Wind Load Resistance of Composite Structural Insulated Panel

MISSING IMAGE

Material Information

Title:
Wind Load Resistance of Composite Structural Insulated Panel
Physical Description:
1 online resource (102 p.)
Language:
english
Creator:
Fernandez, George A
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.E.)
Degree Grantor:
University of Florida
Degree Disciplines:
Civil Engineering, Civil and Coastal Engineering
Committee Chair:
MASTERS,FORREST J
Committee Co-Chair:
PREVATT,DAVID
Committee Members:
GURLEY,KURTIS R

Subjects

Subjects / Keywords:
beam -- composite -- csip -- filled -- foam -- insulated -- panels -- sandwich -- sip -- structural
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre:
Civil Engineering thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The study evaluates the out-of-plane wind load resistance and in-plane shear resistance of a new thermoplastic CSIP developed by the University of Alabama at Birmingham, to address the growing demand for affordable, energy efficient building materials. Uniaxial bending response of these panels under realistic hurricane wind load conditions was evaluated. The panels were subjected to a step- loading pressure sequence and a dynamic wind pressure time history sequence. The dynamic wind pressure time history was derived from boundary layer wind tunnel model experiments. Similar failure mechanisms were observed for both load sequences. Damage was confined primarily to the timber connection plate, however de-bonding between the skin and foam was observed.The pressure step loading function ranged from 0.24-6.96 kPa and an average mid-span deflection of 9.00 cm over a 243.84 cm span was observed. The dynamic pressure loads ranged from 0.24-5.00 kPa with an average mid-span deflection of 7.00 cm over a 243.84 cm span. A load-deflection relationship was established and the use of a theoretical equation, developed by Allen(1969), was used to compare the test data. Investigative testing of in-plane shear resistance was conducted and compared with test conducted by Kermani et al (2006).
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by George A Fernandez.
Thesis:
Thesis (M.E.)--University of Florida, 2013.
Local:
Adviser: MASTERS,FORREST J.
Local:
Co-adviser: PREVATT,DAVID.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0046410:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 WIND LOAD RESISTANCE OF COMPOSITE STRUCTURAL INSULATED PANELS By GEORGE A. FERNANDEZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 George A. Fernandez

PAGE 3

3 To my mom, Claudia, my brother, Kevin and my grandparents, Coco and Papa Mundo

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor, Forrest J. Masters, Ph.D., P.E.; and my committee members, Kurtis R. Gurley, Ph.D. and David O. Prevatt, Ph.D., P.E. for t heir support and guidance throughout this project. I thank my colleagues Dany Romero, Abraham Alende, Carlos Lopez, Scott Bolton, Jimmy Jesteadt, Alex Esposito, Jason Smith, James Austin, Johann Weekes, Sylvia Laboy, Alon Krauthammer, Anthony Chanlee and Justin Henika for their support and assistance throughout this project. I would also like to acknowledge NSF CMMI for supporting this project and Dr. Ed Sutt from Simpson Strong Tie for their assistance.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 TABLE OF CONTENTS ................................ ................................ ................................ .. 5 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURE S ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 LIST OF NOMENCLATURE ................................ ................................ .......................... 13 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 Scope of Research ................................ ................................ ................................ 17 Organization of this Document ................................ ................................ ................ 17 2 BACKGROUND ................................ ................................ ................................ ...... 18 Traditional Light Wood Frame Construction ................................ ............................ 18 Applicable Standards ................................ ................................ ....................... 19 Prior Research ................................ ................................ ................................ 19 Sandwich Panels ................................ ................................ ................................ .... 20 Composite Beam Theory ................................ ................................ .................. 20 Structural Insulated Panels ................................ ................................ ............... 21 Composite structural Insulated Panels ................................ ............................. 21 3 EXPERIMENTAL DESIGN ................................ ................................ ..................... 24 4 DESIGN AND EVALUATIO N OF CSIP WALL CONNE CTION ............................... 26 5 EVALUATION OF CSIPS WALL ASSEMBLY SUBJEC TED TO WIND PRESSURE LOADING ................................ ................................ ........................... 32 Pressure step loading function ................................ ................................ ................ 33 Individual Test Results Static Step Loading 48.05 kg/m 3 (3 pcf) ...................... 33 Test #1 ................................ ................................ ................................ ....... 33 Test #2 ................................ ................................ ................................ ....... 34 Test #3 ................................ ................................ ................................ ....... 35 Test #4 ................................ ................................ ................................ ....... 36

PAGE 6

6 Individual Test Results Static Step Loading 16.02 kg/m3 (1 pcf) ...................... 37 Test #1 ................................ ................................ ................................ ....... 37 Time varying pressur e sequence ................................ ................................ ............ 38 Individual Test Results Dynamic Loading 48.05 kg/m 3 (3 pcf) .......................... 39 Test #1 ................................ ................................ ................................ ....... 39 Test #2 ................................ ................................ ................................ ....... 40 Test #3 ................................ ................................ ................................ ....... 41 Test #4 ................................ ................................ ................................ ....... 42 Test #5 ................................ ................................ ................................ ....... 43 Test #6 ................................ ................................ ................................ ....... 44 Individual Test Results Dynamic Loading 16.02 kg/m 3 (1 pcf) .......................... 44 Test #1 ................................ ................................ ................................ ....... 44 Test #2 ................................ ................................ ................................ ....... 45 Summary of Pressure Loading Results ................................ ................................ ... 46 Comparison of Capacity and Failure Modes ................................ ........................... 48 6 EVALUATION OF CSIP W ALL ASSEMBLY SUBJECT ED TO SHEAR LOADING ................................ ................................ ................................ ................ 85 Methodology ................................ ................................ ................................ ........... 85 Results ................................ ................................ ................................ .................... 85 7 CONCLUSIONS ................................ ................................ ................................ ..... 90 APPENDIX RESUL TS OF LATERAL LOAD T EST ................................ ................................ .......... 93 LIST OF REFERENCES ................................ ................................ ............................... 98 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 101

PAGE 7

7 LIST OF TABLES Table page 3 1 CSIP Panel Inventory ................................ ................................ ........................ 25 3 2 Testing Matrix ................................ ................................ ................................ .... 25 4 1 Fas tener summary ................................ ................................ ............................. 31 4 2 T test results for single fastener at two different temperatures .......................... 31 5 1 Static pressure step and equivalent mean wind speed ................................ ...... 83 5 2 Time varying basic wind speed and mean pressure ................................ .......... 83 5 3 Maximum pressure comparison table ................................ ................................ 83 5 4 Maximum deflection table ................................ ................................ .................. 84 6 1 Racking test results ................................ ................................ ........................... 89

PAGE 8

8 LIST OF FIGURES Figure page 2 1 SIP wall anatomy ................................ ................................ ................................ 23 2 2 CSIP cross sectional illustration ................................ ................................ ......... 23 4 1 Proposed construction assembly ................................ ................................ ........ 28 4 2 Lateral resistance testing. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ................................ ............... 28 4 3 Lateral capacities of nail fasteners ................................ ................................ ..... 29 4 4 Lateral capacities stratified by effective area ................................ ...................... 30 5 1 Two photographs of the pressure testing components A) HAPLA testing apparatus B) exterior view of testing chamber. Photo courtesy of George Fernandez and Forrest Masters, University of Florida. ................................ ....... 50 5 2 Strain gauge location ................................ ................................ .......................... 51 5 3 Reaction frame connection detail ................................ ................................ ....... 51 5 4 Pressure step loading function ................................ ................................ ........... 52 5 5 48.05 kg/m 3 (3 pcf) CSIP Test 1 pressure time history ................................ ....... 52 5 6 48.05 kg/m 3 (3 pcf) CSIP Test 1 de flection time history ................................ ..... 53 5 7 48.05 kg/m 3 (3 pcf) CSIP Test 1 de bonding and failure sketch ......................... 53 5 8 48.05 kg/m 3 (3 pcf) CSIP Test 1 failure. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 54 5 9 48.05 kg/m 3 (3 pcf) CSIP Test 1 summary ................................ ......................... 54 5 10 48.05 kg/m 3 (3 pcf) CSIP Test 2 pressure time history ................................ ....... 55 5 11 48.05 kg/m 3 (3 pcf) CSIP Test 2 deflection time history ................................ ..... 55 5 12 48.05 kg/m 3 (3 pcf) CSIP Test 2 de bonding and failure sketch ......................... 56 5 13 48.05 kg/m 3 (3 pcf) CSIP Test 2 summary ................................ ......................... 56 5 14 48.05 kg/m 3 (3 pcf) CSIP Test 3 pressure time history ................................ ....... 57

PAGE 9

9 5 15 48.05 kg/m 3 (3 pcf) CSIP Test 3 deflection time history ................................ ..... 57 5 16 48.05 kg/m 3 (3 pcf) CSIP Test 3 de bonding and failure sketch ......................... 58 5 17 48.05 kg/m 3 (3 pcf) CSIP Test 3 summary ................................ ......................... 58 5 18 48.05 kg/m 3 (3 pcf) CSIP Test 4 pressure time history ................................ ....... 59 5 19 48.05 kg/m 3 (3 pcf) CSIP Test 4 deflection time history ................................ ..... 59 5 20 48.05 kg/m 3 (3 pcf) CSIP Test 4 summary ................................ ......................... 60 5 21 16.02 kg/m 3 (1 pcf) CSIP Test 1 pressure time history ................................ ....... 60 5 22 16.02 kg/m 3 (1 pcf) CSIP Test 1 deflection time history ................................ ..... 61 5 23 16.02 kg/m 3 (1 pcf) CSIP Test 1 de bonding and failure sketch ......................... 61 5 24 16.02 kg/m 3 (1 pcf) CSIP Test 1 failure. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 62 5 25 16.02 kg/m 3 (1 pcf) CSIP Test 1 summary ................................ ......................... 62 5 26 Complete time varying pressure sequence ................................ ........................ 63 5 27 Sample of the time varying pressure sequence ................................ .................. 63 5 28 48.05 kg/m 3 (3 pcf) CSIP Test 1 time varying pressure time history ................... 64 5 29 48.05 kg/m 3 (3 pcf) CSIP Test 1 time varying pressure deflection time history .. 64 5 30 48.05 kg/m 3 (3 pcf) CSIP Test 1 time varying pressure deflection load plot ....... 65 5 31 48.05 kg/m 3 (3 pcf) CSIP Test 1 de bonding and failure sketch ......................... 65 5 32 48.05 kg/m 3 (3 pcf) CSIP Test 2 time varying pressure time history ................... 66 5 33 48.05 kg/m 3 (3 pcf) CSIP Test 2 time varying pressure deflection time history .. 66 5 34 48.05 kg/m3 (3 pcf) CSIP Test 2 time varying pressure d eflection load plot ...... 67 5 35 48.05 kg/m3 (3 pcf) CSIP Test 2 de bonding and failure sketch ........................ 67 5 36 48.05 kg/m 3 (3 pcf) CSIP Test 3 time varying pressure time history ................... 68 5 37 48.05 kg/m 3 (3 pcf) CSIP Test 3 time varying pressure deflection time history .. 68 5 38 48.05 kg/m3 (3 pcf) CSIP Test 3 time varying pressure deflection load plot ...... 69 5 39 48.05 kg/m3 (3 pcf) CSIP Test 3 de bonding and failure sketch ........................ 69

