Citation
Determination of Wind Out-of-Plane Failure Capacity of Plywood and OSB-Clad Wall Systems

Material Information

Title:
Determination of Wind Out-of-Plane Failure Capacity of Plywood and OSB-Clad Wall Systems
Creator:
Lopez, Jason Martin
Publication Date:
Language:
English

Notes

Abstract:
The wind uplift resistance of wood roof sheathing panels has been the subject of many wind engineering studies due to the summa probability of being subjected to large uplift loads during intense wind occurrences. Meanwhile, few studies focus on wall sheathing resistance, resulting in the assumption that their wind resistance matches the same capacity of roof sheathing panel constructed of similar materials. The study will determine the static out-of-plane wind resistance of wall sheathing on wood framed construction and evaluate whether current assumptions that the capacity is similar to that of a roofing sheathing panel is true. Structural tests were conducted on 20 wall specimens fabricated of ½ in. thick OSB and plywood sheets fastened using nails to Spruce-pine fir wood framing. Using a monotonic-increasing static pressure step-and-hold test sequence, the mean out-of-plane wind resistance capacity attained was 4.21 kPa (88 psf), and we found no statistical difference between plywood and OSB wall panels. When the results are compared against results for roof panels from previous tests conducted by the third author, it was found the out-of-plane capacity of wall panel systems was approximately 17% higher than that of roof panels. ( en )
General Note:
Awarded Bachelor of Science in Civil Engineering, summa cum laude, on May 8, 2018. Major: Civil Engineering
General Note:
College or School: College of Engineering
General Note:
Advisor: David O. Prevatt. Advisor Department or School: Civil and Coastal Engineering

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Jason Martin Lopez. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

UFDC Membership

Aggregations:
UF Undergraduate Honors Theses

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

DETERMINATION OF WIND OUT OF PLANE FAILURE CAPACITY OF PLYWOOD AND OSB CLAD WALL SYSTEMS By JASON MARTIN LOPEZ A THESIS PRESENTED TO THE UNDER GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQU IREMENTS FOR THE SUMMA CUM LAUDE IN ENGINEERING UNIVERSITY OF FLORIDA 2018

PAGE 2

2018 Jason Martin Lopez

PAGE 3

To my sweetie, Sara, to whom has been by my side this entire journey.

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my undergraduate research advisor Dr. David O. Prevatt for the chance to work with him and his Wind Hazard Damage Assessment Group Under his mentorship I was first exposed to the research community and I am greatly indebted to him. This opportun ity has open ed ma n y doors for me and I know it will open many more in the future. I would like to thank Dr. David B. Roueche for all his help at the lab and processing the data from the experiments. His insight to the project was invaluable and much of the work is owed to his expertise. I would like to thank Dr. Kurtis R. Gurley and Dr. David W. Mazyck for agreeing to serve on my thesis defense committee. I would like to thank my mother, Kathy Highley, for always believing in me and encouraging to follow m y passion when others told me not to I would like to thank my girlfriend, Sara Kovachich for being by my side every step of the way. I would like to thank the members of my Research Group Rodrigo Castillo and Anant Jain The ir help has been indispensa ble, and I am proud to call them my friends. I would like to thank Dr. Samesha R. Barnes and the Ronald E. McNair Post Baccalaureate Program for preparing to undertake the task of conducting research and completing an undergraduate thesis. This research would not have been possible without the financial support from the National Science Foundation (NSF) through the a ward 1150975 CAREER: Tornado Resilient Structural Retrofits for Sustainable Housing Communities.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 11 1 INTRODUCTION ................................ ................................ ................................ ........ 12 2 BACKGROUND ................................ ................................ ................................ .......... 15 3 L ITERATURE REVIEW ................................ ................................ .............................. 17 Wall Wind Out of Plane Tests ................................ ................................ ................. 17 Wind Uplift Testing of Roof Panels ................................ ................................ ......... 19 Failure Mechanisms ................................ ................................ ................................ 20 4 EXPERIMENTAL TASKS ................................ ................................ ........................... 21 Test Setup ................................ ................................ ................................ .............. 21 Instrumentation ................................ ................................ ................................ ....... 22 Pressure Measurements ................................ ................................ .................. 22 D isplacement Measurements ................................ ................................ ........... 22 Test Procedure & Pressure Trace ................................ ................................ .......... 24 Wall Section Test Sample ................................ ................................ ....................... 24 Moisture Content & Specific G ravity ................................ ................................ ....... 26 5 RESULTS ................................ ................................ ................................ ................... 31 Structural Sheathing Panels ................................ ................................ ................... 31 Statistical Analysis ................................ ................................ ................................ .. 33 Distribution ................................ ................................ ................................ .............. 34 6 DISCUSSION ................................ ................................ ................................ ............. 36 7 CON CLUSION ................................ ................................ ................................ ........... 40 FULL FAILURE MODE AND LOCATION ................................ ................................ ...... 41 FULL TEST RESULTS FOR P LYWOOD PANELS ................................ ....................... 49 FULL TEST RESULTS FOR OSB PANELS ................................ ................................ .. 54 STIFFNESS OF LINEAR ELASTIC REGION OF SHEATHING AND STUD ................. 59

PAGE 6

6 LIST OF REFERENCES ................................ ................................ ............................... 61

PAGE 7

7 LIST OF TABLES Table page Table 2 1. Materials and nail spacing used for a typical roof and wall system. ........... 16 sheathing tests. ................................ ................................ ................................ ............. 18 Table 3 2. Comparison of the differences in walls samples between current study and that of Kopp and Gavanski (2012). ................................ .............................. 19 Table 4 1. Nail schedule and dimensions used for wall samples tested. ..................... 26

