geotechnical investigation proposed eastgate building 1 · 2018. 10. 18. · 22885 savi ranch...

104
GEOTECHNICAL INVESTIGATION PROPOSED EASTGATE BUILDING 1 SWC Perimeter Road and 99 th Street San Bernardino, California for Hillwood

Upload: others

Post on 27-Jan-2021

6 views

Category:

Documents


1 download

TRANSCRIPT

  • GEOTECHNICAL INVESTIGATIONPROPOSED EASTGATE BUILDING 1

    SWC Perimeter Road and 99th StreetSan Bernardino, California

    forHillwood

  • 22885 Savi Ranch Parkway Suite E Yorba Linda California 92887voice: (714) 685-1115 fax: (714) 685-1118 www.socalgeo.com

    May 23, 2018

    Hillwood901 Via Piemonte, Suite 175Ontario, California 91764

    Attention: Mr. Joshua CoxDevelopment Director

    Project No.: 18G144-1

    Subject: Geotechnical InvestigationProposed Eastgate Building 1SWC Perimeter Road and 99th StreetSan Bernardino, California

    Gentlemen:

    In accordance with your request, we have conducted a geotechnical investigation at the subjectsite. We are pleased to present this report summarizing the conclusions and recommendationsdeveloped from our investigation.

    We sincerely appreciate the opportunity to be of service on this project. We look forward toproviding additional consulting services during the course of the project. If we may be of furtherassistance in any manner, please contact our office.

    Respectfully Submitted,

    SOUTHERN CALIFORNIA GEOTECHNICAL, INC.

    Daniel W. Nielsen, RCE 77915Senior Engineer

    Robert G. Trazo, GE 2655Principal Engineer

    Distribution: (1) Addressee

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    TABLE OF CONTENTS

    1.0 EXECUTIVE SUMMARY 1

    2.0 SCOPE OF SERVICES 3

    3.0 SITE AND PROJECT DESCRIPTION 4

    3.1 Site Conditions 43.2 Proposed Development 4

    4.0 SUBSURFACE EXPLORATION 6

    4.1 Scope of Exploration/Sampling Methods 64.2 Geotechnical Conditions 6

    5.0 LABORATORY TESTING 8

    6.0 CONCLUSIONS AND RECOMMENDATIONS 10

    6.1 Seismic Design Considerations 106.2 Geotechnical Design Considerations 136.3 Site Grading Recommendations 146.4 Construction Considerations 186.5 Foundation Design and Construction 186.6 Floor Slab Design and Construction 206.7 Retaining Wall Design and Construction 216.8 Pavement Design Parameters 23

    7.0 GENERAL COMMENTS 26

    8.0 REFERENCES 27

    APPENDICES

    A Plate 1: Site Location MapPlate 2: Boring Location Plan

    B Boring LogsC Laboratory Test ResultsD Grading Guide SpecificationsE Seismic Design ParametersF Liquefaction Evaluation Spreadsheets

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 1

    1.0 EXECUTIVE SUMMARY

    Presented below is a brief summary of the conclusions and recommendations of this investigation.Since this summary is not all inclusive, it should be read in complete context with the entirereport.

    Geotechnical Design Considerations Mapping performed by the California Geological Survey indicates that the subject site is

    located within a liquefaction hazard zone. Our site-specific liquefaction evaluation indicates that the onsite soils are not subject to

    liquefaction during the design seismic event. No design considerations related to liquefactionare considered warranted for this project.

    Undocumented fill soils were encountered at some of the boring locations, extending fromeither the ground surface or from beneath the existing aggregate base materials to depths of1½ to 3± feet. Boring No. B-2 was drilled at the top of a soil stockpile and encountered fillsoils extending to a depth of about 10± feet, which is roughly equal to the height of thestockpile at that location.

    The fill soils and near surface alluvium possess variable strengths and densities. These soils,in their present condition, are not considered suitable for support of the foundation loads ofthe new structure. These soils are underlain by dense to very dense, high strength alluvialsoils.

    Remedial grading will be necessary to remove the undocumented fill soils in their entirety aswell as the upper portion of the near-surface native alluvium and replace these materials ascompacted structural fill.

    The on-site soils possess varying amounts of oversized materials, including cobbles. Becausegrading will require excavation into these materials, consideration should be given to usingselective grading techniques to remove the cobbles from these soils prior to reuse as fill.Recommendations regarding selective grading and handling of oversized materials areprovided in Section 6.3 and Appendix D of this report.

    Site Preparation Initial site preparation should include stripping of any surficial vegetation. At the time of

    subsurface exploration, localized areas of the site were sparsely vegetated with grass andweeds.

    Stockpiled demolition debris, including asphalt and concrete fragments should be disposed ofoff-site. Alternatively, concrete and asphalt debris may be pulverized to a maximum 2-inchparticle size, well mixed with the on-site soils, and incorporated into new structural fills or itmay be crushed and made into CMB.

    The soils within the proposed building area should be overexcavated to a depth of 3 feetbelow existing grade and to a depth of at least 3 feet below proposed building pad subgradeelevations. The depth of overexcavation should also be sufficient to remove any existing fillsoils. The proposed foundation influence zones should be overexcavated to a depth of at least2 feet below proposed foundation bearing grade.

    After overexcavation has been completed, the resulting subgrade soils should be evaluatedby the geotechnical engineer to identify any additional soils that should be overexcavated.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 2

    The resulting soils should be scarified and thoroughly flooded to achieve a moisture contentof 0 to 4 percent above optimum moisture, to a depth of at least 24 inches. Theoverexcavation subgrade soils should then be recompacted under the observation of thegeotechnical engineer. The previously excavated soils may then be replaced as structural fill,compacted to 90 percent of the ASTM D-1557 maximum dry density.

    The new pavement and flatwork subgrade soils are recommended to be scarified to a depthof 12± inches, thoroughly moisture conditioned and recompacted to at least 90 percent ofthe ASTM D-1557 maximum dry density.

    Foundation Design Recommendations Conventional shallow foundations, supported in newly placed compacted fill. 3,000 psf maximum allowable soil bearing pressure. Reinforcement consisting of at least two (2) No. 5 rebars (1 top and 1 bottom) in strip footings.

    Additional reinforcement may be necessary for structural considerations.

    Building Floor Slab Conventional Slab-on-Grade: minimum 6 inches thick. Modulus of Subgrade Reaction: k = 150 psi/in. Reinforcement is not necessary for geotechnical considerations. The actual thickness and reinforcement of the floor slab should be determined by the

    structural engineer.

    Pavements

    ASPHALT PAVEMENTS (R = 50)

    Materials

    Thickness (inches)

    Auto Parking andAuto Drive Lanes(TI = 4.0 to 5.0)

    Truck Traffic

    TI = 6.0 TI = 7.0 TI = 8.0 TI = 9.0

    Asphalt Concrete 3 3½ 4 5 5½

    Aggregate Base 3 4 5 5 7

    Compacted Subgrade 12 12 12 12 12

    PORTLAND CEMENT CONCRETE PAVEMENTS (R = 50)

    Materials

    Thickness (inches)

    Autos and LightTruck Traffic(TI = 6.0)

    Truck Traffic

    TI = 7.0 TI = 8.0 TI = 9.0

    PCC 5 6 7 8

    Compacted Subgrade(95% minimum compaction)

    12 12 12 12

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 3

    2.0 SCOPE OF SERVICES

    The scope of services performed for this project was in accordance with our Proposal No.18P130R2, dated April 19, 2018. The scope of services included a visual site reconnaissance,subsurface exploration, field and laboratory testing, and geotechnical engineering analysis toprovide criteria for preparing the design of the building foundations, building floor slab, andparking lot pavements along with site preparation recommendations and constructionconsiderations for the proposed development. Based on the location of the subject site, thisinvestigation also included a site-specific liquefaction evaluation. The evaluation of theenvironmental aspects of this site was beyond the scope of services for this geotechnicalinvestigation.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 4

    3.0 SITE AND PROJECT DESCRIPTION

    3.1 Site Conditions

    The site is located on the southwest corner of Perimeter Road and 99th Street in San Bernardino,California. The site is bounded to the north by Perimeter Road, to the west by a parking lot, tothe south by the San Bernardino International Airport, and to the east by 99th Street. The generallocation of the site is illustrated on the Site Location Map, enclosed as Plate 1 in Appendix A ofthis report.

    The subject site consists of multiple parcels which total 52± acres in size. The site is presentlyused as two (2) tractor trailer parking/storage lots which are separated by a chain-link fence. Theeastern ¾± of the site consists of a large tractor trailer parking/storage lot. A portion of a taxiwayfor the San Bernardino International Airport extends from the southwest into the southern portionof this section of the site. With the exception of two large stockpiles, the ground surface coverthroughout the eastern ¾± of the subject site generally consists of aggregate base or gravel.This surface base course appears to have been thinly covered with a bituminous surfacetreatment. A large soil stockpile, which covers an area of about 360,000± ft² is present in thesoutheast corner of the site. The height of this soil stockpile was visually estimated to be about5 to 10± feet in height. A second stockpile, consisting of soil, crushed concrete, and crushedasphalt covers an area of about 100,000 ft² along the southern property line in the central portionof the lot. This stockpile was visually estimated to be about 3 to 4± in height.

