.32 cm) traditional cylinders, six 4 x 4 in. (10.16 x 10.16 cm) cylinders will be filled instead. In order to keep consistent with the other methods, a 12 x 12 x7 in. (30.48 x 30.48 x 17.78 cm) box will be filled as well to provide 4 full height samples.
Proposed procedure for compacting 4 x 4 in. (10.16x10.16 cm) concrete cylinders are as follows:
1. Place cylinder on the scale and set scale to zero.
2. Mark with dark marker a line that represents a 4 in. (10.16 cm) height inside the cylinder.
3. Fill the cylinder until the scale reads a desired weight.
Density (psi) = Weight of concrete (lbs)/Volume of 4 x 4 in.
(10.16 x 10.16 cm) cylinder (inch2).
4. Compact cylinders with compaction tool until it reaches the marked 4in. height.
Note: Cylinders are approximately half full and should weigh the same amount.
Procedure for filling the 12 x 12 x 7 in. (30.48 x 30.48 x 17.78 cm) box:
1. Place box mold on the scale and zero.
2. Fill the box full, no need to weigh the box at this point.
3. Compact the box with compaction tool until the concrete settles 'A inch (1.22 cm) below the top surface in order to make room for additional concrete.
4. Place the box on the scale and fill with additional concrete until the desired calculated weight is reached.
40


5. Take the box off the scale and place on a level surface (such as concrete floor).
6. Compact with compaction tool until the concrete is level with the top edge of box.
Note: Since compaction level can depend on strength of the strike, the number of strikes is not important, only the weight of the box and the volume.
7. Do not strike off. The entire weighted concrete must occupy the volume.
Compact as many times as needed in order to fit the concrete in its volume.
Note: The same method can be followed for the unit weight bucket, or any shape of desired concrete mold.
The procedures in method 4 were developed by studying the ideas of Matthew Offenberg. Matthew Offenberg is the internationally recognized expert in the field of pervious concrete. Offenberg is a chair of the American Concrete Institute (ACI) 522 committee, the founding chair for National Ready Mix Concrete Association (NRMCA), and is currently the secretary for American Society for Testing and Materials (ASTM) committee C09.49 for pervious concrete. Offenberg holds a bachelors and masters degrees in civil engineering from Purdue University and is registered as a professional engineer. Offenbergs test was performed for the purpose of testing durability for pervious concrete, not compaction. In this thesis, the focus is on compaction methods only. Durability is a property that can be directly linked to compaction. The more compact a section is, the higher unit weight it will have, the smaller porosity, and the higher durability. The procedures performed by Mr. Offenberg are listed:
1. 4 x 4 in. diameter cylinder is cast by weighting the amount of concrete that would produce the desired density for that volume.
41


2. The filled cylinder is dropped on the floor ten times from 1 in. height.
3. Cylinder is covered and cured for 7 days.
4. Concrete specimen is removed from cylinder mold at 7 days.
5. Single specimen is placed in L.A. Abrasion machine and rotated 50 times.
6. Mass is determined of remaining specimen.
7. Mass loss is calculated
Mass loss % = (initial mass- Final Mass)/Original Mass) 100
Again, the proposed pervious concrete durability test procedure was not performed in method 4. However, the procedure was the bases of developing the compaction method 4 used for this study.
42


4.7 Compressive Strength
Four boxes were fabricated using each method. A total of 4 drilled cores were removed from each box. In addition, six cylinders were fabricated following methods 1, 2, and 3. Four cylinders were fabricated following method 4 because saw test was no performed on method 4. All cylinders and cores were tested for compressive strength at 28 days of age for batch 1.
Cylinders A and B for each method were testing using Neoprene pads
Cylinder C and D for each method were tested using Sulfur caps
Cylinders E and F for testing methods 1, 2, and 3 were cut into 1/3s. The
porosity will be measured for the bottom 1/3, middle 1/3, and the top 1/3.
Note: During cutting of the specimens, the aggregates unraveled along top and bottom surfaces. This was believed to have created artificially low porosity for top and bottom thirds of the specimen. This data was not used.
The four core samples will be tested using Neoprene pads
Each of the four consolidation methods had a concrete box made. The concrete box was large enough to core four cylinders from. Two of the cores were tested using neoprene pads and two using sulfur caps.
The following is the procedure for making concrete cylinders:
Objective:
Cylinders and cores were tested by following ASTM C39 05: Compressive Strength of Cylindrical Concrete Specimens Predictions:
1. It was expected that all the cylinders from method 1 fail prematurely under the compression testing machine. Since previous concrete has many large voids, it was expected that the failure plane would be along these voids. The
43


axial load would not be evenly distributed because of the void structure. The uneven load distribution and the presence of larger voids were expected to cause the specimen to break prematurely giving it a lower compressive strength value.
2. Methods 2, 3, and 4 were expected to have higher compressive strength values.
3. It was expected that compressive values would be between 400 1500 psi (4.1-10.48 MPa). These values are typical for pervious concrete not containing fine aggregates and chemical admixtures.
Procedure:
These procedures are followed for conventional concrete cylinders. This method
was followed for this study. The procedures that were followed were in
accordance with ASTM C39. The Specimens were compacted with different
compaction methods, but tested for compression following the same procedure.
1. Cylinders and box was taken out of the curing room.
2. Cores were drilled out of the boxes with typical 3inch coring drill.
3. Measured diameter and length for each specimen
4. Place neoprene pads confined by steel end caps on either end of the specimen. Only do this step for specimens that are to be tested with neoprene pads.
5. Place the concrete specimen into the compressive testing machine.
6. Zero the compressive testing machine
44


7. Use the metered advance mode and slowly apply load until the specimen breaks.
8. Calculate the compressive strength per the calculation in ASTM C 39 [2005],
4.8 Neoprene Pads
Objective:
Cylinder samples A, B and core samples A, B, C, and D were tested with neoprene pads in accordance with: ASTM C 1231/C 1231M 09: Use of Unbonded Caps In Determination of Compressive Strength of Hardened Concrete Cylinders.
Predictions:
1. It was expected that cylinders and cores tested with neoprene pads would have lower results than those tested with sulfur caps. Since Sulfur Capping was not successful, this theory was never proved.
2. All the methods are expected to have cores and cylinders with uneven surfaces.
Procedure:
Procedure followed ASTM C 1231 -09.
4.9 Sulfur Capping
Sulfur Capping:
Sulfur capping is a method that caps hardened concrete cylinders or cores with a high strength sulfur mortar. Sulfur capping provides a plane surface on ends of hardened specimens.
45


Objective:
Cylinder samples C and D for each method were sulfur capped. No core samples were sulfur capped. The sulfur capped cylinders were tested for compression. The liquid sulfur will be made in laboratory, and the cylinders capped forming a solid, smooth surface for each specimen.
The level surface will accommodate the pervious concrete void at the top of the cylinder, helping distribute the weight more evenly throughout the cylinder. Predictions:
It is expected that all the sulfur capped cylinders will have higher compressive strength than the cylinder tested with neoprene pads.
Procedure:
The procedures will be in accordance with ASTM 617 and ASHTO T 231
1. Sulfur chips were placed in a crock-pot fdling it % full. The croc-pot was left on high until the chips melted.
2. Capping plate was cleaned and placed on a level surface.
3. Capping plate was filled 14 with liquid sulfur.
4. Pervious concrete specimen was dipped inside and left to cool for 5 to 15 seconds.
5. The capping plate was hit with mallet to loosen the hardened cap from the cap mold.
6. Capped specimen was lifted out of the capping mold.
46


4.10 Porosity Testing
Porosity:
Porosity is taken for concrete specimens at 28 days of age. Porosity is not measured in the mixtures fresh state. Porosity is the percentage of volume of voids in the volume of concrete specimen.
Objective:
All cylinders and drilled cores for all four methods will be tested to obtain porosity.
Predictions:
The higher the porosity, the lower the unit weight and compressive strength. Method 1 is expected to have the highest porosity. Methods 2, 3, and 4 are expected to all have lower porosity than method 1.
Procedure:
1. Obtain a container that can house the specimen. For 4 inch (10.16 cm) wide diameter specimens, a 6 inch (15.24 cm) wide plastic cylinder was used in order to prevent the specimen from getting stuck inside the 4 inch (10.16 cm) mold during testing.
2. Measure the diameter and length of concrete specimen.
3. Place specimen inside the container
4. Slowly pour water into the container until the water surface reaches the top of the concrete specimen. Do not fill the container full with water.
47


5. Leave the concrete specimen in the container for 30 seconds.
6. Drain the water into a graduated cylinder and measure the volume of water.
7. Calculate the porosity as the Volume of Water divided by Volume of Concrete.
48


5 EXPERIMENTAL RESULTS
5.1 Experimental Procedures
One large batch was made to test all four consolidation methods. The mixture was very dry but did not present any of the problems that are typical when dealing with a low w/cm mixture. The mixture did not show any signs of clumping (as common of a low w/cm mixture). Small amounts of water were added to ensure water distribution throughout the concrete batch. The concrete mixer was at approximately half capacity and was tilted during mixing at approximately 35 degrees. Samples for method 1 through 4 were fabricated in chronological order. This was to be noted because method 1 samples were coarser than samples for method 2, 3 or 4. Because the batch was very large in volume, the concrete mixer could not be tilted such that the opening was parallel to the floor. Concrete removed from the mixer first was used for fabrication of method 1 cylinders. The larger aggregates came out of the mixer first, resulting in method 1 cylinders to have larger aggregates that methods 2, 3, or 4. This error was not apparent during cylinder fabrication, but rather at 28 days. It was expected that method 1 would have even lower compressive strength values than expected because of the uneven aggregate gradation. Method 1 and method 2 cylinders were created at the beginning of the batch and two more were created toward the end of the batch. Figure 5.5 shows the variation in the gradation of the aggregates between the beginning and the end of the compacting process. Figure 5.1 shows the drastic difference in aggregate size from sample A and F for method 1.
49


Figure 5.1: Visible Variation of Aggregates
As indicated in Figure 5.1, sample A from method 1 (located on the left side of the photo) has significantly larger aggregates than that of sample F of method 1 (located on the right side of the picture). This was a source of error due to the uneven aggregate distribution that occurred during the large batch. Segregation such as this could be avoided if a smaller volume of concrete was batched.
When following consolidation method 1, it appeared that the rod made voids in the cylinders that were irreversible. The rodding appeared to create additional large voids and did not help consolidate the concrete.
When following consolidation method 2, jigging the concrete did not seem to be beneficial while filling the cylinders. The cylinders were rocked back and forth as stated by the jigging procedure. However, the dry mixture did not shift enough to aid consolidation. The experience was different when jigging the box. Because the concrete
50


box was much heavier, it was difficult to jig the box back and forth as the jigging procedure stated. Instead, each side of the box was lifted (rocked), but due to its awkward size and weight, the comers were dropped on the floor each time. It is likely that the large box was additionally compacted by the floor with each drop. It was expected that the cored samples would be compacted significantly higher than the cylinders for method 2.
When following consolidation method 3, compaction was visually observed as the volume decreased. The compaction tool was dropped 25 times on each cylinder. The cylinders were not compacted in layers. After a cylinder was compacted, the top of the concrete was about 2 (5 cm) below the top of cylinder mold. The cylinder was then topped with additional concrete and compacted again 5 times.
For consolidation method 4, the calculated concrete weight was placed in the cylinder. Each cylinder was compacted using the compaction tool. Some were compacted 10 times, some were compacted 40 times. The point of this procedure was to compact the concrete specimen to the desired weight, disregarding the number of compaction hits. The cylinders were filled and compacted without too much effort. However, when compacting the 12 x 12 x7 in. (30.48 x 30.48 x 17.78 cm) box, method 4 proved to be difficult. It was challenging to place the desired weight into the predetermined volume. It appeared as if the volume was too small. The box for method 4 was compacted for about 8 minutes with an average of over 150 compaction hits. After method 4 cylinders were filled, it was predicted that the core samples would be inconsistent with one another and possibly have lower values than the cylinders.
51


Sulfur Capping
When starting to talk about compressive strength tests, it should be first stated that sulfur capping data was not performed. Due to the difficulty of sulfur capping the cylinders, it was determined that sulfur capping weak pervious concrete is not practical. The concrete mix that was derived with this batch used only large aggregates, no small aggregates and no chemical admixtures. This type of mixture typically results in low compressive and therefore low tensile strength values. With that in mind, it was not possible to sulfur cap the low strength concrete cylinders.
As per the specified procedure in Section 4.9, the melted sulfur was placed in a capping mold. The top of the cylinder was dipped to produce a cap thickness in accordance to ASTM C 617 found in Section 4.9. After the sulfur cap hardened, the specimen was hit with a metal rod to release the cap from the mold. Each time this method was attempted, the top aggregates of the concrete cylinder would break away from the cylinder itself. The bond between top aggregates was very weak. As a result, aggregates would crumble away from the cylinder. Figure 5.2 shows a picture of the sulfur mold containing top aggregates that broke away from the cylinder.
52


Figure 5.2: Failed Sulfur Capping Image
Since the sulfur capping procedure failed, another procedure was tried. Instead of dipping the cylinder in a sulfur mold, the cylinder was placed in its 4in. (10.16 cm) x 8in. (20.32 cm) cylinder in order to have liquid sulfur poured. This pouring method was performed on the low strength concrete in order to prevent the weak bonds between the top aggregates from breaking. Figure 5.3 shows the liquid sulfur being poured over the specimen while in a concrete cylinder.
53


Figure 5.3: Sulfur Poured On Top of Standing Specimen
However, the hot liquid sulfur did not harden instantly. Instead, the sulfur melted between the aggregates filling the top voids of the concrete. The sulfur also trickled down the sides of the specimen. Figure 5.4 shows the liquid sulfur running deep into the voids and down the sides.
54


Figure 5.4: Failed Attempt at Pouring Sulfur
It was determined that this method would alter the porosity and density of the specimen to the point where it would no longer be a true representation of pervious concrete, therefore this method was dismissed from this study.
It was determined that sulfur capping could not be applied to the pervious concrete used in this study. It is to be noted that it is possible to produce a higher strength pervious concrete that would have fewer voids between the aggregates which could be sulfur capped using the original capping mold. It should be investigated to determine if higher strength pervious concrete can be sulfur capped. All the compressive strength tests were done using neoprene pads instead of sulfur caps.
Problems with Preparing Samples
Preparing and gathering samples for compressive testing turned out to be a complex task. First, the concrete cylinders had to be removed from the plastic 4in. x 8in. (10.16 x 20.32
55


cm) cylinder molds. In the process of removing the concrete cylinders samples were subjected to striking and vibrating force. Methods 1 and 2 (thought of as less compacted methods) were easily removed out of their molds, while cylinders in methods 3 and 4 were removed with more difficulty. A few of the concrete cylinders were partially damaged in this procedure and caused the top (ends) of the cylinder surfaces to break and become uneven. Figure 5.5 shows a photograph of a typical uneven concrete cylinder.
Figure 5.5: Tops of Cylinders Not Level
Due to the lack of small aggregates in the mix, a typical pervious concrete cylinder had uneven top (end) and bottom (end) surfaces, which represents a problem because in a standard concrete compression test the load is applied at the ends of the cylinders. A flat even surface is necessary in order to apply the load evenly throughout the specimen. This
56


was not the case with all the porous samples, but it did result in less than satisfactory compressive load values. The more uneven the top (end) surface, the worse the concrete cylinder was expected to perform in the compression test. This proved to be the case for several porous samples. As the compressive load distributed unevenly throughout the sample, it caused premature failure. To solve the problem of uneven surfaces and to ensure proper testing results, specimens were saw-cut with a wet saw. The saw-cut cylinders provided better samples for compressive testing, however, not significantly better. The very high porosity of the samples proved difficult to cut evenly with a wet saw and some samples chose to break and fall apart when subject to the ever slight vibrations of the wet saw.
As previously noted, and as depicted in Figure 5.6, the top (end) of the saw-cut sample was relatively flat, albeit still an unevenness to the cut edge.
Uneven point
Figure 5.6: Tops of Cylinders Not Level After Saw Cutting
57


The box samples were cored with a 3in. (7.62 cm) diameter core bit. Another problem that arose with core drilling was found in methods 1, 3 and 4. Four samples should have been obtained from each box for each compaction method. Instead of drilling 4 cores each, Methods 1, 3, and 4 had less applicable core specimens. Method 1 only produced one valid core because the other three cores broke while drilling. Methods 3 and 4 had only two cores that could be tested. The other two cores broke apart during core removal. The bond between the aggregates was too weak, breaking the cores along the weakest plane. Broken cores separated about one-third a distance from the bottom (end) of each core. Figure 5.7 shows method 3 core samples. Method 2 was the only consolidation method that produced four drilled cores. The reason method 2 cores did not break during drilling is because they had a higher compressive strength than the other methods.
Figure 5.7: Broken Core Samples
Breaking of the core samples occurred because the pervious concrete had a low compressive strength. If the pervious concrete had higher compressive strength, it would
58


not be likely that the drilled cores would break prematurely. Based on batch 1 findings, drilled cores are not likely to break if compressive strength is higher than 1000 psi (6.89 MPa). An interesting observation is that all the cores that broke while in the drill separated approximately one-third distance from the bottom (end).
Based on the success of drilling cores from the boxes, one could predict that method 2 would produce the highest compressive strength values. It is likely the cores with the highest tensile value will not separate while in the drill, therefore, method 2 was predicted to have the highest compressive strength value for cored samples.
When testing cylinders and cores, it was important that the end surfaces be level. Sometimes even with saw cutting, a level surface was not obtained. Figure 5.8 shows a method 2 batch 1 specimen that had a non-level top surface. This is not ideal and should be avoided.
Figure 5.8: Compression Testing of a Non-level Cylinder Surface
59


5.2 Unit Weight/Density Results Batch 2
The calculated unit weight was expected to be 124.6 lb/ft3 (1995.9 kg/m3) It is to be noted that some cores were not tested as originally planned because they broke during the drilling process. The unit weight for all specimens was taken at 28 days. Table 5.1 lists the unit weight obtained for batch 1.
Table 5.1: Unit Weight (lb/ft3) for Cylinders and Cores
Cylinder Method 1 Method 2 Method 3 Method 4
A 101 106 111 116
B 98 104 120 123
C 98 101 101 123
D 96 102 111 123
E 107
F 89
Average 98 103 111 121
Core A 93 109 106 128
B 105 103 114
C 105
D 105
Average 93 106 104 122
Method 1 was the furthest from the desired density. The rodding compaction method produced unfavorable unit weight results as suspected. The rodding only created additional voids in the specimen and did not compact the specimen. Method 4 core and cylinder testing averages were similar. Since method 4 uses volume and desired density to calculate the weight of each cylinder or box, it is likely that method 4 would have similar densities between cores and cylinders.
Figure 5.9 graphs the averaged unit weight/ density for each of the compaction methods. Method 1 value represents the average of 6 cylinders. Method 2, 3, and 4 values
60


represent the average of 4 cylinders. The lowest unit weight is in method 1. The second lowest unit weight was recorded for compaction method 2. Unit weight values for methods 1 and 2 are very low. Method 3 had the second highest unit weight value. Method 4 had the highest unit weight value and the closest to the desired unit weight. The desired unit weight was obtained only by method 4. Compaction methods 1, 2, and 3 had unit weight less than desired. Unit weights for cylinders in methods 1 and 2 were too low to even consider applicable.
Dry Unit Weight of Cylinders
Compaction Method ^Tm3.M4.4
Figure 5.9: Unit Weight of Cylinders for Batch 1 Methods 1, 2,3 & 4
Figure 5.10 shows the unit weight/ density of the cored samples derived from the box for each of the 4 methods. Method 1 value represents the average of 1 core. Method 2 value represents the average of 4 cores. Method 3 and 4 values represent the average of 2 cores. The lowest density is again obtained from method 1 as expected. The second lowest was
61


recorded for method 3. The second highest unit weight was obtained by method 2. Similar to the cylinders, method 4 core samples had the highest density. Method 1 was the only compaction method that produced really low unit weight values. Methods 2 and 3 had lower unit weight values than desired, but they are still considered acceptable.
Dty Unit Weight of Drilled Cores
130 T
12 3 4
# of specimens averaged:
Compaction Method mi-i,M2-4, M3&4-2
Figure 5.10: Unit Weight of Cores for Batch 1 Methods 1, 2,3 & 4
It was predicted that the density would fall in chronological order with method 1 being lowest and method 4 being highest. This can easily be explained by manner in which method 2 (the jigging method) was performed on the box. Like previously discussed in section 5.1, it was found to be very difficult to jig such a heavy 12 x 12 x 7in. (30.48 x 30.48 x 17.78 cm) box. Instead of gently jigging the box back and forth, the box was repetitively dropped on the concrete, giving the method an additional level of compaction. That is most likely the reason why the method 2 box was more compacted and had higher
62


unit weight than the method 3 box. This also explains the significant difference in unit weights between method 2 cylinders and cores. The jigging method is only effective when a larger mold is being fabricated. The compaction level in method 2 box is reached with the help of its self weight.
Unit Weight/Density of Cylinders and Cores
# of specimens averaged (cores):
Ml- 1 M2-4 M3&4- 2
# of specimens averaged (cyl.):
Ml- 6
M2.M3&4- 4 Cylinders
Cores
12 3 4
Compaction Method
Figure 5.11: Unit Weight of Cylinders and Cores for Batch 1 Methods 1, 2,3 & 4
Figure 5.11 shows both the average cylinder and the average core density values for each method. Method 1 cylinder value represents the average of six cylinders. Method 2, 3, and 4 cylinder values represent the average of four cylinders. Method 1 core value represents the average of one core. Method 2 core value represents the average of four cores. Method 3 and 4 core value(s) represents the average of two cores. By looking at the graph, the following observations are made:
Method 4 average density for cylinders and cores are similar. As intended in method 4, a sample can be compacted as many times as needed with the
63


compaction tool in order to fit the calculated weight into a predetermined box. The same compaction is achieved with method 4 regardless of cylinders or boxes. Also, method 4 does not depend on the strength of the person using the compaction tool.
Method 1 provided least compaction for the box and cylinders. The unit weight results were significantly low and furthest away from desired density. The results were so low that method 1 is considered not applicable.
Method 2 produced the second lowest density when averaging cylinders and cores. The jigging method did not produce adequate unit weight values for cylinders. However, the jigging method was successful for cores.
Method 3 ranked third in overall density. Method 3 used a compaction tool during fabrication of concrete cylinders. It compacted the concrete 25% by volume and is believed to accurately represent field conditions. In other words, 8 inches of concrete in a cylinder was compacted into a specimen sample 6 inches tall. Both core and cylinder unit weights had similar values.
Method 4 produced the highest unit weight values and achieved the desired unit weight. Since the specimen was fabricated with a calculated weight being placed into a set volume. This procedure was considered to be successful.
The density values are for the most part in chronological order as expected, with method 1 having the lowest density and method 4 having the highest.
64


