Effects of Through-Sheathing Installation and Test Method on Predicted Nail Withdrawal Capacity

MISSING IMAGE

Material Information

Title:
Effects of Through-Sheathing Installation and Test Method on Predicted Nail Withdrawal Capacity
Physical Description:
1 online resource (74 p.)
Language:
english
Creator:
Kerr, Ashlie M
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:
PREVATT,DAVID
Committee Co-Chair:
MASTERS,FORREST J
Committee Members:
GURLEY,KURTIS R

Subjects

Subjects / Keywords:
hurricane -- nail -- roof -- sheathing -- withdrawal -- wood
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:
Roof to wall connections are important when preventing damages to light framed wood homes during windstorms. Failures in roof sheathing connections, primarily nails, can cause a chain effect leading to the destruction of the entire structure. Nail withdrawal capacities are tested in accordance to the ASTM D1761; ultimate and design values are predicted using equations in the National Design Standard for Wood Construction. A previous study suggested that installation through the sheathing may cause loss in capacity which is not addressed in the ASTM D1761 method. This research seeks to determine whether through-sheathing installation affects withdrawal capacity. The University of Florida developed field investigations to determine the withdrawal capacity of nails installed in wood roof structures in which the sheathing had to be locally removed which motivated this research. The previous study concluded sheathing removal contributed to nail withdrawal strength reduction. This study investigates whether the loss in strength is caused by driving the nail through the sheathing itself. If this is the case the results will have implications on the method which nail strength is typically tested and for the interpretation of ultimate failure wind speeds based on forensic evidence of roof sheathing failures. Based on the reduction in nail withdrawal strengths when driven through-sheathing it was recommended that a reduction factor between 0.6 and 0.8 should be used with the nail withdrawal capacity results per ASTM D1761 to represent a more accurate prediction of nail withdrawal capacities in a wood-framed roof of an existing residential building
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 Ashlie M Kerr.
Thesis:
Thesis (M.E.)--University of Florida, 2014.
Local:
Adviser: PREVATT,DAVID.
Local:
Co-adviser: MASTERS,FORREST J.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2014
System ID:
UFE0046831:00001


This item is only available as the following downloads:


Full Text

PAGE 1

EFFECTS OF THROUGH SHEATHING INSTALLATION AND TEST METHOD ON PREDICTED NAIL WITHDRAWAL CAPACITY By ASHLIE KERR 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 2014

PAGE 2

2014 Ashlie Kerr

PAGE 3

To my mother and Michael Paul

PAGE 4

4 ACKNOWLEDGMENTS First and foremost I would like to thank my advisor Dr. David O. Prevatt for giving me the opportunity and the resources to perform this research. Without his help and wonderful innovative ideas none of this wo uld have been possible. In addition I would like to thank Dr. Forrest Masters and Dr. Kurtis Gurley for serving as members on my thesis committee. To all my colleagues at the University of Florida, David Roueche, Tuan Vo, Xinlai Peng and Craig Dixon, than ks for answering all my questions and helping me when I was stuck trying to figure out my next course of action. To the undergraduate research assi stants Shelly Dean and Jeandona Doreste for assisting me with my testing. To the laboratory managers Dr. Ch ristopher Ferraro, J. Alex Esposito and Scott Bolton who were more than accommodating when I needed to use the laboratories and also to the needed. Lastly and most importan tly I would like to dedicate this paper to both my mother, Rosemarie Kerr, and my boyfriend, Michael Paul Bro wn. Without encouragement from them both I would have never gotten as far as I did today. They were both there for me when I wanted to give up and helped me pull through when I thought I could never get it done.

PAGE 5

5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURE S ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 National Design Standard (Empirical Equation) ................................ ...................... 13 ASTM D1761 Test Protocol ................................ ................................ .................... 14 Nail Withdrawal for In situ Condition ................................ ................................ ....... 14 Objectives ................................ ................................ ................................ ............... 15 2 LITERATURE REVIE W ................................ ................................ .......................... 17 3 MATERIALS AND METHODS ................................ ................................ ................ 26 Materials and Equipment ................................ ................................ ........................ 26 ASTM Standard Test Methods ................................ ................................ ................ 26 Te st Setup ................................ ................................ ................................ .............. 27 Installation and Withdrawal of Nails ................................ ................................ ........ 27 Test Methods and Test Matrix for 6D Nails ................................ ............................. 30 Test Methods and Test Matrix for 8D Nails ................................ ............................. 32 Effects of the Washer on Withdrawal Capacity ................................ ....................... 35 Time Dependent Testing ................................ ................................ ......................... 36 Test Setup ................................ ................................ ................................ ........ 37 Test Matrix ................................ ................................ ................................ ........ 38 4 RESULTS AND DISCUSSION ................................ ................................ ............... 40 6D N ail Results ................................ ................................ ................................ ....... 40 8D Nail Results ................................ ................................ ................................ ....... 51 Time Dependent Testing ................................ ................................ ......................... 57 5 CONCLUSIONS AND RECOMMENDATIONS ................................ ....................... 70 LIST OF REFERENCES ................................ ................................ ............................... 72 BIOGRAPHICAL SKE TCH ................................ ................................ ............................ 74

PAGE 6

6 LIST OF TABLES Table P age 2 1 Pye Data (1995): Withdrawal Capacities for Specimens Stored Indoor Under Ambient Conditions (lb/in (N/mm)) ................................ ................................ ........ 21 3 1 Number of nails tested per board per method (6D nails) ................................ ........ 32 3 2 Test Matrix for 6D nails with specific gravity and moisture content ........................ 32 3 3 Number of nails testing per board per method (8D Nails) ................................ ...... 35 3 4 Test Matrix for 8 D nails with specific gravity and moisture content ........................ 35 3 5 Summary of Washer Test Data ................................ ................................ .............. 36 3 6 Statistical Analysis Summary of Washer Test ................................ ........................ 36 3 7 Nails tested per method per time period ................................ ................................ 38 3 8 Time Dependent Testing Test Matrix ................................ ................................ ..... 39 4 1 Summary of results for 6D Nails ................................ ................................ ............. 40 4 2 Wilcoxon Test p values and Mean & 5% Non Exceedance Percent Differences (6D Nails) ................................ ................................ ................................ .............. 46 4 3 Combined Data Sets Results for 6D Nails ................................ ............................. 47 4 4 Wilcoxon p valu es and Mean and 5% Non Exceedance Percent Differences for Combined Data Sets for 6D Nails ................................ ................................ ......... 47 4 5 Mean Percent Differences comparing to Shreyans et al. (2012) Data ................... 50 4 6 Summary of results for 8D Nails ................................ ................................ ............. 54 4 7 Wilcoxon Test p values and Mean and 5% Non Exceedance Percent Differences (8D Nails) ................................ ................................ ................................ .............. 56 4 8 Summary of results for Time Dependent Testing for 6D Nails ............................... 60 4 9 Wilcoxon Test p values and Mean and Median Percent Differences (6D ASTM D1761) ................................ ................................ ................................ .................. 61 4 10 Wilcoxon Test p values and Mean and Median Percent Dif ferences (6D OSB w/Washer) ................................ ................................ ................................ ............. 61 4 11 Summary of Results for Time Dependent Testing for 8D Nails ............................ 62

PAGE 7

7 4 12 Wilcoxon Test p values and Percent Differences (8D ASTM D1761) .................. 63 4 13 Wilcoxon Test P values and Percent Differences (8D OSB w/Washer) ............... 63 4 14 Wilcoxon P Values and Percent Difference between Test Methods (ASTM D1761 vs. OSB w/Washer) ................................ ................................ .................. 65 4 15 Wilcoxon P values and Percent Difference between Nail Types (6D vs. 8D) ....... 65

PAGE 8

8 LIST OF FIGURES Figure P age 2 1 Graph comparing Ultimate Load (lb/in) of different types of nails to the Specific Gravity of the wood is was withdrawn from. ................................ .......................... 17 2 2 Static Withdrawal Resistance of Various Fasteners over time. ............................ 19 2 3 Average withdrawal for 6D and 8D common bright nails taken from the Kurtenacker results (Kurtenacker 1965b). ................................ ............................. 19 2 4 UTM set up. A) UTM Machine, B) Close up of nail being withdrawn. ..................... 23 2 5 mPNE set up. A) mPNE, B) Close up of nail being withdrawn after sheathing has been removed. ................................ ................................ ................................ ...... 24 2 6 Boxplot of Shreyans et al. Data. ................................ ................................ ............. 24 3 1 Apparatus used to anchor the 2x4 member to base of UTM for testing. ................ 27 3 2 Installation of nail using nail guide.. ................................ ................................ ........ 28 3 3 Nail installed through sheathing. ................................ ................................ ............ 28 3 4 6D Nail Spacing in 2x4 Member.. ................................ ................................ ........... 29 3 5 8D Nail Spacing in 2x4 Members. ................................ ................................ .......... 29 3 6 Direct Withdrawal Method. J ................................ ................................ .................. 29 3 7 Indirect Withdrawal Method.. ................................ ................................ .................. 30 3 8 Slotted Steel Plate set up. ................................ ................................ ...................... 31 3 9 Sheathing being removed using reciprocating blade.. ................................ ............ 33 3 10 Slotted OSB.. ................................ ................................ ................................ ....... 33 3 11 8D Nail installed through center of washer into sheathing. ................................ .. 34 3 12 Test apparatus to anchor 2x4 to UTM for ASTM D1761 Method for Time dependent testing.. ................................ ................................ .............................. 37 3 13 Test apparatus to anchor 2x4 to UTM for OSB w/washer method for Time dependent testing.. ................................ ................................ .............................. 38 4 1 Histogram of ASTM D1761 Data and Probabilistic Modes (6D Nails) .................... 41