    The western ¼± of the site is also utilized as a tractor trailer parking/storage lot. One stockpilecontaining soil, crushed concrete and crushed asphalt coverers an area of about a 25,000 ft² inthe southeast corner of the western ¼± of the site. This stockpile was visually estimated to beabout 3 to 4± in height. Ground surface cover throughout the remainder of the site appears toconsist of gravel and/or aggregate base material.

    Detailed topographic information was not available at the time of this report. However, based onvisual observations, the site topography slopes to the east at an estimated gradient of less than1 percent. There was estimated to be 18 to 20± feet of elevation differential across the site withan additional 3 to 10± feet of elevation differential at stockpile locations.

    3.2 Proposed Development

    A conceptual site plan for the proposed development was provided to our office by the client.Based on this plan, the site will be developed with one (1) warehouse. The warehouse will belocated in the central area of the site and will be 1,038,000± ft² in size. We expect that thebuilding will be surrounded by asphaltic concrete pavements in automobile and/or trailer parkingand Portland cement concrete pavements in the proposed taxiway areas.

    Detailed structural information has not been provided. We assume that the new building will be

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 5

    a single-story structure, no more than 40± feet in height, of tilt-up concrete construction. Basedon the assumed construction, we expect that maximum column and wall loads will be on theorder of 100 kips and 3 to 6 kips per linear foot, respectively.

    No significant amounts of below grade construction, such as basements or crawl spaces, areexpected to be included in the proposed development. Based on the assumed topography, cutsand fills of up to 6 to 7± feet are expected to be necessary to achieve the proposed site grades.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 6

    4.0 SUBSURFACE EXPLORATION

    4.1 Scope of Exploration/Sampling Methods

    The subsurface exploration conducted for this project consisted of fifteen (15) borings, advancedto depths of 5 to 50± feet below currently existing site grades. Four (4) of the borings wereadvanced to a depth of 50± feet, as part of the liquefaction evaluation. All of the borings werelogged during drilling by a member of our staff.

    All borings were advanced with hollow-stem augers, by a conventional truck-mounted drilling rig.Representative bulk and relatively undisturbed soil samples were taken during drilling. Relativelyundisturbed samples were taken with a split barrel “California Sampler” containing a series of oneinch long, 2.416± inch diameter brass rings. This sampling method is described in ASTM TestMethod D-3550. Samples were also taken using a 1.4± inch inside diameter split spoon sampler,in general accordance with ASTM D-1586. Both of these samplers are driven into the ground withsuccessive blows of a 140-pound weight falling 30 inches. The blow counts obtained during drivingare recorded for further analysis. Bulk samples were collected in plastic bags to retain their originalmoisture content. The relatively undisturbed ring samples were placed in molded plastic sleevesthat were then sealed and transported to our laboratory.

    The approximate locations of the borings are indicated on the Boring Location Plan, included asPlate 2 in Appendix A of this report. The Boring Logs, which illustrate the conditions encounteredat the boring locations, as well as the results of some of the laboratory testing, are included inAppendix B.

    4.2 Geotechnical Conditions

    Artificial Fill

    All of the borings, except Boring Nos. B-2 and B-14, encountered aggregate base or gravel at theground surface. The aggregate base surface course possessed thicknesses ranging between of 1to 5± inches at the boring locations. The aggregate base course appeared to be covered with abituminous surface treatment at Boring Nos. B-1, B-3 thru B-7, B-13, and B-15. No surfacetreatment was observed at the other boring locations.

    Artificial fill soils were encountered at the ground surface at Boring No. B-2 and beneath theaggregate base course at Boring Nos. B-4, B-5, B-11, and B-12. The fill soils generally consist ofmedium dense to dense silty sands and well graded sands with varying gravel and cobble content.The fill soils extend to depths of 1½ to 3± feet below the existing site grades at these boringlocations, with the exception of Boring No. B-2 which was drilled through an existing soil stockpile.At Boring No. B-2, the artificial fill soils extend to a depth of about 10± feet, roughly equal to the

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 7

    height of the soil stockpile. The fill soils possess a disturbed appearance, resulting in theirclassification as artificial fill.

    Alluvium

    Native alluvial soils were encountered at the ground surface at Boring No. B-14 and beneath thefill or aggregate base at all of the other boring locations, extending to the maximum depthexplored of 50± feet. Within the upper 5 to 10± feet, the alluvium generally consists of mediumdense to very dense well-graded sands and sandy gravels with occasional silty sand layers. Thenear surface alluvium generally possesses significant cobble content. At greater depths, thealluvium generally consists of dense to very dense well-graded gravelly sands and sandy gravelswith occasional fine sand, silty sand, or sandy silt layers. With the exception of some fine sandand silty sand layers, these soils generally possess occasional to extensive cobble contentthroughout.

    Groundwater

    Free water was not encountered during the drilling of any of the borings. Based on the lack ofany water within the borings, and the moisture contents of the recovered soil samples, the staticgroundwater is considered to have existed at a depth in excess of 50± feet at the time of thesubsurface exploration.

    Research of historic high groundwater levels was performed as a part of the site-specificliquefaction evaluation. USGS Bulletin 1898 (Matti and Carson, 1991) indicates that the minimumhistoric depth to groundwater at the site is 40± feet.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 8

    5.0 LABORATORY TESTING

    The soil samples recovered from the subsurface exploration were returned to our laboratory forfurther testing to determine selected physical and engineering properties of the soils. The testsare briefly discussed below. It should be noted that the test results are specific to the actualsamples tested, and variations could be expected at other locations and depths.

    Classification

    All recovered soil samples were classified using the Unified Soil Classification System (USCS), inaccordance with ASTM D-2488. Field identifications were then supplemented with additional visualclassifications and/or by laboratory testing. The USCS classifications are shown on the BoringLogs and are periodically referenced throughout this report.

    Density and Moisture Content

    The dry density has been determined for selected relatively undisturbed ring samples. Thesedensities were determined in general accordance with the method presented in ASTM D-2937.The results are recorded as dry unit weight in pounds per cubic foot. The moisture contents aredetermined in accordance with ASTM D-2216, and are expressed as a percentage of the dryweight. These test results are presented on the Boring Logs.

    Consolidation

    Selected soil samples have been tested to determine their consolidation potential, in accordancewith ASTM D-2435. The testing apparatus is designed to accept either natural or remoldedsamples in a one-inch high ring, approximately 2.416 inches in diameter. Each sample is thenloaded incrementally in a geometric progression and the resulting deflection is recorded atselected time intervals. Porous stones are in contact with the top and bottom of the sample topermit the addition or release of pore water. The samples are typically inundated with water atan intermediate load to determine their potential for collapse or heave. The results of theconsolidation testing are plotted on Plates C-1 through C-8 in Appendix C of this report.

    Maximum Dry Density and Optimum Moisture Content

    Representative bulk samples have been tested for their maximum dry densities and optimummoisture contents. The results have been obtained using the Modified Proctor procedure, perASTM D-1557 and are presented on Plates C-9 and C-10 in Appendix C of this report. This test isgenerally used to compare the in-situ densities of undisturbed field samples, and for latercompaction testing. Additional testing of other soil types or soil mixes may be necessary at a laterdate.

    Soluble Sulfates

    Representative samples of the near-surface soils were submitted to a subcontracted analyticallaboratory for determination of soluble sulfate content. Soluble sulfates are naturally present in

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 9

    soils, and if the concentration is high enough, can result in degradation of concrete which comesinto contact with these soils. The results of the soluble sulfate testing are presented below, andare discussed further in a subsequent section of this report.

    Sample Identification Soluble Sulfates (%) Severity

    B-3 @ 0 to 5 feet 0.001 Not Applicable (S0)

    B-4 @ 0 to 5 feet 0.001 Not Applicable (S0)

    B-8 @ 0 to 5 feet 0.001 Not Applicable (S0)

    B-10 @ 0 to 5 feet 0.003 Not Applicable (S0)

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 10

    6.0 CONCLUSIONS AND RECOMMENDATIONS

    Based on the results of our review, field exploration, laboratory testing and geotechnical analysis,the proposed development is considered feasible from a geotechnical standpoint. Therecommendations contained in this report should be taken into the design, construction, andgrading considerations.

    The recommendations are contingent upon all grading and foundation construction activitiesbeing monitored by the geotechnical engineer of record. The recommendations are provided withthe assumption that an adequate program of client consultation, construction monitoring, andtesting will be performed during the final design and construction phases to verify compliancewith these recommendations. Maintaining Southern California Geotechnical, Inc., (SCG) as thegeotechnical consultant from the beginning to the end of the project will provide continuity ofservices. The geotechnical engineering firm providing testing and observation services shallassume the responsibility of Geotechnical Engineer of Record.

    The Grading Guide Specifications, included as Appendix D, should be considered part of thisreport, and should be incorporated into the project specifications. The contractor and/or ownerof the development should bring to the attention of the geotechnical engineer any conditions thatdiffer from those stated in this report, or which may be detrimental for the development.