5.3 Porosity Results- Batch 1
The porosity was measured for each compaction method. Table 5.2 lists the porosity values for batch 1.
Table 5.2: Porosity (%) for Cylinders and Cores
Cylinder Method 1 Method 2 Method 3 Method 4
A 37.3 31.2 16.9 19.7
B 30.6 33.5 20.2 23.1
C 33.0 30.7 21.2 19.7
D 32.4 37.0 24.9 24.3
E 46.2
F 38.7
Average 36.3 33.1 20.8 121
Core A 40.9 22.8 20.7 25.9
B 24.4 21.9 16.7
C 20.2
D 24.2
Average 40.9 22.9 21.3 21.3
From the porosity data, method 1 has a very high porosity. This porosity value was most likely caused during the rodding process. The rodding created large voids inside the cylinders and the box. There is a great variation between the average results for the cylinder and cores for method 2. Method 2 box was more compacted than the cylinders, as previously discussed in Section 5.1. Therefore, method 2 cores have fewer voids than method 2 cylinders. Methods 3 and 4 cylinders and cores have porosity readings between 20.8% and 23.8%.
Figure 5.12 shows the porosity of cylinders for batch 1. Compaction method 1 cylinders are most porous. Method 2 cylinders had the second largest amount of voids. Methods 3 and 4 had the lowest porosity values. Based on porosity data, it is expected for methods 3 and 4 to have higher densities and compressive strength values
65


Porosity of Cylinders
40
12 3 4
Compaction Method # of specimens: M1 6 M2, M3, M4 4
Figure 5.12: Porosity of Cylinders for Batch 1 Methods 1, 2, 3 & 4
Figure 5.13 shows the porosity of drilled cores for each method. Method 1 has the highest porosity at 40.9%, making this core sample too porous and not applicable.. Rodding pervious concrete is not an appropriate compaction method because it creates too many voids. After the specimen is rodded at each layer, the voids are not filled with surrounding concrete (as typical behavior for conventional concrete). A practical value for porosity is between 15-30%. Cores in methods 2, 3, and 4 all have acceptable average porosity values between 21.3% and 22.9%.
66


40
Porosity of Drilled Cores
Compaction Method
r Ml- 1, M2- 4, M3&4- 2
Figure 5.13: Porosity of Cores for Batch 1 Methods 1, 2,3 & 4
Figure 5.14 shows porosity percentages for cores and cylinders for all four methods. Method 1 cylinder value represents the average of 6 cylinders. Method 2, 3, and 4 cylinder values represent the average of 4 cylinders. Method 1 core value represents the average of 1 core. Method 2 core value represents the average of 4 cores. Method 3, and 4 core value represents the average of 2 cores.
Method 1 is the only method that can be named impractical due to the overall high porosity values. Again, rodding pervious concrete cylinders does not compact the concrete in order to produce acceptable results. Instead, it creates additional unnecessary voids, making the pervious concrete in method 1 too porous. The more porous a specimen is, the weaker and less durable it will be for practical use. Cylinders for methods 1 and 2 had significantly lower values than the cores. Rodding and jigging
67


methods compact boxes and cylinders at different levels. The jigging method only performs well for larger molds. The concrete self weight inside the box mold compacts itself during the jigging procedure. The heavier specimen (such as the box) was more compacted than the smaller specimen (cylinder) mainly because it was heavier. This level of compaction cannot be achieved for 4 x 4 in. (10.16 x 10.16 cm) cylinder molds with the jigging procedure. Larger cylinder molds such as 6in. (15.2 cm) or 8 in. (20.3 cm) diameter are not expected to produce more favorable values during the jigging procedure because they are 50% heavier.
Porosity of Cylinders and Cores
45
40
35
30
5? 'w' 25
*35 o u 20
£ 15
10
5
0
# of specimens averaged (cores):
Ml- 1 M2-4 M3&4- 2
# of specimens averaged (cyl.): Ml- 6
M2.M3&4- 4
Cylinders
Cores
12 3 4
Compaction Method
Figure 5.14: Porosity of Cylinders and Cores for Batch 1 Methods 1, 2, 3 & 4
5.4 Compressive Strength Results Batch 1
Compressive strength of cylinders and cores alike were tested for all 4 of the compaction methods at 28 days. Rodding compaction method 1 had the lowest compressive strength.
68


Low compressive strength was expected based on low unit weight and high porosity data from Section 5.1
In order to understand the compressive strength values it is important to discuss what is typical of this type of pervious concrete mixture. Pervious concrete will vary in strength from 400 to 4000 psi (2.75 to 27.6 MPa) depending on the application and the pervious concrete mixture. The mixture that was used for all the consolidation methods was from the same batch. This mixture did not include fine aggregate or chemical admixtures. This type of mixture can exhibit compressive strength values below 1000 psi (6.89 MPa), but is more common to see values between 1000 to 1500 psi (6.89 to 10.34 MPa) (Offenberg, 2009). According to an interview with Matthew Offenberg, a more desirable range for pervious concrete compressive strength is between 1500 to 2500 psi (10.34 to 17.23 MPa). Most of the compressive strength values for all the compaction methods turned out somewhat lower than expected.
Table 5.3 shows that cylinders and core samples were equally low in compressive strength, and therefore, impractical for pervious concrete use when using Method 1. The low compressive strength in the rodding method explains all the problems that occurred during the experiment. To summarize from earlier, some of the major problems were: raveling aggregates during cylinder separation from the plastic molds, broken samples during coring, and unexpected raveling while saw cutting. All of these problems are directly related to low compressive strengths and low tensile strengths.
69


Table 5.3: Compression Strength Results for Method 1
Method 1 Cylinder Core
A B c D E F A
Tested Strength (psi) 0.0 0.0 277 272 192 128 3378
Average Strength (psi) 217 338
Table 5.4 states that cylinders compacted with method 2 had very low compressive strengths averaging 350 psi (2.41 MPa). It was expected that the core samples have slightly higher compressive strengths. The core samples had compressive strengths averaging almost four times higher than the cylinders compacted in the very same method. This unexpected difference can be traced back to the jigging method variance from cylinders to box molds. The 4 x 8in. (10.16 x 20.32 cm) plastic cylinders were rocked back and forth as specified in the jigging procedure in ASTM C29. This proved to be more difficult to do with the heavy 12 x 12 x 7in. (30.48 x 30.48 x 17.78 cm) pervious concrete box. Instead of gently rocking the block form, the heavy box would hit the bottom of the floor as it was being lifted up 1 in. (2.54 cm) on each side. The box was compacted in part by its own self weight. Since the cylinders did not have this additional compaction, they were compacted significantly less than the box.
Table 5.4: Compression Strength Results for Method 2
Method 1 Cylinder Core
A B C D A B C D
Tested Strength (psi) 355 398 387 257 1008 1167 896 1354
Average Strength (psi) 349 1106
Table 5.5 states that method 3 cylinders and cores have average compressive strengths within 10% of one another. Method 3, using a compaction tool resulted in relatively
70


consistent compressive strength data. Since the compaction tool was round and slightly smaller in diameter than the 4 x 8 in. (10.16 x 20.32 cm) cylinders, it seemed to work very well for cylinders. However, the round compaction tool did not prove to be a logical shape for a square box. The compaction tool could not get close enough to the edges of the box in order to equally compact the comers. The shape of the compaction tool had a negative outcome on compaction of the box. The drilled cores were compacted less than the cylinders. A compaction tool with a right angled edge should be tried to compact the pervious concrete inside of the box molds.
Table 5.5: Compression Strength Results for Method 3
Method 3 Cylinder Core
A B C D A B
Tested Strength (psi) 979 866 N/A N/A 880 789
Average Strength (psi) 923 834
For method 4, the weight vs. volume method, a calculated weight of pervious concrete was compacted in a predetermined volume. The weight was determined by dividing the desired unit weight by the volume of the box. This method proved to be by far the most consistent method when examining compressive strength. Some cylinders in this method were compacted with 15 strikes and some compacted with 35 strikes. The point of this method is not dependant on the strength of the individuals striking the concrete. All the cylinders weighed the same amount to the nearest hundredth of a pound. The box mold was placed on the scale and compacted until the desired weight was reached. As shown in Table 5.6, the difference between the cylinder and core values was less than 1%. Method 4 produced consistent compressive strength results between cylinder and core samples.
71


Table 5.6: Compression Strength Results for Method 4
Method 4 Cyl inder Core
A B C D A B
Tested Strength (psi) 935 1010 906 985 1005 898
Average Strength (psi) 959 952
Figure 5.15 shows methods 1 and 2 having very low compressive strength results unacceptable for practical use. Methods 3 and 4 produced similar compressive strengths of 923 psi (6.35 MPa) and 959 psi (6.61 MPa), respectively. Method 4 cylinders had compressive strength values that were 4% higher than those of method 3.
. # of specimens: Ml-6
Compaction Method averaged M2, M3, M4 4
Figure 5.15: Compressive Strength of Cylinders for Batch 1 Methods 1, 2, 3 & 4
72


When examining the compressive strength of the cores, the rodding method 1 fell short of the expected value and was too low to consider for the comparison. Method 2, the jigging compaction procedure had the highest compressive strength value of 1106 psi (7.58 MPa). Method 4 produced second highest strength of 952 psi (6.56 MPa). Method 3 had the third lowest value of 834 psi (5.75 MPa). The compaction tool method 3 has 28% less compressive strength than the jigging method 2. Method 1 cylinder value represents the average of 6 cylinders. Method 2, 3, and 4 cylinder values represent the average of 4 cylinders. The method 1 core value represents 1 core. The method 2 core value represents the average of 4 cores. The method 3, and 4 core value represents the average of 2 cores.
Compressive Strength of Drilled Cores
1
3 4
# of specimens averaged:
Ml- 1, M2- 4, M3&4- 2
Compaction Method
Figure 5.16: Compressive Strength of Cores for Batch 1 Methods 1, 2, 3 & 4
73


Compressive Strength (psi)
Figure 5.17 combines the cylinder and core compressive strength data in order conclude
which method is adequate for laboratory compaction of pervious concrete mixtures. Observations made regarding this data include:
Method 1 the rodding method does not produce adequate compressive strength values and is not being compared with the other 3 methods.
Method 2 the jigging method proved to be the best compaction method for the box form which the cores were obtained. However, jigging does not produce adequate compaction in cylinders.
Method 3 produced the third highest compressive strength. Results between cylinders versus cores have a 10% difference.
Method 4 weight vs. volume method produced the most consistent cylinder vs. core results differing less than 1%.
All of the compressive strength results were lower than expected and similar test should be performed with a stronger pervious concrete mixture.
Compressive Strengths of Cores and Cylinders
1200
12 3 4
# of specimens averaged (cores):
Ml- 1 M2-4 M3&4- 2
# of specimens averaged (cyi.): Ml- 6
M2.M3&4- 4
Cylinders
Cores
Compaction Method
Figure 5.17: Compressive Strength of Cylinders and Cores for Batch 1 Methods 1, 2, 3 & 4
74


Compressive Strength (psi)
5.5 Batch 1 Results
Conclusions from the result data shown by Figures 5.18 and 5.19 include:
Method 1 the rodding method produces cylinders with too high of porosity, too low of density and too low of compressive strength. Rodding should not be used as a compaction method for pervious concrete.
Method 2 the jigging method for cylinders produced high porosity, low density and low compressive strength.. These are unacceptably poor results.
Method 3 using a compaction tool produced desired porosity of 21%, 12% lower than expected density, and a compressive strength value of 923 psi (6.36 MPa). While this compressive value is acceptable for pervious concrete, it is still a very low value when being compared to traditional concrete strengths.
Method 4 weight to volume method has highest density, midrange porosity of 24%, and highest compressive strength for the cylinders at 959 psi (6.60 MPa).
Porosity vs. Strength for Cores and Cylinders
1200
1000
Method 2 Core
Method 4 C,vlindat______________________
ITlVtltOU VJtlttUVI
Method 3 Cylinder Method 4 Core
800
600
400
Method 3 Core
Method 1 Method 3 A Method 2
Method 2 Cylinder ^ Method 1 Core ^ Method 4
200
^ Method 1 Cylinder
0
0 10 20 30 40 50
Porosity (%)
Figure 5.18: Porosity vs. Strength for Batch 1 Cylinders and Cores
75


Unit Weight vs. Strength for Cores and Cylinders
1200
1000
% 800
c
600
I 400
E
o
U
200
A Method 2 Core
Method 3 Cylinder Method 3 Core
_ Method 4 Core Method 4 Cylinder
Method 1 Method 3 A Method 2
Method 1 Core A Method 2 Cylinder Method 4 Method 1 Cylinder
0 20 40 60 80 100 120 140
Unit Weight (lb'ftA3)
Figure 5.19: Results for Batch 1 Cores Methods 1, 2,3 & 4
5.6 Lessons Learned and Applied to Batch 2
Another batch is necessary in order to make solid conclusions regarding batch 1 results. Batch 2 was redesigned to incorporate lessons learned from batch 1. First, batch 2 included a revised mixture design with the inclusion of fine aggregate and a hydration stabilizer. Batch 2 only addressed cores. Cylinders were not examined. In addition, a concrete mixture will be batched for each method instead of one single large batch. Curing of the test blocks was better performed.
Chapter 5 has discussed the result data, difficulty that arose performing some procedures, and unfavorable results. Section 5.6 will list some of the major problems that arose during batch 1. A solution will be named for each problem in order to obtain more favorable results in batch 2.
76


List of problems for batch 1 and solutions to be applied toward batch 2:
Problem: Absence of small aggregates made it more likely for large aggregates to unravel from the cylinders and cores. This unraveling of aggregates during regular handling created unnecessary voids and a non-level testing top surface, and additional unnecessary voids. Solution: Use different mixture design that incorporates fine aggregate. In addition, use a smaller batch size when testing.
Problem: Absence of small aggregates contributed to low compressive strength, and thus low tensile strengths. The low tensile strength contributed to breaking the core samples while still in the coring machine. Some specimens were too fragile and broke during coring. Solution: Use different mixture design that incorporates small aggregates. Make a smaller batch to closely monitor the consistency of mixture in order to produce a higher strength pervious concrete.
Problem: Low compressive and tensile strength made it difficult to saw cut cylinders and core samples in order to obtain a level top and bottom surfaces. Solution: Use different mix design that incorporates small aggregates. Perform better curing practice in order to achieve higher strength. Wrap the specimen entirely in plastic, not just the surface of the block.
Problem: Bond between the aggregates was too weak. More moisture was needed to react with the cement paste. Solution: Hydration stabilizer could be added. Better curing practices should be considered. Instead of covering only the top of molds with plastic, cover the entire box with plastic. This will prevent the moisture from evaporating prematurely potentially giving a stronger specimen.
77


Problem: The very large batch was difficult to handle in the laboratory facility. The mixer could mix the necessary volume, but it was apparent that segregation occurred producing different aggregate sizes for the different compaction methods. Solution: Batch 2 should be smaller in volume.
Problem: Due to the large batch, total mixing time was between 15-30 minutes. Solution: To error on the side of caution, batch 2 shall be smaller in volume and total handling time shall not exceed 10 minutes. Another added benefit for this dry mix may be that it would retain more moisture if handling time were reduced.
Problem: Since cylinders are generally considered not applicable for testing pervious concrete. Batch 1 volume was too big. Solution: batch 2 will strictly be focused on cores taken from fabricated boxes.
Problem: Compaction tool used for filling boxes was round, but the box was square. Solution: Consider a rectangular compaction tool in order to better compact the comers.
Problem: There was much room for error in the porosity test. Water was filled to the top of each specimen. Since the top of each cylinder or core may have been uneven, individual judgment was used to determine where the true top is. This created unnecessary room for error because by limitations of the human eye. Solution: New method is to be used for measuring porosity.
Problem: The inverted slump cone did not provide adequate for determining the consistency of the mixture in Batch 1. Solution Repeat the inverted slump cone procedure for batch 2 and compare results.
78


Better procedures for batch 2:
Method 1 -Each layer (layer = 1/3 volume) rodded 10 times for unit weight bucket and 25 times for the box.
Method 2 -Unit weight bucket jigged 10 times for each layer (layer = 1/3 volume) and 25 times for the box.
Method 3 -Unit weight bucket compacted with round compacting tool 10 times for each layer (layer = 1/3 volume). Square compaction tool used for box and compacted 25 times for each layer (layer = 1/3 volume).
Method 4 -Concrete compacted by striking the box with square compaction tool until desired weight was achieved for the 12in. x 12in. x 7in. volume of the box.
Air content was not taken for any of the methods. In batch 1 all air content seemed to be between 18.5% and 19%. The void ratios (porosity readings) varied between 20-40%, but the air content stayed mostly consistent form one compaction method to the other. Clearly, this is incorrect. It is not accurate to measure the air content of pervious concrete.
Each batch 2 compaction method was handled within 10 minutes of batching the concrete.
Hydration stabilizer and fine aggregate meeting ASTM C33 were used.
All boxes were completely wrapped in plastic preventing any air from escaping in order to better the curing process (batch 1 plastic only covered the top surface).
79


New method was used to measure porosity called the modified volume displacement method. A large volumetric measuring bowl is filled with 600mL of water. Each specimen is expected to displace the water based on its volume. The difference between expected volume displacement and actual volume displacement is noted. The difference over the volume of concrete specimen is the void ratio or porosity.
No samples were cut down with a wet saw. Samples appeared to have level top and bottom surfaces.
5.7 Batch 2 Result Data
Four boxes were cast in batch 2, one for each method. Each box was cured until 6 days of age. All specimens were cored on day 6, and compressive strength, unit weight, and porosity was measured at 7 and 28 days of age. The same properties measured during the batch 1 phase were measured during batch 2. The results from batch 2 cannot be directly compared with the results in batch 1 because the two concrete mixtures were different. The four compaction methods were compared to one another for batch 2. In addition, consistency of each compaction method within the batch themselves was compared.
The batch 2 mixture design was very similar to batch 1. One difference is that fine aggregate was included at 6% by weight of total aggregate. Also, 6 fl. oz. of hydration stabilizer were added to every 100 lb/cy (59.3 kg/mA3) of cement. The total volume of concrete needed for one consolidation method was 0.61 ftA3. Batch 2 was repeated four times; one per method. Table 5.7 provides the mixture characteristics for batch 2.
80