PAGE 9

9 4 2 Histogram of Steel Plate Data and Probabilistic Modes (6D Nails) ........................ 41 4 3 Histogram of OSB Indirect Pull Data and Probabilistic Modes (6D Nails) .............. 42 4 4 His togram of Plywood Indirect Pull Data and Probabilistic Modes (6D Nails) ........ 42 4 5 Histogram of OSB Reciprocating Blade Data and Probabilistic Modes (6D Nails) 43 4 6 Histogram of Plywood Reciprocating Blade Data and Probabilistic Modes (6D Nails) ................................ ................................ ................................ ..................... 43 4 7 Mean Withdrawal Capacities (6D Nails) ................................ ................................ 44 4 8 5% Non Exceedance Capacities (6D Nails) ................................ ........................... 44 4 9 Mean Withdrawal Capacities (Combined Test Methods for 6D Nails) .................... 48 4 10 5% Non Exceedance Withdrawal Capacities (Combined Test Methods for 6D Nails) ................................ ................................ ................................ .................... 48 4 11 Comparison of Data to Shreyans et al. (2012) ................................ ..................... 49 4 12 Mean Withdrawal Capacity vs. Specific Gravity per Board (6D Nails) .................. 51 4 13 Histogram of ASTM D1761 Data Set and Probabilistic Modes (8D Nails) ............ 52 4 14 Histogram of OSB Pull Through Direct Pull Data Set and Probabilist ic Modes (8D Nails) ................................ ................................ ................................ ............. 52 4 15 Histogram of OSB Reciprocating Blade Data Set and Probabilistic Modes (8D Nails) ................................ ................................ ................................ .................... 53 4 16 Histogram of OSB w/Washer Data Set and Probabilistic Mode (8D Nails ............ 53 4 17 Histogram of OSB Pull Through Data Set and Probabilistic Modes (8D Nails) .... 54 4 18 Mean Withdrawal Capacities (8D Nails) ................................ ............................... 55 4 19 5% Non Exceedance Capacities (8D Nails) ................................ ......................... 55 4 20 Mean Withdrawal Capacity vs. Specific Gravity per Board (8 D Nails) .................. 57 4 21 Graph of Theory 1 ................................ ................................ ................................ 58 4 22 Graph of Theory 2 ................................ ................................ ................................ 58 4 23 Possible Outcomes of Theory 3 ................................ ................................ ........... 59 4 24 Boxplot of 6D Nail Withdrawal Method per Method Over Time ............................ 60

PAGE 10

10 4 25 Boxplot of 8D Nail Withdrawal Capacities per Method Over Time ....................... 62 4 26 Peak Withdrawal Capacities as a ratio of the Immediate Withdrawal Over Time (6D and 8D nails) ................................ ................................ ................................ 64 4 27 Percent Differences of Mean between ASTM D1761 and OSB w/Washer Test Methods per nail type ................................ ................................ ........................... 65 4 28 Temperature over time ................................ ................................ ......................... 66 4 29 Humidity over time ................................ ................................ ............................... 67 4 30 Temperature and Humidity over time during the 2 day period in which 2 week tests were performed. ................................ ................................ .......................... 67 4 31 Temperature and Humidity over time during the 2 day period in which the 6 week tests were performed. ................................ ................................ .......................... 68 4 32 Temperature and Humidity over time during the 2 day period in which the 12 week tests were performed ................................ ................................ ........................... 68

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Engineering EFFECTS OF THROUGH SHEATHING INSTALLATION AND TEST METHOD ON PREDICTED NAIL WITH DRAWAL CAPACITY By Ashlie Kerr May 2014 Chair: David O. Prevatt Major: Civil Engineering Roof to wall connections are important when preventing damages to light framed wood homes during windstorms. Failures in roof sheathing connections, primarily nails can cause a chain effect leading to the destruction of the entire structure. Nail withdrawal capacities are tested in accordance to the ASTM D1761; ultimate and design values are predicted using equations in the National Design Standard for Wood Construction. A previous study suggested that installation through the sheathing may ca use loss in capacity which is not a ddressed in the ASTM D1761 method. This research seeks to determine whether through sheathing installation affects withdrawal capacity. The University of Florida developed field investigations to determine the withdrawal capacity of nails installed in wood roof structures in which the sheathing had to be locally removed which motivated this research. The previous study concluded sheathing removal contributed to nail withdrawal strength reduction. This study investigates whether the loss in strength is caused by driving the nail through the sheathing itself. If this is the case the re sults will have implications on the method which nail strength is typically tested and for the

PAGE 12

12 interpretation of ultimate failure wind speed s based on forensic evidence of roof sheathing failures. Based on the reduction in nail withdrawal strengths when driven through sheathing it was recommended that a reduction factor between 0.6 and 0.8 should be used with the nail withdrawal capacity resul ts per ASTM D1761 to represent a more accurate prediction of nail withdrawal capacities in a wood framed roof of an existing residential building

PAGE 13

13 CHAPTER 1 INTRODUCTION The damage caused by windstorms on residential structures are of great concern in Florida, especially the damage in older structures. In many occasions, the failure of the roof occurs first, exposing the house to the elements and making it extremely vulnerable to a pl ethora of other types of damage, including complete destruction of the home. In his 1991 paper, ( Sparks 1991 ) stated that about 60% of all wind storm damages occurred in wood residential buildings. In addition 95% of economi c losses to those structures are due to roof failures ( Baskaran and Dutt 1997 ). Many of these roof failures can be attributed to the sheathing to truss fasteners which are primarily nails National Design Standard (Empi rical Equation) The design strength of nails is provided by the American Forest and Paper Association based upon extensive testing done on nail withdrawal capacity According to the National Design Standard (NDS) equation ( AFPA 2 005 ) that predicts nail withdrawal capacities only the specific gravity of the framing member, and the diameter and length of the nail affect the immediate withdrawal capacity of the nails. Through sheathing installation is not accounted for in this equat ion. The equation for the ultimate capacity is shown below: W = 6900DG 2.5 (W = 47.6DG 2.5 ) ( 1 1 ) Where W is the ultimate withdrawal capacity in lb/in (N/mm) (capacity of nail in lb (N) per inch (mm) that is embedded in the framing member), D is the diameter of the nail in inches (mm) and G is the specific gravity of the framing member. This equation was based on research done by Kurtenacker in 1931 which was later revised and publis hed in 1965 ( Kurtenacker 1965a ) The design values are calculated with a Kw

PAGE 14

14 (shown below) coefficient afte r being multiplied by 1/6 for a factor of safety and then multiplied by 1.2 (10% for a change from permanent to normal loading and 10% for experience) ( AFPA 2005 ) K w = 1.2*6900/6 (K w = 1.2*47.6/6) ( 1 2 ) This then gives the equation for design values as follows: W = K w D G 2. 5 ( 1 3 ) ASTM D1761 Test Protocol The current testing procedure used to determine the peak nail withdrawal of nail out of wood is the ASTM D1761 ( ASTM 2006a ) In this testing th e nail is installed directly into a framing member then withdrawn axially at 0.1 in/min (25.4 mm/min) using a gripping device able to fit around the nail head. This is slightly different from the 0.075 in/min withdrawa l rate used in the Kurtenacker research which the NDS values are based on. Nail Withdrawal for In situ Condition Sutt and Rosowsky ( 2000 ) were one of the first to investigate nail withdrawal in the field. They developed the por table nail extractor (PNE) which was able to withdraw nails from the roofs in field This was the gateway to the Shreyans study ( Shreyans et al. 2012 ). Shreyans investigated both in situ conditions and laboratory testing. The goal was to determine a way to use the laboratory testing results to predict how the nail would react in the field; a modified PNE was used for this testing Through sheathing installation was investigated however only direct withdrawal was performed. The sheathing would first be removed locally around the nail to allow the mPNE to grip the nail. This however added another variable, the method that was used t o remove the

PAGE 15

15 sheathing (reciprocating blade, hole saw or circular saw), which later proved to also have an effect on the withdrawal strength It was concluded that the reciprocating blade had the least effect on withdrawal strength. The ASTM D1761 method was non conservative in its withdrawal capacities predictions and it was recommended a 0.5 to 0.7 factor be applied to the ASTM D1761 results for the situ ation of 6D Nails installed in southern yellow pi ne. Objectives The goal of this research is to address the effect that through sheathing installation has on the peak withdrawal capacity of nails. Though the reason for this is not entirely clear it is important to understand how the nail will behave. It is theorized th at when a nail is being installed through sheathing it is either deformed and/or the force at which it enters the framing member is reduced which then reduces the hold it has in the framing member. If this is true then knowing how the nail will perform bas e d on how it is installed in the field is extremely important when determining the forces it will be able to resist in addition to being able to forensically determine failure loads after windstorm s and other natural disasters. Currently there is no standa rd testing procedure that addresses through sheathing installation. The ASTM D1761 as explained above only tests nails installed directly into framing member which is not a realistic representation of what is going on in the field The objectives of this s tudy are as follows: Determine how much of a reduction through sheathing installation has on nail withdrawal capacities. Investigate withdrawal methods that will best isolate the effects of sheathing. Determine whether these effects are constant over time.

PAGE 16

16 Establish a factor to be applied to the ASTM D1761 results that will account for through sheathing installation. The results of this testing will affect the way the ASTM D176 1 values are used when predicting in field strengths. The test method may need t o be modified to better reflect through sheathing installation effects or have a factor applied to the results that more accurately predict the performance of nails in the field.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW The current equation that is found in the Natio nal Design Standard for Wood Construction ( AFPA 2005 ) is based on a technical note written in 1931 by R.S. Kurtenacker. The technical note was revised in 1958 and then released as a research ce in American Kurtenacker, 1965a ). The paper states that the resistance of the nail is directly related to the density or specific gravity of the wood, the diameter of the nail and the depth of penetration. More specifically the withdrawal capacity increases with all three. The relationship between these variables and the ultimate withdrawal load is shown in Figure 2 1 Figure 2 1 Graph comparing Ultimate Load (lb/in) of different types of nails to the Specific Gravity of the wood is was withdrawn from Source: Kurtenacker, Nail holding power of American woods Forest Products Laboratory, U.S. Forest Service Madison, WI.