PAGE 10

10 5 40 48.05 kg/m 3 (3 pcf) CSIP Test 4 time varying pressure time history ................... 70 5 41 48.05 kg/m 3 (3 pcf) CSIP Test 4 time varying pressure deflection time history .. 70 5 42 48.05 kg/m3 (3 pcf) CSIP Test 4 time varying pressure deflection load plot ...... 71 5 43 48.05 kg/m3 (3 pcf) CSIP Test 4 de bonding and failure sketch ........................ 71 5 44 48.05 kg/m 3 (3 pcf) CSIP Test 5 time varying pressure time history ................... 72 5 45 48.05 kg/m 3 (3 pcf) CSIP Test 5 time varying pressure deflection time history .. 72 5 46 48.05 kg/m3 (3 pcf) CSIP Test 5 time varying pressure deflection load plot ...... 73 5 47 48.05 kg/m3 (3 pcf) CSIP Test 5 de bonding and failure sketch ........................ 73 5 48 48.05 kg/m 3 (3 pcf) CSIP Test 6 time varying pressure time history ................... 74 5 49 48.05 kg/m 3 (3 pcf) CSIP Test 6 time varying pressure deflection time history .. 74 5 50 48.05 kg/m3 (3 pcf) CSIP Test 6 time varying pressure deflection load plot ...... 75 5 51 48.05 kg/m3 (3 pcf) CSIP Test 6 de bonding and failure sketch ........................ 75 5 52 16.02 kg/m 3 (1 pcf) CSIP Test 1 time varying pressure time history ................... 76 5 53 16.02 kg/m 3 (1 pcf) CSIP Test 1 time varying pressure deflection time history .. 76 5 54 16.02 kg/m 3 (1 pcf) CSIP Test 1 time varying pressure deflection load plot ....... 77 5 55 16.02 kg/m 3 (1 pcf) CSIP Test 1 de bonding and failure sketch ......................... 77 5 56 16.02 kg/m 3 (1 pcf) CSIP Test 2 time varying pressure time history ................... 78 5 57 16.02 kg/m3 (1 pcf) CSIP Test 2 time varying pressure deflection time history .. 78 5 58 16.02 kg/m3 (1 pcf) CSIP Test 2 time varying pressure deflection load plot ...... 79 5 59 16.02 kg/m 3 (1 pcf) CSIP Test 2 de bonding and failure sketch ......................... 79 5 60 Deflection load results for the 16.02 kg/m 3 (1 pcf) panels ................................ .. 80 5 61 Deflection load results for the 48.05 kg/m 3 (3 pcf) panels ................................ .. 80 5 62 Observed damage Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ................................ ............... 81 5 63 Un filtered mid span deflection versus pressure ................................ ................. 81 5 64 Filtered mid span deflection versus pressure ................................ ..................... 82

PAGE 11

11 5 65 Time varying pressure vs. mid span strain ................................ ......................... 82 6 1 Diagram of racking test setup ................................ ................................ ............. 87 6 2 Racking test load deflection interaction ................................ .............................. 87 6 3 Racking test face sheet de bonding. Photo courtesy of George Fernandez a nd Scott Bolton, University of Florida. ................................ ............................... 88 6 4 Racking test face sheet wrinkling. Photo courtesy of George Fernandez and Scott Bolton, University of Florida. ................................ ................................ ...... 88 A 1 Lateral nail capacity failure results A S 8d. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 93 A 2 Lateral nail capacity failure results B RS 8d. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 94 A 3 Lateral nail capacity failure results C S 6d. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 95 A 4 Lateral nail capacity failure results SST SS 6d. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 96 A 5 Lateral nail capacity failure results SST RS 6d. Photo courtesy of George Fernandez, University of Florida. ................................ ................................ ........ 97

PAGE 12

12 LIST OF ABBREVIATIONS ASCE A MERICAN S OCIETY OF C IVIL E NGINEERS ASTM A MERICAN S OCIETY FOR T ESTING AND M ATERIALS CSIP C OMPOSITE S TRUCTURAL I NSULATED P ANELS FEMA F EDERAL E MERGENCY M ANAGEMENT A GENCY IBHS I NSTITUTE OF B USINESS & H OME S AFETY NOAA N ATIONAL O CEANIC AND A TMOSPHERIC A DMINISTRATION UF U NIVERSITY OF F LORIDA UAB U NIVERSITY OF A LABAMA AT B IRMINGHAM SIP S TRUCTURAL I NSULATED P ANELS

PAGE 13

13 LIST OF NOMENCLATURE E fz Face sheet modulus of elasticity E cz Foam core modulus of elasticity G Shear modulus B Width C Core thickness A Cross sectional area T Face thickness D Distance between center lines of opposite faces q Distributed load per unit length H Overall thickness Displacement at mid span L Span length D Flexural rigidity of beam Integral length scale Frequency U Mean velocity C p Pressure coefficient Density of air v V elocity (mph) q z V elocity pressure (psf) K z 0.85, V elocity pressure coefficient K zt 1.0, T opographic factor K d 0.85, W ind directionality factor

PAGE 14

14 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering WIND LOAD RESISTANCE OF COMPOSITE STRUCTURAL INSULATED PANELS By George A. Fernandez December 2013 Chair: Forrest Masters Major: Civil Engineering The study evaluates the out of plane wind load resistance and in plane shear resistance of a new thermoplastic CSIP developed by the University of Al abama at Birmingham to address the growing demand for affordable energy efficient building materials Uniaxial bending response of these panels under realistic hurricane wind load conditions was evaluated. The panels were subjected to a step loading pre ssure sequence and a dynamic wind pressure time history sequence. The dynamic wind pressure time history was derived from boundary layer wind tunnel model experiments. Similar failure mechanisms were observed for both load sequences. Damage was confined pr imarily to the timber connection plate however d e bonding between the skin and foam was observed. The pressure step loading function ranged from 0.24 6.96 kPa and an average mid span deflection of 9.00 cm over a 243.84 cm span was observed. The dynamic pr essure loads ranged from 0.24 5.00 kPa with an average mid span deflection of 7.00 cm over a 243.84 cm span. A load deflection relationship was established and the use of a theoretical E quation developed by Allen (1969) was used

PAGE 15

15 to compare the test data I nvestigative testing of in plane shear resistance was conducted and compared with test conducted by Kermani et al (2006).

PAGE 16

16 CHAPTER 1 INTRODUCTION Coastal communities have experienced an increase of 39% in shoreline county population from 1970 to 2010 (2012 NOAA). The increase in coastal population requires a higher demand on infrastructure (e.g. homes, apartments, high rises, roads and bridges). C oa stal communities contribute nearly $6.6 trillion dollars or half of the gross domestic product in the United States As engineers strive to provide improved method s and building materials for construction to mitigate damage s arising from hurricane episodes composite materials have received greater attention Composite material form parts of a pre engineered and modul ar building system originally developed for cladding (Davies 1993) and are becoming more widespread in civil, aerospace, and automotive appli cations (Schnabl 2007). There are many benefits for its current use in construction. This innovative resource is stronger than conventional wood fibers and highly resistant to mold, mildew, and decay. Composites provide the opportunity to remove organic ma terial from structural components of the home, (e.g loadbearing walls and floor systems) and make them less vulnerable to natural decay. Another benefit is its energy efficiency T he insulation is not interrupted by studs in the wall and closed cell polystyrene foam s have a high R value (Cathcart 1998) This study focuses on the out of plane bending behavior and in plane shear resistance of the CSIP described in Vaidya (2010) The results bridge the knowledge gap and augment prior w ork in the characterization of CSIP mechanical properties us to recreate such extreme events and quantify the behavior of such components.

PAGE 17

17 S cope of Research The research objectives of this study are as follow s: 1. Quantify the resistance for lateral nail l oads, and temperature effects on CSIP thermal composite face sheet s 2. Design connection det ails for wall to foundation and wall to roof connections 3. Evaluates the performance of the composite panels under uniaxial bending 4. Evaluate the performance of the compo site panels under shear in line with the wall Organization of this Document Chapter 2 provides the background to CSIP and current numerical methods for evaluation of mechanical performance. Chapter 3 describes the experimental methodology for structural testing and the significance of the data being collected. Chapter 4 discusses the design and results of CSIP wall connection detail. Chapter 5 focuses on the results of the quasi static p ressure loading and the time varying pressure loading. Chapter 6 addr esses the preliminary findings on shear wall behavior from racking test. Chapter 7 concludes with recommendations for future research.

