Tensile Strength and Performance of the INDUO®-Heavy-Timber Connector in Combination with Structural Composite Lumber and Douglas Fir by MARKUS STEINIGER Diplom-lngenieur (FH) Holzbau und Ausbau, [B.A.Sc] Fachhochschule Rosenheim, 2001 A thesis submitted in partial fulfi l lment of the requirements for the degree of Master of Science in FACULTY OF GRADUATE STUDIES Department of Wood Science We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 2003 © Markus Steiniger, 2003 In p resen t ing this thesis in part ial fu l f i lment o f t h e requ i remen ts f o r an advanced d e g r e e at the Universi ty o f Brit ish C o l u m b i a , I agree that t h e Library shall make it f reely available f o r re ference and study. I fu r ther agree that pe rmiss ion f o r ex tens ive c o p y i n g o f this thesis f o r scholar ly pu rposes may b e g r a n t e d b y t h e head o f m y d e p a r t m e n t o r by his o r her representat ives. It is u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f th is thesis fo r f inancial gain shall n o t be a l l o w e d w i t h o u t m y w r i t t e n pe rmiss ion . D e p a r t m e n t o f The Univers i ty o f Brit ish C o l u m b i a Vancouver , Canada Date DE-6 (2/88) inoduo Abs t rac t ii Abstract The INDUO-connection is a new steel-to-wood joint for highly loaded heavy-timber structures. Embedded in the end-grain of laminated timber beams, the special INDUO-connector is designed to transfer axial and transverse loads. Due to the weaker strength properties of solid wood perpendicular to the grain direction, the connection's capacity under transverse loading is comparatively small. Parallel to the grain, however, the connection is capable to transfer loads of up to 180kN and is thus predominantly suited for tension applications. Since the introduction to the European market in the Mid-90s, the INDUO-connection system has been mainly deployed in Post-and-Beam structures using softwood timber and Glulam as beam material. This thesis investigates the tensile strength and performance of the INDUO-connector in combination with different beam materials. Static tension tests were performed in two separate test series with a total of 99 specimens of different member cross-sections (100x100mm and 120x120mm), connector types (A and B) and beam materials (Microllam®LVL, Parallam®PSL, TimberStrand®LSL, thick Douglas-Fir plywood and Douglas-Fir lumber). In test series 1 all possible combinations of beam material, connector type and member cross-section were tested with a sample size of up to three, providing trends on strength properties and failure performance. For test series 2 it was decided to focus on fewer combinations with a larger sample size to create statistically more significant results on ultimate tensile strength and connection stiffness for the most advantageous setups. By modeling and calculating the INDUO-connection according to different international timber codes (German DIN1052-2000, European EC5, Canadian CSA 086.1 and US ASCE 16-95) as a bolted or tight-fitting dowel connection, characteristic tensile strength data was computed and compared with the characteristic values derived from the results of series 2. In terms of tensile strength and performance, TimberStrand®LSL presented the best test results, outperforming Microllam®LVL, Parallam®PSL and Douglas-Fir lumber, which showed significantly lower tensile strengths accompanied by brittle failure modes. Thick Douglas-Fir plywood was only examined in test ser ies 1, presenting high tensile strength for larger beam-cross-sections, whereas smaller cross-sections failed brittely. inoduo A b s t r a c t iii The comparison of different design approaches for the INDUO-connection showed that based on the same connection model (dowel-type fasteners with inside steel plate) the Canadian timber code provided