39
REPORT OF SUBSURFACE EXPLORATION AND GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE STATION #6 THREE CHOPT ROAD AND OLD FREDERICKSBURG ROAD GOOCHLAND COUNTY, VIRGINIA FOR Mr. Josh Bennett Moseley Architects, P.C. 3200 Norfolk Street Richmond, Virginia 23230 May 2, 2014 ECS Project No. 03:11203

REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

  • Upload
    lamthu

  • View
    216

  • Download
    1

Embed Size (px)

Citation preview

Page 1: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

REPORT OF

SUBSURFACE EXPLORATION AND GEOTECHNICAL ENGINEERING ANALYSIS

HADENSVILLE FIRE STATION #6

THREE CHOPT ROAD AND OLD FREDERICKSBURG ROAD GOOCHLAND COUNTY, VIRGINIA

FOR

Mr. Josh Bennett Moseley Architects, P.C.

3200 Norfolk Street Richmond, Virginia 23230

May 2, 2014

ECS Project No. 03:11203

Page 2: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

May 2, 2014

Mr. Josh Bennett Moseley Architects, P.C. 3200 Norfolk Street Richmond, Virginia 23230

ECS Project No. 03:11203

Reference: Report of Subsurface Exploration and Geotechnical Engineering Analysis Hadensville Fire Station #6 Goochland County, Virginia Dear Mr. Bennett, ECS Mid-Atlantic, LLC (ECS) has completed this report of subsurface exploration and geotechnical engineering analysis for the referenced project. This study has been completed in general accordance with the agreement in AIA Document C401-2007 dated March 19, 2014. We have enjoyed being of service to Moseley Architects, P.C. during the design phase of this project. If you should have any questions regarding the information and recommendations contained in the accompanying report, or if we can be of further assistance, please do not hesitate to contact us. Respectfully, ECS Mid-Atlantic, LLC David J. Schlotterer, L.P.S.S., A.O.S.E. Robert C. Moss, III, P.E. Senior Project Manager Principal Engineer Copies: (1) client via email [ [email protected] ] (1) Tony Bell via email [ [email protected] ]

Page 3: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

REPORT

Subsurface Exploration and

Geotechnical Engineering Analysis

PROJECT

Hadensville Fire Station #6 Goochland County, Virginia

PREPARED FOR

Mr. Josh Bennett Moseley Architects, P.C.

3200 Norfolk Street Richmond, Virginia 23230

PREPARED BY

ECS Mid-Atlantic, LLC 2119-D North Hamilton Street

Richmond, Virginia 23230

ECS PROJECT NO.

03:11203

DATE

May 2, 2014

Page 4: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

SUBSURFACE EXPLORATION AND GEOTECHNICAL ENGINEERING ANALYSIS

HADENSVILLE FIRE STATION #6 GOOCHLAND COUNTY, VIRGINIA

ECS MID-ATLANTIC, LLC PROJECT NO. 03:11203

TABLE OF CONTENTS SECTION PAGE 1.0 PROJECT OVERVIEW 1

. 1.1 Project Characteristics 1 . 1.2 Purposes of Exploration 1 . 1.3 Scope of Work 1 2.0 EXPLORATION PROCEDURES 2

. 2.1 Subsurface Exploration Procedures 2 . 2.2 Laboratory Testing Program 3

3.0 EXPLORATION RESULTS 4

. 3.1 Site Characteristics 4 . 3.2 Subsurface Conditions 4 4.0 ANALYSIS AND RECOMMENDATIONS 6

. 4.1 Foundation Design 6 . 4.2 Floor Slab Design 7 . 4.3 Pavement Design 8 . 4.4 Drainage Design 10 . 4.5 BMP Design Considerations 10 . 4.6 Subgrade Preparation and Earthwork Operations 11 . 4.7 Construction Considerations 13 5.0 CLOSING 15 APPENDICES I. Boring Location Diagram II. Soil Test Boring Logs and Generalized Subsurface Soil Profile III. Unified Classification System and Reference Notes for Boring Logs IV. Summary of Laboratory Test Data

Page 5: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

1.0 PROJECT OVERVIEW 1.1 Project Characteristics The project will include the following aspects relative to our scope of services: A one-story metal building of steel frame construction with a brick veneer, containing

approximately 12,000 +/- square feet in plan area. Floors will be supported as a slab-on-grade. Maximum column and wall loads are expected to be 35 kips and 3 kips/ft, respectively.

The site will be accessed by entrances from Three Chopt Road and Old Fredericksburg

Road. Parking will be provided for approximately 48 automobiles. Parking areas and drive lanes are to be paved with asphalt, and the emergency vehicle entrance and aprons are to be paved with concrete.

It is anticipated that cuts and fills of less than about 5 feet will be required to establish

design building and pavement area subgrade elevations. Stormwater will be directed to a potential BMP basin located south the proposed building.

The bottom of the basin is expected to be 4 to 6 feet below existing site grades. It is desired to employ an infiltration component to the BMP basins, if feasible.

1.2 Purposes of Exploration The purpose of this exploration was to investigate the soil and groundwater conditions at the site to develop geotechnical engineering recommendations to guide design of building foundations and floor slabs, BMP facility, asphalt pavements and concrete pavements and to develop earthwork specifications for the project. We accomplished these purposes by drilling soil test borings and analyzing the soil samples to evaluate pertinent engineering properties. On this basis we developed our geotechnical engineering recommendations. No warranties are expressed or implied. No investigation for the presence of contamination of the soil, water, or air at or around this site was performed as part of this study. 1.3 Scope of Work To complete this investigation, ECS Mid-Atlantic, LLC (ECS) was provided with a drawing entitled Conceptual Site Plan – Option 2 prepared by Timmons Group and dated 01/23/14. The drawing indicates existing topography and proposed site layout. The conclusions and recommendations contained in this report are based upon a total of seven (7) soil test borings drilled to a depth of 15 feet below the existing grades, a site reconnaissance performed by ECS personnel, and laboratory test results of select boring samples. The soil test borings were located in the field by ECS personnel based on the site plan referenced above and using existing site features for reference.

Page 6: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 2

The recommendations contained herein were developed from the data obtained in the soil borings which indicate subsurface conditions at these specific locations at the time of the field exploration. Conditions may vary between borings. If during the course of construction variations appear evident, the Geotechnical Engineer should be informed so that the conditions can be addressed. Design recommendations were developed based on building and site design criteria considered typical for this type of facility. Should structural loading or other project characteristics differ from those discussed herein, this company should be informed such that a review of these conditions can be performed.