PAGE 8

8 LIST OF FIGURES Figure page Figure 1 1. Shear load transfer due to a lateral wind load. ................................ .......... 13 Figure 3 1. Typical construction for roof u p lift test samples. Nailing pattern shown is typical code minimum 152/305 mm (6/12 in.) This layout results in 33 nails per panel, with 9 nails on edges and 5 nails in the field. ................................ ..... 20 Figure 4 1. Test setup for wall tests wood wall sample installed on steel test chamber. The plastic sheathing draped over sides of steel chamber was placed between sheathing and w ood framing and the edges are taped to chamber wall creating air tight seal. ................................ ................................ ... 22 Figure 4 2. Installation of two string potentio meters to wood framing member and sheathing. The relative displacement of sheathing under suction pressure is obtained as difference of the two measurements. ................................ .............. 23 Figure 4 3. Pressure trace used to test wall. ................................ ............................... 25 Figure 4 4. Typical 152/305 mm (6/12 in.) nai ling pattern (62 nails) used to attach the 1.2 x 2.4 m (4 x 8 ft) wood sheathing panel to the framing members that are spaced at 406 mm (16 in.) on center. ................................ ........................... 27 Figure 4 5. Wood samples used to determine moisture content and specific gravity of wood framing member. Five samples wood member and the average value is used in calculation. ................................ ................................ ................ 29 Figure 4 6. Locations and nomenclature of wood samples nomenclature cut from the framing members. ................................ ................................ ......................... 30 Table 5 1. Summary of ultimate capacity, moisture content, density, and specific gravity for Spruce Pine Fir (SPF) test samples. ................................ .................. 31 Figure 5 1. View of wall panel during testing. The OSB sheathing is pulled away from framing members, with maximum displacements observed at the two middle studs. Initial sep aration of OSB at both center studs at the same time. OSB test 7. ................................ ................................ ................................ ......... 32 Figure 5 3. Typical recorded results (5 th Plywood test). indicates point at which failure occurred. (a) Pressure time history, (b) Time history of the separation between the structural sheathing panel from the stud, displacement time history of sheathing panel, displacement time history of stud. ............................ 34 Figure 5 4. Represents the normalized histograms from the Moisture Content and Specific Gravity data collected with the Probability Distrib ution Function that

PAGE 9

9 best fits the data: (a) Lognormal Distribution for MC, (b) Normal Distribution for SG. ................................ ................................ ................................ ................ 35 Figure A 1. Failu re mode and location for plywood sample test 1. Test statistics from the plywood group, OSB group, and both groups treated as one group. .... 41 Figure A 2. Sample statistics, failure mode, and location for plywood samples 2, 3, and 4. ................................ ................................ ................................ ................. 42 Figure A 3. Sample st atistics, failure mode, and location for plywood samples 5, 6, and 7. ................................ ................................ ................................ ................. 43 Figure A 4. Sample statistics, failure mode, and location for plywood samples 8, 9, and 10. ................................ ................................ ................................ ............... 44 Figure A 5. Sample statistics, failure mode, and location for OSB samples 1, 2, and 3. ................................ ................................ ................................ ................. 45 Figure A 6. Sample statistics, failure mode, and location for OSB samples 4, 5 and 6. ................................ ................................ ................................ ........................ 46 Figure A 7. Sample statistics, fa ilure mode, and location for OSB samples 7, 8, and 9. ................................ ................................ ................................ ................. 47 Figure A 8 Sample statistics, failure mode, and location for OSB sample 10. ........... 48 Figure B 1. Plywood Test Sample 1 Pressure and Displaceme nt Time History. ......... 49 Figure B 2. Plywood Test Sample 2 Pressure and Displacement Time History. ......... 49 Figure B 3. Plywood Test Sample 3 Pressure and Displacement Time History. ......... 50 Figure B 4. Plywood Test Sample 4 Pressure and Displacement Time History. ......... 50 Figure B 5. Plywood Test Sample 5 Pressure and Displacement Time History. ......... 51 Figure B 6. Plywood Test Sample 6 Pressure and Displacement Time History. ......... 51 Figure B 7 Plywood Test Sample 7 Pressure and Displacement Time History. ......... 52 Figure B 8. Plywood Test Sample 8 Pressure and Displacement T ime History. ......... 52 Figure B 9. Plywood Test Sample 9 Pressure and Displacement Time History. ......... 53 Figure B 10. Plywood Test Sample 10 Pressure and Displacement Time History. ..... 53 Figure C 1. OSB Test Sa mple 1 Pressure and Displacement Time History. ............... 54 Figure C 2. OSB Test Sample 2 Pressure and Displacement Time History. ............... 54

PAGE 10

10 Figure C 3. OSB Test Sample 3 Pressure and Displacement Time History. ............... 55 Figure C 4. OSB Test Sample 4 Pressure and Displacement Time History. ............... 55 Figure C 5. OSB Test Sample 5 Pressure and Displacement Time History. ............... 56 Figure C 6. OSB Test Sample 6 Pre ssure and Displacement Time History. ............... 56 Figure C 7. OSB Test Sample 7 Pressure and Displacement Time History. ............... 57 Figure C 8. OSB Test Sample 8 Pressure and Displacement Time History. ............... 57 Figure C 9. OSB Test Sample 9 Pressure and Displacement Time History. ............... 58 Figure C 10. OSB Test Sample 10 Pressure and Displacement Time History. ........... 58 Figure D 1. Plywood test samples 1 10. Red line coincides with stiffness of stud and blue line with sheathing panel. ................................ ................................ ..... 59 Figure D 2. OSB test samples 1 10. Red line coincides with stiffness of stud and Blue line with sheathing panel. ................................ ................................ ........... 60

PAGE 11

11 Abstract of Thesis Presen ted to the Underg raduate School of the University of Florida in Partial Fulfillment of the Requirements for Summa Cum Laude in Engineering DETERMINATION OF WIND OUT OF PLANE FAILURE CAPACITY OF PLYWOOD AND OSB CLAD WALL SYSTEMS By Jason Ma rtin Lopez May 2018 Chair: David O. Prevatt Major: Civil Engineering The wind uplift resistance of wood roof sheathing panels has been the subject of many wind engineering studies due to the high probability of being subjected to large uplift loads duri ng intense wind occurrences. Meanwhile, few studies focus on wall sheathing resistance, resulting in the assumption that their wind resistance matches the same capacity of roof sheathing panel constructed of similar materials. The study will determine the static out of plane wind resistance of wall sheathing on wood framed construction and evaluate whether the current assumption that the capacity of a wall sheat hing panel is similar to that of a roofing sheathing panel is true. Structural tests were conducted on 20 wall specimens fabri cated of 13 mm ( in. ) thick OSB and plywood sheets fastened to s pruce p ine fir wood framing using nails Using a monotonic increasing static pressure step and hold test sequence, the mean out of plane wind resistance capacity attained was 4.21 kPa (88 psf ), and we found no statistical difference between plywood and OSB wall panels. When the results are compared against results for roof panels from previous tests it was found that the out of plane capacity of wall panel systems was approximately 17% hig h er than that of roof panels.