    6.1 Seismic Design Considerations

    The subject site is located in an area which is subject to strong ground motions due toearthquakes. The performance of a site specific seismic hazards analysis was beyond the scopeof this investigation. However, numerous faults capable of producing significant ground motionsare located near the subject site. Due to economic considerations, it is not generally consideredreasonable to design a structure that is not susceptible to earthquake damage. Therefore,significant damage to structures may be unavoidable during large earthquakes. The proposedstructure should, however, be designed to resist structural collapse and thereby providereasonable protection from serious injury, catastrophic property damage and loss of life.

    Faulting and Seismicity

    Research of available maps indicates that the subject site is not located within an Alquist-PrioloEarthquake Fault Zone. Therefore, the possibility of significant fault rupture on the site isconsidered to be low.

    Seismic Design Parameters

    Based on the standards in place at the time of this report, we expect that the proposeddevelopment will be designed in accordance with the 2016 California Building Code (CBC). TheCBC provides procedures for earthquake resistant structural design that include considerationsfor on-site soil conditions, occupancy, and the configuration of the structure including the

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 11

    structural system and height. The seismic design parameters presented below are based on thesoil profile and the proximity of known faults with respect to the subject site.

    The 2016 CBC Seismic Design Parameters have been generated using U.S. Seismic Design Maps,a web-based software application developed by the United States Geological Survey. Thissoftware application, available at the USGS web site, calculates seismic design parameters inaccordance with the 2016 CBC, utilizing a database of deterministic site accelerations at 0.01degree intervals. The table below is a compilation of the data provided by the USGS application.A copy of the output generated from this program is included as Plate E-1 in Appendix E of thisreport. A copy of the Design Response Spectrum, as generated by the USGS application is alsoincluded in Appendix E. Based on this output, the following parameters may be utilized for thesubject site:

    2016 CBC SEISMIC DESIGN PARAMETERS

    Parameter Value

    Mapped Spectral Acceleration at 0.2 sec Period SS 1.959

    Mapped Spectral Acceleration at 1.0 sec Period S1 0.944

    Site Class --- D

    Site Modified Spectral Acceleration at 0.2 sec Period SMS 1.959

    Site Modified Spectral Acceleration at 1.0 sec Period SM1 1.416

    Design Spectral Acceleration at 0.2 sec Period SDS 1.306

    Design Spectral Acceleration at 1.0 sec Period SD1 0.944

    Ground Motion Parameters

    For the liquefaction evaluation, we utilized a site acceleration consistent with maximumconsidered earthquake ground motions, as required by the 2016 CBC. The peak groundacceleration (PGA) was determined in accordance with Section 11.8.3 of ASCE 7-10. Theparameter PGAM is the maximum considered earthquake geometric mean (MCEG) PGA, multipliedby the appropriate site coefficient from Table 11.8-1 of ASCE 7-10. The web-based softwareapplication U.S. Seismic Design Maps (described in the previous section) was used to determinePGAM, which is 0.773g. A portion of the program output is included as Plate E-2 of this report. Anassociated earthquake magnitude was obtained from the USGS Unified Hazard Tool, InteractiveDeaggregation application available on the USGS website. The deaggregated modal magnitude is8.1, based on the peak ground acceleration and soil classification D.

    Liquefaction

    The California Geological Survey (CGS) has not yet conducted detailed seismic hazards mappingin the area of the subject site. The general liquefaction susceptibility of the site was determinedby research of the San Bernardino County Official Land Use Plan, General Plan, Geologic HazardOverlay. Map FH30C for the San Bernardino South Quadrangle indicates that the overall site islocated within a zone of high liquefaction susceptibility. Therefore, the scope of this geotechnicalinvestigation was expanded to include a site-specific liquefaction evaluation.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 12

    Liquefaction is the loss of strength in generally cohesionless, saturated soils when the pore-waterpressure induced in the soil by a seismic event becomes equal to or exceeds the overburdenpressure. The primary factors which influence the potential for liquefaction include groundwatertable elevation, soil type and plasticity characteristics, relative density of the soil, initial confiningpressure, and intensity and duration of ground shaking. The depth within which the occurrenceof liquefaction may impact surface improvements is generally identified as the upper 50 feetbelow the existing ground surface. Liquefaction potential is greater in saturated, loose, poorlygraded fine sands with a mean (d50) grain size in the range of 0.075 to 0.2 mm (Seed and Idriss,1971). Non-sensitive clayey (cohesive) soils which possess a plasticity index of at least 18 (Brayand Sancio, 2006) are generally not considered to be susceptible to liquefaction, nor are thosesoils which are above the historic static groundwater table.

    The liquefaction analysis was conducted in accordance with the requirements of SpecialPublication 117A (CDMG, 2008), and currently accepted practice (SCEC, 1997). The liquefactionpotential of the subject site was evaluated using the empirical method developed by Boulangerand Idriss (Boulanger and Idriss, 2008). This method predicts the earthquake-induced liquefactionpotential of the site based on a given design earthquake magnitude and peak ground accelerationat the subject site. This procedure essentially compares the cyclic resistance ratio (CRR) [thecyclic stress ratio required to induce liquefaction for a cohesionless soil stratum at a given depth]with the earthquake-induced cyclic stress ratio (CSR) at that depth from a specified designearthquake (defined by a peak ground surface acceleration and an associated earthquakemoment magnitude). CRR is determined as a function of the corrected SPT N-value (N1)60-cs,adjusted for fines content. The factor of safety against liquefaction is defined as CRR/CSR. Basedon Special Publication 117A, a factor of safety of at least 1.3 is required in order to demonstratethat a given soil stratum is non-liquefiable. Additionally, in accordance with Special Publication117A, clayey soils which do not meet the criteria for liquefiable soils defined by Bray and Sancio(2006), loose soils with a plasticity index (PI) less than 12 and moisture content greater than85% of the liquid limit, are considered to be insusceptible to liquefaction. Non-sensitive soils witha PI greater than 18 are also considered non-liquefiable.

    The historic high groundwater depth was obtained from Liquefaction Susceptibility in the SanBernardino Valley and Vicinity, Southern California-A Regional Evaluation, USGS Bulletin 1898(Matti and Carson), which indicates a historic high groundwater depth at the subject site ofapproximately 40 feet. Therefore, the historic high groundwater table was considered to be 40feet for the liquefaction evaluation.

    The liquefaction analysis procedure is tabulated on the spreadsheet forms included in AppendixF of this report. As part of the liquefaction evaluation, Boring Nos. B-1, B-5, B-7, and B-11 wereextended to depths of 50± feet. The liquefaction analysis was performed using data obtainedfrom each of these boring locations. The liquefaction potential of the site was analyzed utilizinga PGAM of 0.773g for a magnitude 8.1 seismic event.

    If liquefiable soils are identified, the potential settlements that could occur as a result ofliquefaction are determined using the equation for volumetric strain due to post-cyclicreconsolidation (Yoshimine et. al, 2006). This procedure uses an empirical relationship betweenthe induced cyclic shear strain and the corrected N-value to determine the expected volumetric

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 13

    strain of saturated sands subjected to earthquake shaking. This analysis is also documented onthe spreadsheets included in Appendix F.

    Conclusions and Recommendations

    The liquefaction evaluation indicates that none of the soils below the historic high groundwatertable are subject to liquefaction during the design seismic event. Based on the results of thisanalysis, no design considerations related to liquefaction are considered warranted for thisproject.

    6.2 Geotechnical Design Considerations

    General

    Artificial fill soils were encountered at the ground surface or beneath the existing aggregate basematerial at Boring Nos. B-4, B-5, B-11 and B-12, extending to depths of 1½ to 3± feet belowexisting site grades. Additional fill soils were encountered at boring No. B-2, extending to a depthof 10± feet, but it should be noted that this boring was drilled at the top of a 10± foot high soilstockpile. No documentation regarding the placement or compaction of these fill soils is knownto exist. Based on their characteristics and the lack of documentation, the existing fill materialsare considered to represent undocumented fill. Native alluvium was encountered beneath the fill,at the ground surface, or beneath the aggregate base at all of the borings. The native alluviumin the upper 5 to 10± feet possesses variable strengths with relative densities ranging from looseto very dense. At greater depths, the alluvium possesses relatively high strengths, favorableconsolidation characteristics, dense to very dense relative densities. Based on these conditions,remedial grading is recommended within the proposed building area to remove the existingundocumented fill materials and a portion of the near-surface alluvium and replace thesematerials as compacted structural fill.

    Settlement

    The recommended remedial grading will remove the undocumented fill soils and a portion of thevariable density alluvium. These materials will be replaced as compacted structural fill. The nativesoils that will remain in place below the recommended depth of overexcavation generally possesshigh strengths and favorable consolidation characteristics and will not be subject to significantload increases from the foundations of the new structure. Provided that the recommendedremedial grading is completed, the post-construction static settlement of the proposed structureis expected to be within tolerable limits.

