420 likes | 564 Views
Performance of Pervious Concrete Pavements Marty Wanielista, Manoj Chopra, Matt Offenberg Joshua Spence and Craig Ballock. Stormwater Management Academy University of Central Florida Orlando, FL 32816 wanielis@mail.ucf.edu. Outline of Presentation. Overview Background and Current State
E N D
Performance of Pervious Concrete Pavements Marty Wanielista,Manoj Chopra, Matt Offenberg Joshua Spence and Craig Ballock Stormwater Management Academy University of Central Florida Orlando, FL 32816 wanielis@mail.ucf.edu
Outline of Presentation • Overview • Background and Current State • Objectives of this On-going Project • Research Plan • Progress to Date • UCF Test Site • Field Tests • Discussion
Overview • Pervious or No-fines Concrete – mixture of coarse aggregate, Portland Cement, admixtures and water • Increased Porosity due to limited fines and 15-20% air voids • Strong need for Current and Updated Assessment of Pervious Pavements due to new regulations pending for Stormwater Management
Overview • Issues to be addressed – • Design Section • Construction Methods • Acceptance Criteria • Infiltration Rate Performance • Credit for Replacement of Impervious Area • Our research will initially address – • Design Section • Infiltration Rate Performance • Credit for Replacement of Impervious Areas
Background and Current State • Replacement of Impervious Areas with Properly Designed and Constructed Pervious Paving Surfaces is Desirable • Treating pervious concrete as a system with pavement and sub soil • ACI Committee 522 has been formed to develop Guidelines for the use of Portland Cement Pervious Concrete
Historical and Literature Review • PC Pervious Pavements have been used for past 20 years in Areas of Lower Traffic Loads (parking lots, shoulders, airport taxiways, some state and local roads). • Must have suitable • Subsoil Conditions • Groundwater Locations
Historical and Literature Review • Field et al (1982) Water Resources Bulletin – detailed information on PP. • Florida Concrete and Products Association (FCPA) – Portland Cement Pervious Pavement Manual (No. 605) • EPA (1999) – Stormwater Technology Fact Sheet on Porous Pavements • Several recent articles from USC and Purdue, as well as UK, Japan and China.
FDOT Interests • Need for a permit, or credit (partial or total) for substituting pervious surfaces • Based on Volume of water that can be Stored and allowed to Replenish the Aquifer • Want answers to – • What is design – materials, dimensions, GWT? • What are proper construction methods? • What is the infiltration rate for the system?
Advantages and Disadvantages(EPA, 1999) • Advantages - • Recharge to Local Aquifer • Water budget retention and pollution removal • Less need for Storm Sewers • Disadvantages – • Lack of Construction Experience and Expertise • Clogging • Cold Weather Problems
Construction Specifications • Specifications for contractor certification, materials and mix design, construction practices, and post construction care • Sources from EPA, California-Nevada Cement Promotions Council PC Specs, and PCI Systems, LLC. PC Specs
Construction Specifications • Appropriate mix proportions • +/- 5 lbs/CF of design unit weight • Discrepancies are generally related to water content • Too much water – reject load
Construction Specifications • Concrete should be stricken off ¼ to ½ of an inch about the form boards and compacted to level • Compaction – roll with a 10-inch schedule 40 steel pipe • Curing Time – pavement should be covered a minimum of 7 days
Construction Specifications • Limit frequency of heavy traffic – e.g. construction vehicles, garbage trucks, etc. • Remove or Limit sources of sediment • Signage such as “ADOPT A LOT” • Curbing should be used to direct infiltrating water downward and to prevent erosion at the edges of pervious concrete slabs.
Design and Construction Specifications • Cities of Stuart, Zephyr Hills, Winter Park, and Titusville and the Counties of Citrus, Hernando, Pasco, and Hillsborough have adopted specifications. • Credit is being determined for use by other Cities and WMDs. • A goal is to have 24 cities and counties with pervious concrete code language for pervious concrete. • Contractor Certification will be an Important Factor • Soil Preparation, Curbing, Field Infiltration Tests and Inspections will be Important.