PAGE 12

12 1 INTRODUCTION In the United States, 90% of single family residential structures consist of light framed wood structural systems (LFWS) in which the roof structures and external walls are sheathed with structural wood sheathing panels (van de Lindt and Dao 2009) These panels have dual purposes, 1) to resist in place forces through diaphragm/shear wall action; 2) serving as flexural members in resisting the out of plane forces due to wind and gravity loads on the roofs and wind suctions and p ressures acting on the walls. The exterior walls (sheathed wood stud walls) in LFWS act as columns to support the vertical loads of the roof structure, and as shear wal ls for transferring lateral forces collected by the roof diaphragm to the foundation ( Figure 1 1 ) They also resist out of plane loads induced by wind pressures and provide protection from wind borne debris to the occupants. Thu s, the exterior walls in LFWS are subjected to both in plane and out of plane lateral loads and are critical to maintaining structural integrity and occupant safety during extreme wind events. While the in plane shear capacity of walls (Standohar Alfano et al. 2017) and roof diaphragms (Lee and Rosowsky 2005) have been studied, and several studies have documented the out of plane structural (wind uplift) behavior of roof sheathing panels (Datin et al. 2011; H ill 2009 ; Kopp and Gavanski 2012 ) little work has been done on the out of plane capacity for exterior wood wall systems. In numerical fragility models the default value of the out of plane capacity for exterior walls has been the same as the roof sheathi ng system, despite the fact that the support of the two systems are different and the material properties are not the same.

PAGE 13

13 Figure 1 1 Shear load transfer due to a lateral wind load. The need for a more accurate value of the out of plane failure capacity for wall sheathing arose while the third author was developing a probabilistic computer model used to predict the tornado induced damage ratios to light framed wood residential structures (Peng et al. 2016; Prevatt et al. 2016) A literature review revealed previous numerical models of wood framed buildings (Kasal et al. 1994; Kopp and Gavanski 2012; Li and van de Lindt 2012; Martin et al. 2 011; Thampi et al. 2011) used the same wind uplift capacity for wall systems as for roof sheathing panels. Intuitively this should not be the case, since typical wall framing panels typically have more closely spaced framing (406 mm (16 in.) versus the 61 0 mm (24 in.) spacing used for roof sheathing panels. Further, wall sheathing panels have blocking along the top and bottom edges of panels, which provides additional fastening locations for the panels. Also, in Florida the framing material for wall sheat hing is spruce pine fir (SPF) S G of 0.42, while typically

PAGE 14

14 southern yellow pine (SYP) SG of 0.55, framing is used in roof trusses and rafters (AWC 2017) In their numerical model for tornado damage, Prevatt et al. (2016) used out of plane mean capacity from Henderson et al. (2013) of 4.44 kPa (92.7 psf) and a coefficient of variance (COV) of 0.15. The predicted damage ratios of wall sheathing was about 50% at the tornado center line, a much lower percentage than the observed damage ratios of 7 5% found in the Garland Rowlett tornado (Bhusar 2017) It is the structural failure capacity for wall sheathing is responsible for this anomalous result.

PAGE 15

15 2 BACKGROUND A typical wood structural panel fabricated as plywood or oriented strand board (OSB) measures 2.44 m x 1.22 m (8 ft by 4 ft). For use as roof sheathing the panels are supported transversely on wood framing oriented along the short dimension of the sheets (parallel to the 1.22 m (4 ft) dimension) at 610 mm (24 in.) on center. Assuming a code minimum 150 mm /300 mm (6 in./12 in.) fastener array from the 2 017 Florida Building Code Residential, 6th Edition (FBCR 2017) 33 fasteners are required. In contrast, for wall sheathing systems, the structural panels are installed with their long dimensions oriented vertically and parallel to the direction of framing members. Here, the panels are supported longitudi nally by the wood studs, placed spaced 406 mm (16 in.) apart. In addition, wood structural panels are typically fastened to blocking (wall plates) along the short edges of the panel at the top and bottom of each sheet (assuming the exterior wall is 2.4m (8 ft) in height). As a result, wall sheathing panels in a typical house construction will have nearly twice as many fasteners to attach the wood structural panel to the framing members, which suggests its o ut of plane capacity should be higher that a roof she athing panel, Table 2 1 Given the current drive to develop performance based design approaches the use of these values may be too conservative and therefore provide incorrect results. In particular, for developing tornado resil ient structural design methods, a realistic fragility function developed using an accurate probabilistic model of the wind loads and representative structural capacities of all components involved. To this point, the wind uplift capacity values of wood roo f sheathing has been erroneously used to represent the wind uplift capacity of wall sheathing panels although the structural configurations of these two systems are different.

PAGE 16

16 Table 2 1 Materials and nail spacing used for a t ypical roof and wall system. House Component Structural Framing Exterior Structural Sheathing Nail Spacing Nail Type Number of Nails Per Panel Roof Southern Yellow Pine (SYP) Plywood or OSB 152/305 mm (6/12 in.) 8d common 33 Wall Spruce Pine Fir (SPF) Plywood or OSB 152/305 mm (6/12 in.) 6d common 62