2.0 EXPLORATION PROCEDURES 2.1 Subsurface Exploration Procedures A total of seven (7) soil test borings were performed with an ATV-mounted auger drill rig that utilized continuous-flight, solid-stem augers to advance the boreholes. Drilling fluid was not used in this process. The soil test borings were extended to a depth of 15 feet below existing surface elevations. Drilling services were provided by EarthCon Site Services, Inc. of Lanexa, Virginia. The approximate boring locations are indicated on the Boring Location Diagram included as Appendix I. Representative soil samples were obtained by means of the split-barrel sampling procedure in accordance with ASTM Specification D-1586. In this procedure, a 2-inch O.D., split-barrel sampler is driven into the soil a distance of 24 inches by a 140-pound hammer falling 30 inches. The number of blows required to drive the sampler through a 12-inch interval is termed the Standard Penetration Test (SPT) N-value and is indicated for each sample on the boring logs. This value can be used as a qualitative indication of the in-place relative density of cohesionless soils and, in a less reliable way, the relative consistency of cohesive soils. These indications are qualitative, since many factors such as drill crews, drill rigs, drilling procedures, and hammer-rod-sampler assemblies can affect the Standard Penetration resistance value and skew a direct correlation between blow counts and strength and compressibility of soils. Samples were taken continuously to a 10-foot depth and at 5-foot intervals thereafter. Additionally, a bulk soil sample was obtained for testing in our laboratory. A field log of the soils encountered in the soil test borings was maintained by the drill crew. After recovery, each sample was removed from the sampler and visually classified. Representative portions of each sample were then sealed in glass jars and taken to our laboratory for further visual examination and laboratory testing.

Page 7: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 3

The group symbols for each soil type are indicated in parentheses following the soil descriptions on the soil test boring and hand auger boring logs included in Appendix II. A reference to the boring logs and a description of the Unified Soil Classification System (USCS) are included in Appendix III. The Geotechnical Engineer grouped the various soil types into the major zones noted on the boring logs. The stratification lines designating the interfaces between earth materials on the boring logs are approximate; in-situ, the transitions may be gradual. 2.2 Laboratory Testing Program An experienced Geotechnical Engineer visually classified each soil sample from the test borings on the basis of texture and plasticity in accordance with the USCS and ASTM D2488 (Description and Identification of Soils-Visual/Manual Procedures). Representative soil samples from the test borings were subjected to moisture content, gradation (percent passing #200 sieve), and Atterberg Limits tests to aid in ascertaining pertinent engineering properties. The bulk soil sample was subjected to classification, Standard Proctor, and CBR testing to aid in the pavement design recommendations. The results from this testing were used to substantiate the visual classifications and to evaluate the soil’s pertinent engineering characteristics. The results of all the laboratory testing are included in the summary sheet in Appendix IV of this report. The soil samples will be retained in our laboratory for a period of 60 days, after which they will be discarded unless other instructions are received as to their disposition.

Page 8: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 4

3.0 EXPLORATION RESULTS

3.1 Site Description The project site is an approximately 7.6 acre parcel located immediately north of the intersection of Three Chopt Road and Old Fredericksburg Road in Goochland County, Virginia. The site is open and covered with grassy vegetation. In general, the topography consists of a broad ridge that slopes downward to the south and south east. A drainway is located in the eastern third of the site and traverses through the western portion of the currently proposed building area. Based on available topographic information, site elevations range from about EL 409 feet in the north to EL 396 feet in the south east. 3.2 Subsurface Conditions

Soil Stratification The topsoil thicknesses generally ranged from 3 to 7 inches. Underlying the surface cover, the subsurface materials of the top 15 feet were generally arranged in a two layer configuration as follows: Stratum I (from 0 to 4 or 6 feet): Stratum I consisted of granite residuum soils comprised of orange and tan to brownish red and yellow, moist, Sandy Lean CLAY (CL), trace mica and Clayey, fine to coarse SAND (SC) trace mica. Standard Penetration Test results (N-values) for the cohesive clay soils ranged between 4 and 11 blows per foot (bpf) indicating of a soft to stiff consistency. N-values for the granular clayey sand soils ranged between 7 and 12 bpf indicating a loose to medium dense relative density. Stratum II (4 or 6 feet to boring termination of 15 feet): Soils below Stratum I were comprised of granite residuum consisting of variegated red, orange, tan, and brown, moist to very moist, Silty SAND (SM) and Sandy SILT (ML). N-values for the silt soils ranged between 10 and 16 bpf, indicating a stiff to very stiff consistency. N-values for the sand soils ranged between 8 to 19 bpf indicating a loose to medium dense relative density.

Groundwater Observations for groundwater were made during sampling and upon completion of the drilling operations at each boring location. In auger drilling operations, water is not introduced into the boreholes, and the groundwater position can often be determined by observing water flowing into or out of the boreholes. Furthermore, visual observation of the soil samples recovered during the auger drilling exploration can often be used in evaluating the groundwater conditions. The soil test borings completed in March 2013 encountered a static groundwater table at boring locations B-3, B-4, and BMP-1 at depths of about 10 feet, 10 feet, and 9 feet, respectively. This corresponds to elevations between EL. 387.5 and EL. 390. However, hand

Page 9: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 5

auger borings completed during the preliminary drainfield exploration in January 2014 encountered perched water as shallow as about 3.5 feet in the lower elevations of the site. Perched water originates as rain water which percolates downward to become trapped on less permeable soils below. It appears this perched water is seasonal and present during wetter times of the year. The level of the groundwater table will vary with seasonal and environmental conditions associated with temperature and precipitation.

Soil Survey A review of the USDA soil survey indicates the soils mapped for the site are of the Vance, Sedgefield, and Wedowee soil series. The Vance and Wedowee soils are described as being granite and gneiss residuum, and the Sedgefield soils are descrived as being mixed mafic residuum. The Wedowee soils are described as generally having a low shrink-swell (expansive) potential, and the Vance and Sedgefield soils are described as having a moderate shrink-swell potential. Based on our visual classification of the soils encountered and the laboratory test results, the natural soils on this site are considered to have a low expansive potential. The subsurface conditions are described in more detail on the soil test boring logs included in Appendix II.

Page 10: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 6

4.0 ANALYSIS AND RECOMMENDATIONS 4.1 Foundation Design 1. Allowable Bearing Capacity: Based on the results of our exploration, the proposed

building structure can be supported by shallow spread footings with a floor slab-on-grade. The footings should bear in stable, undisturbed natural soils, properly placed Engineered Fill, or Footing Excavation Backfill. Column (square) and continuous (strip) spread footings having a minimum width of 36 inches and 24 inches, respectively, can be dimensioned for a net allowable bearing capacity of 2,500 psf.

2. Footing Embedment: Footing bottoms should be located at a depth of at least 24 inches

below finished exterior grade for bearing capacity and shrink-swell considerations. 3. Footing Subgrade Preparation: Footing excavations should extend to a depth sufficient

to penetrate any excessively soft or unsuitable materials. Undocumented FILL was not encountered at the boring locations. However, soft clay soils were encountered in the lower elevations of the site and may be unstable where exposed in footing excavations. Where undercutting of footing excavations is required to remove unstable or unsuitable soils, the minimum footing embedment depth of 24 inches can be reestablished by backfilling the excavations with Footing Excavation Backfill (VDOT Size No. 57 Stone).

4. Net Allowable Bearing Capacity: The net allowable soil bearing capacity refers to that

pressure which may be transmitted to the foundation bearing soils in excess of the final minimum surrounding overburden pressure.