PAGE 18

18 CHAPTER 2 BACKGROUND Composite structural insulated panels (CSIPs) are a derivative of sandwich panels, which were first introduced into the consumer market in the form of structural insulated panels (SIP) in 1951 (Cathcart 1998). SIPs gained popularity in the residential cons truction industry due to their competitive material costs energy efficiency and a decrease in the amount of time and labor required for construction Pre fabricated panels allowed for wall assemblies to be brought in assembled and offer time saving optio ns such as electrical pre wiring as well as rough opening for items such as glazing and exterior finishes Options for SIP skin materials include oriented strand board (OSB), plywood, structural insulated panel (CSIP) if the face sheet material is a thermoplastic, carbon fiber, or e glass. This chapter discusses light wood frame construction and sandwich panel construction. Sandwich panel construction will encompass both SIP pane ls and CSIP panels. Traditional Light Wood Frame Construction Light wood frame construction has seen little to no cha nge since the 1950s with the introduction of plywood sheathing for light frame construction Variation in lumber properties provided a dif ficult material to produce standard prediction methods (1983 Gromola) Light frame wood walls presented a unique cha lleng e to early researchers as the varying lumber properties and indefinite combination made it cost prohibitive to test every wall system. T he development of computer based analysis in the early 1970s made it possible to create analytical models that would allow for the prediction of infinite

PAGE 19

19 lumber combinations (1983 Gromola). Additional research demonstrated that a good strength prediction for light frame wood walls was a series of composite I joist. In his early research, Polensek (1984) found that wall studs have the greatest effect on overall wall capacities. Applicable Standards Current wood frame construction is governed by American Wo od Council (AWC) formerly ANSI / AF & PA. The standards designed focus on nail capacity as they have been identified to be the weak point for connection to adequately transfer loads from one member to another. Per AWC Special Design provisions for wind and seismic (SDPWS) wall capacity prior t o design factors for uniform load capacities (psf) for wall sheathing resisting out of plane load s are summarized in table 3.2.1. Sheathing capacity is prescribed according to s tud spacing, sheathing type and sheathi ng thickness. ASTM E 72 is used to conduct strength testing of panels for building construction. ASTM E 72 p section 12 describes the method for the transverse loading of a vertical wall with either a uniform pressur e in a chamber or, a uniform load with a bag or loading the wall specimen at its quarter points. The significance of this testing is to quantify the resistance of the wall subjected to out of plane loads. Section 14 of ASTM E 72 refers to racking test for shear walls. Prior Research The forestry and paper group has been conducti ng and publishing research on wood products as early as the 1970 s Several models presently exist to estimate the capacities of wood wall.

PAGE 20

20 Sandwich Panels Composite Beam Theory The first approaches to composite beam theory were developed in the early 20 th century (e.g, Timoshenko 1921; Newmark 1951). Most studies assume linear elastic behavior and perfect bonding between different layers to estimate mechanical behavior, however, these assumptions are simplification s of complex behavior that was difficult to quantify during this time period Recent research on composite sandwic h panels includes Baehre (1994), Roberts (2002), Girhammar (2008a, 2008b, 2009), Manalo (2010), Uddin (201 0), Fam (2010), Vaidya (2010) and Mohammed (2011). Perhaps the most accepted method to account for an imperfect bond can be found in Newm ark (1951); Girhammer and other researchers developed closed form solutions with the advancement of computer algor ithms and technology. (Girhammer et al., 1993, 2008, 2009). This analysis herein applies classical beam theory to benchmark results. Euler Bernoulli beam theory with a shear deformation t erm is shown to match the load deflection relationship within an acceptabl e tolerance (5%). The closed form solution developed by Allen (1969) E quation 2 1 and 2 2 appear to be good predictor s of load deflection relationship for uniaxial bending of CSIP panels. Ultimately E quation s 2 3 and 2 4 were used to estimate deflection. (2 1) (2 2) (2 3)

PAGE 21

21 (2 4) Structural Insulated Panels Since introduced in the 195 0s SIP research and development was primarily done by private companies and most of the advancement s were considered proprietary. Figure 2 1 shows the cross section of a ty pical SIP wall. SIP panels consist of a foam core sandwiched between two wood fiber sk ins. The wood fiber skins are primarily oriented strand board but are also made out of plywood, gypsum wallboard, or wafer board. These materials are susceptible to organ ic decay and can foster various types of mold if left unattended. Composite structural Insulated Panels The panels in this study ( Figure 2 2 ) consist of a low cost ther moplastics orthotropic glass/ propylene (glass PP) laminate exterior face sheet and expanded polystyrene foam (EPS) for the core material (Mousa 2010). With recent developments in cost effective textile including, hot melt impregnation, direct reinforced thermoplastics (DRIFT), vacuum thermoforming, and long fiber thermoplastics (LFTs), it is now possible to combine various high strain energy fibers with a wide varie ty of thermoplastic polymers to produce very low cost composite product, such as the aforementioned test subject. Advantages of these CSIP panels include: 1. Faster construction: CSIPS can be constructed in one third of the time of conventional light wood fra me construction. 2. Environmental consciousness: The use of this product causes waste reduction and allows the recycling of any remaining material. 3. Design flexibility: These panels can be used for the construction of a simple storage building or even a large office building.

PAGE 22

22 4. High thermal performance: It maintains a high R value ~3.85 5.50/in of thickness 5. Corrosion resistance: E glass and EPS foam exhibit resistance to mold and corrosion. CSIP construction can have a potential impact beyond the coastal United States Its low manufacturing cost and the ease of shipping worldwide make it an excellent candidate to address some of the needs of many underdeveloped coastal communities. Neither c ost effective nor fortified construction form part of the methodology fo r build ing construction for many underdeveloped countries (Prevatt et al., 1994). As a result modular construction can have a positive impact by providing an alternative method of construction and aid in mitigation of home loss in coastal communities arou nd the world. Limitations do exist for this type of composite assembly which were highlighted in the research conducted by Vaidya et al. (2010) He found that delamination occurred between the foam and the exte rior skin during axial loading of the wall assembly. Similar results were observed in this study that applied uniaxial bending. Delamination and its contribution to the findings of this research is further delineated in chapter 5 however it is important to note that delamination may have contributed to the nonlinear strain behavior observed in the compression face sheet Part of the research initiative was to quantify the performance characteristics of such panels and c ompare test result s to analytical results. A classical beam theory will be applied to the analysis of the test results. This research applies closed form composite beam theory approach

PAGE 23

23 Figure 2 1 SIP wall anatomy Figure 2 2 CSIP cross sectional illustration

PAGE 24

24 CHAPTER 3 EXPERIMENTAL DESIGN The principal objective of this study was to develop a testing framework to evaluate CSIP wall assemblies subjected to uniaxial bending caused by pressure loading and in plane shear transferred from the diaphragm to the foundation through the shear wall Table 3 1 provides a summary of CSIP panel inventory provided by UAB. Table 3 2 identifies the testing matrix of the panels that will be subjected to uniaxial bending caused by pressure loading and racking The study was conducted in three phases. Phase one studied the interaction and failure modes of five different nail type fasteners (see Table 4 1 ), subjected to single shear to design the optimal connection. Phase two subjected the completed wall assembly and connection to a pressure step loading function and a time varying pressure sequence. Phase three consisted of the preliminary investigation of shear resistance with two complete wall assemblies. The pressure data was derived from wind load tests conduc ted on a model of a low rise rectangular building at the University of Western Ontario (UWO) Boundary Layer Wind Tunnel. (Ho et al., 2003) The final stage of the research applied in plane shear loads to two wall assemblies using the racking procedure described in ASTM E72. The connection details were identical to the bending test configuration.

PAGE 25

25 Table 3 1 CSIP Panel Inventory Table 3 2 Testing Matrix Panel Type Test type A B C Sum Static 1 4 0 5 Dynamic step loading 2 6 0 8 Racking 0 0 4 4 Total 17 Panel type Panel Dimension Core Density Panels # A 13 9 .7 x 121 9 .2 x 243 8 .4 m m 16.02 kg/m3 (1 pcf) 4 B 13 9 .7 x 121 9 2 x 243 8 .4 m m 48.05 kg/m3 (3 pcf) 11 C 254.0 x 1219.2 x 2438.4 m m 48.05 kg/m3 (3 pcf) 8 D 13 9 .7 x 121 9 .2 x 182 8 .8 m m 48.05 kg/m3 (3 pcf) 1 E 254.0 x 1219.2 x 2438.4 m m 48.05 kg/m3 (3 pcf) 1

PAGE 26

26 CHAPTER 4 DESIGN AND EVALUATIO N OF CSIP WALL CONNE CTION The objective of this phase was to design an easily constructible connection detail for the implementation in the field as depicted in Figure 4 1 Ease of construction will help to promote the use of such panels in under developed regions. The lateral capacities of five nail type fasteners presented in Table 4 1 were quantified us ing a modified lateral resistance ASTM D 1761 test performed on a Tinius properties are temperature dependent, thus face sheet s were tested at two ambient modified to 30 40 times faster than prescribed to minimize temperature effects on cooling of the face sheet s. The thermoplastic composite face sheet s were first separated from the foam core and cut in to 38.1 cm x 8.8 cm (15 in x 3.5 in) specimens, then stored in a temperature controlled room. Samples were assembled in two stages. Stage one prepared a southern yellow pine (SYP) 2 x 4 on edge with a steel bar at one end. Stage two, immediately preceded testing, attached a face sheet specimen to the opposite end of t he SYP 2 x 4 with a single fastener. The single fastener was driven in by a Bostitch Magnesium Model No. F21PL nail gun 620 at 53 kPa (90 psi) The Bostich nail gun was calibrated to a pressure that set the fastener flush. The modified A STM D 1761 loading was required to maintain the face sheet s at prescribed temperature for the Figure 4 shows a photo of a sample and an illustration of the individual components of the test sample. Figure 4 3 and Figure 4 4 provide a summary of the initial test conducted to aid in the selection of a fastene r that would provide the greatest

PAGE 27

27 lateral resistance. A single fastener was chosen based on the results. Thirty add itional tests were performed on the chosen fastener to quantify dependency on the lateral capacity as a function of specific gravity and moisture content. Specific gravity and moisture content were obtained using ASTM standards, ASTM 2395 and ASTM 4442 respectively. From each of the samples tested multiple sections of timber member were taken and evaluated for specific gravity and moisture content and then recorded This provided a direct comparison to NDS lateral capacity of sheathing to main members. Connection Testing Results : Tests were performed under t wo temperatures: and a test was performed to compare the results ( Table 4 2 ) All but one fastener failed to reject the null hypothesis which indicates that temperature did not play a significant role in the determination of the lateral capacit y of a face sheet attached with a single nail The only nail that fail ed to reject the null hypothesis was B RS 8d which is a ring shank common 8d nail. Figure 4 4 groups the results per effective area. All of the fasteners were found to be suitable for assembling the CSIP systems. The stainless steel ring shank 6d common nail ( SST SS) was selected for its resistance to rusting in corrosive coastal environments.

PAGE 28

28 Figure 4 1 Proposed construction assembly Figure 4 2 Lateral resistance testing Photo courtesy of George Fernandez, University of Florida.