    Expansion

    The near-surface soils generally consist of silty sands, well-graded sands, and sandy gravels withvarying cobble content. Based on their composition, these soils have been visually classified asvery low to non-expansive. Therefore, no design considerations related to expansive soils areconsidered warranted for this site.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 14

    Soluble Sulfates

    The results of the soluble sulfate testing indicate that the selected samples of the on-site soils tocorrespond to Class S0 with respect to the American Concrete Institute (ACI) Publication 318-14Building Code Requirements for Structural Concrete and Commentary, Section 4.3. Therefore,specialized concrete mix designs are not considered to be necessary, with regard to sulfateprotection purposes. It is, however, recommended that additional soluble sulfate testing beconducted at the completion of rough grading to verify the soluble sulfate concentrations of thesoils which are present at pad grade within the building area.

    Shrinkage/Subsidence

    Removal and recompaction of the existing fill soils and near-surface alluvium is estimated to resultin an average shrinkage of 8 to 14 percent. The potential shrinkage estimate is based on drydensity testing performed on small-diameter samples taken at the boring locations. If a moreaccurate and precise shrinkage estimate is desired, SCG can perform a shrinkage study involvingseveral excavated test-pits where in-place densities are determined using in-situ testing methodsinstead of laboratory density testing on small-diameter samples. Please contact SCG for detailsand a cost estimate regarding a shrinkage study, if desired.

    Minor ground subsidence is expected to occur in the soils below the zone of removal, due tosettlement and machinery working. The subsidence is estimated to be 0.1 feet.

    These estimates are based on previous experience and the subsurface conditions encountered atthe boring locations. The actual amount of subsidence is expected to be variable and will bedependent on the type of machinery used, repetitions of use, and dynamic effects, all of whichare difficult to assess precisely.

    Grading and Foundation Plan Review

    Grading and foundation plans were not available at the time of this report. It is thereforerecommended that we be provided with copies of the preliminary grading and foundation plans,when they become available, for review with regard to the conclusions, recommendations, andassumptions contained within this report.

    6.3 Site Grading Recommendations

    The grading recommendations presented below are based on the subsurface conditionsencountered at the boring locations and our understanding of the proposed development. Werecommend that all grading activities be completed in accordance with the Grading GuideSpecifications included as Appendix D of this report, unless superseded by site-specificrecommendations presented below.

    Site Stripping and Demolition

    Initial site stripping should include removal of any surficial vegetation. This should include anyweeds and grasses. Based on the conditions at the time of subsurface exploration, stripping of

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 15

    sparse weed growth will be necessary in localized areas of the site. The actual extent of sitestripping should be determined in the field by the geotechnical engineer, based on the organiccontent and stability of the materials encountered.

    Stockpiled debris, including asphalt and concrete should be disposed of off-site. Alternatively,concrete and asphalt debris may be pulverized to a maximum 2-inch particle size, well mixed withthe on-site soils, and incorporated into new structural fills or it may be crushed and made intoCMB.

    Treatment of Existing Soils: Building Pad

    Remedial grading should be performed within the proposed building pad area in order to removethe existing undocumented fill soils and a portion of the near-surface alluvium. Based onconditions encountered at the boring locations, the existing soils within the proposed buildingarea are recommended to be overexcavated to a depth of at least 3 feet below existing gradeand to a depth of at least 3 feet below proposed building pad subgrade elevations, whichever isgreater. The depth of the overexcavation should also extend to a depth sufficient to remove allundocumented fill soils. Undocumented fill soils at the boring locations extend to depths of 1½to 3± feet, with the exception of Boring No. B-2 which was drilled at the top of a soil stockpile.The fill depth at Boring No. B-2 was roughly equal to the stockpile height of about 10± feet.Additional overexcavation should be performed within the influence zones of the new foundations,to provide for a new layer of compacted structural fill extending to a depth of at least 2 feet belowproposed bearing grades.

    The overexcavation area should extend at least 5 feet beyond the building and foundationperimeters, and to an extent equal to the depth of fill below the new foundations. If the proposedstructure incorporates any exterior columns (such as for a canopy or overhang) the area ofoverexcavation should also encompass these areas.

    Following completion of the overexcavation, the subgrade soils within the building area shouldbe evaluated by the geotechnical engineer to verify their suitability to serve as the structural fillsubgrade, as well as to support the foundation loads of the new structure. This evaluation shouldinclude proofrolling and probing to identify any soft, loose or otherwise unstable soils that mustbe removed. Some localized areas of deeper excavation may be required if additional fill materialsor loose, porous, or low density native soils are encountered at the base of the overexcavation.

    After a suitable overexcavation subgrade has been achieved, the exposed soils shouldbe scarified to a depth of at least 12 inches, and thoroughly flooded to raise themoisture content of the underlying soils to at least 0 to 4 percent above optimummoisture content, extending to a depth of at least 24 inches. The moisture conditioningof the overexcavation subgrade soils should be verified by the geotechnical engineer. Thesubgrade soils should then be recompacted to at least 90 percent of the ASTM D-1557 maximumdry density. The previously excavated soils may then be replaced as compacted structural fill.

    Treatment of Existing Soils: Retaining Walls and Site Walls

    The existing soils within the areas of proposed retaining and non-retaining site walls should beoverexcavated to a depth of at least 2 feet below foundation bearing grade and replaced as

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 16

    compacted structural fill as discussed above for the proposed building pad. Any undocumentedfill soils within any of these foundation areas should be removed in their entirety. Theoverexcavation areas should extend at least 3 feet beyond the foundation perimeters, and to anextent equal to the depth of fill below the new foundations. Please note that erection pads areconsidered to be part of the foundation system. These overexcavation recommendations apply toerection pads also. The overexcavation subgrade soils should be evaluated by the geotechnicalengineer prior to scarifying, moisture conditioning, and recompacting the upper 12 inches ofexposed subgrade soils, as discussed for the building areas. The previously excavated soils maythen be replaced as compacted structural fill.

    Treatment of Existing Soils: Parking and Drive Areas

    Based on economic considerations, overexcavation of the existing fill and variable strength alluvialsoils in the new parking areas is not considered warranted, with the exception of areas wherelower strength or unstable soils are identified by the geotechnical engineer during grading.

    Subgrade preparation in the new parking areas should initially consist of removal of any soilsdisturbed during stripping operations. The geotechnical engineer should then evaluate thesubgrade to identify any areas of additional unsuitable soils. The subgrade soils should then bescarified to a depth of 12 inches, moisture conditioned to 0 to 4 percent above optimum, andrecompacted to at least 90 percent of the ASTM D-1557 maximum dry density. Based on thepresence of undocumented fill soils and compressible/collapsible alluvial soils throughout the site,it is expected that some isolated areas of additional overexcavation may be required to removezones of lower strength, unsuitable soils.

    The grading recommendations presented above for the proposed parking and drive areas assumethat the owner and/or developer can tolerate minor amounts of settlement within the proposedparking areas. The grading recommendations presented above do not completely mitigate theextent of undocumented fill and variable strength alluvium in the parking areas. As such,settlement and associated pavement distress could occur. Typically, repair of such distressedareas involves significantly lower costs than completely mitigating these soils at the time ofconstruction. If the owner cannot tolerate the risk of such settlements, the parking and driveareas should be overexcavated to a depth of 2 feet below proposed pavement subgrade elevation,with the resulting soils replaced as compacted structural fill.

    Fill Placement

    Fill soils should be placed in thin (6± inches), near-horizontal lifts, moisture conditionedto 0 to 4 percent above the optimum moisture content, and compacted.

    On-site soils may be used for fill provided they are cleaned of any debris to the satisfactionof the geotechnical engineer.

    All grading and fill placement activities should be completed in accordance with therequirements of the 2016 CBC and the grading code of the City of San Bernardino.

    All fill soils should be compacted to at least 90 percent of the ASTM D-1557 maximum drydensity. Fill soils should be well mixed.

    Compaction tests should be performed periodically by the geotechnical engineer asrandom verification of compaction and moisture content. These tests are intended to aidthe contractor. Since the tests are taken at discrete locations and depths, they may not

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 17

    be indicative of the entire fill and therefore should not relieve the contractor of hisresponsibility to meet the job specifications.

    Selective Grading and Oversized Material Placement

    The native alluvial soils possess significant cobble content. It is expected that large scrapers(Caterpillar 657 or equivalent) will be adequate to move the cobble containing soils. Since theproposed grading will require excavation of cobble containing soils, it may be desirable toselectively grade the proposed building pad area. The presence of particles greater than 3 inchesin diameter within the upper 1 to 3 feet of the building pad subgrade will impact the utility andfoundation excavations. Depending on the depths of fills required within the proposed parkingareas, it may be feasible to sort the on-site soils, placing the materials greater than 3 inches indiameter within the lower depths of the fills, and limiting the upper 1 to 3 feet of soils to materialsless than 3 inches in size. Oversized materials could also be placed within the lower depths of therecommended overexcavations. In order to achieve this grading, it would likely be necessary touse rock buckets and/or rock sieves to separate the oversized materials from the remaining soil.Although such selective grading will facilitate further construction activities, it is not consideredmandatory and a suitable subgrade could be achieved without such extensive sorting. However,in any case, it is recommended that all materials greater than 6 inches in size be excluded fromthe upper 1 foot of the surface of any compacted fills.

    The placement of any oversized materials should be performed in accordance withthe Grading Guide Specifications included in Appendix D of this report. If disposal ofoversized materials is required, rock blankets or windrows should be used and such areas shouldbe observed during construction and placement by a representative of the geotechnical engineer.