5. Bearing Capacity Verification: The bearing capacity at the final foundation elevation

should be verified in the field by an experienced Geotechnical Engineer or his representative prior to concrete placement or backfilling to assure that the in-situ bearing capacity at the bottom of each footing excavation is adequate for the design loads. Where undercutting to remove unstable or unsuitable soils from below the footings is required, grades can be restored by backfilling with Footing Excavation Backfill.

6. Foundation Settlement: Provided the foundation design and construction

recommendations discussed herein are employed, and considering the previously mentioned maximum foundation loading, the total localized settlement for the proposed structure is estimated to be about one (1.0) inch or less with differential settlements of approximately one-half this amount between columns or along equal lengths of wall.

Page 11: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 7

7. Seismic Site Class: The Seismic Site Class, based on International Building Code guidelines, was determined for the site. The Seismic Site Class for the proposed building was determined from soil boring N-Value data and our experience with the soils in the project and surrounding areas. Based on Section 1613 of the International Building Code and our local experience, the project site has a Seismic Site Class of D. It is possible that with additional site testing, the Seismic Site Class could be improved by one order. ECS can provide additional consultation in this regard upon request.

8. Soil Survey Review: A review of the Soil Survey indicates the soils in the site vicinity are

of the Wedowee, Vance, and Sedgefield soils series. These soils are described as having a low to moderate expansive (shrink-swell) potential. Based on this and laboratory testing, the natural Clay soils on this site are considered to have a low shrink-swell potential. Our foundation design recommendations consider this potential.

4.2 Floor Slab Design 1. Underslab Moisture Protection: The lightly loaded building floor slabs-on-grade should

be directly supported by a minimum of 4 inches of Porous Fill materials consisting of VDOT No. 57 Stone or approved alternate. For heavily loaded floor slabs supporting emergency vehicles, the Porous Fill should consist of Aggregate Base Material (VDOT Type I, Size 21A Stone) placed in a minimum 6-inch layer. This densely graded aggregate will help shed water and protect sensitive, underlying soils during wetter, winter weather. The Porous Fill layer will facilitate the fine grading of the subgrade, provide more uniform bearing conditions, and help minimize the rise of water to the bottom of the slab (capillary action). Before placement of floor slab concrete, a polyethylene vapor barrier should be placed on top of the granular material in heated areas to provide additional moisture protection.

2. Subgrade Modulus: Provided the placement of Engineered Fill and Porous Fill is per the

recommendations discussed herein, the floor slab may be designed assuming a Modulus of Subgrade Reaction, Ks, of 150 psi per inch.

3. Thickened Slab Sections: Thickened slab sections which support relatively lightly loaded

walls need not contain the underslab Porous Fill layer provided soils exposed in the excavation bottoms are suitable and stable. However, their excavations should be prepared and, if necessary, undercut to remove FILL or unsuitable or unstable soils in the same manner as the footings.

4. Slab Subgrade Preparation: For the design and construction of all slabs-on-grade for the

proposed building, all topsoil or other deleterious material, debris, and soft or unstable subgrade should be removed from within the 5-foot expanded building limits. The Geotechnical Engineer should be called on to observe exposed subgrades to assure that adequate subgrade preparation has been achieved. A proofroll using a drum roller or loaded dump truck should be performed in his presence at that time. Once subgrades have been approved, subgrades should be properly compacted and new Engineered Fill

Page 12: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 8

can be placed. Existing subgrades to a depth of at least 10 inches and all Engineered Fill should be moisture conditioned to within +/- 3 percentage points of optimum moisture content then be compacted to a dry density at least 95% of that soil’s Standard Proctor maximum dry density (ASTM D698).

4.3 Pavement Design 1. Design Pavement Subgrade Elevations: Soils which will be encountered within the top

several feet will consist primarily of Lean CLAY (CL). The Clay soils will be moisture sensitive and can degrade during wetter times of the year with or without disturbance due to earthwork activities.

2. Pavement Design Assumptions: Drive lanes and parking spaces are to be paved with

asphalt. The emergency vehicle entrance and aprons are to be paved with concrete. For planning purposes, we estimate that the drive lanes will support automobile traffic of 200 vehicles per day or less with daily emergency vehicle traffic. It is assumed that the emergency vehicles will ingress and egress from both the Old Fredericksburg Road and Three Chopt Road access points. Our design assumes that ditches and swales will be provided along the edges of all pavements to prevent water penetration of the subgrade soils.

3. Design CBR: Based on the results of the soil test borings and laboratory testing, it

appears that the soils that will be exposed as pavement subgrades will consist mainly of Sandy Lean CLAY (CL). Laboratory testing resulted in a CBR value of 3.9 percent. Due to potential variability in the subgrade soils a recommend maximum design CBR value of 2 be used in dimensioning the pavement section.

4. Minimum Pavement Sections: The following minimum pavement sections (or their

equivalent) are recommended. The Civil Engineer should review actual traffic patterns and loading to determine whether or not these sections are appropriate.

Entrance Drives and Circulation Lanes Subjected to Emergency Vehicles

(Heavy Duty)

Flexible Pavements

Asphalt Surface - 2.0 inches Asphalt Concrete Surface Mix. Asphalt Base - 3.0 inches Asphalt Concrete Base Mix. Aggregate Base - 8.0 inches untreated Aggregate, Type I, Size 21A. Subgrade - Stable and compacted to a dry density of at least 95% of the soil’s Standard Proctor maximum dry density (ASTM D698) to a depth of 10.0 inches below design subgrade elevation.

Page 13: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 9

Concrete Pavements

Concrete - 6.0 inches Portland Cement Concrete, min. f’c = 4,000 psi with reinforcement. Aggregate Base - 6.0 inches untreated Aggregate, Type I, Size 21A. Subgrade - Stable and compacted to a dry density of at least 95% of the soil’s Standard Proctor maximum dry density (ASTM D698)

to a depth of 10.0 inches below design subgrade elevation. Concrete pavements (slabs) over tanks that have penetrations for underground storage should be designed based on the slab’s structural requirements.

Parking Stalls and Automobile Drive Lanes(Light Duty)

Asphalt Surface - 1.5 inches Asphalt Concrete Surface Mix. Asphalt Base - 2.5 inches Asphalt Concrete Intermediate Mix. Aggregate Base - 6.0 inches untreated Aggregate, Type I, Size 21A. Subgrade - Stable and compacted to a dry density of at least 95% of the soil’s Standard Proctor maximum dry density (ASTM D698) to a depth of 10.0 inches below design subgrade elevation.

4. Dumpster Slab: It is noted that heavy dumpster trucks exert concentrated dynamic wheel loads on pavements during trash pick up. This type of loading typically results in rutting of the base and subgrade materials and ultimately pavement failures. Therefore, we recommend that the pavement in the dumpster pickup and loading area (if applicable), to include the position of the truck’s front wheels, consist of a minimum 6-inch-thick, Welded Wire Fabric or Fibermesh reinforced concrete slab supported on at least a 6-inch layer of Aggregate Base Material, VDOT Type I, Size 21A, compacted to a dry density at least 95% of the soil’s Standard Proctor maximum dry density (ASTM D698).