Table 5.7: Batch 2 Weights
Cement 550 lb/cy
Rock 2477 lb/cy
Sand 160 lb/cy
Water 178 lb/cy
HRWR / AEA 6 fl oz./cwt
Table 5.8 lists the unit weight for each method. All the results were averaged between two specimens.
Table 5. 8: Unit Weights of Fresh and Hardened Concrete
Method Weight (lb) Volume (ft3) Unit Weight (lb/ft3) Dry Unit Weight at 7 days (lb/ft3) % Difference
1 29.95 0.25 120 109.3 8.8
2 30.55 0.25 122 123.1 0.8
3 29.8 0.25 119 117.8 1.2
4 31.55 0.25 126 126.9 0.5
The fresh unit weight taken from the unit weight measure was compared to the dry unit weight of the specimen for each method. The rodding method 1 produced inconsistencies and an overall poor result. Unit weight taken from the fresh concrete and the hardened concrete cylinders was compared for methods 2, 3, and 4. The results have 0.5% to 1.2% difference between the unit weights. Method 4, the weight vs. volume method, has the most consistent measure for unit weights of fresh and hardened concrete.
Table 5.9 summarizes the 7 day compressive strength, porosity, and unit weight data for the four compaction methods. Two of the four cored specimens were tested at 7 days. The remaining two cored cylinders were tested at 28 days of age. Unit weight, porosity, and compressive strengths were taking for each core.
81


Table 5.9 Batch 2 Results at 7 Days
^rosity (%) Unit Weight (lbs/ft3) Compressive (P^i) Strength
Method A B Average A & B A B Average A & B A B Average A & B
1 23.1 29.0 26.0 108.0 110.6 109.3 830 1076 953
2 17.6 17.6 17.6 124.8 121.5 123.1 2257 2130 2193
3 23.4 24.0 23.7 117.8 117.8 117.8 1435 1900 1667
4 15.8 14.9 15.3 126.6 127.2 126.9 4028 4424 4226
In comparison with batch 1, batch 2 had a lower range of values for porosity. In addition, batch 2 had a higher range of values for unit weight, and significantly higher compressive strengths at 7 days.
A quality pervious concrete varies in porosity, unit weight, and strength based on what type of mixture is desired. Different characteristics may be desired depending on the pavements application and environment. Porosity generally ranges from 15-35%, unit weight is typically between 135 to 140 lb/ftA3 (1824 to 2242 kg/mA3), and compressive strength for small aggregate mixtures varies between 1500 to 4000 psi (10.34 to 27.56 MPa). The batch 2 results exceeded the values measured during the batch 1 phase and are more typical for pervious concrete.
Porosity ranged between 15.3% for method 4 to 26% for method 1. Figure 5.20 shows the difference in porosity between each method. All of the values in Figure 5.20 are averaged between two specimens. Method 4 porosity is close to the lower boundary for pervious concrete mixtures Method 4 may not be desired for all applications. Method 2 also produced a lower porosity of 17.6%. After the porosity test was performed, it was expected that method 2 and 4 have higher compressive strength values than method 1 and 3 strictly basing that on porosity results.
82


Porosity of Drilled Cores for Batch 2
40
12 3 4
Averages represent 2 specimens
Compaction Method
Figure 5.20: Porosity of Drilled Cores at 7 Days
Batch 2 unit weights were higher than batch 1 results. Figure 5.21 shows the differences in unit weights between each method. Batch 2 method 1 proved to once again have the smallest unit weight. All of the values in Figure 5.21 are averaged between two specimens. While in batch 1, methods 2 and 3 seemed to have similar unit weights, in batch 2 this is not the case. In batch 2, method 2 has a higher density of 123.1 lb/ft3 (1979.4 kg/m3) (2.5% less than design density) and method 3 has a density of 117.8 lb/ft3 (1886 kg/mA3) (5.5% less than design density). Method 4 density was 2% higher than design density at 126.9 lb/ft3(2040.5 kg/m3).
83


Batch 2 Unit Weight
12 3 4
Averages represent 2 specimens
Compaction Method
Figure 5.21: Unit Weight/Density of Drilled Cores at 7 Days
The most noticeable difference in results between batch 1 and 2 was the compressive strength. Figure 5.22 shows the compressive strength results for batch 2 at 7 days of age. All of the values in Figure 5.22 are averaged between two specimens. Method 1, the rodding method, again proved to be not applicable for pervious concrete. Rodding produced the lowest compressive strength out of the four methods for both batches. Surprisingly, method 2 was much higher than method 3. This difference did not exist in
batch 1 data, where the jigging compaction method produced slightly stronger results. The 7 day compressive strength data for methods 2 and 3 produced a coefficient of variability of only 6.0% and 2.3% respectively. Method 4 produced the highest strength at 7 days of age with an average of 4226 psi (29.12 MPa). Due to the low 15.3% porosity and higher unit weight of 126.9 lb/ft3 (2032 kg/m 3), a higher compressive strength was
84


expected. However, batch 2 method 4 produced a concrete with a compressive strength to conventional concrete mixtures. The outcome is at the low end of porosity readings and high end of strength reading. The application of this particular pervious concrete may be more specific and not widely used. This low porosity pervious concrete may be desirable in areas where greater strength performance is needed and low porosity is not an issue.
Compressive Strength at 7 Days
on
c
v
_
55
v
>
o
u
a
E
o
U
4500
4000
3500
3000
2500
2000
1500
1000
500
0
12 3 4
Averages represent 2 specimens
Compaction Method
Figure 5.22: Compressive Strength at 7 Days
Figure 5.23 shows a picture of method 4 sample failing at 4426 psi (29.12 MPa). This sampled failed in a manner which is typical for conventional concrete A very large pop was heard as ultimate strength was reached. Only method 4 cylinders in batch 2 failed in this manner. The coarse aggregate failed in method 4 cores. Batch land batch 2 methods 1 and 3 failed by the aggregates unraveling and crumbling away from the sample. Again, method 4 of batch 2 had very unique behavior common for conventional concrete.
85


Figure 5.23: Batch 2 Method 4 Compressive Test at 7 Days
Higher readings for method 4 are contributed due to the better curing practices and careful handling of the pervious concrete mixture. All methods in batch 2 produced favorable results, and this is directly linked to proper handling and curing.
While batching, the pervious concrete was carefully handled and cured in a more favorable manner than batch 1. When batch 1 was placed in the molds, only the top was covered with plastic and the block was placed in the curing room for 28 days. However, with batch 2, the entire box was enclosed in a plastic wrap ensuring no moisture could escape. The key to having a stronger pervious concrete is to greatly consider adding hydration stabilizer and enhancing curing. In this laboratory test, it was easy to wrap the
86


boxes in plastic wrap. In the field, it is recommended that the contractor topically coat the pervious concrete for additional hydration and covering the pavement with plastic as soon as possible. Locking in the moisture during curing produces higher compressive strength.
The compressive strength, porosity, and unit weight of batch 2 data was obtained at 28 days of age. Predictions were made about the 28 day test results from other documented strength data for pervious concrete (Mahboub, 2009). Based on 7 day strengths, 28 day strength values are predicted in Table 5.10. Compressive strength is expected to be 15% higher at 28 days than at 7 days of age (Mahboub, 2009).
Table 5.10: Prediction for 28 day Strength Results
Compressive Strength (psi)
Tested at 7 days Predicted for 28 days
Method 1 830 955
Method 2 1462 1681
Method 3 1454 1672
Method 4 4226 4860
One can see that the more compaction is achieved for pervious concrete, the less porosity it will have. This will result in higher unit weight and higher strength.
The actual values obtained for strength at 28 days of age are listed in Table 5.11. Compaction method 1 had the highest increase in compressive strength of 19.3% between 7 and 28 days. Method 3 had the second highest increase in compressive strength of 13.8% between 7 and 28 days. Method 2 had the third highest increase in compressive strength of 10% between 7 and 28 days. Method 4 had the smallest increase in compressive strength between 7 and 28 days. It is to be noted that the increase if compressive strength between 7 and 28 days is related to the ultimate strength value. The
87


highest strength pervious concrete had the smallest increase in compressive strength and the lowest strength pervious concrete had the highest increase between 7 and 28 days of age. Compressive strength increased from 7 to 28 days between 6.8% and 19.3 %.
Table 5.11: Compressive Strength Increase between 7 and 28 days
Compressive Strength (psi) Increas e %
Tested at 28 days Predicted for 28 days Tested at 7 days
Method 1 1180 1096 953 19.3
Method 2 2437 2522 2193 10.0
Method 3 1935 1917 1667 13.8
Method 4 4533 4860 4226 6.8
Figure 5.25 shows the direct relationship between porosity and strength measured at 7 days of age. Figure 5.26 shows the same relationship between porosity and strength at 28 days. This data represents the average between two specimens. Lower porosity produces higher strength. As the porosity increases, the percentage of the voids increases and results in lower strength values. The highest porosity was measured for the rodding method 1. Consequently, this method resulted in the lowest strengths. The weight vs. volume method 4 produced the lowest porosity and the highest compressive strength.
88


Compressive Strength (psi)
Porosity vs. Strength at 7 Days
Method I Method 3
Method 2
Method 4
Averages represent 2 specimens
Porosity (%)
Figure 5.24: Porosity vs. Strength at 7 Days
Porosity vs. Strength at 28 Days
COD
C
0)
L.
t/5
01
c.
E
o
U
Method 1 Method 2
A Method 3
Method 4
0
5 10 15 20 25
30
Porosity (%)
Averages represent 2 specimens
Figure 5.25: Porosity vs. Strength at 28 Days
89


Figure 5.27 shows a direct relationship between unit weight and strength measured at 7 days of age. The lower the unit weight, the less compacted the specimen, and the lower the unit weight, the less the compressive strength. The method with the lowest unit weight and strength was the rodding method 1. The method that resulted in the highest strength and unit weight was method 4, or the weight vs. volume method. The second highest unit weight and strength was produced by method 2, the jigging method. Figure 5.27 shows the same relationship between unit weight and strength at 28 days.
Unit Weight vs. Strength at 7 Days
Method 1 Method 3
Method 2
Method 4
Unit Weight (lbs/ftA3) Averages represent 2 specimens Figure 5.26: Unit Weight vs. Strength at 7 Days
Figure 5.28 shows the same relationship between unit weight and strength at 28 days of age. The higher the unit weight the higher the compressive strength. Method 4 had the highest compressive strength and unit weight.
90


Full Text

PAGE 1

LAB ORA by B.S.,University of Colorado at Boulder, 2003 A thesis submitted to the University of Colorado Denver partial fulfillment Of the requirements for the degree of Masters of Science Civil Engineering May 2010

PAGE 2

(2010) by (Bojana Barovic Brown) rights reserved.

PAGE 3

This thesis for Masters of Science Degree by Bojana Barovic Brown Has been approved By Cheng Li, Ph .D. /:2J. Date

PAGE 4

Brown, Bojana Barovic (M.S. Civil Engineering) Evaluation of Laboratory Compaction Methods for Pervious Concrete Thesis Directed by Assistant Professor Stephan A. Durham As society's infrastructure continues to grow and develop pervious open areas are being depleted and replaced by impervious surfaces such as rooftops, sidewalks, and parking lots. Storm water runoff is an increasing concern of many municipalities as city engineers must design for increased storm drain and sewer system capacities. To address these concerns, pervious concrete has become a popular product that can reduce water runoff by allowing water to permeate through the pavement surface and into the underlying soil. Pervious concrete contains little to no fine aggregate, thus allowing voids of 15% to 35%. Currently, very few testing standards exist for pervious concrete resulting in decreased quality control and in some cases poor pervious concrete pavements. This thesis evaluates laboratory compaction methods for pervious concrete. Many of pervious concrete's properties are related to the amount and type of compaction. Compressive strength, durability, unit weight, and porosity are all properties that can be directly related to compaction of a pervious concrete mixture. This thesis evaluated four methods of laboratory compaction. Results from this study demonstrate that the weight versus volume method produced pervious specimens with superior strength when compared to three other methods. In addition, a common industry method, the jigging method, produced acceptable results The research herein provides recommendations for pervious concrete sample preparation for laboratory testing. iv

PAGE 5

This abstract accurately represents the content of the candidate's thesis. I recommend its publication. Stephan A. Durham, Ph.D

PAGE 6

TABLE OF CONTENTS Figures ............................................................................................... viii Tables .................................................................................................... x 1 INTRODUCTION .................................................................................................... 1 1.1 Overview ............................................................................................................... 1 1.2. Research Objective .............................................................................................. 1 2.1 Description ............................................................................................................ 3 2.2 Consistency ............................................... . ........ .......................... ..................... 6 2.3 Unit Weight .......................................................................................................... 8 2.4 Segregation ......................................................................................................... 11 2.5 Curing ................................................................................................................. 13 2.6 Strength ............................................................................................................... 15 2.7 Permeability ....................................................................................................... 17 2.8 Durability ................................................. ....................... .................................... 18 2.9 Infiltration Rate ................................................................................................. 19 2.10 Research Limitations and Scope ....................................................................... 21 2.11 Pervious Concrete in Colorado ......................................................................... 21 2.12 Compaction ......................................................................................................... 22 3 PROBLEM STATEMENT .................................................................................... 2 4 4 RESEARCH PLAN ................................................................................................ 28 4.1 Concrete Consolidation and Effects on Other Properties .............................. 28 4.2 Mix Design .......................................................................................................... 31 4.3 Consolidation Method 1: Traditional Consolidation ...................................... 33 4.4 Consolidation Method 2: Jigging Method Described In ASTM C29 ............ 35 4.5 Method 3: Compaction As A Percentage OfVolume ..................................... 36 4.6 Method 4: Weight vs. Volume Method ............................................................ 39 4.7 Compressive Strength ........................................................................................ 43 4.8 Neoprene Pads .................................................................................................... 45 4.9 Sulfur Capping ................................................................................................... 45 4.10 Porosity Testing .................................................................................................. 47 5 EXPERIMENTAL RESULTS .............................................................................. 49 5.1 Experimental Procedures .................................................................................. 49 5.2 Unit WeightlDensity Results Batch 2 ............................................................ 60 5.3 Porosity ResultsBatch 1 .................................................................................. 65 5.4 Compressive Strength Results Batch 1 ......................................................... 68 vi

PAGE 7

5.5 Batch 1 Results ................................................................. .... .... ... ... .... ..... 75 5.6 Lessons Learned and Applied to Batch 2 ........................................................ 76 5.7 Batch 2 Result Data ... ... .................................................................... ............ 80 6 CONCLUSION ......... . . ... ... ... .... . ..... ... ...... . .... .... .... ......... ... .... ...... ... 92 6.1 Experiment Conclusions . ........... .... ... ... .... ... . .... ....... ....... . ... .... ... 92 6.2 Recommendations ..................................................... ................... . . ............ 94 BmLIOGRAPHY ...... .. . ........ ...... .... ....... ....... ... ........... ... ..... ....... ... .......... ..... 96 v i i

PAGE 8

FIGURES Figure 2.1: Impervious Layer Zone (Hager, 2009) ......................................... 13 Figure 2.2: Drain Time Testing Apparatus (Hager, 2009) ................................ 20 Figure 2.3: Strike off and Compaction of Pervious Concrete with Steel Roller Screed (Hager, 2009) ........................................................ 23 Figure 3.1: Consistency by Visual Observation (Tennis, 2004) ......................... 25 Figure 4.1: Compaction Tool. .................................................................. 37 Figure 5.1: Failed Sulfur Capping Images ................................................... 50 Figure 5.2: Sulfur Poured On Top of Standing Specimen ............................. .... 53 Figure 5.3: Failed Attempt at Pouring Sulfur ................................................ 54 Figure 5.4: Tops of Cylinders Not Level. ..................................................... 55 Figure 5.5: Visible Variation of Aggregates ........................................... ....... 56 Figure 5.6: Tops of Cylinders Not Level After Saw Cutting .............................. 57 Figure 5.7: Broken Core Samples ............................................................... 58 Figure 5.8: Compression Testing of a Non-level Cylinder Surface ..................... 59 Figure 5.9: Unit Weight of Cylinders for Batch 1 Methods 1,2,3 4 .. .............................................................................. 61 Figure 5.10: Unit Weight of Cores for Batch 1 Methods 1,2,3 4 ................... 62 Figure 5.11: Unit Weight of Cylinders and Cores for Batch 1-Methods 1,2, 3 4 .................................................................................................. 63 Figure 5.12: Porosity of Cylinders for Batch 1 Methods 1,2,3 4 ................................................................................ 66 Figure 5.12: Porosity of Cylinders for Batch 1 Methods 1,2,3 4 ................... 67 Figure 5.14: Porosity of Cylinders and Cores for Batch 1 Methods 1,2, 3 68 Figure 5.15: Compressive Strength of Cylinders Batch 1 Methods 1,2,3 4 ................................................................................ 72 Figure 5.16: Compressive Strength of Cores for Batch 1 Methods 1,2,3 4 ................................................................................ 73 Figure 5.17: Compressive Strength of Cylinders and Cores for Batch 1 Methods 1,2,3 4 ................................................................................ 74 Figure 5.18: Results for Batch 1 Cylinders Methods 1,2,3 4 ........................ 75 Figure 5.19: Results for Batch 1 Cores Methods 1,2,3 4 ............................. 76 Figure 5.20: Batch 2 Unit Weight ............................................................... 83 Figure 5.21: Batch 2 Porosity .................................................................... 84 viii

PAGE 9

Figure 5.22: Compressive Strength at 7 Days ................................................ 85 Figure 5.23: Batch 2 Method 4 Compressive Test at 7 Days .............................. 86 Figure 5.24: Porosity vs. Strength at 7 Days ................................................. 89 Figure 5.25: Porosity vs. Strength at 28 Days ................................................. 89 Figure 5.26: Unit Weight vs. Strength at 7 Days ............................................ 90 Figure 5.27: Unit Weight vs. Strength at 28 Days .......................................... 91 ix

PAGE 10

TABLES Table 4.1: Compaction Factors for estimating in-situ density (Kevern, 2009) .......... 29 Table 4.2: Summary of Specimens to be Created for Methods 1,2 &3 ...................... 31 Table 4.3: Summary of Specimens to be Created for Method 4 ......................... 31 Table 4.4: Summary of Total Specimens Needed ........................................... 32 Table 4.5: Material Properties .................................................................. 32 Table 5.1: Unit Weight (lb/ft3) for Cylinders and Cores ............................................. 60 Table 5.2: Porosity for Cylinders and Cores ........................................... 65 Table 5.3: Compression Strength Results for Method 1. ................................. 70 Table 5.4: Compression Strength Results for Method 2 ................................... 70 Table 5.5: Compression Strength Results for Method 3 ................................... 71 Table 5.6: Compression Strength Results for Method 4 ................................... 72 Table 5.7: Batch 2 Weights (yd 3 ) ............................................................... 81 Table 5.8: Unit Weights of Fresh and Hardened Concrete ............................... 81 Table 5.9: Batch 2 Results ............................................................................................ 82 Table 5.10: Prediction for 28 Day Strength Results ......................................... 87 Table 5.11: Compressive Strength Increase between 7 and 28 days .................... 88

PAGE 11

As our society continues to replace open space with parking lots building roof tops and other impervious areas storm water management has become more challenging. Government jurisdictions restrict the amount of rainwater a new development can add to the existing drainage channels. Storm water retention ponds are costly and take up usable space on a property A way to reduce or eliminate storm water ponds i s to incorporate pervious concrete as the pavement choice for parking lots. Captured stormwater can seep through the pervious concrete and directly flow into the ground replenishing groundwater for existing landscape. By having a porous concrete surface stormwater runoff is reduced. addition, pervious concrete was named a Best Management Practice and is recommended by the Environmental Protection Agency (EPA) making it a concrete worthy of further research and development. The objective of this thesis is to develop and evaluate testing standards for pervious concrete with the main focus being compaction methods. Testing standards of conventional concrete are not appropriate for pervious concrete and therefore should not be adopted as standards. Currently, the American Concrete Institute (ACI) does not have any accepted testing standards for pervious concrete but is in the process of accepting new American Standards for Testing and Materials (ASTM) specifications regarding pervious concrete Recommendations for testing pervious concrete exist but need to be evaluated for their effectiveness and consistency. Testing standards for conventional concrete cannot be used for pervious concrete and need to be rethought and modified in

PAGE 12

order to serve as tests for pervIOus concrete. The level of compaction of pervious concrete is a factor that directly impacts most relative properties of pervious concrete such as: unit weight compressive strength and porosity. The more compacted a specimen is, the less porous it is. As porosity goes down strength and unit weight are increased. Four different methods of compaction are examined and discussed with more detail as noted in the following paragraphs. In method 1 the concrete is compacted by rodding in accordance to the ASTM specification for traditional concrete. Compaction method 2 follows the new C 1688 specification for pervious concrete and uses the jigging procedure for compaction. Compaction method 3 is a new method that uses a compaction tool. Method 4 is the weight versus volume compaction method. Method 4 determ ines the weight of concrete by multiplying the desired density by the volume This quantity of concrete is compacted to the designed volume independent of compaction strikes or individual. This thesis focuses on the effects of four compaction methods and the impact they have on unit weight compressive strength, and porosity Compaction methods were compared not only on laboratory results but also on consistency and repeatability by individuals (presumably in the field) with only a basic knowledge of the subject. In comparing these four compaction methods recommendations were provided for the best suited compaction method or methods given a specific specimen type (ie. cylinder vs block mold). 2