    Imported Structural Fill

    All imported structural fill should consist of very low expansive (EI < 20), well graded soilspossessing at least 10 percent fines (that portion of the sample passing the No. 200 sieve).Additional specifications for structural fill are presented in the Grading Guide Specifications,included as Appendix D.

    Utility Trench Backfill

    In general, all utility trench backfill soils should be compacted to at least 90 percent of the ASTMD-1557 maximum dry density. As an alternative, a clean sand (minimum Sand Equivalent of 30)may be placed within trenches and compacted in place (jetting or flooding is not recommended).It is recommended that materials in excess of 3 inches in size not be used for utility trench backfill.Compacted trench backfill should conform to the requirements of the local grading code, andmore restrictive requirements may be indicated by City of San Bernardino. All utility trenchbackfills should be witnessed by the geotechnical engineer. The trench backfill soils should becompaction tested where possible; probed and visually evaluated elsewhere.

    Utility trenches which parallel a footing, and extending below a 1h:1v plane projected from theoutside edge of the footing should be backfilled with structural fill soils, compacted to at least 90percent of the ASTM D-1557 standard. Pea gravel backfill should not be used for these trenches.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 18

    6.4 Construction Considerations

    Excavation Considerations

    The near surface soils generally consist of silty sands, well-graded sands, and sandy gravels withvarying cobble content. These materials will likely be subject to caving within shallow excavations.Where caving occurs within shallow excavations, flattened excavation slopes may be sufficient toprovide excavation stability. On a preliminary basis, the inclination of temporary slopes should beno steeper than 2h:1v. Deeper excavations may require some form of external stabilization suchas shoring or bracing. Maintaining adequate moisture content within the near-surface soils willimprove excavation stability. All excavation activities on this site should be conducted inaccordance with Cal-OSHA regulations.

    Groundwater

    At the time of this report, the static groundwater table at this site is considered to exist at a depthgreater than 50 feet. Therefore, groundwater is not expected to impact the grading orfoundation construction activities.

    6.5 Foundation Design and Construction

    Based on the preceding grading recommendations, it is assumed that the new building pad willbe underlain by structural fill soils used to replace the existing undocumented fill and near-surfacealluvial soils. These new structural fill soils are expected to extend to depths of at least 3 feetbelow proposed foundation bearing grade, underlain by 1± foot of additional soil that has beendensified and moisture conditioned in place. Based on this subsurface profile, the proposedstructure may be supported on shallow foundations.

    Foundation Design Parameters

    New square and rectangular footings may be designed as follows:

    Maximum, net allowable soil bearing pressure: 3,000 lbs/ft2.

    Minimum wall/column footing width: 14 inches/24 inches.

    Minimum longitudinal steel reinforcement within strip footings: Six : Two (2) No. 5 rebars(1 top and 1 bottom).

    Minimum foundation embedment: 12 inches into suitable structural fill soils, and at least18 inches below adjacent exterior grade. Interior column footings may be placedimmediately beneath the floor slab.

    It is recommended that the perimeter building foundations be continuous across allexterior doorways. Any flatwork adjacent to the exterior doors should be doweled into theperimeter foundations in a manner determined by the structural engineer.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 19

    The allowable bearing pressures presented above may be increased by 1/3 when consideringshort duration wind or seismic loads. The minimum steel reinforcement recommended above isbased on geotechnical considerations; additional reinforcement may be necessary for structuralconsiderations. The actual design of the foundations should be determined by the structuralengineer.

    Foundation Construction

    The foundation subgrade soils should be evaluated at the time of overexcavation, as discussedin Section 6.3 of this report. It is further recommended that the foundation subgrade soils beevaluated by the geotechnical engineer immediately prior to steel or concrete placement. Soilssuitable for direct foundation support should consist of newly placed structural fill, compacted toat least 90 percent of the ASTM D-1557 maximum dry density. Any unsuitable materials shouldbe removed to a depth of suitable bearing compacted structural fill, with the resulting excavationsbackfilled with compacted fill soils. As an alternative, lean concrete slurry (500 to 1,500 psi) maybe used to backfill such isolated overexcavations.

    The foundation subgrade soils should also be properly moisture conditioned to 0 to 4 percentabove the Modified Proctor optimum, to a depth of at least 12 inches below bearing grade. Sinceit is typically not feasible to increase the moisture content of the floor slab andfoundation subgrade soils once rough grading has been completed, care should betaken to maintain the moisture content of the building pad subgrade soils throughoutthe construction process.

    Estimated Foundation Settlements

    Post-construction total and differential settlements of shallow foundations designed andconstructed in accordance with the previously presented recommendations are estimated to beless than 1.0 and 0.5 inches, respectively. Differential movements are expected to occur over a30-foot span, thereby resulting in an angular distortion of less than 0.002 inches per inch.

    Lateral Load Resistance

    Lateral load resistance will be developed by a combination of friction acting at the base offoundations and slabs and the passive earth pressure developed by footings below grade. Thefollowing friction and passive pressure may be used to resist lateral forces:

    Passive Earth Pressure: 300 lbs/ft3

    Friction Coefficient: 0.30

    These are allowable values, and include a factor of safety. When combining friction and passiveresistance, the passive pressure component should be reduced by one-third. These values assumethat footings will be poured directly against compacted structural fill. The maximum allowablepassive pressure is 3,000 lbs/ft2.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 20

    6.6 Floor Slab Design and Construction

    Subgrades which will support new floor slabs should be prepared in accordance with therecommendations contained in the Site Grading Recommendations section of this report.Based on the anticipated grading which will occur at this site, the floor of the new structure maybe constructed as a conventional slab-on-grade supported on newly placed structural fill,extending to a depth of at least 3 feet below proposed finished pad grade. Based on geotechnicalconsiderations, the floor slab may be designed as follows:

    Minimum slab thickness: 6 inches.

    Modulus of Subgrade Reaction: k = 150 psi/in.

    Minimum slab reinforcement: Reinforcement is not considered necessary from ageotechnical standpoint. The actual floor slab reinforcement should be determined by thestructural engineer, based on the imposed slab loading.

    Slab underlayment: If moisture sensitive floor coverings will be used then minimum slabunderlayment should consist of a moisture vapor barrier constructed below the entire areaof the proposed slab where such moisture sensitive floor coverings are expected to beused. The moisture vapor barrier should meet or exceed the Class A rating as defined byASTM E 1745-97 and have a permeance rating less than 0.01 perms as described in ASTME 96-95 and ASTM E 154-88. A polyolefin material such as Stego® Wrap Vapor Barrier orequivalent will meet these specifications. The moisture vapor barrier should be properlyconstructed in accordance with all applicable manufacturer specifications. Given that arock free subgrade is anticipated and that a capillary break is not required, sand belowthe barrier is not required. The need for sand and/or the amount of sand above themoisture vapor barrier should be specified by the structural engineer or concretecontractor. The selection of sand above the barrier is not a geotechnical engineering issueand hence outside our purview. Where moisture sensitive floor coverings are notanticipated, the vapor barrier may be eliminated.

    Moisture condition the floor slab subgrade soils to 0 to 4 percent above the ModifiedProctor optimum moisture content, to a depth of 12 inches. The moisture content of thefloor slab subgrade soils should be verified by the geotechnical engineer within 24 hoursprior to concrete placement.

    Proper concrete curing techniques should be utilized to reduce the potential for slabcurling or the formation of excessive shrinkage cracks.

    The actual design of the floor slab should be completed by the structural engineer to verifyadequate thickness and reinforcement. Additional rigidity may be necessary for structuralconsiderations.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 21

    6.7 Retaining Wall Design and Construction

    Although not indicated on the site plan, the proposed development may require some smallretaining walls (less than 3 to 5± feet in height) to facilitate the new site grades and the in dock-high areas of the building.

    Retaining Wall Design Parameters

    Based on the soil conditions encountered at the boring locations, the following parameters maybe used in the design of new retaining walls for this site. We have provided parameters assumingthe use of on-site soils for retaining wall backfill. The near surface soils generally consist of siltysands, well-graded sands, and sandy gravels. Based on their classifications, these materials areexpected to possess a friction angle of at least 32 degrees when compacted to 90 percent of theASTM-1557 maximum dry density.

    If desired, SCG could provide design parameters for an alternative select backfill material behindthe retaining walls. The use of select backfill material could result in lower lateral earth pressures.In order to use the design parameters for the imported select fill, this material must be placedwithin the entire active failure wedge. This wedge is defined as extending from the heel of theretaining wall upwards at an angle of approximately 60° from horizontal. If select backfill materialbehind the retaining wall is desired, SCG should be contacted for supplementaryrecommendations.

    RETAINING WALL DESIGN PARAMETERS

    Design Parameter

    Soil Type

    On-Site Sands andSilty Sands and Sandy

    Gravels

    Internal Friction Angle () 32

    Unit Weight 130 lbs/ft3

    Equivalent FluidPressure:

    40 lbs/ft3 37 lbs/ft3

    61 lbs/ft3 60 lbs/ft3

    61 lbs/ft3 55 lbs/ft3

    Regardless of the backfill type, the walls should be designed using a soil-footing coefficient offriction of 0.30 and an equivalent passive pressure of 300 lbs/ft3. The structural engineer shouldincorporate appropriate factors of safety in the design of the retaining walls.