PAGE 18

18 After this R.S. Kurtenacker went on to research the effect time had on the withdrawal resistance various types Subjected to Static and Dynamic Wi Kurtenacker 1965b ) Fifteen fastener types were tested, twelve replicates each, from nails to staples, each having different sizes and coatings. T hese nails were installed into dougla s f ir wood through both in. plywood and in white pine. The plywood was used with 2 inch nails and the white pine was used with 2.5 inch nails; the purpose of these was to contro l the embedment depth into the douglas f ir keeping them both at 1.75 inch. Douglas fir w ood according to the NDS has a specific gravity of 0.50 as opposed to s outhern p ine which has a specific gravity of 0.55. Wood was stored indoors at a constant temperature of 73 o F and 50% Humidity except for one set which was stored outdoors. Of the specimens stored indoors tests were performed the same day and then 2, 6.5, 13, 26 and 52 weeks after installation while the specimens stored outdoors were tested only at the 52 week mark. From these tests it was concluded that smooth shank uncoate d fasteners have marked decrease in withdrawal capacity after as little as two weeks except for nylon coated and helically threaded fasteners. Figure 2 2 summarizes the results and Figure 2 3 shows the result for the 6D and 8D common bright nails (graph generated based on the Kurtenacker data for this paper for clarity) As it can be observed within the first two weeks there is as much as a 30% to 70% d ecrease within the first two weeks and more specifically about a 62% decrease for both 6D and 8D common bright nails.

PAGE 19

19 Figure 2 2 Static Withdrawal Resistance of Various Fasteners over time. Source: Kurtenacker, R. S. (1965b). "Performance of Container Fasteners Subjected to Static and Dynamic Withdrawal ." Forest Products Laboratory, USDA Forest Service, Madison, WI. Figure 2 3 Average withdrawal for 6D and 8D common bright nails taken from the Kurtenacker results ( Kurtenacker 1965b )

PAGE 20

20 Today the NDS bases their equation for nail withdrawal strength on this research. The equation to predict the design strength of a nails is as follows: W = K w G 1.5 D ( 2 1 ) where: W = nail or spike design value per inch of penetration in main member, (lb (N)) K w = 1380 (9.52) G = specific gravity of main member based on ovendry weight an d volume, where 0.31< G <0.73 D = shank diameter of the nail or spike (in (mm)), where 0.0999 (2.54) (< D <0.375 (9.53) The equation for K w is as follows: K w = 1.2(6900/6) (K w = 1.2(47.6/6)) ( 2 2 ) The 20% increase was introduced as a part of the World War II emergency increase in wood design values, 10% for change from permanent to normal loading and 10% for experience. Therefore when this and the factor of safety of 6 are taken out the ultimat e load ends up being as follows: W = 6900G 1.5 D (W = 47.6G 1.5 D) ( 2 3 ) After this several studies have tried to isolate each variable that may affect the decrease in withdrawal capacity of nails both in the field and in a laboratory setting ( Herzog and Yeh 2006 ; Pye 1995 ; Shreyans et al. 2012 ; Sutt and Rosowsky 200 0 ; Swane and Vagholkar 1968 ) These included heat, humidity, cyclic loading, racking, time and other environmental factors for in field factors and withdrawal rate for laboratory factors.

PAGE 21

21 In 1995 Pye researched the eff ects of In Service conditions ( Pye 1995 ) These included racking, humidity, heat and a combination of heat and humidity. Various nail coatings and nail shanks were used and the method of driving the nails and type of wood were al so explored. Pye concluded with the following: Power driven fasteners proved to have higher capacities than hand driven fasteners. Ring Shank fasteners had higher withdrawal capacities Heat series had the most dramatic decreases. The heat and humidity se ries also explored withdrawal strength at different times after installation. Under this series there were three subsets; the specimens were stored in a small outdoor shed meant to simulate roof attic conditions, the specimens were stored indoor under cont rolled ambient condition (70 o F and 55% Humidity) and lastly the specimens were rotated weekly between each environment. The ambient conditions have values at 1 week, 3 week and 10 week. Summary of this data can be found in Table 2 1 Table 2 1 Pye Data ( 1995 ): Withdrawal Capacities for Specimens Stored Indoor Under Ambient Conditions (lb/in (N/mm)) Time after Installation Nail Type Wood 1 week 3 week 10 Week 1HSF SYP 169 (29.6) 270 (47.3) 257 (45.0) 1HSF SPF 243 (42.6) 292 (51.1) 2HRF SYP 342 (59.9) 360 (63.0) 388 (67.9) 4PSF SYP 263 (46.1) 289 (50.6) 293(51.3) As shown above for all series in stored in Indoor ambient temperature the nail withdrawal capacity increases over time which is not was has been observed in previous studies. The reason for this was however not the focus of this paper and was not addressed

PAGE 22

22 In the end he determined that the design values were too conservative for ring shank nails, however they overestimated capacity of smooth shank fasteners subjected to heat and the factor used when smooth shank fasteners were subject to change in moistur e content was too conservative Up until this point however there has not been any testing of nails in the field, all testing were laboratory setups trying to imitate in field conditions. Sutt and Rosowsky ( 2000 ) published a paper whic h focused on the effect of in situ conditions on nails withdrawal capacities. For this research he developed what was called a fastener extractor which then evolved into what is now known as a Por table Nail Extractor (PNE). It was a device with a self cont ained load cell that was able to withdraw fasteners in the field. The load was applied manually with levers on either side. One of the concerns of this device was the variability in withdrawal rate between operators, however Sutt stated this was shown to n ot have and effects on 8D nails based on previous studies. Sutt investigated the nails on a home that was constructed between 1972 and 1983 that was subject to a fire and had a renovated area that was constructed in 1996.The testing was performed in 1999. A holesaw was used to remove the sheathing form around the nail in the field. Sutt concluded that there is a reduction in capacity of a nail subjected to in situ conditions therefore design values may be non conservative. In addition the test values were lower than the design values used 1.0 and 1.6 for the load duration factors. In Shreyans et al. (2012) reported a study evaluating the in situ and laboratory withdrawal strength capacities of typical nails used to install plywood and OSB sheathing in roofs of existing Florida homes. In the field sheathing was removed locally

PAGE 23

23 with various methods and withdrawn with a modifi Por table Nail Extractor (mPNE) In addition nails were withdrawn using the ASTM D1761 method using a Universal Testing Machine ( Figure 2 4 ) and compared to installing nails through sheathing in the lab and removing them similarly to those in the field with the mPNE ( Figure 2 5 ) so as to isolate the through sheathing effects on the nails and compare them to the stand ard test method. The results of the laboratory studies are shown in Figure 2 6 A B Figure 2 4 UTM set up. A) UTM Machine B) Close up of nail being withdrawn Source: Shreyans, S., Kerr, A., Prevatt, D. O., and Gurley, K. R. (2012). "In Situ Nail Withdrawal Strengths in Wood Residential Roofs." ATC&SEI Advances in Hurricane Engineering, Miami, FL.

PAGE 24

24 A B Figure 2 5 mPNE set up A) mPNE B) Close up of nail being withdrawn after sheathing has been removed Source: Shreyans, S., Kerr, A., Prevatt, D. O., and Gurley, K. R. (2012). "In Situ Nail Withdrawal Strengths in Wood Residential Roofs. ATC&SEI Advances in Hurricane Engineering, Miami, FL. Figure 2 6 Boxplot of Shreyans et al. Data. Source: Shreyans, S., Kerr, A., Prevatt D. O., and Gurley, K. R. (2012). "In Situ Nail Withdrawal Strengths in Wood Residential Roofs." ATC&SEI Advances in Hurricane Engineering, Miami, FL.

PAGE 25

25 It was concluded that through sheathing installation had an effect on nails withdrawal capacities. The reciprocating blade was determined to be more sui ted for removal of local sheathing. It was recommend that a suggested reduction factor between 0.5 and 0.7 be applied to the ASTM D1761 result to account for installation through sheathing and the results we re non conservative and an improved test method was needed for predicting withdrawal strengths in wood roofs.

PAGE 26

26 CHAPTER 3 MATERIALS AND METHODS Materials and Equipment Two types of Nails were tested: 6D smooth shank, hot dipped galvanized nails (Lengt h 2 in (50.8mm) and diameter 0.113 in (2.87 mm)) and 8D smooth shank, hot dipped galvanized nails ( l ength 2.5 in (63.5 mm) and diameter 0.131 in (3.33 mm). The 6D nails were tested in summer 2012 and the 8D nails were test ed in Spring/Summer of 2013. Nail s were installed using a Stanley Bostitch F21PL Nail Gun. Framing members used were No. 2 Southern Yellow Pine nominal 2 in x 4 in, 8 ft long (38 mm x 89 mm x 2438 mm) and for through sheathing tests both 7/16 in (1.11 mm) oriented strand board and 15/32 i n (11.91 mm) plywood were used for 6D nails and 1 in (2.54 mm) OSB for the 8D nails. Test s were performed on a 33,721 lbf (150 kN) Instron Universal Testing Machine (UTM), Model Number 3384. ASTM Standard Test Methods The control testing method followed t Methods for Mechanical Fasteners in ASTM 2006a ), which states that the fastener should be withdrawn with a gripping device able to fit the base of the fastener head while positioned to a llow true axial loading. In addition it states that nails should be withdrawn at a constant rate of 0.1 in/min (2.54 mm/min). This withdrawal rate was used for all test methods. In addition for each board the specific gravity and moisture content was Gravity of Woods and Wood Based Materials ( ASTM 2006b ) and the ASTM 44 42

PAGE 27

27 Wood Based ASTM 2006c ). The Specific Gravity and Moisture Content Tests were done on five evenly spaced 1 inch segments from the 2x4 s, the first and the last being 6 inches from the edge, for the 6D Nails and nine 1 inch segments from the 2x4s used in the 8D nail testing. Test Setup Tests were set up on the UTM using the ASTM D1761 as a guideline. The 2x4 member was held to the place w ith a fabricated steel apparatus pictured ( Figure 3 1 ). Nails were then withdrawn with a head attachment connected to a load cell following one of the below Figure 3 1 Apparatus used to anchor the 2x4 member to base of UTM for testing July 2012. Courtesy of Ashlie Kerr. Installation and Withdrawal of Nails Na ils were installed two ways for the 6D Nails. First was using a inch (12.7 mm) nail guide which was used to give a consistent embedment depth of 1.5 in (38.1 mm) ( Figure 3 2 ). This depth was chosen because it is the typical thickness of roof sheathing. The second installation method was through a 3.5 in (88.9 mm) by 5 in (127 mm) strips of OSB or Plywood depending on test method ( Figure 3 3 ).