PAGE 29

29 Figure 4 3 Lateral capacities of nail fasteners

PAGE 30

30 Figure 4 4 Lateral capacities stratified by effective area

PAGE 31

31 Table 4 1 Fastener summary ID Shank Shank diameter Head diameter Shank length A eff mm in mm in mm in mm 2 in 2 A S Smooth 3.33 0.131 7.19 0.283 63.50 2.5 00 31.612 0.049 B RS Ring 3.33 0.113 6.99 0.275 60.33 2.375 31.612 0.049 C S Smooth 3.33 0.113 7.19 0.283 50.80 2 .000 34.193 0.053 SST RS Ring 3.33 0.113 6.99 0.275 50.80 2 .000 31.612 0.049 SST SS Spiral 3.33 0.113 6.71 0.264 50.80 2 .000 29.032 0.045 Table 4 2 T test results for single fastener at two differen t temperatures Paired test 32.22 A S 8d B RS 8d C S 6d SST SS 6d SST RS 6d 21.11 A S 8d 0 B RS 8d 1 C S 6d 0 SST SS 6d 0 SST RS 6d 0 0=Failed to reject null hypothesis 1=Rejected null hypothesis

PAGE 32

32 CHAPTER 5 EVALUATIO N OF CSIPS WALL ASSE MBLY SUBJECTED TO WI ND PRESSURE LOADING Phase two experiments used the High Airflow Pressure Loading Actuator (HAPLA Figure 5 1 A ) which is based on the Pressure Loading Actuator system (PLA) developed by UWO ( Kopp et al., 2010) This system was used to apply time varying pressure to surfaces of buildings and other structures. The HAPLA consists of two 75 hp centrifugal blowers connected to a valve with a rotating central d isk actuated by a servo motor. The servo controls the dis pressure with a three chamber design. Custom LabVIEW 8.5 software operating on a National Instruments PXI system controls the pressure in the chamber through a 100 Hz PID loop that receives feedback from an Ashcroft XLdp transducer attached to the test chamber. Currently the HAPLA can simulate up to a 3Hz sinusoidal wave form and a peak pressure of 7 kPa. Time varying pressure sequences with the HAPLA are limited to frequency components as high as 2Hz in th e target pressure time history. Mid span deflections were recorded with a Balluff BOD63M photoelectric distance sensor. The reaction frame consisted of two similar attachment points to represent th e floor and roof connection seen in Figure 4 1 Complete test assembly walls were instrumented with six TML model PFL 30 11 3L strain gauges on the exterior face sheet s. Additionally, three were installed on the compression fa ce and three on the tension face at the third points along the mid section of the wall as illustrated on Figure 5 2 Figure 5 3 is a detail for both the top and bottom connection for the out of plane test frame.

PAGE 33

33 Press ure step loading function The step loading pressure time history was applied to determine the peak static load at failure ( Figure 5 4 ). The function was generated us ing an initial pressure of 0.24 kPa (5 psf) and increasing by 0.48 kPa (10 psf) up to 6.96 kPa (145 psf) with duration of ninety sec onds per pressure step. The pressures were chosen based on the capabilities of the system Table 5 1 shows the fifteen unique pressures tabulated in the center column the right column was determined by solving for the equivalent design wind speed based on ASCE7 10. Equation 5 1 from sect ion 27.3.2 ASCE 7 10 Main Wind Force Resisting System (MWFRS) envel ope was used to find the equivalent wind velocity for the static pressure step. This e quation require d certain assumption s to be made about the velocity pressure coefficient, topographic fa ctors, and the wind directionality factor The equivalent basic wind speeds are presented in Table 5 1 (5 1) Individual Test Results Static Step Loading 48.05 kg/m 3 (3 pcf) Test #1 T est one demonstrated the need for adjustment to the PID parameters at loads greater than 3.60 kPa for subsequent test. Figure 5 5 is a time history plot presenting a target and measured pressure sequence for visual comparison T his figure shows that the PID parameters were not correctly calibrated above 2.5 kPa pressure target. Figure 5 6 represents the deflection time history for this test. The specimen was still able to obtain a max pressure of 6.45 kPa. Figure 5 7 is a sketch of the de bonding that was observed at the conclusion of the test The large de bonding that was observed on the rior factsheet was not caused by mechanical failure and can

PAGE 34

34 be attributed to manufacturing error. T he top timber split longitudinally i n to three segments however, de bonding between the top section of the interior face sheet was measure d to be less than the bottom where the connection remained without fracture. Though the ultimate load of the system was approximately 6.45 kPa the plot on Figure 5 9 shows the limit state was exceeded at approximately at 1 kPa. The blue line represents a deflection limit state of L/180. This pressure corresponds to a wind speed between 40 and 51 m/s (90 and 100 mph). T he theoretical deflection was vali dated with the data collected as can be seen on Figure 5 9 Figure 5 9 confirms t h e equation used accurately predict ed the mid span deflection The strain profile exhibit a linear behavior up to approximately 2 kPa when they begin to show nonlinear trends. The center strain gauge SG5 on the tension face reported the highest strain as expected and SG2 on the compression face had the largest compression strains. The panel however was not performing as a perfectly bonded composite beam evidenced by the unequal opposite strains. At 3 kPa the compression strain gauges be gin to lose their compression leading the author to conclude the face sheet is slipping from the core and no longer behaving as an ideal composite sy stem. Test #2 Test two pressure sequence demonstrates the correction s to the PID parameters have improved the tracking of the sy stem to the pressure sequence. Figure 5 10 presents a target and measure pressure sequence. Figure 5 11 represents the deflection time history for this test. Failure is evident at approximately 4.6 kPa with a sudden drop in the measure pressure; at this point failure of the top timber plate occ urred Figure 5 12 is a sketch of the de bonding that was observed at the end of the

PAGE 35

35 testing. Based on the failure it was difficult to quantify the d e bonding along the inside face. Though the ultimate load of the system was approximately 4.6 kPa the plot on Figure 5 13 shows the limit state was e xceeded approximately at 1 kPa. The blue line represents a deflection limit state of L/180. This pressure corresponds to a wind speed between 40 and 51 m/s (90 and 100 mph). Nonetheless the theoretical deflection was substantiated by the data, as shown in Figure 5 13 As with the initial test, Figure 5 13 validates t he equation used does predict the mid span deflection accurately. The strain profile exhibit ed linear behavior up to approximately 3.5 kPa on the tension face and SG5 and 2 kPa for the compression face SG2 just before failure when they begin to show nonlinear trends. As anticipated, the largest strains for both the tension and compression face were located at the center with SG5 on the tension face and SG2 on the compression face. T he unequal opposite strains prove the panel is not behaving as a perfectly bonded composite beam. At 2.5 kPa the compression strain gauges begin to lose their com pression which suggests the face sheet is slipping from the core and no longer behaving as an ideal composite system. Test #3 Maintaining the same PID parameters from test two the HAPLA was able to maintain the accurate pressure throughout the fifteen different pressure steps. Figure 5 14 presents a target and measure pressure sequence. The panel and timber connection s we re exposed to all fifteen quasi static pressure steps reaching an ultimate pressure of approximately 6.9 kPa without failur e. Figure 5 15 shows the deflection time history for this test. Figure 5 16 depicts the de bonding. At the completion of the

PAGE 36

36 pressure test the panel was examined for de bonding and the inside bottom left corner showed no signs of de bonding between the face sheet s a nd the foam core. T he limit state was exceeded at approximately at 1.5 kPa and that the ultimate pressure was 6.9 kPa ( Figure 5 17 ). The blue line rep resents a deflect ion limit state of L/180. This pressure corresponds to a wind speed between 40 and 51 m/s (90 and 100 mph). However the theoretical deflection in this test overestimated the actual deflection as can be seen in the plot on Figure 5 17 This plot also shows the wall system deflection not behaving linearly with pressure in contrast to the other test s described in this section. The strain profile exhibit ed linear behavior up to approximately 3.5 kPa on the tension face with SG5 and 2 kPa for the compression face with SG2 just before failure which is when they begin to show nonlinear trends. As expected, the center strain gauge SG5 on the tension face reported the highest stra in, yet the SG2 on the compression face had the largest compression strains. The pa nel however is not behaving as a perfectly bonded composite beam and is evident with the unequal opposite strains. At 2.5 kPa the compression strain gauges begin to lose the ir compression strain which leads the author to believe the face sheet is slipping from the core and no longer behaving as an ideal composite system. Test #4 Maintaining the same PID parameters from test two the HAPLA was able to maintain the accurate pre ssur e throughout the eleven different pressure stages. The pressure sequence shown in Figure 5 18 depicts the target and the measured pressure sequence s The panel and timber connection s were able to withstand all fifteen qua si static pressure steps and failed at the eleventh stage of the pressure sequence. Figure

PAGE 37

37 5 19 shows the deflection time history. This panel did not have data for de bonding at the conclusion of testing. Though the ultimate load of the system was approximately 4.9 kPa the plot on Figure 5 20 shows the limit state was exceeded approximately at 1 kPa indicated by the blue line. The blue line represents a deflection limit state of L /180. This pressure corresponds to a wind speed between 40 and 51 m/s (90 and 100 mph). The theoretical deflection in this test was an accurate predictor of the actual deflection as can be seen in the plot on Figure 5 20 This plot also shows the theoretical deflection is underestimating the true wall system defle ction. The strain profile exhibit ed linear behavior up to approximately 3 kPa on the tensio n face and SG5 and 2 kPa for the compression face SG2 when they begin to show nonlinear trends. The center strain gauge SG5 on the tension face reported the highest strain which was expected and SG2 on the compression face had the largest compression str ains. The panel did not behav e as a perfectly bonded composite beam a s proven by the unequal opposite strains. At 2.5 kPa the compression strain gauge s begin to lose their compression strain leading the author to believe the face sheet is slipping from the core and no longer behav es as an ideal composite system. Individual Test Results Static Step Loading 16.02 kg/m3 (1 pcf) Test #1 Maintaining the same PID parameters from test two the HAPLA was able to maintain the accurate pressure throughout the ten different pressure stages. The pressure sequence on Figure 5 21 presents a target and the measured pressure sequence for panel test number one. The m easured pressure sequence begins to show evidence of not achieving the target p ressure at the fifth pressure step. The panel and