PAGE 13

2 LITERATURE REVIEW 2.1 Description Pervious concrete has been used in the United States since the 1970 's, in Europe since the 1910's, and has recently gained popularity in the United States due to its acceptance as a sustainable building material (Offen berg 2008). The most common and likely use of pervious concrete is pavement application. Used as a pavement pervious concrete provides a durable surface that is porous and allows water to pass through into the ground. This water penetrating surface produces less surface runoff, reduces peak water flow into storm sewers and decreases or eliminates the area required for site retention ponds (Paine 1992). As defined by the American Concrete Institute (ACI) the term "pervious concrete typically describes a zero-slump open graded material consisting of portland cement, coarse aggregate, little or no fine aggregate admixtures and water. The combination of these ingredients produces a hardened material with connected pores ranging in size from 0 08 to 0.32 in. (0.20 to 0.81 cm), that allow water to pass through easily The void content ranges from 15% to 35% with typical compressive strengths of 400 to 4000 psi (2 .75 to 27 6 MPa). The drainage rate of pervious concrete pavement varies with aggregate size and density but will generally range between fall in the range between 2 to 18 gal./min /ft2 (1.4 to 12.63 Llminlm 2 ) (ACI 2006). In order for pervious concrete to gam acceptance in the industry standardized test methods need to be recognized and adopted by ACI. This typically comes in the form of specifications and testing standards. ASTM testing standards have recently been developed but not yet accepted. To date no adopted standardized tests (fresh or hardened) exist for pervious concrete. Testing standards are critical for pervious concrete to be widely accepted and utilized Testing standards must be easily performed and 3

PAGE 14

duplicated by any individual with little experience and ba sic knowledge of the subject. The consistency and behavior of pervious concrete is significantly different from conventional concrete. Normal testing procedures (for conventional concrete) based on slump and cylindrical strength are not applicable to pervious concrete (County of Fairfax Virginia, 2007), therefore the same standards should not be used for both types of concrete. New or modified testing methods need to be developed in order to accurately determine pavement performance and quality (Paine 1992) Recommendations from various sources exist on testing fresh and hardened properties of pervious concrete but no standard has been adopted by the ACI. ACI committee 522 addresses pervious concrete but as stated previously ACI 522 does not officially have any recognized tests for pervious concrete. ACI 522 1 currently recommends density testing in accordance with American Society for Testing Materials, ASTM C 13 8 (Standard Method for Testing Density for Conventional Concrete) Standard Test Method for Density (Unit Weight) Yield and Air Content of Concrete. These tests have not been widely accepted or successful for pervious concrete because the tests were developed and aimed at conventional concrete mixtures Fresh concrete properties tests for conventional concrete cannot be performed for pervious concrete because pervious concrete has a different concrete structure and contains large voids Pervious concrete also compacts differently than conventional concrete. Traditional concrete is rodded during consolidation ; however rodding does not provide adequate consolidation for pervious concrete In recognition of pervious concrete ASTM has adopted some newer testing standards In October 2008 ASTM Subcommittee C09.49 released C 1688, Standard Test Method for Density and Void Content of Freshly Mixed Pervious Concrete (Palmer 2009). The consolidation method used in ASTM 1688 is the

PAGE 15

jigging procedure. This is the first step in a series of tests necessary for pervious concrete to become a more predictable quality controlled material. Widely accepted and easily reproduced testing standards provide the industry more confidence that pervious concrete is a quality controlled and desired product. To date, pervious concrete testing standards hav e not proven to be consistent and accurate Additional research is needed to develop standardized tests that can be easily and consistently performed both in the laboratory and in the field environment. The best compaction methode s ) used for these tests is to be de v eloped. To develop standard testing methods applicable to pervIous concrete all concrete properties in the fresh and hardened state are considered and the best curing and handling practices are noted Applicable fresh properties include : consistency and unit weight. A pplicable harden e d properties include: strength durability and penneability (Kosmatka 2002). Density being the most measurable and applicable property for pervious concrete is determined for both plastic and hardened concrete (Kevem 2009). The University of Colorado has tested the performance of pervious concrete pavement systems in Denver Colorado. Recommendations for design and construction were made for pervious concrete pa v ements in Colorado s environment of fluctuating temperatures. Frequent freeze thaw cycles and very low humidity (Hager 2009) A previous concrete test pavement was constructed in a parking lot of the Auraria campus at University of Colorado Denver. extensive laboratory and field examination were included in this study The laboratory phase examined the effects of cementitious content w / cm and sa nd co ntent on th e s tructural and h y drological perfom1ance of pervious concrete mixtures was concluded that the optimum cementitious content w / cm and sand

PAGE 16

content were 525 Ib (238 kg), 0.30 and 7.5 % by total weight of aggregate. In addition it was determined that cement replacement up to 20 % with Class C or Class F fly ash could be used and meet structural and hydrological requirements Additional research demonstrated that air-entraining admixtures increased the freeze / thaw resistance of the pervious concrete mixtures The pervious concrete pavement test section contained 20 % fly ash, crushed recycled concrete as underplaying coar s e aggregate layer, and 10% replacement of sand with crushed glass i n the fine aggregate u nderlayment layer. Field investigations 0 f t he pervious concrete pavement were performed and included monitoring of heat island effects, water quality deterioration clogging and permeability. The study looked into the effects of deicing agents commonly found on Colorado's roads was found that deicing agents strip the top bonds between aggregates and accelerate deterioration of pervious pa v ement (Hager 2009). 2.2 Consistency Consistency of conventional concrete is typically measured by a slump cone test. For a particular mixture the slump should be consistent throughout the concrete placement. In the field a change in slump between batches typically means an undesired change in the ratio of the concrete ingredients ; therefore raising a "red flag" before the concrete is placed. The slump test is not applicable for pervious concrete due to the very low water cement ratios of a typical mixture (Obla 2007). Slump test results for pervious concrete would always measure close to zero (inches) due to the nature of the dry mix (Palmer 2009) therefore a new method of testing is needed to measure pervious concrete con s istency. The only existing method for determining the correct consistency of 6

PAGE 17

pervious concrete is to roll the sample into a ball in the palm of one' s hand and visually judge the consistency (Tennis, 2004). The individual is to look for an even distribution of aggr e gates and for the concrete ball to hold its shape. This testing method could be effecti v e for a single batch but very subjective and inconsistent for several batches of concrete Results will undoubtedly reflect the opinion of the person performing the test and not necessarily reflect the desired outcome. Therefore this method cannot be qualified as the sole testing method for consistency of pervious concrete. Another possible option which has been used for testing the slump of dry conventional concrete mixtures i s the modified vebe apparatus. This method can be reproduced in the laboratory and the field. is applicable for no or low slump concrete having aggregates smaller than 2 inches (Scm.) (U.S. Army Corps of Engineers 2001). Further examination of this method is needed to determine if it is a viable test option for pervious concrete. Perhaps the best current method of testing slump of pervious concrete is the inverted slump cone test. is an altered slump cone method that is suggested for testing the consistency of pervious concrete (Concrete Promotional Group 2009). With conventional concrete a slump cone is the primary quality assurance test for determining the consistency of concrete but because it is not applicable for pervious concrete, the Portland Cement Association (PCA) developed a new method to use the slump cone that c a n b e appli e d to p erv iou s concr e t e (Design of Perviou s Concret e Mixtures PCA Kevem 2009) This method attempts to reproduce the effect of pervious concrete flowing down a concrete truck chute. The test method determines workability and changes in consistency in a mix (MissourilKansas Chapter of the American Concrete Pavement As s ociation 2009). The procedure for applying the inverted slump cone test includes : 7

PAGE 18

Fill the cone with freshly mixed pervious concrete without any compacting or rodding. A full inverted slump cone is then lifted and given a mild shake in order to loosen the mix and initiate flow If the material begins to flow and drip out of the cone this indicates that the material will flow from the truck and place correctly If the mixture stays lodged in the inverted slump cone, this suggests that the pervious mixture will have a difficult time coming out of the truck and is not workable. The mixture not flowing through the inverted cone also suggests that the concrete will have a high porosity and low strength (Kevem, 2009), both undesirable results. Pervious concrete that will not flow through the inverted cone suggests an ineffective mixture that must be corrected prior to placement. is suggested that one method to alter this mixture onsite is to increase the workability by "Add 50% of original dosage of either the water reducer or the hydration stabilizer in addition to 1 to 2 gallons (4 to 8 liters) of water per cubic yard of pervious concrete ." (MissourilKansas Chapter of the American Concrete Pavement Association 2009). Additional testing of this method should be perfom1ed in order to gain a better understanding and develop control measures. 2.3 Recently acceptance of pervious concrete is based on calculated density or the unit weight of the in-place pavement (NRMCA 2004). ASTM does have a testing standard for density of pervious concrete, ASTM CI688 / Cl688M 08: Standard Test Method for Density and Void Content of Freshly Mixed Pervious Concrete. This unit weight test method is the most useful test for pervious concrete. ASTM C 1688 provides a procedure 8

PAGE 19

for determining the in-place density and void content of freshly mixed pervious concrete. The fresh concrete unit weight value is of importance because it has a direct relation to the hardened unit weight value Unit weight of hardened concrete can be directly linked to porosity and compressive strength. The unit weight test is the only standardized test that can be performed on a fresh pervious concrete ASTM C1688 states to take a sample of fresh pervious concrete and place it inside a standard measure (bucket of 0 .25 cubic foot volume or 0.007079 cubic meters) Concrete is then consolidated using a proctor hammer. Proctor hammer is a mechanical weight of 5.51bs (2.5 kg) that is dropped on a specimen (mostly used in soil testing) The density and void are calculated based on measured consolidated concrete mass volume, and total mass (ASTM C 1688 2009). However, it is believed that ASTM C 1688 calculates the density of the sample being tested and varies from field conditions Palmer, 2009). the field t h e p e rviou s c o ncr e t e i s not con s olidated the same manner as the ASTM C 1688 procedure consolidates the test specimen (Palmer, 2009). Different equipment and procedures are followed to consolidate pervious concrete in the field. Contractors typically use rollers truss or laser screeds, all resulting in different compaction levels (Johnston 2009). The ASTM standard states that "the fresh density and void content calculated from this test may differ from the in-place density and void content and this test shall not be used to determine in-place yield" (ASTM C 1688 2009). This test can be easily and consistently reproduced but it is not comparable to field conditions (Palmer 2009) Determining compaction method that can be used in a laboratory testing with similar compaction to field condition is the key to having testing standards. 9

PAGE 20

Prior to the publication of ASTM C 1688, unit weight / density of pervious concrete was tested using ASTM 138 (Method for Testing Density in Conventional Concrete). The method called for classic rodding of the concrete inside the sample in order to obtain the desired consolidation (Palmer, 2009). In normal concrete, the aggregates are evenly grated allowing the paste between the aggregates to flow When conventional concrete is rodded the mixture can consolidate properly With pervious concrete, there are little to no fine aggregate in the mixture. A very thin layer of cement paste bonds the coarse aggregates. The result is a mixture with large voids that will not flow when rodded (Offenberg, 2008). Rodding the pervious concrete does not produce consistent results because of the absence of small aggregates and nature of the dry mixture. is believed that rodding over consolidates the concrete causing the larger aggregates to settle on the bottom. Result became more consistent as water was added to the mixture. The concrete paste has more water available and was able to flow through the pervious concrete voids However, adding additional water to the mixture would create another problem Additional water would make the pervious concrete less permeable (Palmer 2009) Instead of rodding the concrete it was suggested to jig the concrete consolidating the concrete differently inside the sample. Jigging is used as a method in a test for aggregate testing and is specified in ASTM C 29. Jig ging can be used to consolidate concrete when performing both unit weight tests and cylinder tests (Paine 1992). Instead of rodding the concrete 25 times per la yer of concrete sample one would "jig" or "rock" the concrete sample back and forth 25 times in eac h direction (ASTM C29 2009). The unit weight measure if filled in three l ayers with each l ayer equivalent to one-third of the total volume of the measure. Jig ging has proved to be a more effective representation of

PAGE 21

pervious concrete consolidation as it were to appear in the field. According to the County of Fairfax Virginia the suggested quality control test for pervious concrete is the density or unit weight test. Density is detennined using ASTM 138, Test Method for Density (Unit Weight), Yield and Air Content (Gravimetric of Concrete) by following the consolidation procedures in ASTM C29 (County of Fairfax Virginia 2007) other words the County of Fairfax Virginia specifically recommends the density test be perfonned using the jigging method. When detennining the density of hardened concrete core samples can be taken in accordance with ASTM C42 Test Method for Obtaining and Testing Drilled Cores and Sawed Beams of Concrete. More than one core shall be taken in order to represent the entire placement site in accordance with ASTM D 3665 Practice for Random Samp l ing of Construction Materials After core samples have been obtained and the thickness detennined the density shall be tested in accordance with ASTM Cl40 (County of Fairfax, Virginia, 2007). ASTM C140 states the procedure for testing the unit weight / density that can be applied to hardened pervious concrete The trimmed core samples shall be emerged in water for 24 hours, allowed to drain for one minute, surface water would be removed with a cloth, and the weight measured immediately (ASTM C140 2009) Segregation of the components a mixture results a non-unifonn mixture Segregation can occur during mixing transportation placement, or compaction. Typically in conventional concrete segregation occurs due to poor handling For example coarse aggregates can sink to the bottom and separate from the concrete matrix while the paste rises to the top often the result of over vibration One can see that this is

PAGE 22

even more critical with pervious concrete due to the minimal cement paste content and ability to bond large aggregates to one another. A method in which segregation can be remedied in the field is to add water to the mixture. Typically, one gallon ( 4 liters) of water is recommended per yard of concrete. Of course, too much water increases the w / cm, thereby decreasing the compressive strength of the pervious concrete mixture (Obla 2007). Adding water to a segregated mixture should only be a last resort effort in order to save the mixture. If segregation of the aggregates occurs, the necessary paste to aggregate bond will not occur. Because the paste would rise to the surface, it might clog the pores at the top of the slab. If the pores are clogged, water cannot penetrate the pervious concrete slab. this scenario, the entire slab would be inadequate and need to be replaced. Segregation is expected to be overcome solely by careful handling. Research conducted by the University of Colorado Denver, found an impervious layer developed about 1/3 the distance from the bottom of the specimen. See Figure 2.1. This impervious layer was found in specimens produced in the laboratory setting was believed to be caused by improper compaction. The addition of a hydration stabilizer helped to prevent this layer from fom1ing in future mixtures (Hager, 2009).

PAGE 23

PelTiou s ZOllf Figure 2.1: Impervious Layer Zone (Hager, 2009) A test has been proposed to monitor segregation. The test involves 1000g of concrete to be placed through a #30 standard sieve. This sample is then vibrated for 60 seconds. The concrete matter which falls into the bottom pan is then measured. The amount of concrete passing the # 30 sieve is used to determine the allowable segregation level (Offenberg 2009). This test however is onJy mean to be used in a laboratory setting because it is too time consuming. This test is not intended for practical purposes 2.5 Curing Proper curing practice is essential to producing quality pervIOus concrete pavement (Obla, 2007). Curing is an essential step that ensures adequate hydration of cement paste in order to provide sufficient concrete bond and strength (Kosmatka, 2002). Water is a critical varia ble in pervious concrete and often needs to be adjusted in the field. Too little water can lead to improper curing. The porous surface of pervious concrete exposes more

PAGE 24

surface area to air than conventional concrete, thus making over-evaporation of the mixture water more likely to occur. If not properly cured the desired strength will not be reached causing the concrete to ravel and break away (Obla 2007) Most of the problems that have occurred in practice can be contributed to improper curing practices (Palmer 2009). The be s t s cenario in which c o ncrete i s cured would be one that allows all the moisture to remain in the mixture, allowing the concrete to cure slowly. Admixtures have been developed that help pervious concrete cure properly such as a hydration stabilizer added to the water. Topically applied coatings exist that help hydrate the concrete during curing. Effects that admixtures have on pervious concrete are not being covered in the scope of this thesis Currently it is common curing practice to cover the pervious concrete slab with a polyethylene sheathing to retain internal moisture. A study is currently being performed at the University of New Orleans that focuses on a better curing method for pervious concrete The study proposed using a spray-on application of soy bean oil between the surface of pervious concrete slab and the polyethylene cover (Offenberg, 2009). Results from this study have yet to be published and may contribute to better curing practices for pervious concrete. Curing is not a property that can be tested Instead curing is a step in placement of concrete. Best curing practices should be followed by contractors during construction in order to obtain a quality pervious concrete pavement. Hager's findings in Sustainable Design of Pervious Concrete Pavements were that proper curing of pervious concrete was even more critical when the pavement was being placed in a dry climate such as Denver CO. When box specimens were being fabricated for

PAGE 25

laboratory testing all were wrapped with 6mm (0.006 inches) thick plastic and placed in a humidity controlled environment to cure During the placement of the pervious co ncr ete parkin g l o t t es t p ave ment concrete was covered with 6 mm (0.006 inches) thick plastic sheathing. Various weights were placed on top of the sheathing in order to secure the cover from moving. addition water was sprayed on top of the concrete during the initial fourteenth day after placement (Hager, 2009) Also, in an effort to accommodate Denver s very dry climate a hydration stabilizer was added to the pervious concrete mixture used in Hager s study For conventional concrete the most recognized and required hardened property is compressive strength However in pervious concrete strength is considered but is not the property by which pervious concrete is accepted (NRMCA, 2004). Traditional co nc re t e i s typicall y t es t e d w i t h c y lind e r sample s and in ac co rd a nc e with the ASTM C39 Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens. The cylinder strength test following ASTM C39 is not an accurate method of testing compressive strength of pervious concrete. Due to the voids in pervious concrete it is difficult to obtain the proper compaction, and thus difficult to detennine the true strength o f pervious concrete. It is even more difficult to create a high strength pervious mixture by using common materials and is typically not perfonned (Fourtes 2008) Making a cylinder with pervious concrete does not follow the same techniques as those followed when the perviou s concrete is being placed in the field Pervious concrete cannot be consolidated the same way as conventional concrete. Detennining a proper compaction procedure would make it possible to test the compressive strength of perviou s concrete

PAGE 26

addition to compaction difficulty, pervious concrete cylinders have large voids and are likely to fail prematurely The axial load applied by the testing machine will cause failure along lines of larger voids within the concrete cylinder. This premature breaking is typical of dry concrete mixtures such as high strength concrete cylinders. Traditionally compacted and tested cylinders are not an accurate method of testing pervious concrete. Flexural and compressive strength test results greatly depend on the degree of compaction the sample has received (Paine, 1992). A more accurate representation of the compaction found in the field is taking a core sample and testing it for compressive strength Even though a core sample represents the concrete compaction found in the field, it can still produce unreliable results reflecting low strength When the core sample is taken, the paste structure of the hardened concrete can be damaged during the coring process (Kevem 2009). At the University of Colorado parking lot study, cylinders were cored from box specimens for the purpose of compressive strength testing (Hager 2009) Cylinder capping may produce strength results for pervious concrete similar to those in the field. Again, the compressive strength result greatly depends on the compaction technique used when the cylinders are made. But, when looking for a better method to test cylinders (with the best compaction method the user assumes), cylinder capping may be a better alternative to testing cylinders traditionally in accordance with ASTM C39. C y lind e r cappin g h as be e n s hown to be the better method of cylinder testing with dry high strength concrete mixes from 8, 000 to 12,000 psi (Torres, 2006). Cement capping is one of the most widely used methods to determine compressive strength in hollow concrete block masonry. The American Society of Testing Materials recognizes capping as an acceptable testing method for concrete cylinders in ASTM C1231-09 The capped

PAGE 27

pads deform in initial loading of the concrete cylinders. The pads are restrained from excessive lateral spreading by metal rings and provide a uniform distribution of loading from the bearing blocks of the testing machine (ASTM C 1231-09) Same principals can be applied to testing pervious concrete cy l inders a study by Ozyldirim compressive strength data were obtained from two capped cylinder methods; neoprene pads and s ulfur-mortar caps. Th es e two capping methods showed v ery little difference in compressive strength results (Ozyldirim, 1985) More research is needed to determine the compressive strength differences between neoprene pad capping and sulfur capping of pervious concrete cylinders. When examining permeability or porosity there are no applicable test methods that can accurately determine the porosity of pervious concrete in its plastic state. Permeable porosity is a basic measurement of voids in concrete. This can be easily performed once the concrete has hardened as a procedure in the density test ASTM C 1688. Porosity is a percentage of volume of sample divided by a vo l ume of sample with the voids filled with water. Porosity is very difficult to correlate in the field while the concrete is in the plastic state. Hager's study at the University of Colorado Denver, porosity was tested for the pervious concrete cylinders by using the volume displacement method (Hager 2009). This procedure was used for Batch I of this study and is described in detail in Section 4.10 A modified procedure was used for porosity testing for Batch 2 and is listed in Section 5 .6. Currently ASTM is working on a standard test method that will field test permeability titled ASTM WK17606 New Test Method for Field Permeability of Pervious Concrete