PAGE 17

17 3 L ITERATURE REVIEW Wall Wind Out of Plane Tests Kopp and Gavanski (2012) presented an experimental study to determine the ultimate capacities of exterior sheathing on wood framed wall systems. Table 3 1 shows the various wall types tested in the Kopp and Gavanski study. Ten samples of each wal l type were tested with five samples tested under ramp loading and five under fluctuating loading. Only one of five test samples (Wall 2, vertical studs with OSB sheathing) resembles the structural system used in residential walls in the United States hous ing. Even though the Wall 1 sample had a similar mean failure capacity as the Wall 2 sample, Wall 1 is more like a roof section than a wall section. The study used ramp loads to determine a baseline failure pressure to compare against a realistic fluctuating w ind load. The ramp rate loading was 1.92 kPa/min (40 psf/min) with an average test on wood panels lasting about 75 seconds. The full scale fluctuating pressure load time history, P t was created by using ( 1 ) where is the mean hourly wind speed taken at the mean roof height with a flat uniform terrain, is the pressure coefficient taken from an equivalent full scale 15 min sing equation ( 1 ) with an initial wind speed that is likely to not cause damage to a test sample, a 15 minute pressure time history was created. If failure does not occur within the first 15 minutes, the wind speed is scaled up by 2.2 m/s (5 mi/h) to create a new pressure time history. The process is continued until failure has occurred in a continuous cycle with no pause

PAGE 18

18 in between the wind scaling. An OSB test panel under fluctuating load can take 80 minutes before failure occurs. Table 3 1 Wall Configurations u sed by Kopp and Gavanski (2012) in exterior sheathing t ests The differences in methodology used in the current study versus that used by in Kopp and Gavanski (2012) is shown in Table 3 2 The difference in nails in Wall 1 can be accounted for because of a different orientation of the studs. Wall 1 has s tuds that are installed perpendicular to the 2.4 m (8 ft) length where Wall 2 and the current study used studs installed parallel to long edge of sheathing panel The difference in nails in Wall 2 comes from the top and bottom plates not being fully nailed every 152mm (6 in.) o n c enter This study will only use step and hold loading instead of ramp or fluctuating loading. Kopp and Gavanski tested five samples of each loading type (ramp and fluctuating) per wall section. Their results indicated that there was no statistical significanc e between ramp and fluctuating loads, which allows for conclusions to be draw that ramp loading and fluctuating loading can be treated as one group. Both studies use a nailing patter n of 152/305 mm (6/12 in.) on center spacing but this study attaches all edges of the sheathing panel to the studs w ith nails every 152mm (6 in.) on c enter Label Stud framing Orientation Sheathing Type Sheathing Fastener Vinyl Siding Siding Nails Fastener Count Wall 1 Parallel to Short Edge 11 mm OSB 6d common nail No No 43 nails Wall 2 Parallel to Long Edge 11 mm OSB 6d common nail No No 52 nails

PAGE 19

19 Table 3 2 Comparison of the differences in walls samples between current study and that of Kopp and Gavanski (2012). Current Study Kopp and Gavans ki (2012) Panel Material Plywood OSB Wall 1 OSB Wall 2 OSB Top Plate Double Double Single Single Number of Nails 62 62 43 52 Panel Orientation 2.4 m edge Parallel to Studs 2.4 m edge Parallel to Studs 2.4 m edge Perpendicular to Studs 2.4 m edge Parallel to Studs Nail Length 60.3 mm (2.37 in.) 60.3 mm (2.37 in.) 50.8 mm (2 in.) 50.8 mm (2 in.) Number of Specimens per Loading Type 10 Step and Hold 10 Step and Hold 5 Ramp 5 Fluctuating 5 Ramp 5 Fluctuating Wind Uplift Testing of Roof Panels Currently, no consensus exists for determining the wind uplift or out of plane capacity of wood sheathing panels. Hill (2009) developed an approach, used subsequently in Datin et al. (2011) to evaluate the out of plane capacities of wood roof panels. Using a 152 mm x 1.26 m x 2.49 m (6 in. x 4.15 ft x 8.19 ft) steel chamber and pressure loading actuator structural panel test specimens were subjected to either uniform static pressures Method A Static Pressure or to Method B Dynamic Pressure. Figure 3 1 shows the test specimen and fastener arrangement. Sealing of the chamber is achieved by a plastic sheet placed between framing members and the structural panel and taped to the side of the chamber.

PAGE 20

20 Figure 3 1 Typical construction for roof up lift test samples. Nailing pattern shown is typical code minimum 152/305 mm (6/12 in.) This layout results in 33 nails per panel, with 9 nails on edges and 5 nails in the field. Failure Mechanisms An invest igation into the failure mechanism from Datin et al. (2011) Kopp and Gavanski (2012) and Henderson et al. (2013) resulted in a trend various nail failures The common failure mechanisms observed were: nail pull through nail withdraw, nail head failure, shear failure of nail connection between top/bottom plate and stud, fracture in framing member, or sheathing panel fracture. A test sample was not limited to a single failure type, e.g. several experiments resulted in mu ltiple failures occurring in the same panel, but a single fastener connection within the panel is restricted to a single failure type.

PAGE 21

21 4 EXPERIMENTAL TASKS Test Setup A nominal 2.4 m 1.2 m (8 ft x 4 ft) steel chamber used to test the wood specimens is a box with five sides where the test sample is the sixth side (top side) and acts lik e a lid to the chamber ( Figure 4 1 ). A plastic sheet is draped over the sides of the chamber to ensure an airtight seal between the panel and the chamber. The steel chamber has inner dimensio ns of 152 mm x 1.26 m x 2.49 m (6 in. x 4.15 ft x 8.19 ft) which is slightly larger than the sheathing panels being tested; this allows for the wood sheathing panel to fit within the chamber without touching the sides of the steel chamber. The depth of the box allows the test specimen enough room for unimpeded deflection throughout the testing processes of applied suction pressure. The uniform pressure is applied solely to the structural sheathing and is transferred directly to the framing members through t he mechanical connection of the nails. The load on the framing members then proceeds through the bottom and top plates to the outside rim of the chamber where the top and bottom plates rest The steel pressure chamber connected to a Pressure Loading Actua tor (PLA) powered by a 40 HP, 3 phase 460 V centrifugal blower rated at 10 kPa at a peak airflow of 1.44 m 3 /s (3050 CFM). The pressure difference maintained across the sheathing panel is set by the valve position of PLA, which is controlled by a servo moto r. The apparatus is computer controlled using a PID feedback loop via a custom interface developed in LabVIEW. Development and full scale application of the PLAs can be found in Kemp (2008) and Kopp et al. (2012) respectively.