PAGE 38

38 timber failed at the eleventh stage of the pressure sequence. Figure 5 22 represents the deflection time history for this test. Figure 5 23 is a sketch of the de bo nding that results at the end of the testing. At the completion of the pressure test the panel was examined for de bonding Separation of the wall assembly from the ti mber connection hindered the ability to collect de bonding data. Though the ultimate load of the system was approximately 5.05 kPa, the plot on Figure 5 25 shows the l imit state which is the blue line was exceeded approximately at 1 kPa. The blue line represents a deflection limit state of L/180. This pressure corresponds to a wind speed between 40 and 51 m/s (90 and 100 mph). The theoretical deflection in this test w as an accurate predictor of the actual deflection as can be seen in the plot on Figure 5 25 This plot also shows the theoretical deflection is under estimating the tr ue wall system deflection. The strain profile exhibit ed linear behavior up to approximately 3 kPa on the tension face and SG5 and 2 kPa for the compression face SG2 when they begin to show nonlinear trends. The center strain gauge SG5 on the tension face r eported the highest strain, which was expected, and SG2 on the compression face had the largest compression strains. The panel however is not behaving as a perfectly bonded composite beam and is evident with the unequal opposite strains. Time varying pres sure sequence a 12.19 m W x 19.05 m L x 7.32 m H (40 ft W x 62.5 ft L x 24 ft H) building i n the NIST Aerodynamic Database (Ho et al., 2003). The NIST database provides time his tories of Cp values for all taps from a particular wind tunnel model along with other prudent information. For the time varying pressure sequence the C p values taken from the NIST

PAGE 39

39 database must be converted from model scale to full scale values using the r educed frequency relationship E quation 5 2 f is the frequency of the data, l is the integral length scale, U is the mean wind speed. Equation 5 3 was used to obtain the full scale pressure time history from converted C p time history. (5 2) (5 3) Each segment was filtered to 2 Hz using a low pass filter to improve controllability. The mean of the record was increased by 10% to account for internal pressurization (1994 Vickery). For this pressure sequence, seg ments of full records were concatenated to form a full record length of 10394 sec (173.2 min). The basic wind speed was incremented in steps of 4.47 m/s beginning with 44.7 m/s, a total of eleven basic wind speeds were used to formulate the pressure sequen ce. Mechanical limitations of the HAPLA prevented peak pressure in stage 11 from being achieved; Figure 5 26 is a full time history of the target pres sure. Figure 5 27 is a narrow segment of time to show the reduction in peak pressure gain and the fidelity of the HAPLA. Table 5 2 show the basic wind speed used with Equation 5 2 to compute the pressure time history. The mean pressure in Table 5 2 is the measured mean pressure from the data recorded. Individual Test Results Dynamic Loading 48.05 kg/m 3 (3 pcf) Test #1 The pressure sequence in the following figure presents the data that w as recorded from subjecting the CSIP panel to the time varying pressure sequence. The plot on Figure 5 28 is the target and measured pressure vs. time. T he HAPLA system

PAGE 40

40 was unsuccessful at achieving the higher peaks on the final stage of the sequence however the system was able to reproduce smaller amplitude fluctuation s with accuracy. Figure 5 29 is the displacement vs. time; this plot has a disjunction in the data at approximately the 7500 second mark due to the movement of the displacement traducer for safety reasons. Figure 5 30 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previously used in the quasi static pressure results plots. The blue line once again r epresents the deflection limit state of L/180 for light wood frame construction. Equation 2 4 was an accurate predictor of the mean deflection for this CSIP wall. The limit state line unlike that of the static walls is exceeded at approximately 0.4 kPa. Figure 5 31 is a sketch of the de bonding and failure observed after the pressure sequence was completed and the wall removed from the test frame. Test #1 had no visua l de bonding at the conclusion of the test and both the top and bottom plate remain undamaged. Test #2 The pressure sequence in Figure 5 32 presents the data that was recorded from subjecting the second CSIP panel to the same time varying pressure sequence described in the previous chapter. The plot on Figure 5 32 is the target and measured pressure vs. time. The HAPLA system was unsuccessful at achieving the higher peaks on the final stage of the sequence, however the system was able to repr oduce smaller amplitude fluctuation with accuracy. Figure 5 33 is the displacement vs. time, this plot has a disjunction in the data at approximately the 10,500 secon d mark, due to a large movement of the wall assembly. Inspection after completion of the sequence yielded no visual evidence for failure and timber connection remained intact.

PAGE 41

41 Figure 5 34 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previously used in the quasi static pressure results plots. The blue line again represents the deflection limit state of L /180 for light wood frame construction. Equation 2 4 was an accurate predictor of the mean deflection for this CSIP wall, yet it slightly over predicts loads below 1.4 kPa and under predicts deflection for higher loads. The limit state line unlike that of the static walls, is exceeded at just above 0.4 kPa based on the theoretical deflection line and at 0.6 kPa based on the measured data. Figure 5 35 i s a sketch of the de bonding and failure observed after the pressure sequence was completed and the wall removed from the test frame. Test #2 had no visual de bonding at the conclusion of the test and both the top and bottom plate remain undamaged. Test # 3 The pressure sequence in Figure 5 36 presents the data that was recorded from subjecting the third CSIP panel to the same time varying pressure sequence described in this chapter. The plot on Figure 5 36 is the target and measured pressure vs. time. T he HAPLA system was unsuccessful at achieving the higher peaks on the final stage of the sequence, however the system was able to reproduce smal ler amplitude fluctuation with accuracy. Figure 5 37 is a plot of the displacement vs. time. This plot has a jump in the data at approximately 10,000 seconds. After i nspection of the wall assembly, at the completion of the sequence, the cause of the discontinuity was partial cracking of the bottom timber plate. Figure 5 38 plots the deflection vs. pressure. The red line represents the theoretical equation previously used in the quasi static pressure results plots. The blue line represents the deflection limit state of L/180 for light wood frame construction.

PAGE 42

42 Equatio n 2 4 was an accurate predictor of the mean deflection for this CSIP wall, yet it slightly over predicts loads below 1.4 kPa and under predicts deflection for higher loads. The limit state lin e unlike that of the static walls, is exceeded at just above 0. 4 kPa based on the theoretical deflection line and at 0.6 kPa based on the measured data. Figure 5 39 is a sketch of the de bonding and failure observe d after the pressure sequence was completed and the wall removed from the test frame. Test #3 had no visual de bonding at the conclusion of the test and only the top plate remained undamaged. The bottom plate had visual evidence of partial cracking. Test #4 The pressure sequence in Figure 5 40 presents the data that was recorded from subjecting the fourth CSIP panel to the same time varying pressure sequence described in this chapter. The plot on Figure 5 40 is the target and measured pressure vs. time. The HAPLA system was unsuccessful at achieving the higher peaks on the final stage of the sequence, however the system was able to reproduce sm aller amplitude fluctuation with accuracy. Figure 5 41 is the displacement vs. time T his plot shows that the system exhibits a non linear behavior at lower loads tha t was not seen in previous test. This change in slope on Figure 5 41 indicates a change in the load deflection response because it is occurring during the fourth repe tition of the final stage segment. Figure 5 42 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previo usly used in the quasi static pressure results plots. The blue line represents the deflection limit state of L/180 for light wood frame construction. Equation 2 4 was an accurate predictor of the mean deflection for this CSIP wall, there was an over predic tion of loads below 1.4 kPa and under prediction of deflection for loads higher than 1.4 kPa. The limit state line, unlike that of the static

PAGE 43

43 walls, is exceeded at just above 0.4 kPa based on the theoretical deflection line and at 0.6 kPa based on the data Figure 5 43 is a sketch of the de bonding and failure observed after the pressure sequence was completed and the wall removed from the test frame. Te st #4 had no visual de bonding at the conclusion of the test and only the top plate was damaged with visual evidence of partial cracking. Test #5 The pressure sequence in Figure 5 44 presents the data that was recorded from subjecting the fifth CSIP panel to the same time varying pressure sequence described in this chapter. The plot on Figure 5 44 is the target and measured pressure vs. time. This wall sample was successful sustaining loads until the eleventh stage, but only completed two repetitions of the stage before failing. Figure 5 45 which is the deflection vs. time plot, illustrates at approximately 5000 seconds there was either some de bonding or failure bas ed on the CSIP walls deflection response. Figure 5 46 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previously use d in the quasi static pressure results plots. The blue line represents the deflection limit state of L/180 for light wood frame construction. Equation 2 4 was an accurate predictor of the mean deflection for this CSIP wall I t slightly over predicts loads below 1.4 kPa and under predicts deflection for higher loads. The limit state line, unlike that of the static walls, is exceeded at just above 0.4 kPa based on the theoretical deflection line and at 0.6 kPa based on the data. Figure 5 47 is a sketch of the de bonding and failure observed after the pressure sequence was completed and the wall removed from the test frame. Test #5 had no visual de bonding at the conclusion of the test and neither the top plate nor bottom plate had any visual signs of damage.

PAGE 44

44 Test #6 The pressure sequence in Figure 5 48 presents the data that was record ed after subjecting the sixth CSIP panel to the same time varying pressure sequence described in the previous chapter. The plot on Figure 5 48 is the target and measured pressure vs. time. This wall sample was successful sustaining loads through the sixteenth stage before failing. Figure 5 49 which is the deflection vs. time plot, illustrates at approximately 5,000 seconds and 9,000 seconds that there was some form of either de bonding or failure based on the CSIP walls deflection response. Figure 5 50 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previously used in the quasi static pressure results plots. Th e blue line represents the deflection limit state of L/180 for light wood frame construction. Equation 2 4 was an accurate predictor of the mean deflection for this CSIP wall I t does over predict for loads below 1.4 kPa and under predicts deflection for h igher loads. The limit state line, unlike that of the static walls, is exceeded at just above 0.4 kPa based on the theoretical deflection line and at 0.6 kPa based on the data. Figure 5 51 is a sketch of the de bonding and failure observed after the pressure sequence was completed and the wall removed from the test frame. Test number six had visual de bonding on the top of the inside face sheet. At the conclusion of th e test the top plate had no signs of damage, but the bottom plate had a partial crack as a visual sign of damage. Individual Test Results Dynamic Loading 16.02 kg/m 3 (1 pcf) Test #1 The pressure sequence in Figure 5 52 presents the data that was recorded from subjecting the sixth CSIP panel to the same time varying pressure sequence described