PAGE 28

28 Figure 3 2 Installation of nail using nail guide July 2012. Courtesy of Shelly Dean Figure 3 3 Nail installed through sheathing July 2012. Courtesy of Shelly Dean. For the 6D the nails were installed 2 in (50.8 mm) o.c. (on center) on both narrow faces of the 2x4 with first and the last nail being at least 4 in (101.6 mm) from the edge ( Figure 3 4 ).For 8D the nails were also installed 2 in (50.8 mm) o.c. however on one narrow face the first nail was installed 4 in (101.6 mm) from the edge and the opposite face was offset by 1 in (25.4 mm) s o the first nail was installed 5 in (127.0 mm) from the edge ( Figure 3 5 ) Once installed nails were withdrawn either directly by using the nail head ap paratus shown in Figure 3 6 or the indirect method using the C Channel in Figure 3 7.

PAGE 29

29 A B Figure 3 4 6D Nail Spacing in 2x4 Member A) Isometric View, B) Side View. A B Figure 3 5 8D Nail Spacing in 2x4 Members A) Isometric View, B) Side View Figure 3 6 Direct Withdrawal Method July 2012. Courtesy of Shelly Dean.

PAGE 30

30 A B Figure 3 7 Indirect Withdrawal Method. A) Wit hdrawal setup in with C Channel, B) Withdrawal with Styrofoam cube placed in C Channel to prevent bending of sheathing strip. July 2012. Courtesy of Shelly Dean. Test Methods and Test Matrix for 6D Nails Six different Test Methods were explored with the 6D nails each using a different combination of the above installation and withdrawal methods. Each of these methods is detailed below. a) AST M D1761 1. Nail is installed using the nail guide from Figure 3 3 2. Nail is then withdrawn directly from the 2x4 wood member using the Direct Withdrawal M ethod ( Figure 3 6 ) b) Steel Plate 1. Nail is installed using the nail guide from Figure 3 3 2. A slotted Steel plate ( Figure 3 8 ) is then slipped around the nail and withdrawn u sing the C Channel in the Indirect Withdrawal Method.

PAGE 31

31 Figure 3 8 Slotted Steel Plate set up. July 2012. Courtesy of Shelly Dean. c) OSB/Indirect Withdrawal 1. Nail is installed directly through sheathing as shown in Figure 3 3 2. Nail is then withdrawn using the indirect withdrawal with Styrofoam cube in Figure 3 7 B. d) Plywood/Indirect Withdrawal 1. Same as above but with Plywood sheathing instead of OSB e) OSB/Reciprocating Blade 1. Nail is installed directly through the OSB as shown in Figure 3 3 2. Sheathing is removed from around the nail with a reciprocating blade as shown in Figure 3 9 3. Nail is then withdrawn using the direct withdrawal method shown in Figure 3 6 f) Plywood/Reciprocating Blade 1. Same as above but with Plywood sheathing instead of OSB A breakdown of the number of nails tested is given in Table Table 3 1 and the moisture content and specific gravity per board in Table 3 2

PAGE 32

32 Table 3 1 Number of nails tested per board per method (6D nails) Board ASTM D1761 Steel Plate OSB/ Indirect Plywood/ Indirect OSB/Recip. Blade Plywood/ Recip. Blade AA 24 19 19 AB 48 20 18 AC 37 20 20 AD 19 18 15 20 AE 19 18 20 18 AF 18 19 20 20 Table 3 2 Test Matrix for 6D nails with specific gravity and moisture content Board Specific Gravity Moisture Content Test Method a # of Nails per Test Method AA 0.47 14.0% a/b/c 24/19/19 AB 0.48 14.4% a/b/d 48/20/18 AC 0.44 14.7% a/c/d 37/20/20 AD 0.57 13.6% a/b/c/e 19/18/15/20 AE 0.43 13.3% c/d/e/f 19/18/20/18 AF 0.50 14.5% c/d/e/f 18/19/20/20 Average 0.48 14.1% a Test method letter designations are as follows: (a) ASTM D1761 (b) Steel Plate (c) OSB/ Indirect Pull (d) Plywood/ Indirect Pull (e) OSB/ Recip. B l ade (f) Plywood/ Recip. Blade Test Methods and Test Matrix for 8D Nails Initially four different Test Methods were explored with the 8D nails each using a different combination of the above installation and withdrawal methods. Each of these methods is detailed below. a) ASTM D1761 1. Nail is installed using the nail guide from Figure 3 2 2. Nail is then withdrawn directly from the 2x4 wood member using the Direct Withdrawal Method ( Figure 3 6 ) b) OSB/Indirect Withdrawal 1. Nail is installed directly through sheathing as shown in Figure 3 3 2. Nail is then withdrawn using the indirect withdrawal with Styrofoam cube in Figure 3 7 B.

PAGE 33

33 c) OSB/Reciprocating Blade 1. Nail is installed directly through the OSB as shown in Figure 3 3 2. Sheathing is removed from around the nail with a reci procating blade as shown in Figure 3 9 3. Nail is then withdrawn using the direct withdrawal method shown in Figure 3 6 A B Figure 3 9 Sheathing being removed using reciprocating blade. A) Reciprocating Blade in use, B) After Sheathing is removed. July 2012. Courtesy of Shelly Dean. d) Slotted OSB 1. Nail was withdrawn using the nail guide from Figure 3 2 2. A slotted OSB ( Figure 3 10 ) similar to the Steel plate from the 6D nails testing was slipped around the nail and withdrawn using the indirect withdrawal method. Figure 3 10 Slotted OSB April 2013. Courtesy of Ashlie Kerr.

PAGE 34

34 Once the testing for the above testing commenced it was observed that all the slotted OSB testing and a majority of the OSB/Indirect Withdrawal had pull through failures. Pull through occurs when instead of the nail being withdrawn out of the 2x4 member it instead stays an chored in the 2x4 member and rips through the OSB sheathing. The peak load when this occurred was recorded and will be referred to as member using the direct pull method ( Figure 3 6 ); these results will be referred to as modified test metho d was necessary to measure the nail withdrawal capacity. This involved installing a inch diameter and 0.07 in. thick steel washer between the nail head and top surface of sheathing to minimize likelihood of pull though failure which is detailed below. e) O SB/washer 1. Nail installed through sheathing with the nail going through the center of a Figure 3 11 ) 2. Nail withdrawn using the indirect withdrawal method. A breakdown of the number of nails tested is given in Table 3 3 and specific gravity and moisture cont ent per board is given in Table 3 4 Figure 3 11 8D Nail installed through center of washer into sheathing. May 2013. Courtesy of Ashlie Kerr

PAGE 35

35 Table 3 3 Number of nails testing per board per method (8D Nails) Board ASTM D1761 OSB/Indirect Withdrawal (Pull Out/Pull Through) OSB/ Reciprocating Blade Slotted OSB (Pull Out/ Pull Through) OSB Washer BA 26 29 (8/21) 29 BB 21 22 29 (3/26) BC 44 43 BD 31 30 (4/26) 26 BE 27 26 30 BF 44 21 21 BG 25 33 25 BH 26 31 26 BI 23 32 30 BJ 26 33 27 Table 3 4 Test Matrix for 8D nails with specific gravity and moisture content Board Specific Gravity Moisture Content Test Method b # of Nails per Test Method OSB Pull Out/ OSB Pull Through a BA 0.51 11.8% a/b/c 26/29/29 8/21 BB 0.53 12.5% a/b/d 21/22/29 3/26 BC 0.42 12.1% a/c 44/43 BD 0.50 11.5% a/b/c 31/30/26 4/26 BE 0.47 10.8% a/c/e 27/26/30 BF 0.54 11.7% a/c/e 44/21/21 BG 0.45 11.6% a/c/e 25/33/25 BH 0.46 11.8% a/c/e 26/31/26 BI 0.54 11.6% a/c/e 23/32/30 BJ 0.51 11.3% a/c/e 26/33/27 Average 0.49 11.7% a For the OSB method when the nail did not pull out the 2x4 it is said to be a pull through and was then pulled out using the direct pull method. For the slotted OSB all nails were pull through. b Test Method letter designations are as follows: (a). ASTM D1761 (b). OSB/Indirect Pull (c). OSB/Recip. Blade (d). Slotted OSB (e). OSB w washer Effects of the Washer on Withdrawal Capacity When it was decide that a washer would be needed to test with the 8D nails to prevent pull through there was a question of whether or not the washer size would have an effect on the withdrawal capacity; a test matrix was deve loped to determine this. Both the 6D nails and 8D nails were tested because for consistency in the time