5. Pavement Subgrade Preparation: For the construction of new pavements, any soft,

unstable, and/or unsuitable materials should be removed from the pavement areas and pavement subgrades compacted to the requirements of this report. The stripped surface should be proofrolled and carefully observed at the time of construction in order to aid in identifying any localized soft or unsuitable materials. This material, where encountered, should be closely evaluated during construction and should be removed from below the pavement as required or considered necessary by the Geotechnical Engineer. For construction during wet seasonal conditions, undercutting of loose, wet materials from below design subgrade elevations should be anticipated.

6. Pavement Drainage: An important consideration with the design and construction of new

pavements is surface and subsurface drainage. Where standing water develops, either on the pavement surface or within the base course layer, softening of the subgrade and other problems related to the deterioration of the pavement can be expected. Furthermore, good

Page 14: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 10

drainage should minimize the possibility of the subgrade materials becoming saturated over a long period of time. Based upon the results of the soil test borings, the groundwater table should not affect the performance of pavements. However, surface runoff which seeps into base materials could create localized deterioration the soil's bearing capacity. Water that tends to collect within the base course layer may be minimized by installing weep holes in drainage structures and backfilling around these structures and storm sewer pipes with No. 57 Stone wrapped in filter fabric, construction of drainage swales and diversion ditches around the pavement perimeter, and proper backfilling and grading behind curbs to minimize water intrusion from behind the curbs.

4.4 Drainage Design 1. Retaining Walls: No retaining walls are anticipated for this project. 2. Underslab Drainage: No depressed or below grade slabs are anticipated for this project. 3. Building: Positive drainage away from the proposed building is an essential element in

minimizing the adverse effects that water might have on the foundation soil’s bearing quality. Positive drainage should be provided around the perimeter of the building structure to reduce moisture infiltration into the foundation and/or subgrade soils. All parking lots, sidewalks, paved, and landscaped areas should be sloped away from the proposed building. A fall of least six inches for the first ten feet outward from the building should be provided. Gutters should be employed on the building, and gutter effluent discharged away from the building or into the storm sewer system.

4. Pavements: Positive drainage should be provided around the perimeter of the pavement

structures to minimize moisture infiltration into the subgrade soils. Ditches or swales should be employed around the up-hill perimeter of all pavements to intercept runoff and direct it safely away from the pavement structures.

5. Drainfield Areas: It is particularly important that the site grading be designed to direct

surface runoff away from the proposed septic drain field to be located in the north west portion of the site.

4.5 BMP Design Considerations 1. General: It is anticipated that stormwater will be directed to a detention basin located

south of the proposed building location. The bottom of the basin is expected to be 4 to 6 feet below existing site grades. We understand it is desired to employ an infiltration component to the basins, if feasible.

2. Soil Conditions: Soil test boring BMP-1 was completed in the area of the potential pond.

Subsurface materials in this area consist of medium stiff to stiff, Sandy Clay to a depth of 6 feet with medium dense, Silty Sand and stiff, Sandy Silt below. These materials have a low to moderate permeability.

Page 15: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 11

3. Groundwater Table: The soil boring BMP-1 encountered groundwater at a depth of 9 feet

(Approximately EL. 388.5 feet) below the surface elevation. 4. Infiltration: It does not appear that an infiltration component for the stormwater

management facility at that location would be feasible on this site due to the subsurface Clays to 6 feet and relatively shallow ground water (9 feet) below the clay. Alternate filtration methods, such as bio-retention, can be considered for this project.

5. Embankments: Dam embankments incorporated into the stormwater pond design can

consist of homogeneous earth dams; that is, dams which do not contain a defined core and cover. Soils employed to construct the dam embankments should consist of Dam Embankment Fill classified as SC, CL, CH, or MH which contains at least 35% by dry weight Clay. Silts (ML) and relatively clean Sands (SM, SP, SW) are not suitable for this use and should be separated from borrow material intended for use in dam embankments. Dam embankment fill slopes and impoundment cut slopes should be graded to a maximum slope of 3H:1V.

6. Key or Cutoff Trench: For stormwater retention, the dam should be provided with a key

trench (cutoff) which is backfilled with Dam Embankment Fill. For design purposes, the key trench can be taken as 3 feet deep and 6 feet wide. The key trench should extend the full length of those portions of the dam embankment where existing grade in the center of the dam cross section is at or below the planned normal pool. A key is not considered necessary for stormwater detention.

4.6 Subgrade Preparation and Earthwork Operations 1. Stripping: The subgrade preparation should consist of removing all topsoil, rootmat, or

other deleterious material, debris, and soft or unstable subgrade from the 5-foot expanded building and 2-foot expanded pavement limits and from within 2 feet of the toe of embankment fill slopes. The topsoil thicknesses generally ranged from 3 to 7 inches. For planning purposes, a stripping depth of 6 to 8 inches can be anticipated. It will then be necessary to moisture condition and compact the subgrade. Deeper undercutting to remove soft, wet soils may be required, depending on seasonal conditions.

2. Subgrade Restoration: It should be contractually incumbent of the contractor to maintain

or restore subgrades during clearing, stripping, and construction phases. Subgrade soils disturbed by contractor operations shall be recompacted to the specifications of this report. Subgrade soils which are excessively wet and unstable but otherwise consist of suitable materials (inorganic soil) are not considered unsuitable by definition and shall be moisture conditioned and recompacted to the specifications of this report.

3. Subgrade Proofrolling Observation: After stripping and cutting to the desired grade, and

prior to footing construction or Engineered Fill placement, the cut surface should be observed by an experienced Geotechnical Engineer or his authorized representative.

Page 16: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 12

Proofrolling using a 5-ton drum roller or a loaded tandem axle dump truck having an axle weight of at least 10 tons should be performed at this time to aid in identifying localized soft or unsuitable material. Any soft or unsuitable materials encountered during this proofrolling should be moisture conditioned and compacted or removed and replaced with an approved Engineered Fill material compacted to the criteria outlined in the following paragraphs. Existing subgrades should be moisture conditioned and compacted at this time.

4. Subgrade and Engineered Fill Compaction: Existing subgrades as well as subsequent

layers of Engineered Fill should be properly compacted. Existing subgrades to a depth of at least 10 inches and all Engineered Fill shall be moisture conditioned to within +/- 3 percentage points of optimum moisture content then be compacted to a dry density at least 95% of that soil’s Standard Proctor maximum dry density (ASTM D698). Engineered Fill lifts should be a maximum of 8 inches in loose thickness. Field density testing of subgrades and each lift of fill should be performed at a rate of no less than one test per 2,500 square feet in the building area and no less than one test per 10,000 square feet in pavement areas, but not less than 2 tests per lift. Compaction of natural subgrade surfaces may be waived by the Geotechnical Engineer if they are observed to be stable during proofrolling inspection.