PAGE 45

45 in the previous chapter. The plot on Figure 5 52 is the target and measured pressure vs. time. This wall sample successfully sustained loads through the seventh stage before failing. Figure 5 53 the deflection vs. time plot, illustrates at approximately 5500 seconds there was some form of either de bonding or failure based on the CSIP walls deflection response. Figure 5 54 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previously used in the quasi static pressure results plots. The blue line repr esents the deflection limit state of L/180 for light wood frame construction. Equation 2 4 was an accurate predictor of the mean deflection for this CSIP wall ; it under predicts loads above 0.5 kPa. The limit state line, unlike that of the static walls, is exceeded at just above 0.5 kPa as per the theoretical equation and the measured data. Figure 5 55 is a sketch of the de bonding and failure observed after the pressu re sequence was completed and the wall removed from the test frame. Test number one of the light core density did show visual de bonding on the bottom of the inside face sheet and the top right of the outer face sheet. At the conclusion of the test, the to p plate had no signs of damage but the bottom plate was damaged in the process of removal from the test frame. As a result, it could not be determined whether it was a partial crack or a full failure of the member. Test #2 The pressure sequence in Figure 5 56 presents the data that was recorded from subjecting the sixth CSIP panel to the same time varying pressure sequence described in the previous chapter. The plot on Figure 5 56 is the target and measured pressure vs. time. This wall sample was successful sustaining loads through the eighth stage before failing. Figure 5 57 which is the deflection vs. time plot, illustrates at approximately

PAGE 46

46 6,000 seconds there was some form of either de bonding or failure based on the CSIP walls deflection res ponse. Figure 5 58 shows the deflection vs. pressure relationship. The red line represents the theoretical equation previously used in the quasi static pressure results plots. The blue line represents the deflection limit state of L/180 for light wood frame construction. Eq uation 2 4 was not an accurate predictor of the mean deflection for this CSIP wall. The limit state line, unlike that of the static walls, is exceeded at just above 0.4 kPa based on the theoretical deflection line and the data. Figure 5 59 is a sketch of the de bonding and failure observed after the pressure sequence was completed and the wall removed from the test frame. Test number two of the light core density did sh ow visual de bonding on the bottom of the inside face sheet and the top of the outer face sheet. At the conclusion of the test, the top plate had partial cracking but the bottom plate was damaged and had complete failure. Summary of Pressure Loading Result s The pressure step loading sequence provided a reference for load deflection relationship for a complete wall assembly and verification of theo retical equations presented in E quation 2 4 A total of five CSIP wall assemblies were tested, by subjecting the m to the pressure step load sequence. The results of the five CSIP walls are given in Table 5 3 and Table 5 4 for the 16.02 kg/m 3 (1 pcf) and four 48.05 kg/m 3 (3pcf) wall assemblies. The plots on Figures 5 55 and 5 56 indicate the summary of the load vs. deflection data for each of the two density groups tested with the step pressure loading sequence. The summary results show that the wall assemblies have a linear relationship between load and deflection. The solid line was computed using Eq. 2. Though currently only prudent to the state of Florida, both the International Residential

PAGE 47

47 Code (IRC 2009) and Florida Building Code (FBC 2010) Section R613 makes reference to max deflection of SIP type walls to be L/254. The blue line presented in chapter five corresponded to a deflection limit state of L/180 for standard light wood frame stud walls. It was determined from analyzing the load deflection data that the mid span deflection was exceeded at a pressure load of 0.72 kPa and 1 kPa which correspond to the L/254 and L/180 deflection limit state r espectively. When analyzing the strain data, it was observed, as shown in Figure 5 9 Figure 5 13 Figure 5 17 Figure 5 20 and Figure 5 25 a strain vs. load plot, that the compressive strain exhibits a form of strain release. This type of behavior was also observed in research performed by Roberts et al. (2002). Thi s strain behavior was observed in all tested panels, with some being more prominent than others. This strain behavior has a possibility indicative of active de bonding. When de bonding initiates it would cause the compressive sheet to alter its behavior fr om a composite material to a single membrane in tension. Figure 5 62 presents four photographs of failures observed during testing. Within both parts of phase two all but one wall assembly did not fail. Failure was primarily observed in the connection plates w ith splitting of the timber parallel to grain. Morrison (2011) conducted a study focused on fasteners in wood that emphasized the importance of understanding a) the ultimate capacity under static loading and b) the failure load under cyclic loading. This study conducted tests similar to those described by Morrison et al. (2011). The results exhibit a decrease of loading rather than static loading conditions, even thou gh average peak loads reflected

PAGE 48

48 a lower difference of approximately 26%. Figure 5 63 is a load vs. deflection plot of one test with minimal filtering that shows a hig h variability, making it difficult to predict the dynamic behavior. However, as shown in the individual result s discussion, with limited filtering it is possible to predict mean deflection vs mean pressure per stage of the dynamic sequence. Figure 5 64 is a summary of both 48.05 kg/m 3 (3 pcf) and 16.02 kg/m3 (1 pcf) CSIP wall subjected to the dynamic pressure sequence. It was observed that the wall, although flexible, had difficulty returning to a natural state after load was removed. Therefore, it is concluded that this scatter is difficult to predict be cause current equation used do not take previous condition s into consideration. It could be possible to correct some uncertainty using a coefficient of restitution to account for some of the variability Results shown in Figure 5 65 also suggest that shear deformation in the core may have a larger influence in dynamic loading. As first introduced with the pressure step loading, the strain behavior may be a sign of de bonding (not visible during testing). Figure 5 65 is a representation of similar behavior observed on all CSIP walls subjected to dynamic loading. It demonstrates that while tensile behavior is linear with regard to applied load, mid span st rain compression does not follow any particular pattern and almost demonstrates a plateau effect at approximately 100 micro strains. Comparison of Capacity and Failure Modes Table 5 3 tabulates the results for the maximum pressure through the duration of a specific test and provides a mean with one standard deviation. The maximum pressure a 48.05 kg/m 3 (3 pcf) and a16.02 kg/m 3 (1 pcf) CSIP panel was able to resist was 6 .90 kPa 5.05 kPa, respectively, during the Pressure step loading sequence. As

PAGE 49

49 per the results, prior to failure during the dynamic pressure sequence, the mean maximum pressure for both core densities was approximately 25% less than the mean maximum measure d pressure observed in the step loading, whose mean was 5.71 kPa compared to 4.21 kPa Table 5 4 provides a summary of the maximum deflection and compares the results and means of both the step loading test and the dynamic test. When comparing the mean max deflection between step loading and dynamic test results, we find a difference of approximately 30%.

PAGE 50

50 Figure 5 1 Two photographs of the pressure testing components A ) HAPLA testing apparatus B ) exterior view of testing chamber Photo courtesy of George Fernandez and Forrest Masters, University of Florida.

PAGE 51

51 Figure 5 2 Strain gauge location Figure 5 3 Reaction frame connection detail

PAGE 52

52 Figure 5 4 Pressure step loading function Figure 5 5 48.05 kg/m 3 (3 pcf) CSIP Test 1 pressure time history

PAGE 53

53 Figure 5 6 48.05 kg/m 3 (3 pcf) CSIP Test 1 deflection time history Figure 5 7 48.05 kg/m 3 (3 pcf) CSIP Test 1 de bonding and failure sketch

PAGE 54

54 Figure 5 8 48.05 kg/m 3 (3 pcf) CSIP Test 1 failure Photo courtesy of George Fernandez, University of Florida. Figure 5 9 48.05 kg/m 3 (3 pcf) CSIP Test 1 summary

PAGE 55

55 Figure 5 10 48.05 kg/m 3 (3 pcf) CSIP Test 2 pressure time history Figure 5 11 48.05 kg/m 3 (3 pcf) CSIP Test 2 deflection time history

PAGE 56

56 Figure 5 12 48.05 kg/m 3 (3 pcf) CSIP Test 2 de bonding and failure sketch Figure 5 13 48.05 kg/m 3 (3 pcf) C SIP T est 2 summary

PAGE 57

57 Figure 5 14 48.05 kg/m 3 (3 pcf) CSIP Test 3 pressure time history Figure 5 15 48.05 kg/m 3 (3 pcf) CSIP Test 3 deflection time history

PAGE 58

58 Figure 5 16 48.05 kg/m 3 (3 pcf) CSIP Test 3 de bonding and failure sketch Figure 5 17 48.05 kg/m 3 (3 pcf) CSIP Test 3 summary

PAGE 59

59 Figure 5 18 48.05 kg/m 3 (3 pcf) CSIP Test 4 pressure time history Figure 5 19 48.05 kg/m 3 (3 pcf) CSIP Test 4 deflection time history

PAGE 60

60 Figure 5 20 48.05 kg/m 3 (3 pcf) CSIP Test 4 summary Figure 5 21 16.02 kg/m 3 (1 pcf) CSIP Test 1 pressure time history

PAGE 61

61 Figure 5 22 16.02 kg/m 3 (1 pcf) CSIP Test 1 deflection time history Figure 5 23 16.02 kg/m 3 (1 pcf) CSIP Test 1 de bonding and failure sketch

PAGE 62

62 Figure 5 24 16.02 kg/m 3 (1 pcf) CSIP Test 1 failure Photo courtesy of George Fernandez, University of Florida. Figure 5 25 16.02 kg/m 3 (1 pcf) CSIP Test 1 summary

PAGE 63

63 Figure 5 26 Complete time varying pressure sequence Figure 5 27 Sample of the time varying pressure sequence

PAGE 64

64 Figure 5 28 48.05 kg/m 3 (3 pcf) CSIP Test 1 time varying pressure time histo ry Figure 5 29 48.05 kg/m 3 (3 pcf) CSIP Test 1 time varying pressure deflection time history

PAGE 65

65 Figure 5 30 48.05 kg/m 3 (3 pcf) CSIP Test 1 time varying pressure deflection load plot Figure 5 31 48.05 kg/m 3 (3 pcf) CSIP Test 1 de bonding and failure sketch

PAGE 66

66 Figure 5 32 48.05 kg/m 3 (3 pcf) CSIP Test 2 time varying pressure time history Figure 5 33 48.05 kg/m 3 (3 pcf ) CSIP Test 2 time varying pressure deflection time history

PAGE 67

67 Figure 5 34 48.05 kg/m3 (3 pcf) CSIP Test 2 time varying pressure deflection load plot Figure 5 35 48.05 kg/m3 (3 pcf) CSIP Test 2 de bonding and failure sketch

PAGE 68

68 Figure 5 36 48.05 kg/m 3 (3 pcf) CSIP Test 3 time varying pressure time history Figure 5 37 48.05 kg/m 3 (3 pcf) CSIP Test 3 time varying pressure deflection time history