PAGE 36

36 dependent testing both the 6D and 8D nails used washers despite the fact that pull through was not an issue with 6D nails. Table 3 5 shows a brief summary of the results for the 6D Nails. Table 3 5 Summary of Washer Test Data Washer Diameter 6D Nails None 0.5 in (12.7 mm) 0.75 in (19.05 mm) Sample Size 24 29 27 Mean of Withdrawal Capacities lb/in (N/mm) 138 (2 4.2 ) 131 (22. 0 ) 124 (21.8 ) Standard Deviation lb/in (N/mm) 33 (5.8 ) 39 (6. 9 ) 23 (4.1 ) CoV 24.2% 28.9% 19.0% A Wilcoxon Test was performed comparing the data and based on a 5% confidence level we can reject the null hypothesis which assumes that the sample sets come from different populations. The p values are shown in Table 3 6 and based on the previous statement we can conclude that the three test method results can be said to come from the same population. For the 8D nails a majority of the nails had pull t hrough failure with the 0.5 in (12.7mm) diameter washer. This was another reason to stick with the 0.75 in (19.05 mm) washer in addition to the results from the 6D nails. Table 3 6 Statistical Analysis Summary of Washer Test Washer Sizes Compared P value Percent Difference No Washer and 0.5 in (12.7 mm) 0.48 5.2% 0.5 in (12.7 mm) and 0.75 (19.05 mm) 0.48 5.6% No Washer and 0.75 in (19.05 mm) 0.0 6 10.8% Time Dependent Testing The last set of testing was done to compare the withdrawal strength of nails over time. It has already been determined that over time nails lose withdrawal capacity over time as determined in (Kurtenacker 1965b) study nail withdr awal strength is decreased

PAGE 37

37 over time with as much as a 70% decrease over time after two weeks The objective of the time dependent testing was to determine whether through sheathing installation has the same behavior over time. For this testing both 6D an d 8D nails were tested These nails came from the same batch as the ones used in the tests described in the previous sections. From our results it was determined that the ASTM D1761 and the OSB with washer methods would be the ones used for this testing. T est Setup The apparatus for the test setup had to be changed to accommodate all the nails being installed in the 2x4 at one time. The spacing of the nails had to be adjusted as well to accommodate the change in apparatus. The nails tested per the ASTM D176 1 method were spaced 3 in (76.2 mm) apart o.c. and the nails tested per the OSB with washer method were spaced 7 in (177.8 mm) apart o.c. to leave space for the apparatus. Figure 3 1 2 and Figure 3 13 show both these setups. The method of withdrawing the nails remained the same. A B Figure 3 1 2 Test apparatus to anchor 2x4 to UTM for ASTM D1761 Method for Time dependent testing A) Top View, B) Angled Side View. September 2013. Courtesy of Ashlie Kerr.

PAGE 38

38 A B Figure 3 1 3 Test apparatus to anchor 2x4 to UTM for OSB w/washer method for Time d ependent testing A) Front View, B) Angled Side View September 2013. Courtesy of Ashlie Kerr. Test Matrix For the time dependent testing 30 nails were tested per method per time period. The nails were tested immediately, 2 weeks, 6 weeks and will be teste d 12 weeks after installation. Table 3 7 and Table 3 8 show the number of nails tested per test method per time period and the complete text matrix respectively. Table 3 7 Nails tested per method per time period 6D Nails 8D Nails Time after Installation ASTMD1761 OSB with Washer ASTM D1761 OSB with Washer Immediately 30 30 30 30 120 2 weeks 30 30 30 30 120 6 Weeks 30 30 30 30 120 12 weeks 30 30 30 30 120

PAGE 39

39 Table 3 8 Time Dependent Testing Test Matrix Board Specific Gravity Moisture Content Withdrawal Time Nail Type Test Methods # of Nails CA 0.46 11.5% Immediate 6D Both 30 per method CB 0.48 11.4% Immediate 8D Both 30 per method CC 0.40 10.3% 2 week 6D ASTM 30 CD 0.49 11.3% 2 week 6D OSB 12 CE 0.47 11.6% 2 week 6D OSB 12 CF 0.48 11.6% 2 week Both OSB 6 per type CG 0.37 13.4% 2 week 8D OSB 12 CH 0.42 12.0% 2 week 8D OSB 12 CI 0.47 12.1% 2 week 8D ASTM 30 CJ 0.45 12.9% 6 week 6D ASTM 30 CK 0.42 12.5% 6 week 6D OSB 12 CL 0.53 13.1% 6 week 6D OSB 12 CM 0.56 12.0% 6 week Both OSB 6 per nail type CN 0.57 11.6% 6 week 8D OSB 12 CO 0.47 10.9% 6 week 8D OSB 12 CP 0.46 10.9% 6 week 8D ASTM 30 CQ 0.49 9.6 % 12 week 6D ASTM 30 CR 0.50 11.4% 12 week 6D OSB 12 CS 0.46 11.8% 12 week 6D OSB 12 CT 0.56 11.9% 12 week Both OSB 6 per nail type CU 0.44 11.4% 12 week 8D OSB 12 CV 0.57 12.3% 12 week 8D OSB 12 CW 0.39 10.9% 12 week 8D ASTM 30 Average 0.47 11.6%

PAGE 40

40 CHAPTER 4 RESULTS AND DISCUSSION 6D Nail Results For each test the maximum withdrawal capacity was recorded. From there depending on how much of the nail was embedded in the framing member that peak force was divide d by the embedment length to give results in load per unit length as it does in the NDS. A Summary of the results is shown in Table 4 1 Table 4 1 Summary of results for 6D Nails Test Method ASTM D1761 Steel Plate OSB Indirect Pull Plywood Indirect Pull OSB Recip. Blade Plywood Recip. Blade Sample size 134 59 97 76 60 40 Mean a 141 (24.7) 147 (25.7) 124 (21.8) 121 (21.2) 94 (16.4) 83 (14.6) Median a 140 (24.5 ) 146 (25.6) 125 (21.8) 123 (21.5) 9 1 (16.0) 79 (13.8) Std. Dev a 34 (5.9) 39 (6.8) 24 (4.2) 26 (4.62) 29 (5.0) 25 (4.40 5% Non Exceedance a 91 ( 16.0 ) 85 (14.8 ) 8 5 (14.8 ) 78 (13.6 ) 54 (9.4 ) 52 (9.1 ) Minimum a 60 (10.4) 79(13.8) 72 (12.7) 67 (11.7 ) 50 (8.8) 47 (8.3) Maximum a 314 (55.0) 263 (46.1) 181 (31.8) 191 (33.5) 149 (26.1) 169 (29.7) CoV (%) 23.9 26.4 19.4 21.8 30.4 30.2 Mean SG per method b 0.49 0.51 0.48 0.47 0.50 0.47 Boards averaged for SG b AA, AB, AC, AD AA, AB, AD AA, AC, AD, AE, AF AB, AC, AE, AF AD, AE, AF AE, AF a Load per length of nail shank in lb/in (N/mm) b The Specific Gravity is a weighted average of the boards that the methods were tested on (See Table 3 2 ) Histograms are plotted of the data sets to determine if there was a common distribution among them in addition to determining the best way to analyze the data ( Figure 4 1 Figure 4 2 Figure 4 2 Figure 4 4 Figu re 4 3 Figure 4 5 Figure 4 6 ) with common probabilistic modes layered over them In light of the fact that we are dealing with failure loads and design loads are of concern more interest was placed on the

PAGE 41

41 lower values. Therefor e a 5% exclusion values was determined for each data set and plotted as bar graphs ( Figure 4 8 ) as well as the means ( Figure 4 7 ) Figure 4 1 Hist ogram of ASTM D1761 Data and Probabilistic Modes (6D Nails) Figure 4 2 Histogram of Steel Plate Data and Probabilistic Modes (6D Nails)

PAGE 42

42 Figure 4 3 Histogram of OSB Indirect Pull Data and Probabilistic Modes (6D Nails) Figure 4 4 Histogram of Plywood Indirect Pull Data and Probabilistic Modes (6D Nails)

PAGE 43

43 Figure 4 5 Histogram of OSB Reciprocating Blade Data and Probabilistic Modes (6D Nails) Figure 4 6 Histogram of Plywood Reciprocating Blade Data and Probabilistic Modes (6D Nails)

PAGE 44

44 Figure 4 7 Mean Withdrawal Capacities (6D Nails) Figure 4 8 5% Non Exceedance Capacities (6D Nails)

PAGE 45

45 Wilcoxon test was determined to be the most suited because unlike many similar statistical analysis tests it does not assume normally distributed dat a I n general assumes that the data is non parametric (has no distributive characteristics) and can be used to compare data sets of unequal sample sizes ( Wilcoxon, 1945 ). This comparison was done using the statistical softw are R Studio (Version 0.97.449 2009 2012 RStudio, Inc.) which gives a p values. The p value represents the probability that th e null hypothesis can be rejected The null hypothesis in this case being that the data sets compared can be said to come from t wo different populations. For the purposes of this study we reject the hypothesis at a p value above 0.05. Hence if the p value is less than 0.05 we can say the data sets come from separate population and if the p value is above 0.05 we can say the data se ts can be said to come from the same population. The p values are reported in Table 4 2 along with the percent difference between the mean and 5% non exceedance of those data sets.