PAGE 22

22 Figure 4 1 Test setup for wall tests wood wall sample installed on steel test chamber. The plastic sheathing draped over sides of steel chamber was placed between sheathing and wood framing and the edges are taped to chamber wall creating air tight seal. Instrumentation Pressure Measurements The pressure time history was taken using a Pressure Transducer (PT) installe d on the side of the steel chamber. The PT used was an Omega dyne Model # PX243A 2.5BG5V Displacement Measurements Displacement measurements were taken using two Celesco string potentiometers acy of full stroke with repeatability of 0.02%. The two SP were placed side by side to measure the displacement of the local area ( Figure 4 2 ). One SP measured the

PAGE 23

23 displacement of the framing member, and the adjacent SP measured displacement of sheathing panel. For convenience, the SP on the framing member is referred to as SP1 and the SP on the sheathing panel is referred to as SP2. The attachment points of SP1 to the stud and SP2 to the panel is located a distance of 1.22 m (4 ft) from the bottom plate of the test specimen. SP1 was attached to stud B and SP2 was located a projected distance of 15.9 mm (5/8 in.) to side of SP1. The difference between the SP2 and SP1 measurements gives the amount of sheathing separation from frami ng member. All SP measurements are taken simultaneously during the testing phase, therefore the load vs. displacement can be determined for each test sample. Figure 4 2 Installation of two string p otentiometers to wood fr aming member and sheathing. The relative displacement of sheathing under suction pressure is obtained as difference of the two measurements.

PAGE 24

24 T est Procedure & Pressure Trace S ince no testing standard exists for determining the wind uplift or out of plane capacity of wood sheathing panels, the same procedure developed by Hill (2009) was used to test out of plane capacities of the laboratory fabricated wal l sections. For this experiment, UF WRSUT Method A Static pressure is used to determine the out of plane capacity for structural wall sheathing panels. The pressure trace developed for this method is a step and hold approach that increases pressure mono tonically by increments of 0.48 kPa (10 psf) and held for 60 seconds ( Figure 4 3 ). This cycle is repeated until the wall specimen fails or the Pressure Loading Actuator ( PLA ) capacity is reached. Absolute peak pressure values at the time of failure, the type of failure, and location of failure were recorded. Failure is defined as the separation of the panel from the framing member resulting from one of the nail failure mechanisms or from the fracture of specimen, as indicated abo ve. Any type of these failures will cause a sudden spike in pressure (indicating failure) which will cause the PLA to exceed its capacity then shut off. Wall Section Test Sample The wall sections that will be tested are representative of typical wood fram e residential structures found in practice built in Florida in accordance to the current 2017 Florida Residential Building Code (FBCR 2017) Care was taken to use representative construction techniques reflecting typical wood stud wall systems used in Florida. A wall panel consists of 2.32 m (91.5 in.) wall studs of nominal size 51 by 102 mm (2 by 4 in.) s pruce p ine f ir (SPF) and spaced 410 mm (16 in.) apart that are attached to a single bottom plate; 51 mm by 102 mm (2 in by 4 in.) SPF placed flat on one end and a double

PAGE 25

25 top plate (two 51 mm by 102 mm (2 in by 4 in.) ) SPF members placed flat on the other end. The overall height of the wall panel is 2.4m (96 in.). Each stud is fastened to the wall plates using two 76 mm (3 in.) smooth shank full round pneumatic gun nails. Both top and bottom wall plates extend beyond the sides of the panel so they can rest on side walls of the steel chamber, as shown Figure 4 1 In all, ten wall panels were fabricated using plywood sheeting (11.45 mm (0.451 in.) thick sheets) and ten using oriented strand board (OSB) sheathing (10.61 mm (0.418 in.) thick sheets). Figure 4 3 Pressure trace used to test wall The intention of the test is that the wall section being tested is only supported by the top and bottom plates that rest on the outside rim of the steel chamber. Once the framing members are nailed together, an over sized 4 mils thick plastic sheet is laid over the studs before the sheathing is installed. The plastic sheet is used to seal the

PAGE 26

26 assembly by attaching it to the side walls. The methodology used here is similar to methods used in Datin et al. (2011) and Henderson et al. (2013) while testing wind uplift capacities for roof sheathing panels. Figure 4 1 presents a summary of the components used to fabricate the testing samples. The sheathing is attached using a 152/305 mm (6/12 in.) nailing pattern, as shown in Figure 4 4 152/305 mm (6/12 in.) nailing pattern signifies, nails along the edges of the sheathing are spaced 152 mm (6 in.) on center and 305 mm (12 in.) on center on interior studs, i.e., stu ds that land in the field of the sheathing panel. Nails are off set a distance of 9.5 mm (3/8 in.) from the edge of the panel and each panel will have a total of 62 nails per panel ( Figure 4 4 ). Table 4 1 Nail schedule and dimensions used for wall samples tested. Panel Type Number of Samples Nail Type a Number of Nails Nail Use Nailing Pattern b Shank Diameter mm (in) Nail Head Diameter mm (in.) Nail Length mm (in.) OSB 10 6d 62 Sheathing 152/ 305 mm (6/12 in.) 2.87 (0.113) 7.11 (0.280) 60.3 (2.37) 12d 24 Framing 2 per stud 3.05 (0.120) 7.94 (0.313) 76.2 (3.00) Plywood 10 6d 62 Sheathing 152/305 mm (6/12 in.) 2.87 (0.113) 7.11 (0.280) 60.3 (2.37) 12d 24 Framing 2 per stud 3.05 (0.120) 7.94 (0.313) 76.2 (3.00) a Nail types are Bright, Non Coated, Smooth Shank, Full Round Head for Pneumatic Nail Guns b Nailing Pattern indicated spacing on edge/interior framing members. Moisture Content & Specific Gravit y Datin et al. (2011) recommended, when p erforming UF WRSUT testing protocol, to record the moisture content and specific gravity of the wood framing members. For this experiment, the Moisture Content (MC), density, and Specific Gravity were tested dard Test Methods for Density and