5. Utility Trench backfill Compaction: All utility trench backfill should be adequately

compacted. Utility trench backfill, as well as backfill above footings and below slabs, should be moisture conditioned to within +/- 3 percentage points of optimum moisture content then be compacted to a dry density at least 95% of that soil’s Standard Proctor maximum dry density (ASTM D698). Field density testing of trench backfill should be performed at a rate of no less than one test per 50 linear feet of trench, but not less than 1 test per lift. Based on the soil boring data, the groundwater table is not expected to impact utility or other excavations shallower than about 9 feet. However, perched water may be encountered above this depth depending upon weather and seasonal conditions at the time of construction.

6. Engineered Fill: The following Engineered/Structural Fill types are recommended for use

on this project:

On-Site Borrow Engineered Fill: Soil Material classified as CL, SM, SC, SP, or better which is free of organics and debris. Maximum aggregate size should be limited to 4 inches. Maximum Liquid Limit should be 45. Most inorganic soil materials excavated from the top several feet are expected to consist of Sandy Lean CLAY (CL) and Clayey SAND (SC) which would be suitable for reuse as Engineered Fill but may be difficult to work and compact during adverse seasonal conditions. However, all on-site soils should be evaluated through laboratory testing before use as Engineered Fill. Imported Engineered Fill: Soil Material classified as SM, SC, SP, or better containing less than 35% by weight Silt or Clay and free of organics and debris. Maximum aggregate size should be limited to 4 inches.

Page 17: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 13

Porous Fill: A minimum 4-inch layer of VDOT Size No. 57 Stone or approved alternate for lightly loaded slabs and a minimum 6-inch layer of Aggregate Base Material, VDOT Type I, Size 21A for heavily loaded slabs. Footing Excavation Backfill or backfill in areas which compaction equipment cannot access: VDOT Size No. 57 Stone. Aggregate Base: Aggregate Base Material, VDOT Type I, Size 21A.

All materials to be used for Engineered Fill should be analyzed and approved by the Geotechnical Engineer prior to their use on the site.

7. Sensitive Soils: Clay soils on this site, which would comprise most of the cut and fill materials available, are moisture sensitive. Compaction of these soils would be difficult during adverse seasonal conditions when moisture contents could not be readily manipulated. The contractor should consider this in time and work budgeting. Soil modification could be employed using hydrated lime (Quick Lime) or Calciment® to more readily reduce soil moisture contents.

8. Unit Rates: We recommend that favorable unit rates be established in the construction

contract for undercutting and backfilling and subgrade stabilization. Unit rates could be established as follows:

a. Undercut and backfill with Imported Engineered Fill, per cubic yard in place; b. Undercut and backfill with On-site Borrow Engineered Fill, per cubic yard in place; c. Undercut and backfill with Aggregate Base Material, per cubic yard in place; d. Undercut and backfill with No. 57 Stone (wet areas and below footings), per cubic yard

in place; e. Dispose of undercut material off site, per cubic yard in place, f. Place medium duty, woven and non-woven geotextile fabrics, per square yard. Suitable

non-woven fabric for use in stabilization and separation would include Mirafi 160N or equivalent. Suitable woven fabric for use in stabilization would include Mirafi 600X or equivalent.

g. Place medium duty geogrid, per square yard. Suitable geogrid for use in stabilization would include Tensar TX160 or equivalent.

The Geotechnical Engineer should be called on to recommend and/or approve material type and placement procedures where subgrade remediation is required.

4.7 Construction Considerations 1. Subgrade Protection: The contractor should exercise care during clearing and stripping

to reduce subgrade disturbance and deterioration. The use of tracked equipment should help reduce subgrade deterioration during wet conditions. As the shallower soils of this

Page 18: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 14

site are sensitive to deterioration in the presence of water, with or without exposure to construction traffic, exposure to the environment and construction activity may weaken the subgrade soils if stripped subgrade surfaces or excavations are exposed to inclement weather or trenches remain open for too long a time. Therefore, Engineered Fill, Aggregate Base Material, and foundation/slab concrete should be placed as early in the construction stages as practical to minimize subgrade exposure. With regard to foundation construction, concrete should be placed the same day that excavations are made (or as soon as practicable). If bearing soils are softened by surface water intrusion or exposure, the softened soils must be dried out and recompacted or removed and replaced as deemed necessary by the Geotechnical Engineer. We recommend the site be proofrolled immediately after clearing and stripping to document the condition of the subgrade. Repair of subgrade deterioration caused by contractor operations during clearing and stripping or which occurs thereafter should be the responsibility of the contractor.

2. Drainfield Area Protection: The contractor should exercise care during clearing and

stripping to avoid disturbance of the drainfield area. The drainfield area should be cordoned off to all earthwork equipment and construction activities prior to stripping of the site and through completion of the facility. Failure to protect the drainfield area could damage/compact the drainfield soils rendering them unusable, potentially causing delay in obtaining the certificate of occupancy and increasing cost.

3. Unsuitable Subgrades: While we expect existing subgrades which are left undisturbed to

be generally suitable for slab and pavement support, localized pockets of soft soils or of unsuitable materials may be present, particularly in low lying site areas. A contingency for undercut and replacement beyond the design subgrade cut elevations should be included in the construction contract.

4. Site Drainage: Positive site drainage should be maintained during earthwork operations

so as to help maintain the stability of the subgrade. Construction drainage should not be directed towards the drainfield area. The contractor should use additional depths of stone or employ geotextile fabrics or geogrids to protect subgrades from damage due to temporary construction traffic, if necessary.

5. Dewatering: Based on the soil boring data, the groundwater table is not expected to

impact excavations less than about 9 feet deep. However, seepage from shallow perched water could still occur, requiring sump pumping.

6. Borrow Sensitivity: Most soils available as borrow on this site will consist of Clays which

may be difficult to work and compact when not near their optimum moisture content. The contractor should be prepared to moisture condition on site soils prior to compaction.

Page 19: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 15

7. Engineered Fill Compaction: Engineered Fill materials should be placed, compacted, and tested in accordance with the recommendations contained in this report. We recommend that all cut and fill operations be observed full-time by a representative of the Geotechnical Engineer to determine if minimum earthwork and compaction requirements are being met.

8. Undercutting of Foundation Excavations: All footing excavations must extend to a

depth sufficient to penetrate any other unsuitable materials. Where undercutting of footing excavations is required, the minimum recommended footing embedment depth can be reestablished by backfilling the excavations with Footing Excavation Backfill (VDOT Size No. 57 Stone.

9. Foundation Subgrade Evaluation: All foundation excavations should be observed by a

representative of the Geotechnical Engineer to determine that existing the bearing soils have adequate bearing and are free of organic matter and deleterious material prior to the placement of reinforcing steel or backfilling with Footing Excavation Backfill.

10. Seasonal Considerations: It would be most desirable to perform initial earthwork

operations on this site during the drier seasonal conditions that typically extend from May through October. In this manner, material that otherwise will require removal or drying and reworking could possibly be left in place below new fills.