PAGE 28

Pavements. Pervious concrete contains a porosity ranging between 15-25 % (ACI 522 2009). This method is a quality control test that can determine the level of maintenance that the pervious concrete requires. All pervious concrete is recommended to be cleaned and inspected on yearly bases. ASTM WKI7606 monitors clogging and long term permeability of pervious pavements. (ASTM WK 17606) ASTM is developing a standard to test durability. ASTM WK23367 is a new test method that evaluates the surface durability potential of a pervious concrete mixture. According to ASTM WK23367 the concrete producers need a tool to assess the impact of using different raw materials to make pervious concrete. Additionally raw material suppliers need a way to access the impact of their raw materials in pervious concrete. This test is not intended to be used for acceptance. A factor of durability is with concrete resistance to freeze thaw cycles. ASTM C666 provide s a method for testing resistance of conventional concrete to freezing and thawing cycles. The procedure calls for the concrete specimen to be submerged in water for a period of time and subjected to freeze and thaw cycles (ASTM C666, 2003). This testing standard, however is not recommended for pervious concrete because it is not believed to be accurate. The repeated freezing and thawing stresses are to be resisted by the cement paste alone. Typically, pervious concrete does not contain air entrainment like conventional concrete to help resist the freeze thaw cycles. The paste thickness that surrounds the aggregate is very low causing a rapid deterioration of the specimen. In fact research conducted by Bass showed damage was found to be quite dramatic when specimens were subjected to ASTM C666 test (Bass, 2008). Pervious concrete is never likely to be submerged fully

PAGE 29

with water. Rainwater would drain through the pervious concrete and the soil below At the University of Colorado parking lot study, ASTM C666 was used to determine freeze thaw resistance. Due to porous nature of pervious concrete, the transfer frequency could not be measured and it was recommended that an alternate method was developed to determine the freeze thaw resistance of pervious concrete (Hager, 2009). In this study, Hager recorded the mass loss of the freeze / thaw beams after each 28 cycles. Thus, mass loss versus number of cycles was used to predict the pervious concrete's resistance to freeze-thaw. Though the ASTM C666 test is an aggressive test for pervious concrete, the method does provide an indication of performance when compared to other pervious concrete mixtures. For example, if two mixtures are subjected to this test method and one mixture experiences less mass loss at an extended number of freeze / thaw cycles than another mixture, the overall perfonnance can be examined. A modified version of ASTM C666 is recommended for deternlining the durability of pervious concrete. A modified version may include usin g Procedure A (freezing and thawing the pervious concrete when the sample is not submerged). Additional research is needed in this area. Since pervious concrete is expected to allow rainwater to pass through it, infiltration rate is another factor that is of importance when rating pervious concrete. ASTM C170l/C1701M 09: Standard Test Method for Infiltration Rate of In-Place Pervious Concrete tests the infiltration rate of hardened pervious concrete but does not address the flow of water through the plastic mixture This test method does not provide a method to predict infiltration rate of fresh concrete. This test is designed to identify the need for maintenance of an existing slab. All pervious concrete needs to be cleaned and

PAGE 30

monitored yearly and this test provides guidance for that. One of the limits to this test is that the infiltration rate obtained by this method is valid only for the area of the pavement being tested. If the entire pavement was to be tested multiple test locations are necessary and the results in that situation are averaged. This test method does not, however measure the influence on in-place infiltration rate due to clogging of voids near the bottom of the pervious concrete slab. is recommended that visual inspection of concrete cores is the best approach for determining whether there sealing of voids is present (ASTM 1701). At the University of Colorado Denver pervious concrete pavement test section the drain time of a pavement was measured using the clogging test (Hager, 2009 ; Delatte et. al. 2007). The cogging test measured the length of time for a given vo lume of water to drain thorough the pervious concrete pavement. The test equipment is shown in Figure 2.2. this procedure, a typical concrete compressive strength cylinder mold with a hole and stopper in the bottom was utilized The mold was filled with water, the stopper was removed and the time required to empty the mold was recorded. 2.2: 20

PAGE 31

This thesis focuses on compaction methods in laboratory testing for pervious concrete. Current ASTM pervious concrete standards can be successful in providing valid data if a compaction method is determined that is similar to field conditions. All pervious concrete ha s a site specific need and function depending on the area of the country its purpose and soil characteristics. This research does not address the pervious concrete system as a whole but the pervious concrete material itself. For example a clay soil will not allow the water to penetrate as quickly as a sand soil and therefore, the pervious concrete being placed over clay soil may require additional sub grade preparation. addition the sandy soil may be very dry and would require proper measurements to ensure the pervious concrete does not lose moisture in the bottom layers of the pavement. Different results will occur in the laboratory then the field due to varying field conditions Adjustment should be made to the conclusions of this paper before applying it to any field condition. Pervious concrete has been use d as a pavement material in Colorado and partial failures ha v e occurred at these specific locations : Safeway Grocery store parking lot located at SE corner of 13th Ave. Kremeria St., Denver constructed in 2005: Walmart Super Center parking lot located at NW corner of Tower Rd. & 1-70, Denver constructed in 2006 ; and Vitamin Cottage pa r king lot, loca t ed NW corner of Colorado Blvd & Evans Ave., Denver constructed 2007 (Hager 2008) June of2008, the Urban Drainage and Floor District of Denver (UFFCD) set in p l ace a temporary moratorium on pervious concrete placement. The UDFCD stated that the moratorium was set in place due to partial failures of pervious concrete pavements around D enver metropolitan area s u ch as

PAGE 32

the Safeway, Walmart, and Vitamin Cottage previously named. Forensic investigation was launched by UDFCD to determine the cause of these failures Speculation to the cause of failures was chemical reaction of magnesium chloride (deicing agent common in Denver) abrasions due to traffic freeze / thaw conditions, improper mix design and placement and/or curing techniques. This moratorium was removed later in 2009 shortly after the publication of Specifier's Guide for Pervious Concrete Pavement Design (Bush et. ai, 2009). Pervious concrete is inherently difficult to test and verify in-situ properties from field placed samples or cores" (Kevern, 2009) The existing unit weight testing standard ASTM 1688 are planned to be adopted by the ACI. Even the new ASTM 1688 standard only addressed one compaction method. The ASTM prescribed jigging will be evaluated as compaction method 2 (procedure explained in Section 4.4). The compaction method used governs the level of consolidation in a fabricated cylinder or other concrete mold Field consolidation and laboratory consolidation need to be as close as possible in order to obtain valid data. Performance of pervious concrete can be predicted by using unit weight of the fresh mix (Kevern, 2009) However in order to achieve the desired unit weight the best and most appropriate compaction method should be identified Figure 2.3 shows the compaction of the pervious concrete pavement at the University of Colorado Denver. In this parking lot test pavement the compaction method that was chosen included the use of a steel roller screed During fabrication of cylinders for laboratory testing the compaction method used was the rodding method (Hager 2009). 22

PAGE 33

The rodding method is the compaction method used for traditional concrete and IS referred to as method I in this study. (Hager, 2009) Compaction of pervious concrete during fabrication is directly related to its properties such as porosity, unit weight, and compressive strength. If the specimen receives a high level of compaction, it is expected to have lower porosity and higher unit weight and compressive strength. If a design unit weight is reached while concrete is still in its fresh state, it will govern the unit weight of the pervious concrete in its hardened state If one can predict hardened unit weight, he also can predict the porosity and compressive strength. Using a correct compaction method controls the outcome of the final pervious concrete pavement. Fabricated samples or cores can be used to determine hardened concrete properties. Different compaction methods shall be examined to determine unit weight, porosity, and compressive strength. 23

PAGE 34

3 One way to reduce or eliminate storm water ponds is to incorporate pervious concrete as the pavement choice for parking lots Captured storm water can seep through the pervious concrete and directly into the ground replenishing groundwater for existing landscape. Pervious concrete is a material that some owners or contractors shy away from because of the l ack of testing standar ds needed for quality control. Even though pervious concrete has been used sparingly in the United States since the 1970 's, it has recently gained popularity due to its acceptance as a sustainable building material. Pervious concrete can be found as an environmentally responsible material in projects ranging from small sidewalks and driveways to large parking lot in commercial complexes. Economically pervious concrete can be a more affordable choice than conventional concrete, particularly if it were to replace for example, a collection pond. Though the installation of pervious concrete typically requires more soil preparation good placement techniques and proper curing, the overall labor involved in the installation process is similar or lower that of conventional concrete because it does not require the finishin g steps necessary of conventional concrete. Pervious concrete has a concrete matrix with little or no fine aggregates. Primarily pervious concrete contains coarse aggregates cement and water. Minimal sand up to 10% by weight of total aggregate can be added to improve strength. Typically pervious concrete contains a void ratio between 15% -25 %. This void structure can allow three to five gallons of water per minute for every square foot (1.0 1.75 Liters / minlm2 ) of pavement. Unlike conventional concrete pervious concrete is not specified or accepted 24

PAGE 35

based on strength. Currently, there are no standardized test methods to provide the tools for quality control and assurance. The best means to ensure goo d performance of pervious concrete is through the use of a well qualified and experienced contractor. Testing standards are not in place to ensure consistency and accuracy in the field. The only recommendations in place for quality pervious concrete placement are those judged by visual observation of an individual. Figure 3.1 shows the variance in consistency that a contractor would typically use to visually judge pervious concrete mix. The criterion for just right" concrete varies with the individual and is subjective, therefore an easily reproduced standardized test needs to be developed in order to ensure quality placement. Many different test methods have been proposed for testing pervious concrete in the fresh and hardened state. However standardized tests need to be able to easily measure all the properties that are essential for a quality pavement such as porosity, density, and 25

PAGE 36

compressive strength. These tests need to be easi l y reproduced and consistent across the board regardless of testing technician. Because pervious concrete is a low slump concrete with water to cement ratios (w / cm) between 0 .25 -0.35, the slump test does not apply as a valid test method. A slump test is typically a field quality control item that tests consistency during placement. Compressive strength of concrete is tested usmg standard molded cylindrical specimens. When pervious concrete cylinders are tested, the result is an inaccurate representation of the strength of concrete in the field. Pervious concrete aggregate bonds greatly depend on the confinement within the slab and this bond cannot be found along the perimeter of a cylindrical specimen. The highly porous cylinder shear prematurely along one of its voids and would not resemble that of a core sample. Cylindrical compression tests are not recommended because of the dependency of the result on the compaction of the concrete. is believed that the compaction in cylindrical specimen does not resemble the same compaction achieved in the field with heavy machinery. A compressive test should exist that resembles the field conditions as closely as possible. Field tests are not in place for verifying the density of pervious concrete Unit weight is a test performed only as part of post construction inspection. The most important criterion for pervious concrete is the void ratio. Currently there are no methods to test the void ration until the mixture has already hardened. E valuation of e x isting and development of new testing standards is necessary for pervious concrete The compaction method is directly linked to concrete properties such as unit weight compressive strength and porosity The more compaction a specimen receives the higher the unit weight higher the compressive strength and lower the 2 6

PAGE 37

porosity. Unit weight is the only property that can be tested while pervious concrete is still in the fresh state. Being able to test concrete before placement allows an opportunity for field monitoring and correction. Unit weight of pervious concrete greatly varies depending on the compaction method used. Unit weight of fresh concrete is linked to the unit weight of hardened concrete, compressive strength and porosity. A compaction method should be developed that is best suited for pervious concrete. Determining a compaction method for pervious concrete will improve the accuracy of the standardized unit weight test. accurate unit weight value can be used to monitor the concrete during placement and predict hardened concrete properties 2 7

PAGE 38

Consolidation is a process by which concrete is compacted and decreases in volume. Consolidation occurs when stress is applied to the concrete causing the paste to pack together tightly and shrink in volume Proper consolidation is a step in measuring unit weight and fabricating cylinders for compressive strength tests. As noted previously pervious concrete consolidates differently from conventional concrete For this reason traditional testing methods cannot be applied. Consolidation is directly related to the unit weight strength porosity, and overall performance of pervious concrete. Since compaction of the pervious concrete greatly influences concrete properties an acceptable method of laboratory compaction must be developed. This research evaluated the following consolidation techniques: Method I : Traditional consolidation following ASTM C 138 where cylinder is rodded 25 times M e th o d 2 : J iggi n g m e th o d de sc ribed in ASTM C29 Method 3 : Compaction as a percentage of volume. Compaction percentage is determined based on the thickness of the slab. Compaction tool is used Method 4: Weight vs. Volume Method order to create a consistent consolidation method for testing pervious concrete in the laboratory it is necessary to understand how concrete i s consolidated in the field Pervious concrete is consolidated in the field by using compacting equipment different than that used for conventiona l concrete 2 8

PAGE 39

Two of the most common field consolidation methods are: Pervio u s conc r ete s l a b is ro u g hl y fin i s hed slightly a b ove desired elevation. A weighted roller is driven on top of the concrete unti l a desired final density is reached Pervious concrete s l ab is compacted using a roller screed. Both of these common methods compact the concrete according to Table 4.1 (Kevem, 2009) Slab thickness (in.) Compaction 4 25% 17% 8 1 3% When pervious concrete is being compacted for any test (unit weight or cylinder strength), it shou l d represent the fie l d con d ition. A specific test does not exist that can correlate the cylinder compaction to the field compaction. t h e field, the pavement percentage of compaction is listed in Figure 4.1. Compaction method 3 follows this percentage chart as a guide for compacting cylinders for laboratory testing. The compaction method used to fabricate concrete specimens for laboratory testing should simulate field condition. A recommendation for future research is to examine the difference in compaction between field condition and laboratory specimens. To avoid unnecessary variance that could occur from one batch to the next, one large batch was made. Four different compaction methods were used to concrete cylinders and a concrete box. This thesis focuses on consol i dation methods in order to test unit weight, compressive strength, and porosity. The concrete box was used to create additional specimens Four drilled cores were obtained from each box Cored samples 29

PAGE 40

were compared to cylinders for each compaction method. The cores and cylinders were compared based on measured unit weight porosity, and compressive strength. To get consistent data each consolidation method needed to fill 6 cylinders and 1 box. Unit weight was taken for each method. Unit weight for all methods was taken for all specimens by following ASTM C 138 with some modifications. The difference of unit weight procedures was the method of compaction. Method I uses the traditional rodding compaction method and is explained in Section 4.3. Method 2 follows the ASTM 1688 jigging compaction procedure and i s explained in Section 4.4. Method 3 uses a compaction tool and is explained in Section 4.5. Method 4 is the weight vs. volume compaction method and is explained in Section 4 6. After the concrete cured for 28 days, unit weight / density porosity and compressive strength were measured. Unit weight was calculated once more at 28 days for all specimens and compared to fresh unit weight. Unit weight of hardened concrete was calculated by dividing the weight of specimen by the measured volume. Porosity was calculated as a percentage of concrete voids and the procedures are described in Section 4.10. Tables 4 2 and 4.3 lists the amount of cylinders and boxes needed for each consolidation method. Samples within each method were named A B C, etc. Each sample is listed under the test that was perfonned. is to be noted that method 4 requires a different number of cylinders. Additional cylinders were fabricated for methods 1 2 and 3 in order to saw cut the cylinder and examine density as a function of cylinder height. Since method 4 specimen are 4in. x 4in. (J 0 .16 x 10. 16 cm) cylinders (half size) the saw cut test was not performed on method 4 The wet saw used for cutting cylinders into 1I3's i s not intended for cutting I in. (2.54 cm) strips. 30

PAGE 41

4.2 Mix De s i gn Batch I was made large enough to have concrete for all the methods. Each method consisted of 6 cylinders, I box from which 4 core samples were drilled, and one unit weight. additional 1 5% was mixed in ord er to accommodate waste. Table 4.4 provides the necessary volume for batch 1. Total volume of concrete batched was 3.41 Table 4.2 : Summary of S pecimen s to be Created for Consolidation Methods 1 2 &3 Saw C ut Total Neoprene Su l fur 1 / 3 UW Porosity Compressive Made Pads Capped height Test Test Test A,B, A,B, C,D, C,D, A,B,C, Cylinders E,F A,B C,D E,F E,F D,E,F A,B,C,D of 1 box Boxes made Cores from A,B, A, B,C, each box C,D D A,B,C,D S a e ummary o f S tbC C ,peClmens 0 e rea e or ons o I a Ion e 0 Saw Cut Tota l Neoprene Sulfur Porosity Compressive Made Pads Capped height Test Test Test A,B, A,B, A,B,C, Cylinders C,D, A,B C,D C,D, D, A,B,C,D # of I box Boxes made Cores from A B, A,B,C, each box C D D A ,B,C,D

PAGE 42

a e Testing Specimens Required # Volume (ft3) Compressive cylinders 24 1.40 Boxes filled for cores 4 1.37 Total 3.02 x 1.15 3.47 The remained constant at 0.30. This value is within the range of typical pervious concrete. The design density was at 124 Ib/ft3 (1986 kg/m 3). Fine aggregate was not used for the mixture From the above criteria, the mix proportions were calculated and are shown in Table 4.5 0.30 Unit Weight 124.6 Cement 550 Rock 2636 lb / cy Water 179 Inverted slump cone test was used prior to fabricating specimens for each method The inverted slump cone procedure follows: Fill the cone with freshly mixed pervious concrete without any compacting or rodding., A full inverted slump cone is then lifted and given a mild shake in order to loosen the mix and initiate flow. 32

PAGE 43

Consolidation Method Method I follows the consolidation used for conventional concrete. The concrete was rodded. Objective: Objective was to record unit weight for method I of the perviOus concrete mixture Consolidation method used for method ] was in accordance with ASTM C138 each procedure step, a note may have been added on slight differences procedure Prediction : was expected that the unit weight will be v ery low. The large voids in the specimens were expected to low unit weight v ery high porosity and very low strength. This prediction pertains to cylinders and drilled cores. Terminology : Yield v olume of concrete produced per batch cubic yard or cubic meter Air content percentage of air voids by volume of concrete Apparatus: Measure -a cylindrical metal watertight measure a yield bucket (0 .25 ft3 or .007079 m3 ) Tamping Rod (5/ 8in or 1.58 cm diameter) Mallet rubber weighing approximately 1 .25 Ib (2.75 kg) Flat plate Procedure : 1 Weigh the empty measure. 33

PAGE 44

2 Fill the measure with freshly mixed concrete in three layers of approximately equal volume Rod each layer 25 times with the tamping rod. Note: This procedure step was altered for batch 2 and is listed in Section 5 6 under title Bettering Procedures for batch 2. is to be noted that cylinders the unit weight bucket and the box were all rodded 25 times at each for batch 1. 3 After each layer is rodded tap the sides of the measure 10 -15 times with the mall e t (this proc e dure is re quired to relea s e any large trapped air bubbles). After consolidation the measure must not contain any excess of concrete protruding above (approximately 1/8 inch) the top of the yield bucket. Note : Striking off was difficult to do with pervious concrete instead the best level surface was created with a flat board. is to be noted that it is not possible to strike off pervious concrete. 4 Strik e off the top s urface with a sawing motion of the flat trowel (using little vertical pressure) 5 Clean all e x cess concrete from the exterior o f the measure (use a dampened towel if necessary and then dry). Note: Pervious concrete needs significantly more cleaning than regular concrete. Wet towel was used 6 Wei g h th e m eas ure w ith concrete. 7. Calculate the unit weight of concrete as the ratio between weight of concrete and measure volume: yconcrete = Wconcrete l Ymeasure where: yconcrete unit weight of concrete fe) W c oncrete net weight of concrete (lb) Vmeasure volume of measure (ft 3 ) 34

PAGE 45

Consolidation Method 2 Consolidation method 2 was in accordance with the jigging procedure. Objective: The objective was to record unit weight for method 2 of pervious concrete mixture. Consolidation method used for method 2 was in accordance with ASTM C29 using the jigging method. each procedure step a note may have been added on slight differences in procedure. Prediction: was expected that the unit weight will be more consistent and higher than unit weight in method I. Consolidation method 2 was also expected to have lower porosity and higher strength than method I. Apparatus Measure a c y lindrical metal watertight measure (0 .25 ft3 or 0.007079 m3 ) Procedure: I. Fill measure in three approximately equal layers (by volume) compacting each layer by placing the measure on a firm base such as a concrete floor, raising the opposite sides about 2 inches (5 cm) allowing the measure to drop in a such manner as to hit with a sharp slapping blow 2. Compact each l ayer by dropping the measure 50 times 25 times on each side Note: The procedure was followed exactly when filling the unit weight bucket and when consolidating the box. This procedure was also followed exactly when filling the cylinders. However it is to be stated that it was apparent that dropping a heavy 12xl2x7 inch (30.48 x 30.48 x 17.78 cm) box 2 inches (5 cm) 25 times each side of each layer produced a much higher compaction than when following that method for cylinders. 35

PAGE 46

3. Level the surface of concrete with fingers or a straight edge. Note: It was difficult to level off pervious concrete. A straight edge was attempted, but ultimately, a flat square plate was used to level off the box and unit weight bucket. While plate leve l ing may produce additional compaction, this was performed consistently for all the methods. Calculation I Calculate the unit weight. ybulk (G T) V where: ybulk unit weight of the aggregate, G mass of the mixture plus the measure Ib T mass of the measure, Ib V volume of the measure fe 4 .5 Method 3: C o m paction As A P e rc e ntage Of V olum e Consolidation Method 3 Consolidation method 3 was similar to the traditional compaction method 1. The only difference in the procedures was that method I uses a rod and method 3 used a compaction tool. Picture of a compaction tool is in Figure 4 .1. Objective: The objective was to record unit weight for compaction method 3. Consolidation method 3 used a compaction tool seen in Figure 4.1. Compaction method 3 follows ASTM C 138, with the exception listed in the procedure. each procedure step, a note may have been added on slight differences in procedure Compaction tool was defined as solid piece of steel tubing (without the interior 36

PAGE 47

void) that has a level surface on the strik i ng side. Compaction tool is not weight dependant. Compaction Too l Prediction: is expected that the unit weight will be higher than that for methods 1 and 2 Procedure: Follow method 1. Instead of rodding the sample during compaction use the compaction tool. Same procedure applied for cylinders compacted with the compaction tool. The only difference is the tool being used; compaction tool vs. rod. Figure 4.1 shows a typical compaction tool. This particular tool has a round plate with a diameter slightly smaller than that of a cylinder so it can easily be used for cylinder compaction. 1. Weigh the empty measure. 2 Fill the measure with freshly mixed concrete in three layers of approximately equal volume Tamp each layer 25 times with the compaction tool.