    The active earth pressure may be used for the design of retaining walls that do not directlysupport structures or support soils that in turn support structures and which will be allowed todeflect. The at-rest earth pressure should be used for walls that will not be allowed to deflectsuch as those which will support foundation bearing soils, or which will support foundation loadsdirectly.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 22

    Where the soils on the toe side of the retaining wall are not covered by a "hard" surface such asa structure or pavement, the upper 1 foot of soil should be neglected when calculating passiveresistance due to the potential for the material to become disturbed or degraded during the lifeof the structure.

    Seismic Lateral Earth Pressures

    In accordance with the 2016 CBC, any retaining walls more than 6 feet in height must be designedfor seismic lateral earth pressures. If walls 6 feet or more are required for this site, thegeotechnical engineer should be contacted for supplementary seismic lateral earth pressurerecommendations.

    Retaining Wall Foundation Design

    The retaining wall foundations should be supported within newly placed compacted structural fill,extending to a depth of at least 3 feet below the proposed bearing grade. Foundations to supportnew retaining walls should be designed in accordance with the general Foundation DesignParameters presented in a previous section of this report.

    Backfill Material

    On-site soils may be used to backfill the retaining walls. However, all backfill material placedwithin 3 feet of the back wall face should have a particle size no greater than 3 inches.Some sorting and/or crushing operations may be required due to the presence of cobblesthroughout the site. The retaining wall backfill materials should be well graded.

    It is recommended that a properly installed prefabricated drainage composite such as theMiraDRAIN 6000XL (or approved equivalent), which is specifically designed for use behindretaining walls be used. If the drainage composite material is not covered by an impermeablesurface, such as a structure or pavement, a 12-inch thick layer of a low permeability soil shouldbe placed over the backfill to reduce surface water migration to the underlying soils. The drainagecomposite should be separated from the backfill soils by a suitable geotextile, approved by thegeotechnical engineer.

    All retaining wall backfill should be placed and compacted under engineering controlled conditionsin the necessary layer thicknesses to ensure an in-place density between 90 and 93 percent ofthe maximum dry density as determined by the Modified Proctor test (ASTM D1557). Care shouldbe taken to avoid over-compaction of the soils behind the retaining walls, and the use of heavycompaction equipment should be avoided.

    Subsurface Drainage

    As previously indicated, the retaining wall design parameters are based upon drained backfillconditions. Consequently, some form of permanent drainage system will be necessary inconjunction with the appropriate backfill material. Subsurface drainage may consist of either:

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 23

    A weep hole drainage system typically consisting of a series of 4-inch diameter holes inthe wall situated slightly above the ground surface elevation on the exposed side of thewall and at an approximate 8-foot on-center spacing. The weep holes should include aone cubic foot gravel pocket surrounded by a suitable geotextile at each weep holelocation.

    A 4-inch diameter perforated pipe surrounded by 2 cubic feet of gravel per linear foot ofdrain placed behind the wall, above the retaining wall footing. The gravel layer should bewrapped in a suitable geotextile fabric to reduce the potential for migration of fines. Thefooting drain should be extended to daylight or tied into a storm drainage system.

    6.8 Pavement Design Parameters

    Site preparation in the pavement area should be completed as previously recommended in theSite Grading Recommendations section of this report. The subsequent pavementrecommendations assume proper drainage and construction monitoring, and are based on eitherPCA or CALTRANS design parameters for a twenty (20) year design period. However, thesedesigns also assume a routine pavement maintenance program to obtain the anticipated 20-yearpavement service life.

    Pavement Subgrades

    It is anticipated that the new pavements will be primarily supported on a layer of compactedstructural fill, consisting of scarified, thoroughly moisture conditioned and recompacted existingsoils. The near-surface soils generally consist of silty sands, well-graded sands, and sandy gravels.These soils are considered to possess good to excellent pavement support characteristics withestimated R-values of 50 to 70. The subsequent pavement design is based upon an R-value of50. Any fill material imported to the site should have support characteristics equal to or greaterthan that of the on-site soils and be placed and compacted under engineering controlledconditions. It is recommended that R-value testing be performed after completion of roughgrading. Depending upon the results of the R-value testing, it may be feasible to use thinnerpavement sections in some areas of the site.

    Asphaltic Concrete

    Presented below are the recommended thicknesses for new flexible pavement structuresconsisting of asphaltic concrete over a granular base. The pavement designs are based on thetraffic indices (TI’s) indicated. The client and/or civil engineer should verify that these TI’s arerepresentative of the anticipated traffic volumes. If the client and/or civil engineer determine thatthe expected traffic volume will exceed the applicable traffic index, we should be contacted forsupplementary recommendations. The design traffic indices equate to the following approximatedaily traffic volumes over a 20 year design life, assuming six operational traffic days per week.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 24

    Traffic Index No. of Heavy Trucks per Day

    4.0 0

    5.0 1

    6.0 3

    7.0 11

    8.0 35

    9.0 93

    For the purpose of the traffic volumes indicated above, a truck is defined as a 5-axle tractor trailerunit with one 8-kip axle and two 32-kip tandem axles. All of the traffic indices allow for 1,000automobiles per day.

    ASPHALT PAVEMENTS (R = 50)

    Materials

    Thickness (inches)

    Auto Parking andAuto Drive Lanes(TI = 4.0 to 5.0)

    Truck Traffic

    TI = 6.0 TI = 7.0 TI = 8.0 TI = 9.0

    Asphalt Concrete 3 3½ 4 5 5½

    Aggregate Base 3 4 5 5 7

    Compacted Subgrade 12 12 12 12 12

    The aggregate base course should be compacted to at least 95 percent of the ASTM D-1557maximum dry density. The asphaltic concrete should be compacted to at least 95 percent of theMarshall maximum density, as determined by ASTM D-2726. The aggregate base course mayconsist of crushed aggregate base (CAB) or crushed miscellaneous base (CMB), which is arecycled gravel, asphalt and concrete material. The gradation, R-Value, Sand Equivalent, andPercentage Wear of the CAB or CMB should comply with appropriate specifications contained inthe current edition of the “Greenbook” Standard Specifications for Public Works Construction.

    Portland Cement Concrete

    The preparation of the subgrade soils within concrete pavement areas should be performed aspreviously described for proposed asphalt pavement areas. The minimum recommendedthicknesses for the Portland Cement Concrete pavement sections are as follows:

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 25

    PORTLAND CEMENT CONCRETE PAVEMENTS

    Materials

    Thickness (inches)

    Autos and LightTruck Traffic(TI = 6.0)

    Truck Traffic

    TI = 7.0 TI = 8.0 TI = 9.0

    PCC 5 6 7 8

    Compacted Subgrade(95% minimum compaction)

    12 12 12 12

    The concrete should have a 28-day compressive strength of at least 3,000 psi. Any reinforcementwithin the PCC pavements should be determined by the project structural engineer. The maximumjoint spacing within all of the PCC pavements is recommended to be equal to or less than 30times the pavement thickness.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 26

    7.0 GENERAL COMMENTS

    This report has been prepared as an instrument of service for use by the client, in order to aid inthe evaluation of this property and to assist the architects and engineers in the design andpreparation of the project plans and specifications. This report may be provided to thecontractor(s) and other design consultants to disclose information relative to the project.However, this report is not intended to be utilized as a specification in and of itself, withoutappropriate interpretation by the project architect, civil engineer, and/or structural engineer. Thereproduction and distribution of this report must be authorized by the client and SouthernCalifornia Geotechnical, Inc. Furthermore, any reliance on this report by an unauthorized thirdparty is at such party’s sole risk, and we accept no responsibility for damage or loss which mayoccur. The client(s)’ reliance upon this report is subject to the Engineering Services Agreement,incorporated into our proposal for this project.

    The analysis of this site was based on a subsurface profile interpolated from limited discrete soilsamples. While the materials encountered in the project area are considered to be representativeof the total area, some variations should be expected between boring locations and sampledepths. If the conditions encountered during construction vary significantly from those detailedherein, we should be contacted immediately to determine if the conditions alter therecommendations contained herein.

    This report has been based on assumed or provided characteristics of the proposed development.It is recommended that the owner, client, architect, structural engineer, and civil engineercarefully review these assumptions to ensure that they are consistent with the characteristics ofthe proposed development. If discrepancies exist, they should be brought to our attention toverify that they do not affect the conclusions and recommendations contained herein. We alsorecommend that the project plans and specifications be submitted to our office for review toverify that our recommendations have been correctly interpreted.

    The analysis, conclusions, and recommendations contained within this report have beenpromulgated in accordance with generally accepted professional geotechnical engineeringpractice. No other warranty is implied or expressed.

  • Proposed Eastgate Building 1 – San Bernardino, CAProject No. 18G144-1

    Page 27

    8.0 REFERENCES

    California Division of Mines and Geology (CDMG), “Guidelines for Evaluating and MitigatingSeismic Hazards in California,” State of California, Department of Conservation, Division of Minesand Geology, Special Publication 117A, 2008.