PAGE 27

27 Specific Gravity (Relative Density) of Wood and Wood (ASTM 2014) (ASTM 2014) was used because the test specimens are regular in shape, smooth, and sufficient measurements can be made to easily determine the volume. Figure 4 4 Typical 152/305 mm (6/12 in.) nailing pattern (62 nails) used to attach the 1.2 x 2.4 m (4 x 8 ft) wood sheathing panel to the framing members that are spaced at 406 mm (16 in.) on center. To obtain a representative value of MC for the wood frame, density, and SG, five samples of 25.4 mm (1 in.) in length were cut from each of the framing members ( Figure 4 5 ). The samples were taken roughly in equally spaced intervals. Because wood has a high variability of defects, samples cannot be taken at exact intervals; ASTM D2395 requires the wood samples be free of knots or other infiltrates. Also, to ensure a sample

PAGE 28

28 was not affected b y exterior conditions near end grains, the samples taken near the ends of the framing members were a minimum of 152 mm (6 in.) away from the edge. Each framing member in a wall section was individually labeled from A D and each wood specimen cut from a fra ming member was labeled 1 5, with1 being closet to the top plane and 5 near the bottom plate ( Figure 4 6 ). The samples were labeled with the wall section type (Ply (short for Plywood), OSB), test number (1 10), a designated fram ing member letter (e.g. A D), and a position number in the framing member (1 5), The moisture content in the wood was determined from equation ( 2 ): ( 2 ) where: = the moisture content in the wood specimen as a percentage. = the initial mass of the wood specimen = the mass of the oven dried wood specimen content was found by using equation ( 3 ). ( 3 ) where: = the density of the wood specimen at the moisture content = is the volume of the wood specimen at th e moisture content The specific gravity was determined by using equation ( 4 ) ( 4 ) where:

PAGE 29

29 = is the specific gravity of the wood specim en at the moisture content = is a constant which is determined by the units in which the mass and volume are measured. Figure 4 5 Wood samples used to determine moisture content and specific gravity of wood framing member. Five samples wood member and the average value is used in calculation.

PAGE 30

30 Figure 4 6 Locations and nomenclature of wood samples nomenclature cut from the framing members.

PAGE 31

31 5 RESULTS Structural Sheathing Panels Table 5 1 provides a summary of results for the 10 plywood and 10 OSB wall panels tested. Figure 5 1 details observations of failure progression typically seen throughout testing. The failure in test pa nels always initiated along either a single center stud or both center studs together. 90% of the samples had complete failure of both center members which then progressed to failure of the t op/ b ottom p late and side framing members, or various combinations of the aforementioned. Two of the test panels failed because of nail withdrawal from the stud to bottom plate connection ( Figure 5 2 ). Table 5 1 Summary of u ltimate capacity, moisture content den sity, and specific gravity for Spruce Pine Fir (SPF) test samples. Panel Type --Failure Pressure kPa (psf) COV Moisture Content of lumber COV of lumber kg/m 3 (lb/ft 3 ) COV Specific Gravity of lumber COV OSB Mean 4.11 (85.9) 0.109 10.8 % 0.065 484 (30.2) 0.087 0.436 0.084 Max 4.79 (100.1) 12.4% 586 (36.6) 0.523 Min 3.31 (69.2) 9.5% 401 (25.0) 0.363 Plywood Mean 4.31 (90.1) 0.130 11.3% 0.064 488 (30.5) 0.065 0.439 0.062 Max 5.21 (108.7) 13.2% 557 (34.8) 0.502 Min 3.31 (69.1) 9.6% 407 (25.4) 0.367 OSB & Plywood --4.21 (88.0) 0.106 11.10% 0.073 486 (30.4) 0.08 0.438 0.078 The performance of the OSB panels were tested using monotonic step and hold pressure trace with a static wind pressure ( Figure 5 3 (a) ). Displacement measurements of the sheathing panel and the stud ( Figure 5 3 (b) ) were taken simultaneously with pressure reading using two string potentiometers side by side ( Figure 4 2 ). The relative displacement of the sheathing panel (e.g., the separation of

PAGE 32

32 the sheathing panel from the stud) was determine by taking the difference between the structural sheathing panel displacement and the stud displacement ( Figure 5 3 (b)). The peak failure pressures were determined to be the max absolute value obtained in each test prior to failure of panel. After the experimental tests were completed the failure mechanisms and l ocations of failure were recorded. It was observed that o f the ten OSB specimens tested, seven of the tests resulted in at least one nail pull through and a max of four nails pulled through in one specimen. Whereas, out of the ten plywood test specimens th ere was only one nail pull through Figure 5 1 View of wall panel during testing. The OSB she athing is pulled away from framing members, with maximum displacements observed at the two middle studs. Initial separation of OSB at both center studs at the same time. OSB test 7.

PAGE 33

33 Subsequently f ive 25.4 mm (1 in.) samples were cut from each stud ( Figure 4 5 Figure 4 6 ). The samples were weighed and measured to determ ine their volume. The samples were dried in and oven at 102 C for 24 hours and weighed again. The samples were dried again for another 24 hours and reweighed to ensure practical equilibrium is reached (such that no more than 0.2% change in mass occurs ove r the drying period). The values obtained for the MC, density, and SG fall in the range of typical values that would be expected for SPF lumber. Figure 5 2 Failure of wall panel by nail wi thdrawal from along the bottom wall plate Statistical Analysis The Wilco xon Rank Sum (WRS) test was used to determine if the distribution s of the failure capacities between plywood and OSB tests were equal WRS was chosen because of the relatively small sample size of each group The WRS is a nonparametric

PAGE 34

34 test that assumes the two samples (e.g. plywood group and OSB group) are independent of one ano ther and have equal variance. Using a confidence level of 95%, the WRS tested the null hypothesis of no difference between the two sample means. The analysis showed, with a p value = 0.52, that the mean failure pressures for the plywood and OSB panels were not significantly different. This implies that there is a failure to reject the null hypothesis, that the mean failure pressures come from the same distribution. Table 5 1 gives a summary of the results of the plywood and OSB tests a s one group. Figure 5 3 Typical reco rded results (5 th Plywood test). indicates point at which failure occurred. (a) Pressure time history, (b) Time history of the separation between the structural shea thing panel from the stud, d isplacement time h istory of sheathing panel, d isplacement time history of stud. Distribution The Akaike information criterion (AIC) (Sakamoto et al. 1986) was used to rank the relative quantality of different statistical distribution models (e.g. normal, lognormal, Weibull, etc.) for the given data set. This method estimates the quality of the different