PAGE 69

69 Figure 5 38 48.05 kg/m3 (3 pcf) CSIP Test 3 time varying pressure deflection load plot Figure 5 39 48.05 kg/m3 (3 pcf) CSIP Test 3 de bonding and failure sketch

PAGE 70

70 Figure 5 40 48.05 kg/m 3 (3 pcf) CSIP Test 4 time varying pressure time history Figure 5 41 48.05 kg/m 3 (3 pcf) CSIP Test 4 time varying pressure deflection time history

PAGE 71

71 Figure 5 42 48.05 kg/m3 (3 pcf) CSIP Test 4 time varying pressure deflection load plot Figure 5 43 48.05 kg/m3 (3 pcf) CSIP Test 4 de bonding and failure sketch

PAGE 72

72 Figure 5 44 48.05 kg/m 3 (3 pcf) CSIP Test 5 time varying pressure time hi story Figure 5 45 48.05 kg/m 3 (3 pcf) CSIP Test 5 time varying pressure deflection time history

PAGE 73

73 Figure 5 46 48.05 kg/m3 (3 pcf) CSIP Test 5 time varying pressure deflection load plot Figure 5 47 48.05 kg/m3 (3 pcf) CSIP Test 5 de bonding and failure sketch

PAGE 74

74 Figure 5 48 48.05 kg/m 3 (3 pcf) CSIP Test 6 time varying pressure time history Figure 5 49 48.05 kg/m 3 (3 pcf ) CSIP Test 6 time varying pressure deflection time history

PAGE 75

75 Figure 5 50 48.05 kg/m3 (3 pcf) CSIP Test 6 time varying pressure deflection load plot Figure 5 51 48.05 kg/m3 (3 pcf) CSIP Test 6 de bonding and failure sketch

PAGE 76

76 Figure 5 52 16.02 kg/m 3 (1 pcf) CSIP Test 1 time varying pressure time history Figure 5 53 16.02 kg/m 3 (1 pcf) CSIP Test 1 time varying pressure deflection time history

PAGE 77

77 Figure 5 54 16.02 kg/m 3 (1 pcf) CSIP Test 1 time varying pressure deflection load plot Figure 5 55 16.02 kg/m 3 (1 pcf) CSIP Test 1 de bonding and failure sketch

PAGE 78

78 Figure 5 56 16.02 kg/m 3 (1 pcf) CSIP Test 2 time varying pressure time history Figure 5 57 16.02 kg/m3 (1 pcf) CSIP Test 2 time varying pressure deflection time history

PAGE 79

79 Figure 5 58 16.02 kg/m3 (1 pcf) CSIP Test 2 time varying pressure deflection load plot Figure 5 59 16.02 kg/m 3 (1 pcf) CSIP Test 2 de bonding and failure sketch

PAGE 80

80 Figure 5 60 Deflection load results for the 16.02 kg/m 3 (1 pcf) panels Figure 5 61 Deflection load results for the 48.05 kg/m 3 (3 pcf ) panels

PAGE 81

81 Figure 5 62 Observed damage Photo courtesy of George Fernandez, University of Florida. Figure 5 63 Un f iltered mid span deflection versus pressure

PAGE 82

82 Figure 5 64 Filtered mid span d eflection versus pressure Figure 5 65 Time varying pressure vs. mid span strain

PAGE 83

83 Table 5 1 Static pressure step and equivalent mean wind speed Step # Mean Pressure kPa (psf) Mean Wind Velocity m/s (mph) 1 0.24 (5) 23.24 (52) 2 0.72 (15) 40.23 (90) 3 1.20 (25) 51.85 (116) 4 1.68 (35) 61.24 (137) 5 2.16 (45) 69.73 (156) 6 2.64 (55) 76.89 (172) 7 3.12 (65) 83.59 (187) 8 3.60 (75) 89.85 (201) 9 4.08 (85) 95.66 (214) 10 4.56 (95) 101.03 (226) 11 5.04 (105) 106.39 (238) 12 5.52 (115) 111.31 (249) 13 6.00 (125) 115.78 (259) 14 6.48 (135) 120.70 (270) 15 6.96 (145) 125.17 (280) Table 5 2 Time varying basic wind speed and mean pressure Step # Basic Wind Velocity m/s (mph) Mean Pressure kPa (psf) 1 44.71 (100) 0.409 (8.5) 2 49.17 (110) 0.496 (10.4) 3 53.69 (120) 0.593 (12.4) 4 58.11 (130) 0.697 (14.5) 5 62.58 (140) 0.811 (16.9) 6 67.05 (150) 0.933 (19.48) 7 71.52 (160) 1.064 (22.22) 8 75.99 (170) 1.204 (25.24 9 80.46 (180) 1.352 (28.23) 10 84.93 (190) 1.509 (31.51) 11 16 89.40 (200) 1.675 (34.98) Table 5 3 Maximum pressure comparison table Foam core density Maximum step loading pressure (kPa) Maximum dynamic pressure (kPa) Sample Mean Std Sample Mean Std 48.05 kg/m 3 (3 pcf) 6.45 5.71 1.13 4.42 4.21 0.28 4.60 4.63 6.90 4.22 4.90 3.89 --4.13 --3.96 16.02 kg/m 3 (1 pcf) 5.05 5.05 3.78 3.81 --3.84

PAGE 84

84 Table 5 4 Maximum deflection table Foam core density Maximum quasi steady deflection (m m) Maximum dynamic deflection ( m m) Sample Mean Std Sample Mean Std 48.05 kg/m 3 (3 pcf) 10 1 0 9 0 4 22 2 5 3 .0 6 2 .1 1 7 1 6 4 0 7 7 1 11 4 7 6 9 8 8 1 9 3 6 4 --5 5 .0 --8 1 5 16.02 kg/m 3 (1 pcf) 9 5 0 9 5 0 4 6 1 7 6 5 --10 6 9

PAGE 85

85 CHAPTER 6 EVALUATION OF CSIP W ALL ASSEMBLY SUBJECT ED TO SHEAR LOADING Methodology Racking loads were applied to two separate CSIP wall assemblies in accordance with ASTM E72. The panels tested had a core thickness of 25.4 cm (10 in) however their density was 48.05 kg/m3 (3 pcf). Two separate tests were conducted. Each test used two CSI P wall segments of dimension 25 4 .0 x 121 9 .2 x 243 8 .4 m joined along the top and bottom with a reduced timber member. The bottom connection was the same as the connecti on detail in the pressure loading wall assembly. The top connections consisted of a timber member affixed to the wall and resting on fixed roller to prevent vertical translation. Loads were applied using a n Enerpac 10,000 psi hydraulic hand pump connected to a 12 ton Enerpac hydraulic bottle jack model number RCH123 The loads were measured using an Omega 2 267.96 Kg (5,000 lb) load cell model number LCH 5k. The data was recoded using a NI PXI at 50 Hz. Displacement was measured using t wo Unimeasure Linear Position Transducers (JX P420 25 N11 11S 31N and JX PA 80 N11 11S 321) and o ne Balluff BOD63M photoelectric distance sensor measured displacements. Figure 6 1 is a diagram of the test set up per ASTM E72. The location of instrumentation, supports, and restrains of the wall assembly are labeled in Figure 6 1 Results The ra cking tests were completed at the University of Florida (UF) Weil Hall structur al engineering test laboratory. Table 6 1 presents a summary of the dat a collected during these two tests. Figure 6 2 Panel 1 performed with a 38% higher capacity than panel two. This difference in load capacity between both wall assembl ies

PAGE 86

86 highlights the requirement for further testing. It is evident from Figure 6 3 that de bonding could be a possible explanation for the difference between the perfor mances of the two test setups. Figure 6 3 and Figure 6 4 present the failure observed at the completion of test 1. Comparing the data from this study with results observed by Kermani et al. (2006) our study can be recognized as conservative. The test configuration exceeded SIP panels with a core of 95 mm with no vertical load by 77% and performed equivalent to a solid OSB SIP wall with 25 kN (5620 .22 lb) of v ertical load. The panel assemblies tested at the UF were only fastened with a top plate and bottom plate.

PAGE 87

87 Figure 6 1 Diagram of racking test setup Figure 6 2 Racking test load deflection interaction

PAGE 88

88 Figure 6 3 Racking test face sheet de bonding Photo courtesy of George Fernandez and Scott Bolton, University of Florida. Figure 6 4 Racking test face sheet wrinkling Photo courtesy of George Fernandez and Scott Bolton, University of Florida.

PAGE 89

89 Table 6 1 Racking test results Test # Vertical Loads kN (lb) Test Ultimate Load kN (lb) Maximum displacement cm (in) Failure observed 1 0 27.9 (6271.4) 7.06 (2.78) Wrinkling of the face sheet 2 0 20.1 (4519.8) 7.80 (3.07)

PAGE 90

90 CHAPTER 8 CONCLUSION S An experimental study was performed to characterize the out of plane bending and shear capacity of composite structural insulated panels. This experimental work was conducted to evaluate adequacy of a connection detail and to improve the understanding of C SIP wall assemblies subjected to uniaxial bending caused by pr essure loading and shear response to rac king loads Phase one of the testing investigated the design of the connection detail. The investigation of the connection design began with the selection of five nail type, though all fasteners were found to be suitable for the connection ultimately a stainless steel 6d nail was chosen. A conservative approach was taken when specifying the nail spacing of 76.2 mm (3 in) on center for interior nails and 38. 1 mm (1.5 in) for the end nails. Unlike timber construction nails are only place d along the top and bottom edge of face sheets to connect to the timber members With the elimination of studs the quantity of nails required diminishes. Typical perimeter spac ing for timber walls may vary from 51 mm (2 in) to 153 mm (6 in) and 305 mm (12 in) for field nails. (Lindt et al., 2005) Following the design of the connection, the two core densities of CSIP wall s w ere subjected to wind pressures u sing the HAPLA a tota l of 19 walls were tested The wall assemblies were tested with both increasing step loading and time varying pressures. The walls with a step loading condition were able to obtain an average peak load capacity of 5.71 kPa (119 psf) In comparison the wall tested with a time varying pressure averaged a maximum load of 4.21 kPa (88 psf) These pressure s correspond to instantaneous wind velocities of approximately 110 m/s (246 Mph) and 100 m/s (223 Mph) respectively, exceeding the highest wind observed by onl y 20 m/s (46 Mph).