    Idriss, I. M. and Boulanger, R. W., “Soil Liquefaction During Earthquakes”, EarthquakeEngineering Research Institute, 2008.

    National Research Council (NRC), “Liquefaction of Soils During Earthquakes,” Committee onEarthquake Engineering, National Research Council, Washington D. C., Report No. CETS-EE-001,1985.

    Seed, H. B., and Idriss, I. M., “Simplified Procedure for Evaluating Soil Liquefaction Potential usingfield Performance Data,” Journal of the Soil Mechanics and Foundations Division, American Societyof Civil Engineers, September 1971, pp. 1249-1273.

    Sadigh, K., Chang, C. –Y., Egan, J. A., Makdisi. F., Youngs, R. R., “Attenuation Relationships forShallow Crustal Earthquakes Based on California Strong Motion Data”, Seismological ResearchLetters, Seismological Society of America, Volume 68, Number 1, January/ February 1997, pp.180-189.

    Southern California Earthquake Center (SCEC), University of Southern California, “RecommendedProcedures for Implementation of DMG Special Publication 117, Guidelines for Analyzing andMitigating Liquefaction in California,” Committee formed 1997.

    Tokimatsu K., and Seed, H. B., “Evaluation of Settlements in Sands Due to Earthquake Shaking,”Journal of the Geotechnical Engineering Division, American society of Civil Engineers, Volume113, No. 8, August 1987, pp. 861-878.

    Tokimatsu, K. and Yoshimi, Y., “Empirical Correlations of Soil Liquefaction Based on SPT N-valueand Fines Content,” Seismological Research Letters, Eastern Section Seismological Society OfAmerica, Volume 63, Number 1, p. 73.

    Youd, T. L. and Idriss, I. M. (Editors), “Proceedings of the NCEER Workshop on Evaluation ofLiquefaction Resistance of Soils,” Salt Lake City, UT, January 5-6 1996, NCEER Technical ReportNCEER-97-0022, Buffalo, NY.

  • S

    IT

    E

    EASTGATE BUILDING 1

    SCALE: 1" = 2400'

    DRAWN: JLH

    CHKD: RGT

    SCG PROJECT

    18G144-1

    PLATE 1

    SITE LOCATION MAP

    SAN BERNARDINO, CALIFORNIA

    SOURCE: THOMAS GUIDE, 2013

  • B-9

    B-12

    B-1

    B-2

    B-3

    B-8

    B-4

    B-5

    B-6

    B-11

    B-10

    B-7

    B-13

    B-14

    B-15

    3 R D S T R E E T

    P E R I M E T E R R O A D

    E X I S T I N G C A N A L

    1 0 0 T

    H

    S

    T

    R

    E

    E

    T

    N.A.P.

    N.A.P.

    SCALE: 1" = 240'

    DRAWN: PM

    CHKD: RGT

    PLATE 2

    SCG PROJECT

    18G144-1

    SAN BERNARDINO, CALIFORNIA

    EASTGATE BUILDING 1

    BORING LOCATION PLAN

    APPROXIMATE BORING LOCATION

    GEOTECHNICAL LEGEND

    NORTH

    So

    Ca

    lG

    eo

    NOTE: BASE SITE PLAN PREPARED BY HPA.

    AutoCAD SHX TextBUILDING AREA

    AutoCAD SHX Text1,038,000 S.F.

    AutoCAD SHX Text40' CLEAR

    AutoCAD SHX Text116 DOCK DOORS

    AutoCAD SHX TextGROUP IV TAXILANE

    AutoCAD SHX TextPOTENTIAL PLANE EXPANSION

    AutoCAD SHX Text115 DOCK DOORS

    AutoCAD SHX TextGROUP IV TAXILANE

    AutoCAD SHX TextOFFICE AREA

    AutoCAD SHX Text10,000 S.F.

    AutoCAD SHX TextOFFICE AREA

    AutoCAD SHX Text90,000 S.F.

  • BORING LOG LEGEND SAMPLE TYPE GRAPHICAL SYMBOL SAMPLE DESCRIPTION

    AUGER

    SAMPLE COLLECTED FROM AUGER CUTTINGS, NO FIELD MEASUREMENT OF SOIL STRENGTH. (DISTURBED)

    CORE ROCK CORE SAMPLE: TYPICALLY TAKEN WITH A

    DIAMOND-TIPPED CORE BARREL. TYPICALLY USED ONLY IN HIGHLY CONSOLIDATED BEDROCK.

    GRAB SOIL SAMPLE TAKEN WITH NO SPECIALIZED EQUIPMENT, SUCH AS FROM A STOCKPILE OR THE GROUND SURFACE. (DISTURBED)

    CS CALIFORNIA SAMPLER: 2-1/2 INCH I.D. SPLIT BARREL

    SAMPLER, LINED WITH 1-INCH HIGH BRASS RINGS. DRIVEN WITH SPT HAMMER. (RELATIVELY UNDISTURBED)

    NSR

    NO RECOVERY: THE SAMPLING ATTEMPT DID NOT RESULT IN RECOVERY OF ANY SIGNIFICANT SOIL OR ROCK MATERIAL.

    SPT STANDARD PENETRATION TEST: SAMPLER IS A 1.4 INCH INSIDE DIAMETER SPLIT BARREL, DRIVEN 18 INCHES WITH THE SPT HAMMER. (DISTURBED)

    SH SHELBY TUBE: TAKEN WITH A THIN WALL SAMPLE TUBE, PUSHED INTO THE SOIL AND THEN EXTRACTED. (UNDISTURBED)

    VANE VANE SHEAR TEST: SOIL STRENGTH OBTAINED USING

    A 4 BLADED SHEAR DEVICE. TYPICALLY USED IN SOFT CLAYS-NO SAMPLE RECOVERED.

    COLUMN DESCRIPTIONS DEPTH: Distance in feet below the ground surface.

    SAMPLE: Sample Type as depicted above.

    BLOW COUNT: Number of blows required to advance the sampler 12 inches using a 140 lb hammer with a 30-inch drop. 50/3” indicates penetration refusal (>50 blows) at 3 inches. WH indicates that the weight of the hammer was sufficient to push the sampler 6 inches or more.

    POCKET PEN.: Approximate shear strength of a cohesive soil sample as measured by pocket penetrometer.

    GRAPHIC LOG: Graphic Soil Symbol as depicted on the following page.

    DRY DENSITY: Dry density of an undisturbed or relatively undisturbed sample in lbs/ft3.

    MOISTURE CONTENT: Moisture content of a soil sample, expressed as a percentage of the dry weight.

    LIQUID LIMIT: The moisture content above which a soil behaves as a liquid.

    PLASTIC LIMIT: The moisture content above which a soil behaves as a plastic.

    PASSING #200 SIEVE: The percentage of the sample finer than the #200 standard sieve.

    UNCONFINED SHEAR: The shear strength of a cohesive soil sample, as measured in the unconfined state.

  • SM

    SP

    COARSEGRAINED

    SOILS

    SW

    TYPICALDESCRIPTIONS

    WELL-GRADED GRAVELS, GRAVEL -SAND MIXTURES, LITTLE OR NOFINES

    SILTY GRAVELS, GRAVEL - SAND -SILT MIXTURES

    LETTERGRAPH

    POORLY-GRADED GRAVELS,GRAVEL - SAND MIXTURES, LITTLEOR NO FINES

    GC

    GM

    GP

    GW

    POORLY-GRADED SANDS,GRAVELLY SAND, LITTLE OR NOFINES

    SILTSAND

    CLAYS

    MORE THAN 50%OF MATERIAL ISLARGER THANNO. 200 SIEVE

    SIZE

    MORE THAN 50%OF MATERIAL ISSMALLER THANNO. 200 SIEVE

    SIZE

    MORE THAN 50%OF COARSEFRACTION

    PASSING ON NO.4 SIEVE

    MORE THAN 50%OF COARSEFRACTION

    RETAINED ON NO.4 SIEVE CLAYEY GRAVELS, GRAVEL - SAND -

    CLAY MIXTURES

    FINEGRAINED

    SOILS

    SYMBOLSMAJOR DIVISIONS

    SOIL CLASSIFICATION CHART

    PT

    OH

    CH

    MH

    OL

    CL

    ML

    CLEAN SANDS

    SC

    SILTY SANDS, SAND - SILTMIXTURES

    CLAYEY SANDS, SAND - CLAYMIXTURES

    INORGANIC SILTS AND VERY FINESANDS, ROCK FLOUR, SILTY ORCLAYEY FINE SANDS OR CLAYEYSILTS WITH SLIGHT PLASTICITY

    INORGANIC CLAYS OF LOW TOMEDIUM PLASTICITY, GRAVELLYCLAYS, SANDY CLAYS, SILTY CLAYS,LEAN CLAYS

    ORGANIC SILTS AND ORGANICSILTY CLAYS OF LOW PLASTICITY

    INORGANIC SILTS, MICACEOUS ORDIATOMACEOUS FINE SAND ORSILTY SOILS

    INORGANIC CLAYS OF HIGHPLASTICITY

    ORGANIC CLAYS OF MEDIUM TOHIGH PLASTICITY, ORGANIC SILTS

    PEAT, HUMUS, SWAMP SOILS WITHHIGH ORGANIC CONTENTS

    SILTSAND

    CLAYS

    GRAVELS WITHFINES

    SANDAND

    SANDYSOILS (LITTLE OR NO FINES)