PAGE 35

35 models tested and ranks them fr om best to worst based on the goodness to fit statistic and the number of parameters associated with the model. The AIC method was also used to determine the best fit distribution for Moisture Content (MC) and SG was lognormal and normal respectively. Figure 5 4 shows the normalized histogram of the data collected for MC and SG with the Probability Density Function (PDF) that best represents the distribution of the data collected form the experiments. (a) (b) Figure 5 4 Represents the normalized histograms from the Moisture Content and Specific Gravity data collected with the Probability Distribution Function that best fits the data: (a) Lognormal Distribution for MC, (b) Normal Distributio n for SG

PAGE 36

36 6 DISCUSSION When comparing the results of tests performed at the University of Florida versus the tests conducted at the University of Western Ontario of perpendicularly orientated studs with respect to the long edge of the sheathing and studs oriented parallel to the long edge, there is clear relationship of an increase of nails results in an increase in pressure ( Figure 6 1 ). Note: PLY, OSB Current Study A Datin et al. (2011) Performed at the University of F lorida W1D, W1R, W2D, W2R Kopp and Gavanski (2012) Performed at the University of Western Ontario A, PLY, OSB W1D W1R, W2D, W2R Figure 6 1 Relationship of the increase in failure capacity of a test sample with respect to the increase in the number of nails used to attach the sheathing panel the studs.

PAGE 37

37 The out of plane capacities obtained for the structural wood sheathing panels were not in agreement to previous wall capacity studies that also examined structural wood sheathing panels (Kopp and Gavanski 2012) ( Figure 6 2 ) An increase in ultimate capacities found in this study versus the previously mentioned study could be accounted for by the following: The increase in the number of nails. The use of a slightly longer nail to connect the sheathing to the framing member; [60.4 mm (2 3/8 in.) current study vs. 50.8 mm (2 in.) Kopp and Gavanski (2012)]. Differ ent support end conditions: This study used double top plates where the previous study only used a single top plate. The bottom and top plates for this study spanned the 1.2 m (4 ft) direction of the steel chamber and were simply supported at the ends. Thi s study used a monotonic step and hold approach to apply pressure to the structural panels whereas as the previous study used ramp and fluctuating loads. The mean failure capacities for plywood and OSB panels were 4.31 kPa (90.1 psf) with a COV of 0.123 a nd 4.11 kPa (85.9 psf) with a COV of 0.103 respectively. The next task, set out to determine if there was a significant difference between the capacities of the plywood and OSB sheathing types. Using the Wilcoxon Rank Sum analysis of the mean values from e ach of experimental sample groups, it was determined that there was no significant difference between the two sheathing types. This also imp l ies that the two test groups come from the same distribution. Therefore, the mean failure capacity of the entire population was determined to be 4.21 kPa (88.0 psf) with a COV of 0.101. The wall sheathing panels tested used a 152/305 m m (6/12 in.) nailing pattern, with the orientation of the sheathing parallel to the framing members, therefore the wall

PAGE 38

38 sheathing con tained a total of 62 nails. Whereas, r oof sheathing, generally has a 152/305 mm (6/12 in.) nailing patter as well, but it will have nine nails on the outer edges and five nails on the inner members giving it a total of 33 nails per panel. This change in or ientations leads to a 61% increase in nails on the wall sheathing panels. In spite of this relatively large increase in nails, there is only a 17 % percent increase in mean failure capacity for wall sheathing panels. This suggests that the tributary area o f each fastener is more important to the capacity of the panel than the total number of nails. The failure mechanism is controlled by the local behavior of the sheathing around a fastener.

PAGE 39

39 Figure 6 2 Box Plot to compare mean capacity failures.

PAGE 40

40 7 CONCLUSION F or this experiment, two samples types, plywood and OSB sheathing, were tested in order to determine the ultimate capacities of each sheathing type. The pressure is increased monotonically in a step and hold approach. The pressure is stepped in increments of 0.48 kPa (10 psf) and held for 60 s econds and repeated until failure of test specimen. Failure occurs once the panel separates from the wood framing and the blower can no longer maintain a constant air pressure. The failure pressure is taken as the maximum absolute pressure maintained by the system prior to the panel giving way and the system losing pressure. Because of the failure to reject the null hypothesis, it was determined that the ultimate failure capacity is not significantly affected by the sheathing type. However failure type between the plywood and OSB sheathing was noticeably different. OSB test specimens are more prone to pull through failures than plywood test specimens. The most common failure mechani sm of both test samples was nail withdraw in the two center studs. This failure mechanism eventually led to ultimate failure of the panel at one or multiple edges. The ultimate capacities from the plywood and OSB sheathing were tested under a monotonic ste p and hold approach using static wind pressures. 18 of the 20 tests conducted, initial failure (e.g. separation of sheathing from framing member) began with either one or both center two framing members. In two of the test specimens there was a shear failu re connection of the nails between the bottom plate and the studs.

PAGE 41

41 APPENDIX A FULL FAILURE MODE AND LOCATION Figure A 1 Failure mode and location for plywood sample test 1. Test statistics from the plywood group, OSB group, and both groups treated as one group.

PAGE 42

42 Figure A 2 Sample statistics, failure mode, and location for plywood samples 2, 3, and 4.

PAGE 43

43 Figure A 3 Sample statistics, failure mode, and loc ation for plywood samples 5, 6, and 7.

PAGE 44

44 Figure A 4 Sample statistics, failure mode, and location for plywood samples 8, 9, and 10.

PAGE 45

45 Figure A 5 Sample statistics, failure mode, and locat ion for OSB samples 1, 2, and 3.

PAGE 46

46 Figure A 6 Sample statistics, failure mode, and location for OSB samples 4, 5 and 6.

PAGE 47

47 Figure A 7 Sample statistics, failure mode, and location for OS B samples 7, 8, and 9.

PAGE 48

48 Figure A 8 Sample statistics, failure mode, and location for OSB sample 10.

PAGE 49

49 APPENDIX B FULL TEST RESULTS FOR PLYWOOD PANELS Figure B 1 Plywood Test Sample 1 Pr essure and Displacement Time History. Figure B 2 Plywood Test Sample 2 Pressure and Displacement Time History.