PAGE 46

46 Table 4 2 Wilcoxon Test p values and Mean & 5% Non Exceedance Percent Differences (6D Nails) T est Method ASTM D1761 Steel Plate OSB Indirect Pull Plywood Indirect Pull OSB Recip.Blade Plywood Recip. Blade ASTM D1761 1 0.24 (3.9%) (7.6%) 3.2E 5 (12.7%) (8.5%) 5.6E 6 (15.4%) (19.2%) 5.6E 16 (40.4%) (50.7%) 7.1E 16 (49.8%) (58.9%) Steel Plate 1 5.2E 5 (16.6%) (0.9%) 2.3E 5 (19.3%) (11.7%) 5.5E 12 (44.2%) (43.6%) 1.8E 12 (55.2%) (51.9%) OSB/ Indirect Pull 1 0.38 (2.8%) (10.8%) 1.7E 9 (28.1%) (42.7%) 1.2E 12 (39.5%) (8.5%) Plywood/ Indirect Pull 1 3.3E 7 (23.4%) (32.3) 5.0E 10 (36.8%) (40.9%) OSB/ Recip. Blade 1 0.085 (11.7%) (8.8%) Plywood/ Recip. Blade 1 According to the p values the 6 data sets above can be combined and reduced to 3 data sets: 1. ASTM D1761 and Steel Plate 2. OSB Indirect Pull and Plywood Indirect Pull 3. OSB Reciprocating Blade and Plywood Reciprocating Blade. Based on this the following can be said: The Steel Plate Test Method has no effect on the withdrawal strength of the nails. Installing through sheathing reduces the withdrawal strength of the nails from as little as 13 % (ASTM D1761 and OSB Indir ect Pull) and to as much as 55 % (Steel Plate and Plywood Reciprocating Blade) when comparing means. The method used to withdraw the nails once the sheathing is installed (Indirect Pull vs. Reciprocating Blade ) has an effect on the withdrawal strength. The type of sheathing (OSB vs. Plywood) has no effect on withdrawal strength.

PAGE 47

47 Table 4 3 show the new resul ts once the data sets are combined and Table 4 4 shows the p values and percent differences between these new combined sets to further iterate the differences. In addition to the bar graphs comparing the mean and 5% non exceedance values ( Figure 4 9 and Figure 4 10) Table 4 3 Combined Data Sets Results for 6D Nails Test Method AS TM D1761 & Steel Plate OSB & Plywood Indirect Pull OSB & Plywood Recip.Blade Sample size 193 193 100 Mean a 143 (25.0) 123 (21.5) 90 (15.7) Median a 141 (24.7) 124 (21.7) 84 (14.7) 5% Non Exceedance a 87 (15.2) 78 (13.6) 53 (9.2) Std. Dev a 35 (6.2) 25 (4.4) 28 (4.8) Minimum a 60 (10.4) 67 (11.7) 47 (8.3) Maximum a 314 (55) 191 (33) 169 (30) CoV (%) 24.7 20.5 30.8 Mean SG per method b 0.49 0.48 0.50 Boards Averaged for SG b AA, AB, AC, AD AA, AB, AC, AD, AE, AF AD, AE, AF a Load per length of nail shank in lb/in (N/mm) b The Specific Gravity is a weighted average of the boards that the methods were tested on ( Table Table 3 2 ) Table 4 4. Wilcoxon p values and Mean and 5% Non Exceedance Percent Differences for Combined Data Sets for 6D Nails Test Method ASTM D1761 & Steel Plate OSB & Plywood Indirect Pull OSB & Plywood Recip. Blade ASTM D1761 & Steel Plate 1 2.0E 9 (15.1%) (10.7%) <2.2E 16 (45.9%) (48.6%) OSB & Plywood Indirect Pull 1 <2.2E 16 (31.3%) (38.4%) OSB & Plywood Recip. Blade 1

PAGE 48

48 Figure 4 9 Mean Withdrawal Capacities (Combined Test Methods for 6D Nails) Figure 4 10 5% Non Exceedance Withdrawal Capacities (Combined Test Methods for 6D Nails)

PAGE 49

49 As seen above there is a larger difference between the ASTM D1761 Test Results and the Reciprocating Blade results than between the ASTM and the Indirect Pull Results. It is speculated that the vi brations form the reciprocating blade as the sheathing is being removed may affect the nail thus being the reason for the reduction in the withdrawal capacity. The theory was suggested in a previous study as the same phenomena occurred in ( Shreyans et al 2012 ) The 6D nail data was compared to Shreyans data ( Figure 4 11 ). The only similarities found was when comparing the mPNE nail guide from Shreyans test results to the current ASTM D1761 test method which had a p value of 0.9 (Table 4 5) Figure 4 11 Co mparison of Data to Shreyans et al. (2012)

PAGE 50

50 Table 4 5 Mean Percent Differences comparing to Shreyans et al (2012) Data Mean Percent Difference ASTM D1761 (Shreyans) and ASTM D1761/Steel Plate 33.7% mPNE/Nail Guide and ASTM D1761/Steel Plate 1.3% mPNE/Recip. Blade and OSB/Plywood Recip. Blade 33.8% The last thing that was investigated was the effects of specific gravity on the withdrawal results. As mentioned in the Methods and Materials section above five 1 inch (25.4 mm) sections were cut from each board to determine specific gravity and moisture c ontent. This was averaged to determine the specific gravity of each board and these values were plotted against the mean peak withdrawal capacity for each method on that board. These results are show Figure 4 12 along with the ultimate withdrawal strength calculated according to the NDS u lt imate withdrawal e quation ( Equation 1 1 ) Based on the graph above there is no observable pattern between the peak nail withdrawal capacities and the specific gravity. In addition the majority of the mean peak nail withdrawal capacities for the results through sheathing are below the expected ultimate withdrawal capacity.

PAGE 51

51 Figure 4 1 2 Mean Withdrawal Capacity vs. Specific Gravity per Board (6D Nails) 8D Nail Results Histograms with probabilistic modes were plotted just as with the 6D nails ( Figure 4 13 Figure 4 1 4 Figure 4 1 5 Figure 4 1 6 and Figure 4 17 ) The theory that installing nails through sheathing r educes withdr awal capacity is further reinforced in the 8D nail withdrawal results ( Table 4 6 Figure 4 18 Figure 4 19 ). However unlike the 6D nails the 8D nails do not seem to be affected by the use of the reciprocating blade t o remove the local sheathing. In fact the reciprocating blade results are higher than the washer method, which is comparable to the 6D nails indirect pull method, however it is only a 5.6% difference in the means and are the same for the 5% non exceedance values In addition when pull th rough occurred and the nails were then directly withdrawn from the

PAGE 52

52 framing member there was only a 4.78% difference in means between these results and the ASTM D1761. Figure 4 1 3 Histogram of ASTM D1761 Data Set and Probabilistic Modes (8D Nails) Figure 4 1 4 Histogram of OSB Pull Through Dire ct Pull Data Set and Probabilistic Modes (8D Nails)

PAGE 53

53 Figure 4 1 5 Histogram of OSB Reciprocating Blade Data Set and Probabilistic Modes (8D Nails) Figure 4 1 6 Histogram of OSB w/Washer Data Set and Probabilistic Mode (8D Nails

PAGE 54

54 Figure 4 17 Histogram of OSB Pull Through Data Set and Probabilistic Modes (8D Nails) Table 4 6 Summary of results for 8D Nails Test Method ASTM D1761 OSB Pull Through Direct Pull c OSB Recip. Blade OSB w/Washer OSB Pull Through Sample size 291 62 274 158 73 Mean a 173 (30.4) 165 (28.9) 149 (26.1) 141 (24.7) 160 (28.1) Median a 171 (30.0) 165 (28.9) 145 (25.4) 141 (24.6) 133 (23.2) 5% Non Exceedance 11 2 (19.6 ) 12 2 (21.4 ) 98 (17.2 ) 96 (16.8 ) 67 (12.7 ) Std. Dev a 35 (6.1) 27 (4.7) 31 (5.3) 27 (4.8) 73 (12.7) Minimum a 89 (15.6) 116.6 (20.4) 76.6 (13.4) 69.9 (12.3) 43.8 (7.7) Maximum a 269 (47.0) 227 (39.7) 272 (47.7) 224 (39.2) 360 (63.0) CoV (%) 20.1 16.4 20.5 19.5 45.1 Mean SG per method b 0.49 0.52 0.49 0.49 0.51 Boards Averaged for SG BA, BI BA, BB, BD BA, BC BI BE BI BA BB, BD a Load per length of nail shank in lb/in (N/mm) b The Specific Gravity is a weighted average of the boards that the methods were tested on ( Table 3 4 )

PAGE 55

55 Figure 4 18 Mean Withdrawal Capacities (8D Nails) Figure 4 19 5% Non Exceedance Capacities (8D Nails)

PAGE 56

56 In addition to this the Wilcoxon test rejected the null hypothesis meaning that they can be said to come from the same population ( Table 4 7). These small differences could be a result in the increase diameter of the nails. T he vibrations of the reciprocating blade and the stress from the pull through are not enough to reduce the withdrawal capacity drastically because the 8D nail has a stronger hold in the wood. However based on testing it is more likely that the 8D nails wil l fail in pull though before any pull out from the framing member occurs. In comparison to the ASTM D1761 pull through failure loads were lower by 8%. Table 4 7 Wilcoxon Test p values and Mean and 5% Non Exceedance Percent Differences (8D Nails) Test Method ASTM D1761 OSB Pull Through Direct Pull OSB Recip . Blade OSB w/Washer OSB Pull Through ASTM D1761 1 0.06 (4.8%) (8.7%) <2.2E 16 (15.2%) (13.1%) <2.2E 16 (20.7%) (13.2%) 0.0005 (7.8%) (49.1%) OSB Pull Through/Direct Pull 1 8.8E 5 (10.4%) (21.8%) 5.9E 8 (16.0%) (21.9%) 0.02 (3.0%) (57.2%) OSB/ Recip. Blade 1 0.01 (5.6%) (0.1%) 0.34 (7.5%) (36.5%) OSB w/Washer 1 0.9 (13.0%) (36.5%) OSB Pull Through 1 Figure 4 20 shows the results plotted against the specific gravity. The results are calculated and plotted with the same method as the 6D nails. The only difference is for each board nine 1 inch (25.4m m) samples were taken instead of five. Once again there is no observable pattern with regards to specific gravity. In addition above a specific