    SANDS WITHFINES

    LIQUID LIMITLESS THAN 50

    LIQUID LIMITGREATER THAN 50

    HIGHLY ORGANIC SOILS

    NOTE: DUAL SYMBOLS ARE USED TO INDICATE BORDERLINE SOIL CLASSIFICATIONS

    GRAVELAND

    GRAVELLYSOILS

    (APPRECIABLEAMOUNT OF FINES)

    (APPRECIABLEAMOUNT OF FINES)

    (LITTLE OR NO FINES)

    WELL-GRADED SANDS, GRAVELLYSANDS, LITTLE OR NO FINES

    CLEANGRAVELS

  • 15

    18

    39

    40

    65

    54

    55

    58

    49

    4± inches BaseALLUVIUM: Light Brown fine to medium Sand, trace coarseSand, little Silt, fine Gravel, medium dense-moist

    Light Gray Brown fine to coarse Sand, some fine to coarseGravel, Cobbles, medium dense to dense-damp

    Light Brown fine to coarse Sand, trace fine Gravel,dense-damp to moist

    Light Gray Brown fine to coarse Sand, some fine Gravel, someCobbles, very dense-damp

    Light Brown fine to coarse Sand, trace fine to coarse Gravel,very dense-damp

    Light Brown Gravelly fine to coarse Sand to fine to coarseSandy Gravel, some Cobbles, very dense-damp to moist

    Light Brown fine Sand, dense-moist

    12

    4

    3

    4

    3

    4

    5

    3

    6

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-1

    PLATE B-1a

    DRILLING DATE: 5/2/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    SURFACE ELEVATION: --- MSL

    WATER DEPTH: DryCAVE DEPTH: 42 feetREADING TAKEN: At Completion

    5

    10

    15

    20

    25

    30

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 50

    71

    71/11"

    Light Brown fine Sand, dense-moist

    Light Brown fine to medium Sand, trace coarse Sand, fineGravel, dense to very dense-damp

    Light Brown fine Sand, trace medium Sand, very dense-dampto moist

    Boring Terminated at 50'

    4

    7

    3

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-1

    PLATE B-1b

    DRILLING DATE: 5/2/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    (Continued)

    WATER DEPTH: DryCAVE DEPTH: 42 feetREADING TAKEN: At Completion

    40

    45

    50

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 33

    33

    22

    20

    8

    22

    28

    32

    36

    87/9"

    FILL: Light Brown fine to medium Sand, little Silt, somecoarse Sand, fine to coarse Gravel, medium dense todense-damp to moist

    ALLUVIUM: Light Gray Brown fine to coarse Sand, some fineGravel, loose to medium dense-damp

    Brown Silty fine Sand, medium dense-dry to damp

    Light Gray Brown Gravelly fine to coarse Sand to fine tocoarse Sandy Gravel, abundant Cobbles, very dense-damp

    Boring Terminated at 25' due to refusal

    99

    97

    96

    98

    106

    3

    3

    6

    6

    3

    2

    4

    5

    2

    3

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-2

    PLATE B-2

    DRILLING DATE: 5/2/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    SURFACE ELEVATION: --- MSL

    WATER DEPTH: DryCAVE DEPTH: 10 feetREADING TAKEN: At Completion

    5

    10

    15

    20

    25

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 30

    50

    23

    50/3"

    63

    1± inch BaseALLUVIUM: Light Gray Brown fine to coarse Sand, some fineGravel, some to abundant Cobbles, medium dense todense-dry to damp@ 0 to 2 feet, very moist

    @ 7 to 8 feet, very dense

    Boring Terminated at 10½'

    101

    120

    115

    123

    107

    17

    2

    2

    1

    3

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-3

    PLATE B-3

    DRILLING DATE: 5/4/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    SURFACE ELEVATION: --- MSL

    WATER DEPTH: DryCAVE DEPTH: 7 feetREADING TAKEN: At Completion

    5

    10

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 21

    16

    17

    53

    41

    44

    90/9"

    2± inches BaseFILL: Brown fine to medium Sand, trace coarse Sand, fineGravel, medium dense-damp to moistALLUVIUM: Light Brown fine to coarse Sand, some toabundant Cobbles, some fine to coarse Gravel, mediumdense to dense-dry to damp

    @ 13½ to 15 feet, very dense

    Light Brown fine to medium Sand, trace coarse Sand, little fineGravel, occasional Cobbles, very dense-damp

    Boring Terminated at 20'

    104

    101

    105

    119

    113

    5

    2

    2

    3

    3

    2

    3

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-4

    PLATE B-4

    DRILLING DATE: 5/4/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    SURFACE ELEVATION: --- MSL

    WATER DEPTH: DryCAVE DEPTH: 13 feetREADING TAKEN: At Completion

    5

    10

    15

    20

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 35

    32

    35

    68/8"

    61

    85/9"

    77/9"

    78/11"

    95/9"

    2± inches BaseFILL: Brown Silty fine Sand, trace medium Sand, fine Gravel,dense-damp

    ALLUVIUM: Light Brown fine to medium Sand, some fine tocoarse Gravel, Cobbles, dense-dry to damp

    Light Brown Gravelly fine to coarse Sand to fine to coarseSandy Gravel, abundant Cobbles, very dense-dry to damp

    No SampleRecovered

    7

    2

    2

    2

    3

    4

    2

    3

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-5

    PLATE B-5a

    DRILLING DATE: 5/3/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    SURFACE ELEVATION: --- MSL

    WATER DEPTH: DryCAVE DEPTH: 39 feetREADING TAKEN: At Completion

    5

    10

    15

    20

    25

    30

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 31

    50

    77

    Light Brown Gravelly fine to coarse Sand to fine to coarseSandy Gravel, abundant Cobbles, very dense-dry to damp

    Light Brown fine to medium Sand, trace coarse Sand, fineGravel, some Cobbles, dense-damp

    Light Brown Silty fine Sand, dense to very dense-very moist

    Light Brown fine to medium Sand, trace coarse Sand, verydense-dry to damp

    Boring Terminated at 50'

    4

    14

    2

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-5

    PLATE B-5b

    DRILLING DATE: 5/3/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    (Continued)

    WATER DEPTH: DryCAVE DEPTH: 39 feetREADING TAKEN: At Completion

    40

    45

    50

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 38

    31

    28

    50/4"

    43

    57

    61

    36

    57

    3± inches BaseALLUVIUM: Light Brown fine to coarse Sand, some fine tocoarse Gravel, medium dense-damp to moist

    Light Brown Gravelly fine to coarse Sand to fine to coarseSandy Gravel, some Cobbles, medium dense-dry to damp

    Brown fine to coarse Sand, some fine to coarse Gravel, littleSilt, medium dense-damp

    Light Brown Gravelly fine to coarse Sand, some Cobbles, verydense-dry to damp

    Light Brown fine to medium Sand, some coarse Sand, someCobbles, very dense-damp

    Brown fine Sand, trace Silt, dense-very moist

    Brown fine to coarse Sand, some fine to coarse Gravel, verydense-damp

    Boring Terminated at 30'

    No SampleRecovered

    110

    112

    106

    118

    6

    2

    2

    3

    2

    3

    17

    3

    JOB NO.: 18G144PROJECT: Eastgate Building 1LOCATION: San Bernardino, California

    BORING NO.B-6

    PLATE B-6

    DRILLING DATE: 5/4/18DRILLING METHOD: Hollow Stem AugerLOGGED BY: Jason Hiskey

    FIELD RESULTS LABORATORY RESULTS

    CO

    MM

    EN

    TS

    SURFACE ELEVATION: --- MSL

    WATER DEPTH: DryCAVE DEPTH: 11 feetREADING TAKEN: At Completion

    5

    10

    15

    20

    25

    30

    GR

    AP

    HIC

    LOG

    PA

    SS

    ING

    #200

    SIE

    VE

    (%)

    TEST BORING LOG

    DESCRIPTION

    PO

    CK

    ET

    PE

    N.

    (TS

    F)

    DR

    YD

    EN

    SIT

    Y(P

    CF)

    DE

    PTH

    (FE

    ET)

    MO

    ISTU

    RE

    CO

    NTE

    NT

    (%)

    LIQ

    UID

    LIM

    ITP

    LAS

    TIC

    LIM

    IT

    SA

    MP

    LE

    BLO

    WC

    OU

    NT

    OR

    GA

    NIC

    CO

    NTE

    NT

    (%)

    TBL

    18G

    144.

    GPJ

    SOC

    ALG

    EO.G

    DT

    5/23

    /18

  • 25

    24

    26

    37

    50/2"

    84/11"

    85/11"

    43

    48

    2± inches BaseALLUVIUM: Light Brown fine to coarse Sand, some Cobbles,some fine to coasre Gravel, medium dense-dry to damp

    @ 13½ to 27 fee