PAGE 50

50 Figure B 3 Plywood Test Sample 3 Pressure and Displacement Time History. Figure B 4 Plywood Test Sample 4 Pressure and Displacement Time History.

PAGE 51

51 Figure B 5 Plywood Test Sample 5 Pressure and Displacement Time History. Figure B 6 Plywood Test Sample 6 Pressure and Displacement Time History.

PAGE 52

52 Figure B 7 Plywood Test Sample 7 Pressure and Displacement Time History. Figure B 8 Plywood Test Sample 8 Pressur e and Displacement Time History.

PAGE 53

53 Figure B 9 Plywood Test Sample 9 Pressure and Displacement Time History. Figure B 10 Plywood Test Sample 10 Pressure and Displacement Time History.

PAGE 54

54 A PPENDIX C FULL TEST RESULTS FOR OSB PANELS Figure C 1 OSB Test Sample 1 Pressure and Displacement Time History. Figure C 2 OSB Test Sample 2 Pressure and Displacement Time History.

PAGE 55

55 Figure C 3 OSB Test Sample 3 Pressure and Displacement Time History. Figure C 4 OSB Test Sample 4 Pressure and Displacement Time History.

PAGE 56

56 Figure C 5 OSB Test Sample 5 Pressure and Displacement Time History. Figure C 6 OSB Test Sample 6 Pressure and Displacement Time History.

PAGE 57

57 Figure C 7 OSB Test Sample 7 Pressure and Displacement Ti me History. Figure C 8 OSB Test Sample 8 Pressure and Displacement Time History.

PAGE 58

58 Figure C 9 OSB Test Sample 9 Pressure and Displacement Time History. Figure C 10 OSB Test Sample 10 Pressure and Displacement Time History.

PAGE 59

59 APPENDIX D STIFFNESS OF LINEAR ELASTIC REGION OF SHEATHING AND STUD Figure D 1 Plywood test samples 1 10. Red line coincides with stiffness of s tud and blue line with sheathing panel.

PAGE 60

60 Figure D 2 OSB test samples 1 10. Red line coincides with stiffness of stud and Blue line with sheathing panel

PAGE 61

61 LIST OF REFERENCES ASTM (2014). "Standard Te st Methods for Density and Specific Gravity (Relative Density) of Wood and Wood Based Materials." ASTM International. AWC (2017). "National Design Specifications (NDS) Supplement: Design Values for Wood Construction 2018 Edition." Bhusar, A. A. (2017). "De velopment and Validation of a Tornado Risk Assessment Tool for Residential Structures." Masters Thesis, University of Florida. Datin, P. L., Prevatt, D. O., and Pang, W. (2011). "Wind uplift capacity of residential wood roof sheathing panels retrofitted wi th insulating foam adhesive." Journal of architectural engineering 17(4), 144 154. FBCR (2017). "Florida Building Code 6th Edition (2017) Residential." Henderson, D., Williams, C., Gavanski, E., and Kopp, G. (2013). "Failure mechanisms of roof sheathing u nder fluctuating wind loads." Journal of Wind Engineering and Industrial Aerodynamics 114, 27 37. Hill, K. M. (2009). "Development of time varying wind uplift test protocols for residential wood roof sheathing panels." Masters Thesis, University of Florid a, Gainesville, Florida. Kasal, B., Leichti, R. J., and Itani, R. Y. (1994). "Nonlinear finite element model of complete light frame wood structures." Journal of Structural Engineering 120(1), 100 119. Kemp, G. (2008). "New hurricane test facility allows repeatable full scale building testing." Test 70(4), 10. Kopp, G. A., and Gavanski, E. (2012). "Effects of Pressure Equalization on the Performance of Residential Wall Systems under Extreme Wind Loads." Journal of Structural Engineering 138(4), 526 538. Kopp, G. A., Morrison, M. J., and Henderson, D. J. (2012). "Full scale testing of low rise, residential buildings with realistic wind loads." Journal of Wind Engineering and Industrial Aerodynamics 104, 25 39. Lee, K. H., and Rosowsky, D. V. (2005). "Fragility assessment for roof sheathing failure in high wind regions." Engineering Structures 27(6), 857 868. Li, Y., and van de Lindt, J. W. (2012). "Loss based formulation for multiple hazards with application to residential buildings." Engineering Structures 38(0), 123 133.

PAGE 62

62 Martin, K. G., Gupta, R., Prevatt, D. O., Datin, P. L., and van de Lindt, J. W. (2011). "Modeling System Effects and Structural Load Paths in a Wood Framed Structure." Journal of Architectura l Engineering 17(4), 134 143. Peng, X., Roueche, D. B., Prevatt, D. O., and Gurley, K. R. (2016). "An Engineering Based Approach to Predict Tornado Induced Damage." Multi hazard Approaches to Civil Infrastructure Engineering P. Gardoni, and M. J. LaFave, eds., Springer International Publishing, Cham, 311 335. Prevatt, D. O., Ph.D, PE, Bhusar, A. A., and Roueche, D. B. (2016). "A Predictive Tool for Estimation Tornado Induced Damage to Residential Structures Application to the 2015 Garland/Rowlett, TX Torn ado." 1st International Conference on Natural Hazards & Infrastructure Chania, Greece. Sakamoto, Y., Ishiguro, M., and Kitagawa, G. (1986). "Akaike information criterion statistics." Standohar Alfano, C. D., van de Lindt, J. W., and Ellingwood, B. R. (2017) "Vertical Load Path Failure Risk Analysis of Residential Wood Frame Construction in Tornadoes." Journal of Structural Engineering 04017045. Thampi, H., Dayal, V., and Sarkar, P. P. (2011). "Finite element analysis of interaction of tornados with a low r ise timber building." Journal of Wind Engineering and Industrial Aerodynamics 99(4), 369 377. van de Lindt, J. W., and Dao, T. N. (2009). "Performance based wind engineering for wood frame buildings." Journal of structural engineering 135(2), 169 177.