5.0 CLOSING If changes are made in the overall site layout or design or location of the structures or if the previously described project characteristics differ significantly from the actual design characteristics, the recommendations presented in this report must not be considered valid unless the actual project characteristics are reviewed by ECS and our recommendations are modified or verified in writing. We request the opportunity to review the foundation plan, grading plan, and applicable portions of the project specifications when the design is finalized. This review will allow us to ascertain whether these documents are consistent with the intent of our recommendations. This report has been prepared for the exclusive use of the client for specific application to the project described herein. Our conclusions and recommendations have been rendered in a manner consistent with the level and skill ordinarily exercised by members of the geotechnical engineering profession in the Commonwealth of Virginia. No warranty is expressed or implied. Our conclusions and recommendations are based on subsurface information obtained from the borings. This information does not necessarily reflect variations in the subsurface conditions which can occur between borings or in unexplored areas of the site due to geologic characteristics of the region or past land use. Should such variations become apparent during construction, it will be necessary to reevaluate our conclusions and recommendations based upon site observations and additional investigation, if required.

Page 20: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Hadensville Fire Station #6 Goochland County, Virginia ECS Project No. 03:11203 Page 16

Field observations, monitoring, and quality assurance testing during earthwork and foundation installation are an extension of and integral to the geotechnical design recommendation. We recommend that the owner retain these services and we be allowed to continue our involvement throughout these phases of construction. ECS is not responsible for the conclusions, opinions, or recommendations of others based on the data in this report.

Page 21: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX

I. Boring Location Diagram

II. Soil Test Boring Logs and Generalized Subsurface Soil Profile

III. Unified Soil Classification System and Reference Notes For Boring Logs IV. Summary of Laboratory Test Data

Page 22: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX I

BORING LOCATION DIAGRAM

Page 23: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX I BORING LOCATION DIAGRAM

Hadensville Fire Station #6 Goochland County, Virginia

ECS Project No. 03:11203

Approximate Soil Test Boring Location

B-1

B-2

B-4

B-3

P-2

BMP-1

P-1

Approximate Boundaries of Proposed Drainfields (Actual locations are subject to change)

Page 24: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX II

SOIL TEST BORING LOGS AND GENERALIZED SUBSURFACE SOIL PROFILES

Page 25: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

400

395

390

385

380

375

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

21

18

17

24

20

20

Topsoil Depth [4"]Granite residuum - Clayey fine to coarse SAND,mottled tan and orange, moist, loose to mediumdense (SC)

Granite residuum - Silty fine to coarse SAND,variegated orange, brown, pink, and tan, moist,loose (SM)Granite residuum - Sandy SILT, trace mica,variegated pink, white, brown, orange, and tanto variegated pink, white, and red, very moist,stiff (ML)

END OF BORING @ 15.00'

34545456566754674688

4565

9

9

12

10

14

11

CLIENT

Moseley Architects

JOB #

11203

BORING #

B-1

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL DRY WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT401 FT

1 OF 1

Page 26: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

395

390

385

380

375

370

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

20

19

19

18

19

20

Topsoil Depth [3"]Granite residuum - Sandy Lean CLAY, orangishtan, moist, medium stiff (CL)Clayey fine to coarse SAND, mottled reddishorange and tan, moist, loose (SC)

Granite residuum - Silty fine to coarse SAND,variegated white, orange, and tan, moist,medium dense (SM)

Granite residuum - Sandy SILT, trace mica,variegated orange, brown, and tan, very moist,stiff (ML)

END OF BORING @ 15.00'

33433444434556786676

6787

7

8

7

13

13

15

CLIENT

Moseley Architects

JOB #

11203

BORING #

B-2

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL DRY WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT397.5 FT

1 OF 1

Page 27: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

400

395

390

385

380

375

370

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

18

21

20

20

19

18

Topsoil Depth [4"]Granite residuum - Sandy Lean CLAY, tracemica, mottled orange and tan, moist, mediumstiff to stiff (CL)

Granite residuum - Sandy SILT, trace mica,variegated red, orange, and tan to variegatedred, orange, tan, and brown, moist to verymoist, stiff (ML)

Granite residuum - Silty fine to coarse SAND,trace mica, variegated orange, brown, tan, andwhite, very moist, medium dense (SM)

END OF BORING @ 15.00'

34344555556856786689

5688

7

10

11

13

14

14

CLIENT

Moseley Architects

JOB #

11203

BORING #

B-3

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL 10.0' WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT400 FT

1 OF 1

Page 28: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

395

390

385

380

375

370

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

18

21

18

21

21

21

Topsoil Depth [6"]Granite residuum - Sandy Lean CLAY, tan toorangish tan, very moist to moist, soft tomedium stiff (CL)

Granite residuum - Sandy SILT, variegated red,orange, brown, and tan, moist, stiff (ML)

Granite residuum - Silty fine to coarse SAND,trace mica, variegated tan, white, brown, andorange, very moist, medium dense (SM)

END OF BORING @ 15.00'

22222323344445454556

4566

4

5 321820.0

8

9

10

11

CLIENT

Moseley Architects

JOB #

11203

BORING #

B-4

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL 10.0' WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT397.5 FT

1 OF 1

Page 29: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

395

390

385

380

375

370

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

22

21

20

20

19

18

Topsoil Depth [7"]Granite residuum - Sandy Lean CLAY, brown,very moist to moist, medium stiff to stiff (CL)

Granite residuum - Silty fine SAND, trace mica,variegated red, orange, tan, and brown, moist,medium dense (SM)Granite residuum - Sandy SILT, trace mica,variegated red, orange, tan, and brown, moist,stiff (ML)

Granite residuum - Silty fine to coarse SAND,trace mica, variegated black, white, tan, andorange, very moist, medium dense (SM)

END OF BORING @ 15.00'

23333445454555665678

4566

620.4

8

931.5

11 32.9

13 31.7

11

CLIENT

Moseley Architects

JOB #

11203

BORING #

BMP-1

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL 9.0' WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT397.5 FT

1 OF 1

Page 30: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

405

400

395

390

385

380

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

21

23

19

18

18

21

Topsoil Depth [5"]Granite residuum - Lean CLAY, with fine sand,trace mica, brownish red with yellow mottles,moist, medium stiff to stiff (CL)

Granite residuum - Sandy Lean CLAY, tracemica, variegated yellowish red, tan, and yellow,moist, stiff (CL)Granite residuum - Silty fine SAND, trace mica,variegated, brownish red, pink, and tan, moist,loose (SM)

Granite residuum - Silty fine to medium SAND,trace mica, variegated white and tan, verymoist, medium dense (SM)

END OF BORING @ 15.00'

32333454456844454555

4565

5

9

11

8 26.0

10

11

CLIENT

Moseley Architects

JOB #

11203

BORING #

P-1

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL DRY WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT408 FT

1 OF 1

Page 31: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

0

5

10

15

20

25

30

400

395

390

385

380

375

S-1

S-2

S-3

S-4

S-5

S-6

SS

SS

SS

SS

SS

SS

24

24

24

24

24

24

20

19

20

18

19

19

Topsoil Depth [5"]Granite residuum - Sandy Lean CLAY, brown tomottled orange and brown, very moist to moist,medium stiff (CL)

Clayey fine to coarse SAND, mottled tan andorange, moist, loose (SC)

Granite residuum - Silty fine to coarse SAND,variegated yellow, orange, tan, and grayishbrown to variegated white, tan, and pink, moist,medium dense (SM)

END OF BORING @ 15.00'

232233444556791010891012

6788

5

7

10

19

19

15

CLIENT

Moseley Architects

JOB #

11203

BORING #

P-2

SHEET

PROJECT NAME

Hadensville Fire Station #6 Geotech

ARCHITECT-ENGINEER

SITE LOCATION

Goochland County, VirginiaNORTHING EASTING STATION

THE STRATIFICATION LINES REPRESENT THE APPROXIMATE BOUNDARY LINES BETWEEN SOIL TYPES. IN-SITU THE TRANSITION MAY BE GRADUAL.