PAGE 57

57 gravity above 0.5 majority of the results fall below the predicted NDS Ultimate Withdrawal Capacity values, incl uding the ASTM D1761 values. Figure 4 20 Mean Withdrawal Capacity vs. Specific Gravity per Board (8D Nails) Time Dependent Testing Based on the results so far it can be said than installation through sheathing affects the immediate peak nail withdrawal strength. However the next question is: How does time affect the peak nail withdrawal capacities? From this three outcomes were theori zed: 1. The peak nail withd rawal capacity would decrease at the same rate over time. Hence if the difference between the immediate peak withdrawal capacities with and without sheathing starts out at x% then it would continue to be a difference of that same x % over time ( Figure 4 21 )

PAGE 58

58 Figure 4 21 Graph of Theory 1 2. The difference of the peak nail withdrawal capacity of between nails installed with and without sheathing would decrease until they leveled out at the same value. Hence if the difference between the immediate peak nail withdrawal capacities with and without sheathing starts out at x% over time the difference would dec rease to 0% or close to it, therefore over time the withdrawal capacities between the two installation methods converge ( Figure 4 2 2 ) Figure 4 22 Graph of Theory 2 3. The last possible outcome is that the withdrawal would decrease at noticeable different rates so in the end the percent difference would be different whether positive of negative. Hence if the difference between the immediate peak withdra wal capacities with and without sheathing was x% in the end it would be a different y% with the reciprocating blade decreasing faster and having a larger percent difference at the end or the slower and being higher than the ASTM values in the end (Figure 4 23)

PAGE 59

59 Figure 4 23 Possible Outcomes of Theory 3 As discussed in the Chapter 3 for the indirect pull the 8D nails would pull through for the majority of the tests. The washer was used to prevent this in the 8D Nails. Further testing was done to determine whether the washers affected the withdrawal strength and it was determined it d id not according to statistical analysis. Although pull through was not a problem with 6D Nails for the time dependent testing washers were used for the sake of consistency. In the case of the 6D nails there is a decrease over time in withdrawal capacity, with the exception of the 6 week data ( Table 4 8 and Figure 4 24 ). The withdrawal capacity however does not decrease as rapidly as in the Kurtenacker 1965 paper which states the biggest loss in strength happens in the first two w eeks. On average there is a 11% and 3 % difference in the first two wee ks for the ASTM D1761 ( Table 4 9 ) and OSB ( Table 4 1 0 ) test methods respectively and after 12 wee ks there is a difference of 44% and 29 %. It can then be said that the loss in strength over time is lower when installed through sheathing for 6D Nails.

PAGE 60

60 Table 4 8 Summary of results for Time Dependent Testing for 6D Nails Test Method ASTM D1761 OSB w/Washer Time of Withdrawal Immediate 2 wk. 6 wk. 12 wk. Immediate 2 wk. 6 wk. 12 wk. Sample size 30 30 30 30 30 30 30 30 Mean a 126 (22.0 ) 113 (19.8 ) 108 (19.0 ) 80 (14.0 ) 106 (18.5 ) 84 (14.7 ) 117 (20.4 ) 79 (13.8 ) Median a 121 (21.1 ) 118 (20.6 ) 103 (18.0 ) 77 (13.5 ) 89 (15.6 ) 86 (15.1 ) 119 (20.8 ) 82 (14.3 ) Std. Dev a 30 (5.2 ) 26 (4.5 ) 18 (3.2 ) 26 (4.6 ) 42 (7.3 ) 31 (5.4 ) 28 (4.8 ) 12 (2.2 ) Minimum a 85 (14.8 ) 48 (8.4 ) 87 (15.2 ) 38 (6.7 ) 71 (12.4 ) 39 (6.8 ) 53 (9.3 ) 53 (9.3 ) Maximum a 182 (31.9 ) 159 (27.9 ) 150 (26 .3 ) 170 (29.8 ) 282 (49.3 ) 151 (26.5 ) 170 (29.8 ) 109 (19.1 ) CoV (%) 23.6 22.7 16.9 33.1 39.5 36.7 23.6 15.6 Mean SG per method b 0.46 0.40 0.45 0.49 0.46 0.48 0.50 0.51 Boards Averaged for SG CA CC CJ CQ CA CD, CE, CF CK, CL, CM CR, CS, CT a Load per length of nail shank in lb/in (N/mm) b The Specific Gravity is a weighted average of the boards that the methods were tested on (See Table 3 8 ) Figure 4 24 Boxplot of 6D Nail Withdrawal Method per Method Over Time

PAGE 61

61 Table 4 9 Wilcoxon Test p values and Mean and Median Percent Differences (6D ASTM D1761) Immediate 2 Weeks 6 Weeks 12 Weeks Immediate 1 0.18 (10.8 %) (2.5 %) 0.03 (15.0 %) (16.1 %) 1.7 1E 7 (44.9 %) (43.8 %) 2 Weeks 1 0.24 (4 .2 %) (13.6% ) 9.8 E 6 (34.5 %) (41.4 %) 6 Weeks 1 3.0 E 6 (30.4 %) (28.2% ) 12 Weeks 1 Table 4 10 Wilcoxon Test p values and Mean and Median Percent Differences (6D OSB w/Washer) Immediate 2 Weeks 6 Weeks 12 Weeks Immediate 1 0.08 (23.0 %) (3.4 %) 0.01 ( 10.1 %) ( 28.7 ) 1 .0 E 3 (29.2 %) (8.5 %) 2 Weeks 1 7.6 E 5 ( 32.9 %) ( 32.0 %) 0.37 (6.3%) (5.1 %) 6 Weeks 1 1.0 E 7 (40.0 %) (40.0 %) 12 Weeks 1 The same phenomenon is observed with the 8D OSB w/Washer test series ( Table 4 11 and Figure 4 2 5 ) at 6 weeks as with the 6D nails. The mean withdrawal capacity is higher than the other test series in the time line. This also happens at the 2 week mark for the ASTM D1761 for 8D Nails. After further analysis of the 8D ASTM D1761 test series the Immediate, 6 week and 12 week can all be said to come from the same population according to the p values obtained f rom the Wilcoxon Test ( Table 4 12) The 2 week results for the 8D ASTM D1761 h as an average difference of 21 % from the means of the Immediate, 6 week and 12 week

PAGE 62

62 Table 4 11 Summary of Results for Time Dependent Testing for 8D Nails Test Method ASTM D1761 OSB w/Washer Time of Withdrawal Immediate 2 wk. 6 wk. 12 wk. Immediate 2 wk. 6 wk. 12 wk. Sample size 30 29 30 30 30 29 29 30 Mean a 148 (25.9 ) 179 (31.3 ) 146 (25.6 ) 141 (24.7 ) 133 (23.2 ) 132 (23.0 ) 135 (23.6 ) 99 (17.3 ) Median a 145 (25.4 ) 176 (30.7 ) 145 (25.4 ) 140 (24.5 ) 127 (22.3 ) 124 (21.7 ) 134 (23.5 ) 97 (17.1 ) Std. Dev a 20 (3.5 ) 18 (3.2 ) 17 (2.9 ) 22 (3.9 ) 20 (3.5 ) 3 0 (5.3 ) 33 (5.7 ) 24 (4.3 ) Minimum a 113 (19.8 ) 136 (23.7 ) 110 (19.3 ) 90 (15.8 ) 107 (18.7 ) 88 (15.4 ) 75 (13.1 ) 55 (9.6 ) Maximum a 215 (37.6 ) 211 (36.9 ) 175 (30.6 ) 202 (35.4 ) 178 (31.2 ) 193 (33.8 ) 192 (33.7 ) 154 (27.0 ) CoV (%) 13.3 10.1 11.3 15.7 15.1 23.0 24.2 24.6 Mean SG per method b 0.48 0.47 0.46 0.39 0.48 0.42 0.53 0.53 Boards Averaged for SG CB CI CP CW CB CF, CG, CH CM, CN, CO CT, CU, CV a Load per length of nail shank in lb/in (N/mm) b The Specific Gravity is a weighted average of the boards that the methods were tested on (See Table 3 8 ) Figure 4 25 Boxplot of 8D Nail Withdrawal Capacities per Method Over Time

PAGE 63

63 Table 4 12. Wilcoxon Test p values and Percent Differences (8D ASTM D1761) Immediate 2 Weeks 6 Weeks 12 Weeks Immediate 1 5.4 E 7 ( 18.6 %) ( 18.9 %) 0.79 (1. 5%) (0.2 %) 0.14 (4.9 %) (3.7 %) 2 Weeks 1 6.2 E 8 (20.1 %) (19.0 %) 6.7 E 8 (23.4 %) (22.5 %) 6 Weeks 1 0.21 (3.4 %) (3.5 %) 12 Weeks 1 In the case of the 8D OSB w/Washer results ( Table 4 13 ) the Immediate, 2 week and 6 week results can be said to come from the same population. The 12 week is lower with 30 % average difference fr om the means of the Immediate, 2 week and 6 week results. The mean ratio of the withdrawal capacity over time in comparison to the immediate withdrawal per method is graphed below in Figure 4 26. Table 4 13 Wilcoxon Test P values and Percent Differences (8D OSB w/Washer) Immediate 2 Weeks 6 Weeks 12 Weeks Immediate 1 0.64 (0.9 %) (2.8 %) 0.44 ( 1.4 %) (5.3 %) 8.5 E 7 (29.5 %) (26.6 %) 2 Weeks 1 0.59 ( 2.3 %) ( 0.1 %) 6.5 E 5 (29.6%) (23.8 %) 6 Weeks 1 9.7 E 5 (30.9 %) (%31.8 ) 12 Weeks 1

PAGE 64

64 Figure 4 26 Peak Withdrawal Capacities as a ratio of the Immediate Withdrawal Over Time (6D and 8D nails) As shown in the graph and after statistical analysis ( Table 4 14 and Figure 4 27 ) it can be seen that ASTM D1761 and OSB w/Washer values converge starting at 6 weeks, however for the 8D nails converge at 6 weeks but separate again at 12. Weeks. Another observation was t hat over time the reduction of the through sheathing method had the same percent reduction over time between 6D and 8D nails, which was not observed for ASTM D1761 test methods over time.