by far the most conservative design values, whereas all other codes presented significantly higher numbers relative to the Canadian code . The comparison of characteristic tensile strength properties generated from the test results and values derived from the different design strengths indicate all four timber codes have more or less similar results. inoduo T a b l e of C o n t e n t s iv Table of Contents Abstract ii Table of Contents iv List of Figures vii List of Tables xi Dedicat ion xii Acknowledgements xiii 1 Int roduct ion 1 1.1 Research Objective 5 1.2 Scope 6 2 Non-v is ib le heavy- t imber j o in ts 7 2.1 Non-visible mechanical connections 7 2.1.1 The BSB-system (Tight-fitting dowels) 7 2.1.2 The SpikeTec-system (Embedded nail plate) 9 2.2 Non-visible composite connections 10 2.2.1 Glued-in steel rods 11 2.2.2 The TiSCo®-system 14 2.2.3 The BVD®-system 15 3 Materials and Methods 19 3.1 Materials 19 3.1.1 Microllam®LVL 19 3.1.2 Parallam®PSL 21 3.1.3 TimberStrand®LSL 23 3;1.4 Douglas-Fir plywood 24 3.1.5 Douglas-Fir lumber 26 3.1.6 INDUO-connector (type A) 27 3.1.7 INDUO-connector (type B) 28 inoduo T a b l e of C o n t e n t s v 3.2 Methods 29 3.2.1 Scope of the test series 29 3.2.2 Connection design 31 3.2.3 Fabrication of test specimens 34 3.2.4 Experimental setup 37 3.2.5 Test procedure 39 4 Results 40 4.1 Test series I 41 4.1.1 Group I (TimberStrand®LSL & Douglas-Fir plywood) 43 4.1.2 Group II (Douglas Fir & Microllam®LVL) 47 4.1.3 Group III (Parallam® PSL) 50 4.2 Test series II 52 4.2.1 Performance 54 4.2.1.1 TimberStrand®LSL 54 4.2.1.2 Douglas Fir 56 4.2.1.3 Parallam®PSL 58 4.2.1.4 Microllam®LVL 60 4.2.2 Strength and stiffness 62 5 Discussion 68 5.1 " Evaluation of test results 68 5.2 Comparison of characteristic strength values 71 5.2.1 Connection Model 71 5.2.2 Determination of code design values 72 5.2.2.1 DIN 1052-2000 (draft) 75 5.2.2.2 EC5 (Eurocode 5) 76 5.2.2.3 CSA 086.1 77 5.2.2.4 ASCE 16-95 78 5.2.3 Characteristic strength values 79 5.3 Evaluation of connection stiffness with structural model 83 6 Conc lus ion and Recommendat ions 86 inoduo T a b l e of C o n t e n t s vi 7 List of References 88 8 Appendices 92 8.1 Photographic documentation 92 8.1.1 Manufacturing steps of test specimen 92 8.1.2 Test procedure 95 8.1.3 Failure modes 100 8.2 Calculations 110 incduo List of F igures vii List of Figures Figure 1: P lan of c o n n e c t i o n 2 Figures 2 & 3: A s s e m b l y of a s ing le - fami l y h o u s e c o n s t r u c t e d w i th the I N D U O - s y s t e m 2 Figure 4: 3 - d i m e n s i o n a l roof t russ 3 Figure 5: I N D U O - s y s t e m a p p l i e d in 3 - d i m e n s i o n a l t russ s y s t e m 3 Figure 6: I N D U O - c o n n e c t o r a n d t i m b e r ha lves be fo re a s s e m b l y 3 Figure 7: E m b e d d e d c o n n e c t o r ; b e a m ready to be p r e s s e d in hydrau l i c p ress 3 Figure 8: I N D U O - c o n n e c t o r t ype A 4 Figure 9: I N D U O - c o n n e c t o r t y p e B 4 Figure 10: M a n u f a c t u r e of I N D U O - Q u a r t e r L o g s 4 Figure 1 1 : BSB- jo in t s in h e a v y t i m b e r t russ 8 Figure 12: S c h e m e of B S B - c o n n e c t i o n 8 Figure 13: B S B - s y s t e m app l i ed in a roof s t ruc tu re of a s p a 8 Figure 14: F o o t b r i d g e c o n s t r u c t e d w i th B S B - c o n n e c t i o n 8 Figure 15: C l o s e - u p of S p i k e T e c - c o n n e c t i o n 9 Figure 16: Nai l p la tes a re se t in p lace 9 Figure 15: S p i k e T e c - s y s t e m app l i ed in the roof s t ruc tu re of a s u p e r m a r k e t 10 Figure 16: Roof t r uss s y s t e m c o n s t r u c t e d w i th S p i k e T e c - c o n n e c t i o n 10 Figure 17: G l u e d - i n s tee l rod c o n n e c t i o n ( M a d s e n ) 12 Figure 18: App l i ca t i ons fo r g lued- in s tee l rod c o n n e c t i o n s ( M a d s e n ) 12 Figure 