WL DRY WS WD BORING STARTED 03/29/14

WL(BCR) WL(ACR) BORING COMPLETED 03/29/14 CAVE IN DEPTH None

WL RIG Badger Model 1 FOREMAN DRILLING METHOD SSA

DE

PT

H (

FT

)

SA

MP

LE

NO

.

SA

MP

LE

TY

PE

SA

MP

LE

DIS

T.

(IN

)

RE

CO

VE

RY

(IN

)

SURFACE ELEVATION

DESCRIPTION OF MATERIAL

WA

TE

R L

EV

ELS

EL

EV

AT

ION

(F

T)

BL

OW

S/6

"

10 20 30 40 50+

20% 40% 60% 80% 100%

1 2 3 4 5+

ENGLISH UNITS

BOTTOM OF CASING LOSS OF CIRCULATION

CALIBRATED PENETROMETER TONS/FT2

PLASTICLIMIT %

WATERCONTENT %

LIQUIDLIMIT %

ROCK QUALITY DESIGNATION & RECOVERY

RQD% REC.%

STANDARD PENETRATIONBLOWS/FT403 FT

1 OF 1

Page 32: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

B-1

9

9

12

10

14

11

END OF BORING@ 15'

SC

SM

ML

B-2

7

8

7

13

13

15

END OF BORING@ 15'

CL

SC

SM

ML

B-3

7

10

11

13

14

14

END OF BORING@ 15'

CL

ML

SM

B-4

4

5

8

9

10

11

END OF BORING@ 15'

CL

ML

SM

BMP-1

6

8

9

11

13

11

END OF BORING@ 15'

CL

P-1

5

9

11

8

10

11

END OF BORING@ 15'

CL

SM

P-2

5

7

10

19

19

15

END OF BORING@ 15'

CL

SC

SM

410

405

400

395

390

385

380

Elev

ation

in F

eet

410

405

400

395

390

385

380

Elevation in Feet

GENERALIZED SUBSURFACE SOIL PROFILENOTES:1 SEE INDIVIDUAL BORING LOG AND GEOTECHNICAL REPORT FOR ADDITIONAL INFORMATION.2 PENETRATION TEST RESISTANCE IN BLOWS PER FOOT (ASTM D1586).3 HORIZONTAL DISTANCES ARE NOT TO SCALE.

Hadensville Fire Station #6 GeotechMoseley Architects

Goochland County, VirginiaPROJECT NO.: 11203 DATE: 5/2/2014 VERTICAL SCALE: 1"=5'

Page 33: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX III

UNIFIED CLASSIFICATION SYSTEM AND REFERENCE NOTES FOR BORING LOGS

Page 34: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

UNIFIED SOIL CLASSIFICATION SYSTEM (ASTM D-2487)

Major Divisions Group

Symbols Typical Names Laboratory Classification Criteria

GW

Well-graded gravels, gravel-sand mixtures, little or no fines

Cu = D60/D10 greater than 4 Cc = (D30)2/(D10xD60) between 1 and 3

Cle

an g

rave

ls

(Litt

le o

r no

fines

)

GP

Poorly graded gravels, gravel-sand mixtures, little or no fines

Not meeting all gradation requirements for GW

d

GMa

u

Silty gravels, gravel-sand mixtures

Atterberg limits below “A” line or P.I. less than 4

Gra

vels

(M

ore

than

hal

f of c

oars

e fra

ctio

n is

la

rger

than

No.

4 s

ieve

siz

e)

Gra

vels

with

fine

s (A

ppre

ciab

le a

mou

nt o

f fin

es)

GC

Clayey gravels, gravel-sand-clay mixtures

Atterberg limits below “A” line or P.I. less than 7

Above “A” line with P.I. between 4 and 7 are borderline cases requiring use of dual symbols

SW

Well-graded sands, gravelly sands, little or no fines

Cu = D60/D10 greater than 6 Cc = (D30)2/(D10xD60) between 1 and 3

Cle

an s

ands

(L

ittle

or n

o fin

es)

SP

Poorly graded sands, gravelly sands, little or no fines

Not meeting all gradation requirements for SW

d

SMa

u

Silty sands, sand-silt mixtures Atterberg limits above “A” line or P.I. less than 4

Coa

rse-

grai

ned

soils

(M

ore

than

hal

f of m

ater

ial i

s la

rger

than

No.

200

Sie

ve s

ize)

San

ds

(Mor

e th

an h

alf o

f coa

rse

fract

ion

is

smal

ler t

han

No.

4 s

ieve

siz

e)

San

ds w

ith fi

nes

(App

reci

able

am

ount

o

fines

)

SC

Clayey sands, sand-clay mixtures

Det

erm

ine

perc

enta

ge o

f san

d an

d gr

avel

from

gra

in-s

ize

curv

e.

Dep

endi

ng o

n pe

rcen

tage

of f

ines

(fra

ctio

n sm

alle

r tha

n N

o. 2

00 s

ieve

siz

e), c

oars

e-gr

aine

d so

ils

are

clas

sifie

d as

follo

ws:

Le

ss th

an 5

per

cent

GW

, GP

, SW

, SP

M

ore

than

12

perc

ent

G

M, G

C, S

M, S

C

5 to

12

perc

ent

Bor

der 4

line

cas

es re

quiri

ng d

ual s

ymbo

ls b

Atterberg limits above “A” line with P.I. greater than 7

Limits plotting in CL-ML zone with P.I. between 4 and 7 are borderline cases requiring use of dual symbols

ML

Inorganic silts and very fine sands, rock flour, silty or clayey fine sands, or clayey silts with slight plasticity

CL

Inorganic clays of low to medium plasticity, gravelly clays, sandy clays

Silt

s an

d cl

ays

(Liq

uid

limit

less

than

50

)

OL Organic silts and organic silty clays of low plasticity

MH

Inorganic silts, micaceous or diatomaceous fine sandy or silty soils, elastic silts

CH

Inorganic clays of high plasticity, fat clays

Silt

s an

d cl

ays

(Liq

uid

limit

grea

ter t

han

50)

OH

Organic clays of medium to high plasticity, organic silts

Fine

-gra

ined

soi

ls

(Mor

e th

an h

alf m

ater

ial i

s sm

alle

r tha

n N

o. 2

00 S

ieve

)