How are Pervious Systems Working? • Develop New Embedded Single Ring TestMethod to Measure Infiltration rates • Laboratory Testing – Build Two Test Cells at the UCF Stormwater Laboratory Site and a Control Chamber • Field Testing – Four field sites in Central Florida and one in Tallahassee
Preparation of Test Cells • Stormwater Laboratory Field Sites • Two 6 ft.x 6ft. x 4 ft. deep Chambers • 5 inch thick pervious concrete pavement • One cell has a “reservoir” of 3/8 inch coarse aggregate to increase storage • Soils were Sandy (Type A hydrological) compacted in 8 inch lifts to 92% Standard Proctor to about 104 lb/ft3
Development of Embedded Single Ring Infiltrometer • Double Ring Infiltrometer on the surface of Pervious Pavement not Suitable due to Preferred Lateral Migration of Water • Led to Concept of Single Embedded Infiltrometer • Depth of Embedment is an Important Parameter (Initial Assumption = 14 inches including the 6 inches of pavement) • 12 inch Diameter (11-5/8” ID) with 11-Guage Steel
Pervious Concrete Core 6” 20” Subsoil 11-5/8” 11-Guage Steel Embedded Single Ring Infiltrometer
Pervious Concrete Core 6” 20” Subsoil 11-5/8” 11-Guage Steel Embedded Single Ring Infiltrometer • Advantages • One dimensional flow (no horizontal flow between pavement and soil) • Representative of site existing conditions assuming same soil types, • and concrete conditions.
Results at Test Cells • Using ASTM D3385-03 (Double Ring) procedure adapted to embedded Single Ring • Initial Double Ring Tests on Bare Subsoil before Concrete Placement have yielded infiltration rate of 2.6 in/hr • Without compaction, the rate for the soil was 12-20 in/hr
Results of UCF Embedded Ring Tests * System or concrete core plus soil infiltration rates
Preliminary Observations from UCF Test Chambers • Pervious Concrete Pavement and Subsoil System displays Infiltration Rates nearly equal to Subsoil Alone • Infiltration rates of the system are greater than the minimum rates of 1 in/hr used for the design of FDOT retention areas.
Laboratory Control Chamber • Better “Control” • Address issues such as Clogging and Water Table Impact • The Chamber was Filled with Sandy Soils from UCF Stormwater Lab. (Type A Hydrologic Group ) • Filled in 8” lifts to 92% Standard Proctor
Pervious Concrete 12” 20” Plastic Tank 4’ ½” PVC Pipe Outlets @ 4’, 3’ & 2’ from top of tank Subsoil 2’ Laboratory Control Chamber
Field Site Reconnaissance • Vet Office in Sanford • FCPA Office in Orlando • Sunray StoreAway – Lake Mary • Strang Communications – Lake Mary • FDEP Office - Tallahassee
Field Testing Progress • Six cores at Sunray Storaway, Three at Strang Communications, Three at FCPA, Six in Tallahassee, and Three at Murphy Vet Clinic. • Field infiltration tests completed at all locations • Laboratory tests using Control Chamber on-going
Field Testing Process • 12-in diameter cores • Run field tests • Collect soil samples • Lab work on soil samples • Lab test on core infiltration rates
Field Test Results *Age of concrete varies from 10 to 20 years (except for Site 4 – Area 1, which is 1 yr).
Conclusions • Proper construction is important (water in mix, curing); Certification program is needed. • Specifications need to be followed for design and construction; Good design practices (curbing, pavement thickness). • Pavement and Subsoil must be treated as a SYSTEM.
Conclusions • Infiltration rates are comparable to Stormwater Retention Ponds. • Water storage is directly proportional to the porosity and the depth to the water table. Modeling efforts currently underway
Thank You!For additional information:Please see www.stormwater.ucf.eduor call 407.823.4143Questions?