PAGE 48

Note: This procedure step was altered for batch 2 and is listed in Section 5 6 under title Bettering Procedures for batch 2. 3. After each layer is tamped tap the sides of the measure 10 -15 times with the mallet (this procedure is required to release any large trapped air bubbles) After c o n so lid a ti o n th e m e asur e mu st n o t cont ain any e x ce ss of concre te protruding above (approximately 1I8in. or .317 cm) the top of the yield bucket. 4 Strike off the top surface with a sawing motion of the flat plate. Note: Striking off was difficult to do with pervious concrete instead the best level s urface was created with a flat board is to be noted that it is not possible to strike off pervious concrete. 5 Clean all excess concrete from the exterior of the measure (use a dampened towel if necessary and then dry) Note : Pervious concrete needs significantly more cleaning than regular concrete. Wet towe l was used. 6. Weigh the measure with concrete. 7. Calculate the unit weight of concrete as the ratio between weight of concrete and measure volume : yconcrete Wconcrete / Ymeasure where: yconcrete unit weight of concrete (\b / ft 3 ) Wconcrete net weight of concrete (Ib) Ymeasure vo l ume of measure (ft 3 ) 38

PAGE 49

vs Consolidation Method 4 Consolidation method 4 is the weight vs. volume method. A weighted amount of concrete was placed into a predetermined volume. Concrete was compacted as much as necessary to be able to be placed inside the volume. Objective: The objective was to record unit weight for method 4 of pervious concrete mix Four cylinders and one box were filled per compaction method 4. The basic concept of method 4 was to compact a predetermined weight of concrete into a volume by striking it as many times as necessary. The concrete weight was determined by taking the desired density value and multiplying it by the volume. Then, the weighted concrete is place into the volume and compacted with compaction tool as necessary Prediction : was expected that the unit weight for method 4 will be higher than method 1 and 2 It is uncertain how it will compare to method 3. The porosity for the method 4 was expected to be less than any of the other methods. addition since method 4 proposes 4 x 4in. (10.16 x 10.16 cm) cylinders instead of 4 x 8in (10.16 x 20 32 cm) cylinders it was expected that the concrete tests higher in strength than any of the other methods Procedure: This method was proposed by Matthew Offenberg, U.S. Technical Service Manager at R Grace & Co. (Offenberg 2009) According to his experience 4 x 4 in. (10.16 x 1 0.16 cm) cylinde r s are a best representation of the field 39

PAGE 50

conditions Instead of filling six 4 x 8 in (10 .16 x 20 32 cm) traditional cylinders six 4 x 4 in. (10.16 x 10.16 cm) cylinders will be filled instead. order to keep consistent with the other methods a 12 x 12 x7 in. (30.48 x 30.48 x 17.78 cm) box will be filled as well to provide 4 full height s amples. Proposed procedure for compacting 4 x 4 in. (10.16xlO .16 cm) concrete cylinders are as follows: 1 Place cylinder on the scale and set scale to zero. 2. Mark with dark marker a line that represents a 4 in. (10.16 cm) height inside the cylinder. 3. Fill the cylinder until the scale reads a desired weight. Density (psi) Weight of concrete (lbs)/volume of 4 x 4 in. (10.16 x 10.16 cm) cylinder (inch 4. Compact cylinders with compaction tool until it reaches the marked 4in height. Note: Cylinders are approximately half full and should weigh the same amount. Procedure for filling the 12 x 12 x 7 in. (30.48 x 30.48 x 17.78 cm) box: 1 Place box mold on the scale and zero. 2. Fill the box f ull, no need to weigh the box at this point. 3. Compact the box with compaction tool until the concrete settles 12 inch (1.22 cm) below the top surface in order to make room for additional concrete 4 Place the box on the scale and fill with additional concrete until the desired calculated weight is reached 40

PAGE 51

5. Take the box off the scale and place on a level surface (such as concrete floor). 6. Compact with compaction tool until the concrete is level with the top edge of box. Note: Since compaction level can depend on strength of the strike the number of strikes is not important only the wei g ht of the box and the volume. 7. Do not strike off. The entire weighted concrete must occupy the volume. Compact as many times as needed in order to fit the concrete in its volume. Note: The same method can be followed for the unit weight bucket or any shape of desired concrete mold The procedures in method 4 were developed by studying the ideas of Matthew Offenberg. Matthew Offenberg is the internationally recognized expert in the field of pervious concrete Offenberg is a chair of the American Concrete Institute (ACI) 522 committee the founding chair for National Ready Mix Concrete Association (NRMCA) and is currently the secretary for American Society for Testing and Materials (ASTM) committee C09.49 for pervious concrete. Offenberg holds a bache lor's and master s degrees in civil engineering from Purdue University and is registered as a professional engineer. Offenberg's test was performed for the purpose of testing durability for pervious concrete, not compaction. In this thesis the focus is on compaction methods only. Durability is a property that can be directly linked to compaction. The more compact a section is, the higher unit weight it will have, the smaller porosity and the higher durability The procedures performed by Mr Offen berg are l isted: 1 4 x 4 in. diameter cylinder is cast by weighting the amount of concrete that would produce the desired density for that volume

PAGE 52

2. The filled cylinder is dropped on the floor ten times from I in. height. 3. Cylinder is covered and cured for 7 days. 4. Concrete specimen is removed from cylinder mold at 7 days. 5 Single specimen is placed in L.A. Abrasion machine and rotated 50 times. 6. Mass is determined of remaining specimen. 7. Mass loss is calculated Mass loss % (initial massFinal Mass) / Original Mass) 100 Again the proposed pervious concrete durability test procedure was not performed in method 4 However, the procedure was the bases of developing the compaction method 4 used for this study. 42

PAGE 53

Four boxes were fabricated using each method. A total of 4 drilled cores were removed from each box addition, six cylinders were fabricated following methods I, 2, and 3. Four cylinders were fabricated following method 4 because saw test was no performed on method 4. All cylinders and cores were tested for compressive strength at 28 days of age for batch I. Cylinders A and B for each method were testing using Neoprene pads Cylinder C and D for each method were tested using Sulfur caps Cylinders E and F for testing methods I, 2, and 3 were cut into 113' s. The porosity will be meas u red for the bottom 113, middle 113, and the top 113. Note : During cutting of the specimens, t he aggregates unraveled along top and bottom surfaces. This was be l ieved to have created artificially low porosity for top and bottom thirds of the specimen. This data was not used. The four core samp l es will be tested u sing Neoprene pads Each of the four consolidation methods had a concrete box made. The concrete box was large enough to core four cy l inders from. Two of the cores were tested using neoprene pads and two using sulfur caps. The following is the procedure for making concrete cylinders: Objective: Cylinders and cores were tested by following ASTM 05: Compressive Strength of Cylindrical Concrete S p ecimens Predictions: I. was expected that all the cylinders from method I fail prematurely under the compression testing machine. Since previous concrete has many large voids, it was expected that the failure plane would be along these voids. The 43

PAGE 54

axial load would not be evenly distributed because of the void structure. The uneven load distribution and the presence of larger voids were expected to cause the specimen to break prematurely giving it a lower compressive strength value. 2. Methods 2, 3 and 4 were expected to have higher compressive strength values 3. was expected that compressive values would be between 400 1500 psi (4.1-10.48 MPa). These values are typical for pervious concrete not containing fine aggregates and chemical admixtures. Procedure: These procedures are followed for conventional concrete cylinders. This method was followed for this study. The procedures that were followed were in accordance with ASTM C39. The Specimens were compacted with different compaction methods, but tested for compression following the same procedure. Cylinders and box was taken out of the curing room. 2 Cores were drilled out ofthe boxes with typical 3inch coring drill. 3. Measured diameter and length for each specimen 4. Place neoprene pads confined by steel end caps on either end of the specimen. Only do this step for specimens that are to be tested with neoprene pads 5. Place the concrete specimen into the compressive testing machine. 6 Zero the compressive testing machine 44

PAGE 55

7. Use the metered advance mode and slowly apply load until the specimen breaks 8. Calculate the compressive strength per the calculation in ASTM C 39 [2005]. 4.8 Neoprene Pads Objective: Cylinder samples A B and core samples A B C, and D were tested with neoprene pad s accordance with: ASTM C 123lM 09: Use of Unbonded Caps Determination of Compressive Strength of Hardened Concrete Cylinders. Predictions : 1 It was expected that cylinders and cores tested with neoprene pads would have lower results than those tested with sulfur caps. Since Sulfur Capping was not successful, this theory was never proved. 2. All the methods are expected to have cores and cylinders with uneven surfaces Procedure : Procedure followed ASTM C 1231 09. 4.9 Capping Sulfur Capping: Sulfur capping is a method that caps hardened concrete cylinders or cores with a high strength sulfur mortar. Sulfur capping provides a plane surface on ends of hardened specimens. 45

PAGE 56

Objective: Cylinder samples C and for each method were sulfur capped. No core samples were sulfur capped. The sulfur capped cylinders were tested for compression. The liquid sulfur will be made in laboratory, and the cylinders capped forming a solid smooth surface for each specimen The level surface will accommodate the pervious concrete void at the top of the cylinder helping distribute the weight more evenly throughout the cylinder. Predictions: is expected that all the sulfur capped cylinders will have higher compressive strength than the cylinder tested with neoprene pads. Procedure : The procedures will be in accordance with ASTM 617 and ASHTO T 231 1. Sulfur chips were placed in a crock-pot filling it full. The croc-pot was left on high until the chips melted. 2 Capping plate was cleaned and placed on a level surface. 3. Capping plate was filled W' with liquid sulfur. 4. Pervious concrete specimen was dipped inside and left to cool for 5 to 15 seconds. 5. The capping plate was hit with mallet to loosen the hardened cap from the cap mold. 6. Capped specimen was lifted out of the capping mold. 46

PAGE 57

Porosity: Porosity is taken for concrete specimens at 28 days of age. Porosity is not measured in the mixture s fresh state. Porosity is the percentage of volume of voids in the volume of concrete specimen. Objective: All cylinders and drilled cores for all four methods will be tested to obtain p o rosit y Predictions : The higher the porosity, the lower the unit weight and compressive strength Method I is expected to have the highest porosity. Methods 2 3 and 4 are expected to all have lower porosity than method 1. Procedure : 1. Obtain a container that can house the specimen. For 4 inch (10.16 cm) wide diameter specimens, a 6 inch (15 24 cm) wide plastic cylinder was used in order to prevent the specimen from getting stuck inside the 4 inch (10 .16 cm) mold during testing. 2. Measure the diameter and length of concrete specimen. 3. Place specimen inside the container 4 Slowly pour water into the container until the water surface reaches the top of the concrete specimen. Do not fill the container full with water.

PAGE 58

5. Leave the concrete specimen in the container for 30 seconds. 6. Drain the water into a graduated cylinder and measure the volume of water. 7. Calculate the porosity as the Volume of Water divided by Volume of Concrete. 48

PAGE 59

One large batch was made to test all four consolidation methods. The mixture was v ery dry but did not present any of the problems that are typical when dealing with a low mixture. The mixture did not show any signs of clumping (as common of a low mixture) Small amounts of water were added to ensure water distribution throughout the concrete batch. The concrete mixer was at approximately half capacity and was tilted during mixing at approximately 35 degrees Samples for method through 4 were fabricated in chronological order This was to be noted because method samples were coarser than samples for method 2, 3 or 4. Because the batch was very large in volume, the concrete mixer could not be tilted such that the opening was parallel to the floor Concrete removed from the mixer first was used for fabrication of method cylinders. The larger aggregates came out of the mixer first resulting in method cylinders to have larger aggregates that methods 2 3 or 4. This error was not apparent during cylinder fabrication but rather at 28 days. It was expected that method would have even lower compressive strength values than expected because of the uneven aggregate gradation Method 1 and method 2 cylinders were created at the beginning of the batch and two more were created toward the end of the batch. Figure 5 5 shows the variation in the gradation of the aggregates between the beginning and the end of the compacting process. Figure 5.1 shows the drastic difference in aggregate size from sample A and F for method 1. 49

PAGE 60

As indicated in Figure 5 .1, sample A from method I (located on the left side of the photo) has significantly larger aggregates than that of sample F of method I (located on the right side of the picture). This was a source of error due to the uneven aggregate distribution that occurred during the large batch. Segregation such as this could be avoided if a smaller volume of concrete was batched. When following consolidation method 1 it appeared that the rod made voids in the cylinders that were irreversible. The rodding appeared to create additional large voids and did not help consolidate the concrete. When following consolidation method 2 jigging the concrete did not seem to be beneficial while filling the cylinders. The cylinders were rocked back and forth as stated by the jigging procedure However the dry mixture did not shift enough to aid consolidation The experience was different when jigging the box. Because the concrete 50

PAGE 61

box was much heavier it was difficult to jig the box back and forth as the jigging procedure stated. Instead, each side of the box was lifted (rocked), but due to its awkward size and weight the comers were dropped on the floor each time. is likely that the large box was additionally compacted by the floor with each drop. was expected that the cored samples would be compacted significantly higher than the cylinders for method 2. When following consolidation method 3 compaction was visually observed as the volume decreased. The compaction tool was dropped 25 times on each cylinder The cylinders were not compacted in layers After a cylinder was compacted the top of the concrete was about 2 (5 cm) below the top of cylinder mold. The cylinder was then topped with additional concrete and compacted again 5 times For consolidation method 4 the calculated concrete weight was placed in the cylinder Each cylinder was compacted using the compaction tool. Some were compacted 10 times, some were compacted 40 times The point of this procedure was to compact the concrete specimen to the desired weight disregarding the number of compaction hits The c y linders were filled and compacted without too much effort. However, when compacting the 12 x 12 x7 in. (30.48 x 30.48 x 17.78 cm) box method 4 proved to be difficult. was challenging to place the desired weight into the predetermined volume. appeared as if the volume was too small. The box for method 4 was compacted for about 8 minutes with an average of over 150 compaction hits. After method 4 cylinders were filled it was predicted that the core s amples would be inconsistent with one another and possibly ha v e lower values than the cylinders.

PAGE 62

Sulfur Capping When starting to talk about compressive strength tests it should be first stated that sulfur capping data was not performed. Due to the difficu l ty of sulfur capping the cylinders it was detem1ined that sulfur capping weak pervio u s concrete is not practical. The concrete mix that was derived with this batch used only large aggregates no small aggregates and no chemical admixtures. This type of mixture typically results in low compressive and therefore low tensile strength values. With that in mind it was not possible to s ulfur cap the low strength concrete cylinders. As per the specified procedure in Section 4.9 the melted sulfur was placed in a capping mold The top of the cylinder was dipped to produce a cap thickness in accordance to ASTM C 617 found in Section 4 9 After the sulfur cap hardened, the specimen was hit with a metal rod to release the cap from the mold Each time this method was attempted the top aggregates of the concrete cylinder wou l d break away from the cylinder itself. The bond between top aggregates was very weak As a result, aggregates would crumble away from the cylinder. Figure 5.2 shows a picture of the sulfur mold containing top aggregates that broke away from the cy l inder. 52

PAGE 63

Since the sulfur capping procedure failed, another procedure was tried. Instead of dipping the cylinder in a sulfur mold the cylinder was placed in its 4in. (10.16 cm) x 8in. (20.32 cm) cylinder in order to have liquid sulfur poured. This pouring method was performed on the low strength concrete in order to prevent the weak bonds between the top aggregates from breaking. Figure 5 3 shows the liquid sulfur being poured over the specimen while in a concrete cylinder. 5 3

PAGE 64

However the hot liquid sulfur did not harden instantly. Instead, the sulfur melted between the aggregates filling the top voids of the concrete The sulfur also trickled down the sides of the specimen. Figure 5.4 shows the liquid sulfur running deep into the voids and down the sides. 54

PAGE 65

w as determined that this method would alter the porosit y and density of the s pecimen to the point where it would no longer be a true representation of pervious concrete therefore this method was dismissed from this study. was determined that sulfur capping could not be applied to the pervious concrete used in this study. is to be noted that it is possible to produce a higher strength pervious concrete that would have fewer voids between the aggregates which could be s ulfur capped using the original capping mold should be investigated to determine if higher strength pervious concrete can be sulfur capped. All the compressive strength tests were done using neoprene pads instead of sulfur caps Problems with Preparing Samples Preparing and gathering samples for compressive testing turned out to be a complex task. First the concrete cylinders had to be removed from the plastic 4in. x Sin 0 .16 x 20.32 55

PAGE 66

cm) c y linder mold s the process of removing the concrete cylinders samples were subjected to striking and vibrating force. Methods I and 2 (thought of as less compacted methods) were easily removed out of their molds, while cylinders in methods 3 and 4 were removed with more difficulty A few of the concrete cylinders were partially damaged in this procedure and caused the top (ends) of the cylinder surfaces to break and become uneven. Figure 5 5 shows a photograph of a typical uneven concrete cylinder. Due to the lack of small aggregates in the mix a typical pervious concrete cylinder had uneven top (end) and bottom (end) surfaces, which represents a problem because in a standard concrete compression test the load is applied at the ends of the cylinders. A flat even surface is necessary in order to apply the load evenly throughout the specimen. This 5 6

PAGE 67

was not the case with all the porous samples, but it did result in less than satisfactory compressive load values. The more uneven the top (end) surface, the worse the concrete cylinder was expected to perform in the compression test. This proved to be the case for several porous samples. As the compressive load distributed unevenly throughout the sample, it caused premature failure. To solve the problem of uneven surfaces and to ensure proper testing results specimens were saw-cut with a wet saw. The saw-cut cylinders provided better samples for compressive testing, however, not significantly better. The very high porosity of the samples proved difficult to cut evenly with a wet saw and some samples chose to break and fall apart when subject to the ever slight vibrations of the wet saw. As previously noted and as depicted in Figure 5.6 the top (end) of the saw-cut sample was relatively flat albeit still an unevenness to the cut edge. Uneven point 57

PAGE 68

The box samples were cored with a 3in. (7.62 cm) diameter core bit. Another problem that arose with core drilling was found in methods 1,3 and 4. Four samples should have been obtained from each box for each compaction method. Instead of drilling 4 cores each, Methods 1 3, and 4 had less applicable core specimens. Method 1 only produced one valid core because the other three cores broke while drilling. Methods 3 and 4 had only two cores that could be tested The other two cores broke apart during core removal. The bond between the aggregates was too weak, breaking the cores along the weakest plane. Broken cores separated about one-third a distance from the bottom (end) of each core. Figure 5.7 shows method 3 core samples. Method 2 was the only consolidation method that produced four drilled cores. The reason method 2 cores did not break during drilling is because they had a higher compressive strength than the other methods. 5.7: Core Breaking of the core samples occurred because the pervIOus concrete had a low compressive strength If the pervious concrete had higher compressive strength it would 5 8