PAGE 65

65 Table 4 14 Wilcoxon P Values and Percent Difference between Test Methods (ASTM D1761 vs. OSB w/Washer) 6D Nails 8D Nails Time of Test P Value Percent Difference P Value Percent Difference Immediate 0.0008 17.6 % 2.5E 3 11.0% 2 Weeks 0.0004 29.7 % 2.9 E 8 30.3 % 6 Weeks 0.0574 7.5 % 0.19 8.2 % 12 Weeks 0.9000 1.3 % 2.9 E 11 35.5 % Table 4 15 Wilcoxon P values and Percent Difference between Nail Types (6D vs. 8D) ASTM D1761 OSB w/Washer Time of Test P Value Percent Difference P Value Percent Difference Immediate 3 0 E 3 16.3 % 3.2 E 5 22.8 % 2 Weeks 1.0 E 10 45.0 % 1.0 E 6 44.3 % 6 Weeks 3.1 E 8 22.7 % 0.03 14.2 % 12 Weeks 1.2 E 9 55.6 % 3 .0 E 11 22.5 % Figure 4 27 Percent Differences of Mean between ASTM D1761 and OSB w/Washer Test Methods per nail type

PAGE 66

66 Temperature was also a concern when performing these tests. Temperature and Humidity data were collected by a LogTag Analyzer Device and imported using the LogTag Analyzer Software (Version 2.3 2002 2013 LogTag Recorders). The device was placed beside th e test specimens to get a log of the temperature and humidity in the room they were stored in over time. The data was graphed over time and compared to atmospheric temperatures obtained from the National Weather Service Website and shown below ( Figure 4 28 and Figure 4 29 ). There are also graphs whe re the data over the time is graphed over the two day period in which both the s ix week and twelve week test groups were performed ( Figure 4 3 0 Figure 4 3 1 and Figure 4 3 2 ). There is not data for Immediate and 2 week as the LogTag Device was not available at that time. Figure 4 28 Temperature over time

PAGE 67

67 Figure 4 29 Humidity over time Figure 4 3 0 Tem perature and Humidity over time during the 2 day period in which 2 week tests were performed. (Atmospheric Only)

PAGE 68

68 Figure 4 31 Temperature and Humidity over time during the 2 day period in which the 6 week tests were performed (Line represents mean value of log tag data over time period). Figure 4 32 Temperature and Humidity over time during the 2 day period in which the 12 week tests were performed (Line rep resents mean value of log tag data over time period).

PAGE 69

69 As shown above the temperature remains fairly s table over time averaging at 75.3 o F with a minimum of 67.7 o F and a maximum of 86.1 o F. The humidity however has large spikes with a minimum of 28.7% an d a maximum of 86.4% averaging at 55.2%. On e of these large spike occurs during the time the 6 week testing was being performed which is when the withdrawal capacities for OSB increase instead of decreasing as expected. It could be said that the lower humi dity caused the withdrawal to be higher however there is not enough data to substantiate this observation.

PAGE 70

70 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS The main objective o f this study was to determine the nature in which through sheathing installation affects nail withdrawal capacities and to develop an accurate method f or testing nails installed through sheathing in the laboratory. As show n in the results above there is a definite difference in the results obtained using the standard ASTM D1761 test method and the modified test method in which nails were installed through sheathing The following conclusions can be made from the above results: Driving a nail through wood sheathing ( regardless of sheathing type), reduces its withdrawal capa city as compared with similar nails driven directly into the wood framing member. This is evident by the difference in withdrawal capacity of nails between the ASTM D1761 test method and methods in which the nails are driven through sheathing. The reciproc ating blade and indirect pull test methods (both through sheathing installation methods) caused a reduction in nail withdrawal capacity. These were 46% and 16% reduction respectively for 6D Nails and 15% and 21% reduction respectively for 8D Nails. The ind irect pull m ethod is determined as a more appropriate test method to isolate the effects of sheathing as it causes less of a reduction in capacity in 6D nails. Over time the difference between the ASTM D1761 and through sheathing installation method s is re duced for 6D nails but there is almost no change for 8D nails. However due to t he unexpected jump in mean capacity at the 6 week mark no definite conclusion can be made This jump however could be attributed to the drop in humidity during that time period. A reduction factor ranging from 0.6 to 0.85 should be used with the immediate nail withdrawal capacity determined by the ASTM D1761 procedure in estimating the strength of 6D nails (through sheathing installed) in situ. The reduction factor for 8D nail s is 0.85. Directly using the NDS specified values of nail withdrawal strength (without the above reduction factors) to estimate the failure loads and wind speeds from roof damage observations will over estimate the failure wind speed and is non conservat ive.

PAGE 71

71 In this testing it was determined that through sheathing installation causes a decrease in nail withdrawal capacities, however the reason for this loss in strength is unclear. One theory is that as the nail goes through the sheathing there may be a gr eater chance for deformation of the tip of the nails however on inspection this did not seem to be the case. Another theory is that the nail is resisted and slowed as it goes through the sheathing so the force and speed at which it enters the framing membe r is less than when there is no obstruction however there is no way to test this at this point. The results of this study affect the way the results of the ASTM D1761 test method are currently interpreted when determining how a nail will behave in the fie ld. Having a better understanding of this is he lpful when it comes to forensic engineers as often they may need to back calculate to determine failure wind speed If the data gathered for the nails is not accurate this could lead to incorrect assumptions i n the field. However on the design side despite these findings the NDS design predictions should not be affected due to the various reduction factors applied to the ultimate prediction to before arriving at a the design value. It is recommended that more testing be conducted to verify the strength reduction factor, using a wider sample of nails, and wood species. The conditioning of the wood samples was insufficient to fully eliminate fluctuations in temperature and humidity over the 12 week period. Stil l, these results do not support the substantial (over 50%) loss in nail withdrawal strength observed in the Kurtenacker ( 1965 a, b ) studies of the late 1960s, as reported earlier

PAGE 72

72 LIST OF REFERENCES American Forest and Paper Association ( AFPA ) (2005). "National Design Specification for Wood Construction. In Dowel Type Fasteners ." American Forest and Paper Association Washington, D.C. American Soc iety for Testing and Materials ASTM (2006a). "Standard Test Methods Mechanical Fasteners in Wood, D 1761 06." American Society for Testing and Materials ( ASTM ) (2006b). "Standard Test Methods for Specific Gravity of Wood and Wood Based Materials, D2395 02". American Society for Testing and Materials ASTM (2006c). "Standard Test Methods for Direct Moisture Content Measurement of Wood and Wood Based Materials, D4442 0 6 ." fa steners under simulated loading con Journal of Wind Engineering and Industrial Aerodynamics, 72, 389 400. Herzog, B., and Yeh, B. (2006). "Nail Withdrawal and Pull through Strength of Structural Use Panels." 9th World Conference on Timber Engineering, Portland, Oregon, USA. Kurtena Nail holding power of American woods Forest Products Laboratory, U.S. Forest Service Madison, WI. Kurtenacker, R. S. (1965b). "Performance of Container Fasteners Subjected to Static and Dynamic Withdrawal ." Forest Products Laboratory USDA Forest Service Madison, WI. Pye, S. J., Jr. (1995). "Effect of in Service Conditions on the Withdrawal Capacity of Roof Sheathing Fasteners." M.S. Thesis, Clemson Universit y ,. Clemson, SC. Shreyans, S., Kerr, A., Prevatt, D. O., and Gurley, K. R. ( 2012). "In Situ Nail Withdrawal Strengths in Wood Residential Roofs." ATC&SEI Advances in Hurricane Engineering, Miami, FL. Sparks, P. R. (1991). "Damages and lessons learned from Hurricane Hugo." NIST Special Publication, (820), 435 450. Sutt, E., Reinhol d, T., and Rosowsky, D. (2000). "The Effect of in Situ Conditions on Nail Withdrawal Capacities." World Conference of Timber Engineering, Whistler, BC. Swane, R. A., and Vagholkar, M. K. (1968). "Effect of heat on static withdrawal resistance of plain shan k nails in some Australian timbers." Building Science 3(1), 51 63.

PAGE 73

73 Biometrics Bulletin 1(6), 80 83

PAGE 74

74 BIOGRAPHICAL SKETCH Ashlie Kerr started the University of Florida in 2007 as a civil e ngineering major. During her time as an undergraduate student she was highly involved in various organizations including the Ameri can Society of Civil Engineers student c hapte r, Society of Women Engineers, the American Concr ete Institute s tude nt c hapter and the Jamaican American Student Association. Ashlie became involved in research as an undergraduate and gained a position as a research assistant in her final year of her undergraduate degree program. Ashlie graduated with her Bachelor of Sci ence in civil e ngineering in December 2011 after which she immediate began her graduate degree program in structural e ngineering. Ashlie continued with research through the graduate program and completed her graduate degree program in M ay 2014 earning her Master of Engineering in civil e ngineering. In February of 2014 Ashlie joined the consulting engineering firm Botkin, Parssi and Associates of Lake Worth, FL as a Structural E ngineer, where she is working on the restoration and remodeling of condominiums and custom homes.