PAGE 91

91 T hough the likelihood of exposure to sustained winds of that magnitude due to a tropical cyclone is minimal the possibility does seem to be plausible in a tornado. Ultimate loads comprise only part of the criteria for building material s, another facet of the material designers must concern themselves with would be the serviceability. The walls subjected to step pressure and time varying pressure failed to meet deflection limit states of L/254 at the design pressure. (FBC, 2010) The wall assemblies exceeded the deflection limit state at approximately 0.72 kPa equivalent to a wind velocity of 40 m/s ( 90 m ph) during the step loading. The walls exceeded the deflection limit state at 0.4 kPa during the time varying pressure sequence, which would be equivalent to a mean 60 sec wind speed of 44 m/s (100 mph). However, m ost coastal communities require a design wind speed of 58 m/s (130 mph) or greater. (ASCE 7 2010) Delamination was observed a t the conclusion of 7 of the 14 pressure tests. The step loaded walls experience delamination in 4 of the 5 test in contrast to the time varying pressure loaded walls experiencing delamination in 3 of the 9 walls. The difference in the observed delaminatio n suggests that the walls may vary in performance based on load time exposure and perhaps be susceptible to creep. In plane loads are also a consideration for design. Based on a limited sample size only two wall assemblies were tested for their in plane sh ear capacities. Consequent to the vertical timber end members and a joint between two individual wall segments not being installed, the racking test setup can be considered to be a conservative approach to the racking capacities of such wall assemblies. It is the

PAGE 92

92 timber connection around the perimeter and having the two panels joined at the center. However, the two shear walls tested did exceed comparable timber shear wal ls tested by 77% with no vertical load. (Kermani et al., 2006) CSIP walls have been introduced to aid in the advancement of building materials and technologies. Based on the s mall sample size the ability to conduct statistical comparison between core dens ities and loading conditions for this study was limited. T he recommendation for future testing are as followed: Select one core density and one thickness to allow for larger sample size, enabl ing probabilistic modeling. Alternative adhesives should be exp lored to prevent delamination between face sheets and core to maximize potential composite behavior. Testing of shear walls with varying location and size of openings and vertical load to compar e with equivalent timber wall studies

PAGE 93

93 APPENDIX A RESULTS OF LATERAL LOAD TEST Figure A 1 Lateral nail capacity failure results A S 8d Photo courtesy of George Fernandez, University of Florida.

PAGE 94

94 Figure A 2 Lateral nail capacity failure results B RS 8d Photo courtesy of George Fernandez, University of Florida.

PAGE 95

95 Figure A 3 Lateral nail capacity failure results C S 6d Photo courtesy of George Fernandez, University of Florida.

PAGE 96

96 Figure A 4 Lateral nail c apacity failure results SST SS 6d Photo courtesy of George Fernandez, University of Florida.

PAGE 97

97 Figure A 5 Lateral nail capacity failure results SST RS 6d Photo courtesy of George Fernandez, University of Florida.

PAGE 98

98 LIST OF REFERENCES 1. Allen, H. G. (1969). Analysis and design of structural sandwich panels Oxford: Pergamon Press. 2. ASTM D1761 06 (2006) Standard Test Methods for Mechanical Fasteners in Wood, West Conshohocken, PA: American Society for Testing and Materials, 1 00 Barr Harbor Drive, PO Box C700, 1942 3. Baehre, R., & Ladwein, T. (1994). Diaphragm action of sandwich panels. Journal of Constructional Steel Research, 31 (2), 305 316. 4. Daniel, I. M., & Abot, J. L. (2000). Fabrication, testing and analysis of composite s andwich beams. Composites Science and Technology, 60 (12), 2455 2463. 5. Davies, J. M. (1993). Sandwich Panels. Thin WalledStructures ,16, 179 198. 6. Fam, A., Sharaf, T., & Shawkat, W. (2010). Structural performance of sandwich wall panels with different foam core densities in one way bending. Journal of Composite Materials, 44 (19), 2249 2263. 7. Florida Building Commission and International Code Council (2010) Florida building code 2010 International Code Council, Country Club Hills, IL 8. Florida Building Commiss ion and International Code Council (2009) International Residential Code 2009 International Code Council, Country Club Hills, IL 9. Girhammar, U. A. (2008). Composite beam columns with interlayer slip Approximate analysis. International Journal of Mechanical Sciences, 50 (12), 1636 1649. 10. Girhammar, U. A. (2009). A simplified analysis method for composite beams with interlayer slip. International Journal of Mechanical Sciences, 51 (7), 515 530. 11. Girhammar, U. A. (2008). Composite beam columns with interlayer slip approximate analysis. 12. Hartsock, J. A. (1969). Design of foam filled structures Stamford, Conn: Technomic Pub. Co.

PAGE 99

99 13. Heuer, R., Adam, C., & Ziegler, F. (2003). Sandwich panels with interlayer slip subjected to thermal loads. Journal of Thermal Stresse s, 26 (11 12), 1185 1192. 14. T.C. E. Ho, D. Surry, and D.P. Morrish (2003). BLWT SS20 2003 NIST/TTU Cooperative Agreement Windstorm Mitigation Initiative: Wind Tunnel Experiments on Generic Low Buildings. 15. T.C. E. Ho, D. Surry, and M. Nywening (2003). BLWT SS21 2003 NIST/TTU Cooperative Agreement Windstorm Mitigation Initiative: Further Experiments on Generic Low Buildings. 16. Kermani, A., Hairstans, R. (2006). Racking Performance of Structural Insulated Panels. Journal of Structural Engineering 132, 1806 1812. 17. Kopp, G., Morrison, M., Gavanski, E., Hender son, D., and Hong, H. (2010). Three Project: Hurricane Risk Mitigation by Integrated Wind Tunnel a nd Full Scale Laboratory Tests. Natural Hazards Review 11(4), 151 161. 18. Manalo, A. C., Arav inthan, T., & Karunasena, W. (2010). Flexural behaviour of glue laminated fibre composite sandwich beams. Composite Structures, 92 (11), 2703 2711. 19. Mohammed A. M., Nasim U. (2011) Global buckling of composite structural insulated wall panels, Materials & D esign 32(2), 766 772. 20. Morrison, M. J., & Kopp, G. A. (2011). Performance of toe nail connections under realistic wind loading. Engineering Structures, 33 (1), 69 76. 21. Mullens, M., & Arif, M. (2006). Structural insulated panels: Impact on the residential construction process. Journal of Construction Engineering and Management ASCE, 132 (7), 786 794. 22. Prevatt, D. O. (1994). Improving the cyclone resistance of traditional Caribbean house construction through ratio nal structural design criteria. Journal of Wind Engineering and Industrial Aerodynamics 52 305 319 23. Roberts, J. C., Boyle, M.P., Wienhold, P. D., Ward, E. E. and White, G. J. (2002). Strains and deflections of GFRP sandwich panels due to uniform out of plane pressure. Marine Technology and SNAME Ne ws, 39 (4), 223.

PAGE 100

100 24. Schnabl, S., Saje, M., Turk, G., & Planinc, I. (2007). Locking free two layer Timoshenko beam element with interlayer slip. Finite Elements in Analysis & Design, 43 (9), 705 714. 25. Uddin, N. and Mousa, M. A. (2010). Debonding of composites s tructural insulated sandwich panels. Journal of Reinforced Plastics and Composites, 29 (22), 3380 3391. 26. Vaidya, A, et al. (2010). Structural Characterization of Composite Structural Insulated Panels for Exterior Wall Applications Journal of Composites for Construction Vol. 14, No. 4, 464 469. 27. Viswanathan, A. V., Tamekuni, M., Boeing Commercial Airplane Company., & Langley Research Center. (1973). Elastic buckling analysis for composite stiffened panels and other structures subjected to biaxial inplane lo ads Washington, D.C: National Aeronautics and Space Administration. 28. Wood, L. W., Forest Products Laboratory (U.S.), & University of Wisconsin. (1958). Sandwich panels for building construction Madison, Wis: U.S. Dept. of Agriculture, Forest Service, For est Products Laboratory. 29. Yoders, J. (2005). Re inventing Modular Construction. Building Design & Construction Aug 2005, 46 50.

PAGE 101

101 BIOGRAPHICAL SKETCH George Fernandez was born in 1986 in Jacksonville, North Carolina. At a young age he moved to Miami, Florid a, where he remained until 2005 when he graduated from dedication served to motivate George in pursuing his academic endeavors. Always intrigued by dismantling and reassembling household items, George was afforded a unique opportunity to enter an engineering magnet program at his high school. During his time at Miami Coral Park Sr. High George was actively involved in student government and the robotics team. He was his contribution and dedication in the field of technology. He later moved to Winston Salem, N orth C arolina where he went on to complete the majority of the requirements for his Associates of Arts degree. George also had the privilege of becoming the first Hispanic to join a local volunteer fire department and be a registered firefighter in that state. In the fall of 2006 George returned to Florida and enrolled in Santa Fe College. Upon completion of his Associate s degree he was admitted into the University of began his academic career by teaching incoming freshmen AutoCAD for the dept artment His first exposure to research was a position assisting professors with their individual projects at various labs within the civil engineering department. It was during this time he found his passion for Hurricane research, to which he would invest the following sev en years of his life. Under the guidance and tutelage of Dr. Forrest Masters and Dr. Kurt Gurley, George was welcomed into the Hurricane research group civil engineering department. For the duration of his

PAGE 102

102 undergraduate degree George was afforded t he opportunity to work with some of the leading experts and professors in the fields of wind engineering and civil engineering. He became a part of the Florida Coastal Monitoring Program (FCMP), a unique joint multi university venture established in 1996 t hat focuses on experimental methods to quantify tropical cyclone wind behavior and the resulting loads on residential structures. He assisted in preparation of field deployable instruments to quantify near surface cyclone winds for all named storms from fa ll 2006 through fall 2012. After completing his Bachelor of Arts degree in the fall of 2009, George was extended an offer to continue in his research as a graduate student under the mentorship of Dr. Forrest J. Masters. His graduate work was focused on wi nd load resistance Composite Structural Insulated Panels (CSIP), retrofitting FCMP field observation towers, and serving as the student lead for the Hurricane Hazard Immersion Program. The latter is an outreach program focused on providing current event in formation and hands on research experience to middle school Science, Technology, Engineering, and Mathematics (STEM) teachers for the development of a hurricane mitigation curriculum. George Fernandez is currently a student member of the American Association for Wind Engineering and American Society of Civil Engineers. Following his graduation with his Master of Engineering degree, George will join Florida International University as a research scientist.