Hig

hly

Org

anic

so

ils

Pt

Peat and other highly organic soils

Plasticity Chart

0

10

20

30

40

50

60

0 10 20 30 40 50 60 70 80 90 100

Liquid Limit

Plas

ticity

Inde

x

"A" line

CH

MH and OH

CL

ML and OLCL-ML

a Division of GM and SM groups into subdivisions of d and u are for roads and airfields only. Subdivision is based on Atterberg limits; suffix d used when L.L. is 28 or less and the P.I. is 6 or less; the suffix u used when L.L. is greater than 28. b Borderline classifications, used for soils possessing characteristics of two groups, are designated by combinations of group symbols. For example: GW-GC,well-graded gravel-sand mixture with clay binder. (From Winterkorn and Fang. 1975)

Page 35: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

REFERENCE NOTES FOR BORING LOGS I. Drilling Sampling Symbols:

SS Split Spoon Sampler ST Shelby Tube Sampler RC Rock Core, NX, BX, AX PM Pressuremeter DC Dutch Cone Penetrometer RD Rock Bit Drilling BS Bulk Sample of Cuttings PA Power Auger (no sample) HAS Hollow Stem Auger WS Wash sample

II. Correlation of Penetration Resistances to Soil Properties:

Standard Penetration (blows/ft) refers to the blows per foot of a 140 lb. hammer falling 30 inches on a 2-inch OD split-spoon sampler, as specified in ASTM D-1586. The blow count is commonly referred to as the N value.

A. Non-Cohesive Soils (Silt, Sand Gravel and Combinations)

Density Relative Properties Under 4 blows/ft Very Loose Adjective Form 12% to 49% 4 to 10 blows/ft Loose With 5% to 12%

11 to 30 blows/ft Medium Dense 31 to 50 blows/ft Dense Over 51 blows/ft Very Dense

Particle Size Identification

Boulders 8 inches or larger Cobbles 3 to 8 inches Gravel Coarse 1 to 3 inches Medium ½ to 1 inch Fine ¼ to ½ inch Sand Coarse 2.00 mm to ¼ inch (dia. of lead pencil) Medium 0.42 to 2.00 mm (dia. of broom straw) Fine 0.074 to 0.42 mm (dia. of human hair) Silt and Clay 0.0 to 0.074 mm (particles cannot be seen)

B. Cohesive Soils (Clay, Silt, and Combinations)

Blows/ft Consistency Unconfined

Comp. Strength Qp (tsf)

Degree of Plasticity

Plasticity Index

Under 2 Very Soft Under 0.25 None to slight 0 – 4 2 to 4 Soft 0.25-0.49 Slight 5 – 7 5 to 8 Medium Stiff 0.50-0.99 Medium 8 – 22

9 to 15 Stiff 1.00-1.99 High to Very High Over 22 16 to 30 Very Stiff 2.00-3.00 Over 30 Hard Over 4.00

III. Water Level Measurement Symbols:

WL Water Level BCR Before Casing Removal DCI Dry Cave-In WS While Sampling ACR After Casing Removal WCI Wet Cave-In WD While Drilling Est. Groundwater Level Est. Perched Water Level The water levels are those levels actually measured in the borehole at the times indicated by the symbol. The measurements are relatively reliable when augering, without adding fluids, in a granular soil. In clay and plastic silts, the accurate determination of water levels may require several days for the water level to stabilize. In such cases, additional methods of measurement are generally applied.

Page 36: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX IV

SUMMARY OF LABORATORY TEST DATA

Page 37: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

APPENDIX IV

SUMMARY OF LABORATORY TEST DATA

Hadensville Fire Station #6 Goochland County, Virginia

ECS Project No. 03:11203

SPT Boring samples

Boring No.

Sample No.

Sample Depth

(ft)

Natural Moisture Content

(%)

Silt and/or Clay Fraction

(%)

Atterberg Limits

(LL/PL/PI)

Unified Soil Classification

B-4 S-2 2-4 20.0 61 32/18/14 CL

BMP-1 S-1 0-2 20.4 55 - CL

BMP-1 S-3 4-6 31.5 61 - CL

BMP-1 S-4 6-8 32.9 45 - SM

BMP-1 S-5 8-10 31.7 51 - ML

P-1 S-4 6-8 26.0 37 - SM

Bulk CBR Sample

Sample Number

Sample Location

Sample Depth

(ft)

Natural Moisture Content

(%)

Silt and/or Clay Fraction

(%)

Atterberg Limits

(LL/PL/PI)

Unified Soil Classification

CBR-1 P-1 0-2 20.2 54 40/20/20 CL

Sample Number

Sample Depth

(ft)

Optimum Moisture

(%)

Max. Dry Density (pcf)*

Soaked CBR*

Swell (%)

CBR-1 0-2 16.0 111.3 3.9 0

*See enclosed Proctor and CBR plots.

Page 38: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

BEARING RATIO TEST REPORTASTM D 1883-07

Project No: 11203

Project: Hadensville Fire Station #6 Geotech

Source of Sample: CBR-1 P-1 Depth: 0.00-2.00

Sample Number: D4S-43

Date: 5/1/2014

Orange/Brown Sandy Lean Clay

Test Description/Remarks:

Sample was retested to confirm thehigher value at 0.2"

Figure

111.3 16.0 40 20CL

Material DescriptionUSCS

Max.Dens.(pcf)

OptimumMoisture

(%)LL PI

MoldedDensity

(pcf)Percent ofMax. Dens.

Moisture(%)

SoakedDensity

(pcf)Percent ofMax. Dens.

Moisture(%)

CBR (%)

0.10 in. 0.20 in.

LinearityCorrection

(in.)

Surcharge(lbs.)

Max.Swell(%)

1 111.2 99.9 16.1 111.2 99.9 16.9 3.2 3.9 0.000 10 0

2

3

Pene

trat

ion

Res

ista

nce

(psi

)

0

40

80

120

160

200

Penetration Depth (in.)0 0.1 0.2 0.3 0.4 0.5

Page 39: REPORT OF SUBSURFACE EXPLORATION AND … GEOTECHNICAL ENGINEERING ANALYSIS HADENSVILLE FIRE ... Report of Subsurface Exploration and Geotechnical Engineering ... boring and hand auger

Tested By: MD

COMPACTION TEST REPORTD

ry d

ensi

ty, p

cf

104

106

108

110

112

114

Water content, %

10.5 12 13.5 15 16.5 18 19.5

16.0%, 111.3 pcf

ZAV forSp.G. =2.65

Test specification: ASTM D 698-07 Method A Standard

0.00-2.00 CL 20.2 2.65 40 20 9.7 54

Orange/Brown Sandy Lean Clay

11203 Moseley Architects

4/24/14

Elev/ Classification Nat.Sp.G. LL PI

% > % <Depth USCS AASHTO Moist. #4 No.200

TEST RESULTS MATERIAL DESCRIPTION

Project No. Client: Remarks:Project:

Date:Source of Sample: CBR-1 P-1 Sample Number: D4S-43

Figure

Maximum dry density = 111.3 pcf

Optimum moisture = 16.0 %

Hadensville Fire Station #6 Geotech