PAGE 69

not be likel y that the drilled cores would break prematurely Based on batch I fmdings drilled cores are not likely to break if compressive strength is higher than 1000 psi (6 .89 MPa) interesting observation is that all the cores that broke while in the drill separated approximately one-third distance from the bottom (end). Based on the success of drilling cores from the boxes one could predict that method 2 would produce the highest compressive strength values. is likely the cores with the highest tensile value will not separate while in the drill therefore, method 2 was predicted to have the highest compressive strength value for cored samples. When testing cylinders and cores it was important that the end surfaces be level. Sometimes even with saw cutting a level surface was not obtained. Figure 5.8 shows a method 2 batch 1 specimen that had a non-level top surface This is not ideal and should be avoided 5 9

PAGE 70

5.2 Unit WeightIDensity Results -Batch 2 The calculated unit weight was expected to be 124.6 lb /ft3 (1995.9 kg/m 3 ) is to be noted that some cores were not tested as originally planned because they broke during the drilling process. The unit weight for all specimens was taken at 28 days. Table 5.1 lists the unit weight obtained for batch 1. T bl 5 1 U t W ht (lb/ff) C d d C a e : elgl or n ers an ores Method 1 Method 2 Method 3 Method 4 Cylinder A 101 106 116 98 104 120 123 C 98 101 101 123 96 102 123 E 107 F 89 Average 98 103 121 Core A 93 109 106 128 105 103 114 C 105 105 Average 93 106 104 122 Method 1 was the furthest from the desired density. The rodding compaction method produced unfavorable unit weight results as suspected. The rodding only created additional voids in the specimen and did not compact the specimen. Method 4 core and cylinder testing averages were similar. Since method 4 uses volume and desired density to calculate the weight of each cylinder or box, it is likely that method 4 would have similar densities between cores and cylinders. Figure 5.9 graphs the averaged unit weight / density for each of the compaction methods Method 1 value represents the average of 6 cylinders. Method 2, 3, and 4 values 60

PAGE 71

represent the average of 4 cylinders The lowest unit weight is in method 1. The second lowest unit weight was recorded for compaction method 2. Unit weight values for methods 1 and 2 are very low. Method 3 had the second highest unit weight value. Method 4 had the highest unit weight value and the closest to the desired unit weight. The desired unit weight was obtained only by method 4. Compaction methods 1, 2, and 3 had unit weight less than desired. Unit weights for cylinders in methods 1 and 2 were too low to even consider applicable Dry Unit Weight of Cylinders 130 120 110 100 90 --..c 80 70 -60 -50 -= 40 30 20 10 0 1 2 3 4 # of s p ec imen s : M 1 6 M2. M3. M4 4 ComPletion Method Figure 5.9: Unit Weight of Cylinders for Batch 1 Methods 1,2,3 4 Figure 5.10 shows the unit weight! density of the cored samples derived from the box for each of the 4 methods Method 1 value represents the average of 1 core. Method 2 value represents the average of 4 cores. Method 3 and 4 values represent the average of 2 cores. The lowest density is again obtained from method 1 as expected. The second lowest was

PAGE 72

recorded for method 3 The second highest unit weight was obtained by method 2 Similar to the cylinders method 4 core samples had the highest density. Method I was the only compaction method that produced really low unit weight values. Methods 2 and 3 had lower unit weight values than desired, but they are still considered acceptable. 11\ .Q 130 120 110 100 90 ;: 80 -60 50 == 40 30 c: :::l 20 10 0 Dry Unit Weight of Drilled Cores 1 3 # of specime n s a v eraged : MI-I,M2-4,M3&4-2 Figure 5.10: Unit Weight of Cores for Batch 1 Methods 1,2,3 4 It was predicted that the density would fall in chronological order with method I being lowest and method 4 being highest. This can easily be explained by manner in which method 2 (the jigging method) was performed on the box. Like previously discussed in section 5.1, it was found to be very difficult to jig such a heavy 12 x 12 x 7in (30.48 x 30.48 x 17.78 em) box Instead of gently jigging the box back and forth, the box was repetitively dropped on the concrete giving the method an additional level of compaction. That is most likely the reason why the method 2 box was more compacted and had higher 62

PAGE 73

unit weight than the method 3 box This also explains the significant difference in unit weights between method 2 cylinders and cores. The jigging method is only effective when a larger mold is being fabricated. The compaction level in method 2 box is reached with the help of its self weight. 140 120 -m 100 g 80 c: 111 60 .c 111 40 3: 20 0 Unit WeightJDensity of Cylinders and Cores 1 2 3 4 # of specimens averaged (cores): Ml-l M2-4 M3&4-2 # of specimens averaged (cyl.) : M2, M3&4-4 Figure 5.11: Unit Weight of Cylinders and Cores for Batch 1 Methods 1, 2,3 4 Figure 5.11 shows both the average cylinder and the average core density values for each method. Method 1 cylinder value represents the average of six cylinders. Method 2, 3, and 4 cylinder values represent the average of four cylinders. Method 1 core value represents the average of one core. Method 2 core value represents the average of four cores. Method 3 and 4 core value(s) represents the average of two cores. By looking at the graph, the following observations are made: Method 4 average density for cylinders and cores are similar. As intended in method 4, a sample can be compacted as many times as needed with the

PAGE 74

compaction tool in order to fit the calculated weight into a predetermined box. The same compaction is achieved with method 4 regardless of cylinders or boxes Also method 4 does not depend on the strength of the person using the compaction tool. Method 1 provided least compaction for the box and cylinders. The unit weight results were significantly low and furthest away from desired density. The results were so low that method I is considered not applicable. Method 2 produced the second lowest density when averaging cylinders and cores. The jigging method did not produce adequate unit weight values for cylinders. However, the jigging method was successful for cores. Method 3 ranked third in overall density Method 3 used a compaction tool during fabrication of concrete cylinders. compacted the concrete 25% by volume and is believed to accurately represent field conditions. other words 8 inches of concrete in a cylinder was compacted into a specimen sample 6 inches tall. Both core and cylinder unit weights had similar values. Method 4 produced the highest unit weight values and achieved the desired unit weight. Since the specimen was fabricated with a calculated weight being placed into a set volume. This procedure was considered to be successful. The density values are for the most part in chronological order as expected, with method 1 having the lowest density and method 4 having the highest. 64

PAGE 75

The porosity was measured for each compaction method. Table 5.2 lists the porosity values for batch 1. Method 1 Method 2 Method 3 Method 4 Cylinder A 37.3 31.2 16.9 19.7 B 30.6 33.5 20 2 23.1 C 33.0 30.7 21.2 19. 7 32.4 37.0 24.9 24.3 E 46.2 F 38 7 Average 36.3 33.1 20.8 121 Core A 40.9 22.8 20.7 25.9 B 24.4 21.9 16.7 C 20.2 24.2 Average 40.9 22.9 21.3 21.3 From the porosity data, method 1 has a very high porosity. This porosity value was most likely caused during the rodding process. The rodding created large voids inside the cylinders and the box. There is a great variation between the average results for the cylinder and cores for method 2. Method 2 box was more compacted than the cylinders, as previously discussed in Section 5.1. Therefore, method 2 cores have fewer voids than method 2 cylinders. Methods 3 and 4 cylinders and cores have porosity readings between 20.8% and 23.8 %. Figure 5.12 shows the porosity of cylinders for batch 1. Compaction method I cylinders are most porous. Method 2 cylinders had the second largest amount of voids. Methods 3 and 4 had the lowest porosity values. Based on porosity data, it is expected for methods 3 and 4 to have higher densities and compressive strength values 65

PAGE 76

of Cylinders 30 ,-.. e -.;;; Q ... Q o # of s pecime ns: M 1 6 M2, M3, M4 4 Figure 5.13 shows the porosity of drilled cores for each method. Method 1 has the highest porosity at 40. 9%, making this core sample too porous and not applicable .. Rodding pervious concrete is not an appropriate compaction method because it creates too many voids. After the specimen is rodded at each layer the voids are not filled with surrounding concrete (as typical behavior for conventional concrete). A practical value for porosity is between 15-30%. Cores methods 2 3 and 4 all have acceptable average porosity values between 2l.3% and 22 .9%. 66

PAGE 77

Porosity of Drilled Cores 40 :::R '-' 0 0 0 Q., o 2 3 # of specimens averaged : MI-I M24 M3&42 Com(Xlction Method Figure 5 13: Porosity of Cores for Batch 1 Methods 1, 2,3 4 Figure 5 .14 shows porosity percentages for cores and cylinders for all four methods. Method I cylinder value represents the average of 6 cylinders. Method 2 3 and 4 cylinder values represent the average of 4 cylinders. Method I core value represents the average of I core. Method 2 core value represents the average of 4 cores Method 3 and 4 core value represents the average of 2 cores Method I is the only method that can be named impractical due to the overall high porosity values. Again, rodding pervious concrete cylinders does not compact the concrete in order to produce acceptable results Instead it creates additional unnecessary voids, making the pervious concrete in method 1 too porous The more porous a specimen is, the weaker and less durable it will be for practical use. Cylinders for methods I and 2 had significantly lower values than the cores. Rodding and jigging 67

PAGE 78

methods compact boxes and cylinders at different levels. The jigging method only performs well for larger molds The concrete self weight inside the box mold compacts itself during the jigging procedure. The heavier specimen (such as the box) was more compacted than the smaller specimen (cylinder) mainly because it was heavier. This level of compac ti o n ca nnot b e achie ve d for 4 x 4 in. (10 .16 x 10.16 cm) cylind e r molds with the jigging procedure Larger cylinder molds such as 6in (15.2 cm) or 8 in. (20.3 cm) diameter are not expected to produce more favorable values during the jigging procedure because they are 50 % heavier. of Cylinders ---c 0 0 3 4 o f s p ecime n s averaged (cores) : M 2-4 M3& 42 of s p eci m ens averaged (cyl.) : M2 M 3 &4 4 Compressive strength of cylinders and cores alike were tested for all 4 of the compaction methods at 28 days. Rodding compaction method I had the lowest compressive strength 68

PAGE 79

Low compressive strength was expected based on low unit weight and high porosity data from Section 5 1 order to understand the compressive strength values it is important to discuss what is typical of this type of pervious concrete mixture Pervious concrete will vary in strength from 400 to 4000 psi (2.75 to 27. 6 MPa) depending on the application and the pervious concrete mixture. The mixture that was used for all the consolidation methods was from the same batch. This mixture did not include fme aggregate or chemical admixtures. This type of mixture can exhibit compressive strength values below 1000 psi (6.89 MPa), but is more common to see values between 1000 to 1500 psi (6.89 to 10.34 MPa) (Offenberg, 2009). According to an interview with Matthew Offenberg, a more desirable range for pervious concrete compressive strength is between 1500 to 2500 psi (l0.34 to 17.23 MPa). Most of the compressive strength values for all the compaction methods turned out somewhat lower than expected. Table 5.3 shows that cylinders and core samples were equally low in compreSSive strength, and therefore impractical for pervious concrete use when using Method 1. The low compressive strength in the rodding method explains all the problems that occurred during the experiment. To summarize from earlier, some of the major problems were: raveling aggregates during cylinder separation from the plastic molds, broken samples during coring, and unexpected raveling while saw cutting. All of these problems are directly related to low compressive strengths and low tensile strengths 69

PAGE 80

5.3: Method 1 Cylinder Core A C A Tested 0.0 0.0 277 272 192 128 3378 Strength (psi) Average 217 338 Strength (psi) Table 5.4 states that cylinders compacted with method 2 had very low compressive strengths averaging 350 psi (2.41 MPa). was expected that the core samples have slightly higher compressive strengths. The core samples had compressive strengths averaging almost four times higher than the cylinders compacted in the very same method. This unexpected difference can be traced back to the jigging method variance from cylinders to box molds. The 4 x 8in. (10.16 x 20.32 cm) plastic cylinders were rocked back and forth as specified in the jigging procedure in ASTM C29 This proved to be more difficult to do with the heavy 12 x 12 x 7in. (30.48 x 30.48 x 17.78 cm) pervious concrete box Instead of gently rocking the block form, the heavy box would hit the bottom of the floor as it was being lifted up I in. (2.54 cm) on each side. The box was compacted in part by its own self weight. Since the cylinders did not have this additional compaction, they were compacted significantly less than the box. Method I Cylinder Core A C A C Tested 355 398 387 257 1008 1167 896 1354 Strength (psi) Average 349 1106 Strength (psi) Table 5.5 states that method 3 cylinders and cores have average compressive strengths within 10% of one another. Method 3, using a compaction tool resulted in relatively 70

PAGE 81

consistent compressive strength data Since the compaction tool was round and slightly smaller in diameter than the 4 x 8 in. (10.16 x 20.32 cm) cylinders it seemed to work very well for cylinders However the round compaction tool did not prove to be a logical shape for a square box The compaction tool could not get close enough to the edges of the bo x in order to equally compact the comers. The shape of the compaction tool had a negative outcome on compaction of the box The drilled cores were compacted less than the cylinders. A compaction tool with a right angled edge should be tried to compact the pervious concrete inside of the box molds 3 Method 3 Cylinder Core A C A Tested 979 866 880 789 Strength (psi) A v erage 923 834 Strength (psi) For method 4 the weight vs. volume method a calculated weight of pervious concrete was compacted in a predetermined volume. The weight was determined by dividing the desired unit weight by the volume of the box This method proved to be by far the most consistent method when examining compressive strength Some cylinders in this method were compacted with 15 strikes and some compacted with 35 strikes. The point of this method is not dependant on the strength of the individuals striking the concrete All the cylinders weighed the same amount to the nearest hundredth of a pound. The box mold was placed on the scale and compacted until the desired weight was reached As shown in Table 5 6 the difference between the cylinder and core values was less than 1 % Method 4 produced consistent compressive strength results between cylinder and core samples.

PAGE 82

Method 4 Cylinder Core A C A Tested Strength (psi) 935 1010 906 985 1005 898 Average 959 952 Strength (psi) Figure 5.15 shows methods 1 and 2 having very low compreSSIve strength results unacceptable for practical use. Methods 3 and 4 produced similar compressive strengths of 923 psi (6.35 MPa) and 959 psi (6.61 MPa), respectively Method 4 cylinders had compressive strength values that were 4% higher than those of method 3. c. 0.0 ... .:! ... c. e = u 0 2 3 4 # of s pecimens: av eraged

PAGE 83

When examining the compressive strength of the cores, the rod ding method 1 fell short of the expected value and was too low to consider for the comparison. Method 2, the jigging compaction procedure had the highest compressive strength value of 1106 psi (7.58 MPa). Method 4 produced second highest strength of952 psi (6.56 MPa) Method 3 had the third lowest value of 834 psi (5.75 MPa). The compaction tool method 3 has 28% less compressive strength than the jigging method 2. Method 1 cylinder value represents the average of 6 cylinders. Method 2, 3, and 4 cylinder values represent the average of 4 cylinders. The method 1 core va lue represents 1 core. The method 2 core value represents the average of 4 cores. The method 3, and 4 core value represents the average of 2 cores. c. .. c J.. .. ell J.. e 0 0 2 3 4 # of s pecimen s averaged: Ml-l,M2-4,M3&4-2 73

PAGE 84

Figure combines the cylinder and core compressive strength data in order conclude which method is adequate for laboratory compaction of pervious concrete mixtures Observations made regarding this data include: Method 1 the rodding method does not produce adequate compressive strength values and is not being compared with the other 3 methods. Method 2 the jigging method proved to be the best compaction method for the box form which the cores were obtained. However, jigging does not produce adequate compaction in cylinders Method 3 produced the third highest compressive strength. Results between cylinders versus cores have a 10% difference. Method 4 weight vs. volume method produced the most consistent cylinder vs. core results differing less than 1 %. All of the compressive strength results were lower than expected and similar test should be performed with a stronger pervious concrete mixture Compressive Strengths of Cores and Cylinders 1200 1000 800 +-------------1 600 400 -----------1 200 2 3 Compaction Method 4 # of s pecime n s averaged (cores): MI-I M2-4 M3&4 -2 # of s pecimens averaged (cyl.) : MI-6 M2,M3&4-4 Cylinders Cores Figure 5.17: Compressive Strength of Cylinders and Cores for Batch 1 Methods 1,2,3 4 74

PAGE 85

Conclusions from the result data shown b y Figures 5.18 and 5.19 include: Method 1 the rodding method produces cylinders with too high of porosity, too low of density and too low of compressive strength. Rodding should not be used as a compaction method for pervious concrete. Method 2 the jigging method for cylinders produced high porosity, low density and low compressive strength.. These are unacceptably poor results. Method 3 using a compaction tool produced desired porosity of 21%, 12% lower than expected density and a compressive strength value of 923 psi (6.36 MPa). While this compressive value is acceptable for pervious concrete, it is still a very low value when being compared to traditional concrete strengths. Method 4 weight to volume method has highest density, midrange porosity of 24%, and highest compressive strength for the cylinders at 959 psi (6 60 MPa). ,-. .;;; ....., CJI c (/) .;;; 0 0 ... Method 2 Core f-------Melhed-4-GyliooeF'---=----------Method 3 Cylinder Method 4 Core 0 M ethod 3 -------- Method Method 3 Method 2 Method 2 Cylinder ... Method I Core Method 4 ___ ------'+. Method I Cylinder 75

PAGE 86

Q. .c C 00 Q. e 0 0 0 ... Method 2 Core ----------=--Method-4 Core Method 4 C y linder Method 3 Cylinder -.Method 3 Core Method Method 3 .6. Method 2 M e thod I Core ... M e thod 2 C y linder Another batch is necessary in order to make solid conclusions regarding batch 1 results Batch 2 was redesigned to incorporate lessons learned from batch 1. First, batch 2 included a revised mixture design with the inclusion of flne aggregate and a hydration stabilizer. Batch 2 only addressed cores. Cylinders were not examined. addition, a concrete mixture will be batched for each method instead of one single large batch. Curing of the test blocks was better performed. Chapter 5 has discussed the result data, difflculty that arose performing some procedures and unfavorable results. Section 5.6 will list some of the major problems that arose during batch 1 A solution will be named for each problem in order to obtain more favorable results in batch 2. 7 6

PAGE 87

List of problems for batch 1 and solutions to be applied toward batch 2: Problem : Absence of small aggregates made it more likely for large aggregates to unravel from the cylinders and cores. This unraveling of aggregates during regular handling created unnecessary voids and a non-level testing top surface and additional unnecessary voids. Solution: Use different mixture design that incorporates fine aggregate. In addition use a smaller batch size when testing. Problem: Absence of small aggregates contributed to low compressive strength and thus low tensile strengths The low tensile strength contributed to breaking the core samples while still in the coring machine. Some specimens were too fragile and broke during coring Solution: Use different mixture design that incorporates small aggregates. Make a smaller batch to closely monitor the consistency of mixture in order to produce a higher strength pervious concrete. Problem: Low compressive and tensile strength made it difficult to saw cut cylinders and core samples in order to obtain a level top and bottom surfaces. Solution : Use different mix design that incorporates small aggregates Perform better curing practice in order to achieve higher strength. Wrap the specimen entirely in plastic not just the surface of the block. Problem : Bond between the aggregates was too weak. More moisture was needed to react with the cement paste. Solution: Hydration stabilizer could be added Better curing practices should be considered. Instead of covering only the top of molds with plastic cover the entire box with plastic This will prevent the moisture from evaporating prematurely potentially giving a stronger specimen.

PAGE 88

Problem: The very large batch was difficult to handle in the laboratory facility. The mixer could mix the necessary volume but it was apparent that segregation occurred producing different aggregate sizes for the different compaction methods Solution: Batch 2 should be smaller in volume. Problem : Due to the large batch total mixing time was between 15-30 minutes Solution: To error on the side of caution, batch 2 shall be smaller in volume and total handling time shall not exceed 10 minutes. Another added benefit for this mix may be that it would retain more moisture if handling time were reduced. Problem: Since cylinders are generally considered not applicable for testing pervious concrete Batch 1 volume was too big. Solution: batch 2 will strictly be focused on cores taken from fabricated boxes. Problem: Compaction tool used for filling boxes was round, but the box was square. Solution: Consider a rectangular compaction tool in order to better compact the corners. Problem : There was much room for error in the porosity test. Water was filled to the top of each specimen Since the "top of each cylinder or core may have been uneven individual judgment was used to determine where the true top is This created unnecessary room for error because by limitations of the human eye Solution: New method is to be used for measuring porosity Problem : The inverted slump cone did not provide adequate for determining the consistency of the mixture in Batch 1 Solution Repeat the inverted slump cone procedure for batch 2 and compare results. 78

PAGE 89

Better procedures for batch 2: Method 1 -Each layer (layer = 1/3 volume) rodded lO times for unit weight bucket and 25 times for the box. Method 2 Unit weight bucket jigged 10 times for each layer (layer 1/3 volume) and 25 times for the box. Method 3 Unit weight bucket compacted with round compacting tool lO times for each layer (layer 1/3 volume). Square compaction tool used for box and compacted 25 times for each layer (layer 1/3 volume) Method 4 Concrete compacted by striking the box with square compaction tool until desired weight was achieved for the 12in. x 12in. x 7in. volume of the box Air content was not taken for any of the methods. batch 1 all air content seemed to be between 18.5% and 19%. The void ratios (porosity readings) varied between 20-40%, but the air content stayed mostly consistent form one compaction method to the other. Clearly, this is incorrect. is not accurate to measure the air content of pervious concrete. Each batch 2 compaction method was handled within 10 minutes of batching the concrete. Hydration stabilizer and fme aggregate meeting ASTM C33 were used. All boxe s were completely wrapped in plastic preventing any air from escaping in order to better the curing process (batch 1 plastic only covered the top surface). 79

PAGE 90

New method was used to measure porosity called the modified volume displacement method. A large volumetric measuring bowl is filled with 600mL of water. Each specimen is expected to displace the water based on its volume. The difference between expected volume displacement and actual volume displacement is noted. The difference over the volume of concrete specimen is the void ratio or porosity. No samples were cut down with a wet saw. Samples appeared to have level top and bottom surfaces. Four boxes were cast in batch 2, one for each method. Each box was cured until 6 days of age. All specimens were cored on day 6, and compressive strength, unit weight and porosity was measured at 7 and 28 days of age The same properties measured during the batch 1 phase were measured during batch 2. The results from batch 2 cannot be directly compared with the results in batch 1 because the two concrete mixtures were different. The four compaction methods were compared to one another for batch 2. In addition consistency of each compaction method within the batch themselves was compared The batch 2 mixture design was very similar to batch 1. One difference is that fine aggregate was included at 6 % by weight of total aggregate. Also, 6 oz of hydration stabilizer were added to every 100 (59.3 kg/m 3) of cement. The total volume of concrete needed for one consolidation method was 0.61 ft"3. Batch 2 was repeated four times; one per method. Table 5.7 provides the mixture characteristics for batch 2 80

PAGE 91

0 a eo: eigl Cement 550 lb / cy Rock 2477 Ib/ cy Sand 160 Ib/ cy Water 178 lb / cy 6 oz.!cwt Table 5.8 lists the unit weight for each method All the results were averaged between two specimens a e 0 OJ elgl Unit Volume Weight Dry Unit Weight % Method Weight (lb) (fe) at 7 days (lb /ft3 ) Difference 1 29 .95 0.25 120 109.3 8.8 2 30 55 0 .25 122 123. 1 0.8 3 29 8 0.25 119 117.8 1.2 4 31.55 0 .25 126 126. 9 0.5 The fresh unit weight taken from the unit weight measure was compared to the dry unit weight of the specimen for each method. The rodding method 1 produced inconsistencies and an overall poor result. Unit weight taken from the fresh concrete and the hardened concrete cylinders was compared for methods 2, 3, and 4. The results have 0 5% to 1 2% difference between the unit weights. Method 4, the weight vs. volume method has the most consistent measure for unit weights of fresh and hardened concrete. Table 5.9 summarizes the 7 day compressive strength porosity and unit weight data for the four compaction methods. Two of the four cored specimens were tested at 7 days. The remaining two cored cylinders were tested at 28 days of age. Unit weight porosity and compressive strengths were taking for each core.

PAGE 92

5 9 B h 2 7 D a e a c esu sa ays Compressive Strength Porositv (%) Unit Weight lbs / ft3 ) (psi Average Average Average Method A B A&B A B A&B A B A&B 1 23.1 29.0 26.0 108.0 110.6 109.3 830 1076 953 2 17.6 17.6 17.6 124.8 121.5 123.1 2257 2130 2193 3 23.4 24.0 23.7 117.8 117.8 117.8 1435 1900 1667 4 15. 8 14.9 15.3 126 6 127.2 126.9 4028 4424 4226 In comparison with batch 1, batch 2 had a lower range of values for porosity. In addition, batch 2 had a higher range of values for unit weight, and significantly higher compressive strengths at 7 days. A quality pervious concrete varies in porosity unit weight, and strength based on what type of mixture is desired. Different characteristics may be desired depending on the pavements application and environment. Porosity generally ranges from 15-35%, unit weight is typically between 135 to 140 (1824 to 2242 kg/m 3), and compressive strength for small aggregate mixtures varies between 1500 to 4000 psi (10.34 to 27.56 MPa). The batch 2 results exceeded the val ues measured during the batch 1 phase and are more typical for pervious concrete. Porosity ranged between 15.3% for method 4 to 26% for method 1. Figure 5 20 shows the difference in porosity between each method. All of the values in Figure 5.20 are averaged between two specimens. Method 4 porosity is close to the lower boundary for pervious concrete mixtures Method 4 may not be desired for all applications. Method 2 also produced a lower porosity of 17.6%. After the porosity test was performed, it was expected that method 2 and 4 have higher compressive strength values than method 1 and 3 strictly basing that on porosity results 82

PAGE 93

Porosity of Drilled Cores for Batch 2 40 1 30 26.0 ,-. 20 0 0 c.. 10 o 1 2 3 Compaction Method A v era ges repr es ent 2 s pecim e n s Figure 5.20: Porosity of Drilled Cores at 7 Days Batch 2 unit weights were higher than batch 1 results. Figure 5.21 shows the differences in unit weights between each method Batch 2 method 1 proved to once again have the smallest unit weight. All of the v alues in Figure 5.21 are averaged between two specimens While batch 1 methods 2 and 3 seemed to have similar unit weights in batch 2 this is not the case batch 2 method 2 has a higher density of 123.1 (1979.4 kg/m 3 ) (2 5 % less than design density) and method 3 has a density of 117.8 Ib/ft3 (1886 kg/m 3) (5.5 % less than design density). Method 4 density was 2 % higher than design density at 126. 9 Ib/ft\2040. 5 kg/m 3). 83

PAGE 94

---123.-1-----ct: .c c C = ...... ... .c 0 2 3 verage s represent specimens The most noticeable difference in results between batch 1 and 2 was the compressive strength. Figure 5.22 shows the compressive strength results for batch 2 at 7 days of age. All of the values in Figure 5 22 are averaged between two specimens Method 1 the rodding method again proved to be not applicable for pervious concrete Rodding produced the lowest compressive strength out of the four methods for both batches. Surprisingly method 2 was much higher than method 3. This difference did not exist in batch 1 data where the jigging compaction method produced slightly stronger results. The 7 day compressive strength data for methods 2 and 3 produced a coefficient of variability of only 6.0 % and 2 3 % respectively. Method 4 produced the highest strength at 7 days of age with an average of 4226 psi (29 .12 MPa) Due to the low 15. 3 % porosity and higher unit weight of 126.9 Ib/ft3 (2032 kg/m3) a higher compressive strength was 8 4

PAGE 95

expected However batch 2 method 4 produced a concrete with a compressi ve strength to conventional concrete mixtures The outcome is at the low end of porosity readings and high end of strength reading The application of thi s particular pervious concrete may be more specific and not widely used. This low porosity pervious concrete may be desirable in areas where greater strength performance is needed and low porosity is not an i s sue. 4500 Co '-' ..c:: = 100 -------100 Co E3 0 0 2 3 verages r e pr ese nt s pecim e n s Figure 5.23 shows a picture of method 4 sample failing at 4426 psi (29 .12 MPa) This sampled failed in a manner which is typical for conventional concrete -A very large pop was heard as ultimate strength was reached Only method 4 cylinders in batch 2 failed in this manner The coarse aggregate failed in method 4 cores. Batch land batch 2 methods 1 and 3 failed by the aggregates unraveling and crumbling away from the sample Again method 4 of batch 2 had v ery unique behavior common for conventional concrete. 8 5

PAGE 96

Higher readings for method 4 are contributed due to the better curing practices and careful handling of the pervious concrete mixture. All methods in batch 2 produced fav orable results, and this is directly linked to proper handling and curing. While batching the pervious concrete was carefully handled and cured in a more favorable manner than batch 1 When batch 1 was placed in the molds only the top was covered with plastic and the block was placed in the curing room for 28 days However with batch 2 the entire box was enclosed in a plastic wrap ensuring no moisture could escape. The key to having a stronger pervious concrete is to greatly consider adding hydration stabilizer and enhancing curing this laboratory test it was easy to wrap the 8 6

PAGE 97

boxes in plastic wrap. the field, it is recommended that the contractor topically coat the pervious concrete for additional hydration and covering the pavement with plastic as soon as possible. Locking in the moisture during curing produces higher compressive strength. The compressive strength, porosity, and unit weight of batch 2 data was obtained at 28 days of age. Predictions were made about the 28 day test results from other documented strength data for pervious concrete (Mahboub, 2009) Based on 7 day strengths, 28 day strength values are predicted in Table 5.10 Compressive strength is expected to be 15% higher at 28 days than at 7 days of age (Mahboub, 2009). Compressive Strength (psi) Tested at 7 days Predicted for 28 days Method 1 830 955 Method 2 1462 1681 Method 3 1454 1672 Method 4 4226 4860 One can see that the more compaction is achieved for pervious concrete, the less porosity it will have. This will result in higher unit weight and higher strength. The actual values obtained for strength at 28 days of age are listed in Table 5.11. Compaction method 1 had the highest increase in compressive strength of 19. 3% between 7 and 28 days. Method 3 had the second highest increase in compressive strength of 13.8% between 7 and 28 days. Method 2 had the third highest increase in compressive strength of 10% between 7 and 28 days Method 4 had the smallest increase in compressive strength between 7 and 28 days It is to be noted that the increase if compressive strength between 7 and 28 days is related to the ultimate strength val ue. The 87

PAGE 98

highest strength pervious concrete had the smallest increase in compressive strength and the lowest strength pervious concrete had the highest increase between 7 and 28 days of age Compressive strength ipcreased from 7 to 28 days between 6.8% and 19.3 %. Compressive Strength (psi) Tested at 7 Increas Tested at 28 days Predicted for 28 days days e% Method 1 1180 1096 953 19.3 Method 2 2437 2522 2193 10. 0 Method 3 1935 1917 1667 13.8 Method 4 4533 4860 4226 6.8 Figure 5.25 shows the direct relationship between porosity and strength measured at 7 days of age. Figure 5 26 shows the same relationship between porosity and strength at 28 days. This data represents the average between two specimens. Lower porosity produces higher strength. As the poro sity increases, the percentage of the voids increases and results in lower strength val ues. The highest porosity was measured for the rodding method 1 Consequently, this method resulted in the lowest strengths. The weight vs. volume method 4 produced the lowest porosity and the highest compressive strength 88

PAGE 99

.;; t c ell CIJ ell .;; '" ell .. c. <5 4500 4000 3500 3000 2500 2000 1500 1000 500 0 5000 4500 4000 :: 3500 c J,., -rJ) 3000 2500 0 Porosity vs. Strength at 7 Days 5 10 15 20 25 30 Method I Method 3 Method 2 Method verages represent specime n s Poros ity Figure 5.24: Porosity vs. Strength at Days Porosity vs. Strength at 28 Days r------------------------.---------------2000 + Method I Method 2 "'Method 3 Method 4 c. 1500 1000 500 o o 5 10 15 20 25 Porosity 30 Figure 5.25: Porosity vs. Strength at 28 Days 89

PAGE 100

Figure 5 27 shows a direct relationship between unit weight and strength measured at 7 days of age. The lower the unit weight the less compacted the specimen, and the lower the unit weight the less the compressive strength. The method with the lowest unit weight and strength was the rodding method I The method that resulted the highest strength and unit weight was method 4, or the weight vs. volume method. The second highest unit weight and strength was produced by method 2 the jigging method. Figure 5 27 shows the same relationship between unit weight and strength at 28 days 4500 --4000 3500 ..c: CiI 3000 c ... 2500 U5 .. 2000 ;; 1500 ... Co E 1000 0 500 0 105 Unit Weight vs. Strength at 7 Days 110 115 120 125 130 Method 1 Method 3 Method 2 Method 4 Unit Weight (Ibs/ft 3) A v erage s repre s ent 2 s pecim e n s Figure 5.26: Unit Weight vs. Strength at 7 Days Figure 5.28 shows the same relationship between unit weight and strength at 28 days of age. The higher the unit weight the higher the compressive strength Method 4 had the highest compressive strength and unit weight. 90

PAGE 101

5000 4500 c. 4000 .c 3500 3000 Q,j Q,j 2500 2000 "iii Q,j 1500 c. 1000 0 500 0 1--Unit Weight vs. Strength at 28 Days __ -c+ Method 1 Method 2 Method 3 eMethod 4 105 120 125 130 3 Ave r ages r e pr ese nt 2 s p eci m e n s Unit Weight (lbs/ft ) Figure 5.27: unit Weight vs. Strength at 28 Days

PAGE 102

These conclusions were derived from both batch 1 and 2 results. Batch 2 was a higher strength pervious concrete mixture that followed better handling and curing practices. addition, batch 2 contained 6% ftne aggregate by weight of total aggregate and a hydration stabilizer. Rodding pervious concrete produced unfavorable results for both batch I and 2. The higher the porosity, the lower the density compressive strength. Unit weight is a valid test that can be used to predict performance of the hardened pervious concrete mixture The jigging procedure can be used as a valid consolidation method for larger specunens ; however is may not b e an appropriate method when fabricating cylinders. Method 4, weight vs. volume method, produced the most consistent results regardless of the individuals experience or strength, and regardless of the shape and size of specimen being molded is the recommended method for cylinders. Method 3, the compaction method with a tool produced a lower level of compaction than the jigging method and has more variability than data produced using the jigging method. Pervious concrete handling time should be further investigated. Batch 2 was hand l ed in less than half of the time as recommended and produced dramatically better results Curing pervious concrete proved to be a bigger issue than originally expected. Batch 2 had a better curing method and outperformed batch 1 signiftcantly. Only 9 2

PAGE 103

the top surface of batch I boxes was covered with plastic. Batch 2 boxes were wrapped all around wit h plastic and produced far better results Compaction is a significant factor that affects a pervious concrete s porosity unit weight, and compressive strength. Using a concrete saw to cut specimens proved only to be a valid option when the concrete strength was greater than 1000 psi (6.89 MPa). Shape of the compaction tool should match that of the specunen mold A rounded compaction tool should not be used to compact pervious concrete inside a box. The inverted slump cone did not provide an adequate indication of mixture consistency like the traditional slump cone Batch I and batch 2 appeared to have similar inverted slump cone behavior, yet, they performed very differently. 93

PAGE 104

Recommendations for pervious concrete procedures can be found in Section 5.6. The following recommendations are made toward bettering pervious concrete testing practices. Also, this section will answer questions previously raised in this thesis. When calculating porosity of cores or cylinders, the displacement volume method is to be used. provided more accurate results and is not dependant on an individual's discretion. This method proved to be favorable for cylinders and for cores removed from boxes. Batch 2 method 4 was able to produce a higher level of compaction Thus this method produced higher compressive strength and unit weight values when compared to the other methods Based on the research herein method 4 is the recommended method for laboratory compaction of pervious concrete mixtures. Drilled cores provide more consistent and accurate results than concrete cylinders. Compaction method 2 proved to produce favorable results for cores but not for cylinders. Sulfur capping is not a practical method capping the surface of pervious concrete cores or cylinders when the compressive strength of the pervious concrete is low. Aggregates will break away from the specimen during sulfur capping. Proper curing is even more critical for pervious concrete than originally stated Extensive efforts should be taken to ensure the pervious concrete retains as much moisture as possible. Hydrations stabilizers should be added to pervious pavements. However a typical pavement will have different curing procedures. Use of existing curing products 94

PAGE 105

and development of new products (like soy bean oil discussed in the literature review) should be highly encouraged. The lower the units weight of the spec lIDen, the higher the porosity and lower the compressive strength. Porosity is governed by the design value derived from a site specific examination of the soils and the pavement's needs. order to obtain the desired porosity, one can alter the design unit weight of the pervious concrete. The fresh unit weight is directly related to the fmal dry unit weight taken from the sample at 7 and 28 days of age Measurements taken for unit weight of fresh concrete is the only test that can monitor the pervious concrete as it is being placed in the field. Unit weight is the most relevant test that can be directly linked to the compressive strength and porosity of hardened concrete The measured value obtained from the fresh mixture is governed by the compaction method used to fill the unit weight bucket. The compaction method which produced the most consistent results from fresh to hardened unit weight is method 4 or the weight vs. volume method The second most effective compaction method was method 2, or the jigging method which is specified by ASTM 1688. order to widely accept pervious concrete, standardized tests need to exist that are adopted by the ACI. The standardized tests need to be performed in the field during concrete placement to ensure consistency and good handling and curing practices. Fabricated test samples for laboratory testing need to truly represent the pervious pavement. The ASTM Cl688 standard specifying the jigging method of compaction should be only used for unit weight measures and box samples. The jigging compaction method is not recommended for cylinder testing. is recommended to core drill fabricated boxed samples for pervious concrete and not use conventional cylinders. 95

PAGE 106

ACI Committee 522 Pervious Concrete 522R-06 American Concrete Institute. Farmington Hills, MI. ASTM C1231 C1231M 09: Standard Practice for Use of Unbounded Caps in Determination of Compressive Strength of Hardened Concrete Cylinders ASTM C138 C138M 09: Standard Test Method for Density (Unit Weight), Yield, and Air Content (Gravimetric) of Concrete. ASTM C C 1688 M-09: Standard Test Method for Density and Void Content of Freshly Mixed Pervious Concrete. ASTM C 1701/ C 1701 M -09: Standard Test Method for Infiltration Rate of Place Pervious Concrete. ASTM C29 C29M 07: Standard Test Method for Bulk Density ("Unit Weight") and Voids in Aggregate. ASTM Standard Test for Compressive Strength of Cylindrical Concrete Specimens. ASTM C42: Test Method for Obtaining and Testing Drilled Cores and Sawed Beams of Concrete ASTM D 3665: Practice for Random Sampling of Construction Materials. ASTM C666 C666M 03: Standard Test Method for Resistance of Concrete to Rapid Freezing and Thawing. Bass, W. (2008). Read y Mix Concrete Association. County of Fairfax, Virginia. (2007, December 21). Retrieved from www.fairfaxcounty.gov / dpwes/publications/lti / 08 01.pdf Delatte, N., Miller, D., Mrkajic, (2007). RMC Research & Education Foundation. Fourtes, R.M., Merighi, J.V., Bandeira (2008). Laboratory Studies on Performance of Porous Concrete. Hager, S. (2009) Denver CO: University of Colorado Denver. 96

PAGE 107

Johnston, K. (2009, August). Pervious Concrete: How to Properly Place and Cure. Retrieved October 9,2009, from website: http://www.allbusiness.com/construction/specialty-trade-contractorsIl2938813-I.html Kevern, IT., Schaefer, V.R., Wang, K. (2009, November). Predicting Performance of Pervious Concrete using Fresh Unit Weight. Kosmatka, S., Kerkhoff, B., Panarese, W.C. (2002). 14th edition, Skokie Illinois USA: Portland Cement Association. Mahboub, C., Canter, J Rathbone R., Robl, T., Davis B. (2009, November December) American Concrete Institute. ACI Materials Journal VOL 106 NO 6, pg. 526. MissourilKansas Chapter of the American Concrete Pa vement Association. The Concrete Promotional Group. NRMCA. (2004). 38 Retrieved October 4,2009, from National Ready Mix Concrete Association website: http://www.nrmca.org/aboutconcrete/cips/38p.pdf Obla, K.H. (2007 September). Silver Spring Maryland, U.S: National Ready Mix Association. Offenberg, M. (2008, February). Is Pervious Concrete Ready for Structural Applications. 48-49. Retrieved September 10,2009, from website: http://www.strllctllremag org!article.aspx?articleTD=532 Offenberg, M. (2009, September 25). Ozyldirim C. (1985). Neoprene Pads for Capping Concrete Cylinders. Paine J. (1992). Portland Cement Association. Palmer, W. D. (2009, April). Concrete Construction Magazine Retrie ved from http://www.Concreteconstruction.net. Tennis, P.D., Leming, M L., & Akers, D J (2004). Skokie, lllinois; Portland Cement Association. Torres, S (2006). Capping Systems for High-Strength Concrete. 45-53 Retrieved October 5,2009, from the Transportation Research Board we bsite: http: // dx.doi.org!10.314111979-08 97

PAGE 108

United States Environmental Protection Agency. (1999, September). Office of Water Washington, DC: United States Environmental Protection Agency. U.S Army Corps of Engineers. (2001, December) American Concrete Institute. Yang 1., Jiang G. (2002). Department of Civil Engineering Tsinghua University, Beijing, Peoples Republic of China. 9 8