19: R e s e a r c h c o n d u c t e d o n g lued- in s tee l rod c o n n e c t i o n s 13 Figure 20: S k e t c h of tes t s p e c i m e n s (be fo re g rou t ing ) 13 Figure 2 1 : T i S C o - c o n n e c t o r : S a n d b l a s t e d a n d g r o o v e d v e r s i o n 14 Figure 22: T i S C o - c o n n e c t o r inser ted in e n d gra in 14 Figure 23: M a n u f a c t u r i n g s t e p s of t he T i S C o - c o n n e c t i o n 15 Figure 24a: C o m p o n e n t s of t h e B V D - s y s t e m 16 Figure 24b: D i f fe ren t h a n g e r s izes in va r ious leng th c o n f i g u r a t i o n 16 Figure 25: M a n u f a c t u r i n g the B V D - c o n n e c t i o n ' 16 Figure 26: K n e e jo in t w i th B V D - c o n n e c t i o n 17 Figure 27: T o w e r c o l u m n w i th B V D - c o n n e c t i o n to f o u n d a t i o n 18 Figure 28: E X P O - R o o f w i th t o w e r s s u p p o r t i n g t h e roof s t ruc tu re 18 Figure 29: M a n u f a c t u r e of L V L 20 Figure 30: C l o s e - u p of L V L 20 Figure 3 1 : M a n u f a c t u r e of P S L 21 Figure 32: C l o s e - u p of P S L 22 Figure 33: P S L a p p l i e d in h e a v y - t i m b e r s t ruc tu re (Fores t ry bu i ld ing , U B C ) 22 inoduo List of F igures viii Figure 34: M a n u f a c t u r e of L S L 23 Figure 35: C l o s e - u p of LSL 24 Figure 36: M a n u f a c t u r e of p l y w o o d 2 5 Figure 37: C l o s e - u p of th ick D o u g l a s Fir p l y w o o d 25 Figure 38: C l o s e - u p of D o u g l a s Fir 2 6 Figure 39: M a n u f a c t u r e of s a w n l u m b e r 2 6 Figure 40: C o n n e c t o r t ype A 2 7 Figure 41: C o n n e c t o r t ype B 28 Figure 42: C o m p o n e n t s of lower s u p p o r t 3 3 Figure 43: L o w e r s u p p o r t a s s e m b l e d 3 3 Figure 44: R ing S ide Plate 33 Figure 45: U p p e r s u p p o r t 3 4 Figure 46: P lan of t i m b e r ha l ves : a) t ype A - 1 0 0 c o m b i n a t i o n ; b) type A - 120 c o m b i n a t i o n 36 Figure 47: P lan of t i m b e r ha l ves : a) t ype B - 1 0 0 c o m b i n a t i o n , s c r e w - b o n d ; b) type B - 120 c o m b i n a t i o n 36 Figure 48: E x p e r i m e n t a l s e t u p : M T S 8 1 0 w i th cont ro l s y s t e m a n d tes t m e m b e r 37 Figure 49: T e s t m e m b e r 38 Figure 50: U p p e r s u p p o r t f ea tu r i ng a D C D T m e a s u r i n g d e v i c e 38 Figure 51: C o n n e c t i o n fa i lu re m o d e s 40 Figure 52: G r o u p I: No fa i lu re fo r c o n n e c t o r t ype B c o m b i n a t i o n s 43 Figure 53: G r o u p I: S e v e r e b e a r i n g ; a typ ica l fa i lu re m o d e for c o n n e c t o r t ype A c o m b i n a t i o n s 4 3 Figure 54: L o a d - d i s p l a c e m e n t c u r v e s of T B - c o m b i n a t i o n s 4 5 Figure 55: L o a d - d i s p l a c e m e n t c u r v e s of T A - c o m b i n a t i o n s 45 Figure 56: L o a d - d i s p l a c e m e n t c u r v e s of X B - c o m b i n a t i o n s 46 Figure 57: L o a d - d i s p l a c e m e n t c u r v e s of X A - c o m b i n a t i o n s 4 6 Figure 58: G r o u p I I : Spl i t t ing a l o n g the rows of ho les ; a typ ica l fa i lu re m e c h a n i s m of c o n n e c t o r t ype B c o m b i n a t i o n s 47 Figure 59: G r o u p I I : B e a r i n g a n d re la t ive ly la rge d i s p l a c e m e n t s d u e to s h e a r fa i lu re of c o n n e c t o r t ype A c o m b i n a t i o n s 47 Figure 60: L o a d - d i s p l a c e m e n t c u r v e s of D B - c o m b i n a t i o n s 48 Figure 61: L o a d - d i s p l a c e m e n t c u r v e s of D A - c o m b i n a t i o n s 4 8 Figure 62: L o a d - d i s p l a c e m e n t c u r v e s of M B - c o m b i n a t i o n s 49 Figure 63: L o a d - d i s p l a c e m e n t c u r v e s of M A - c o m b i n a t i o n s 49 Figure 64: G r o u p I I I : B e a r i n g a n d tea r -ou t of c o n n e c t o r t ype A c o m b i n a t i o n s 50 Figure 65: G r o u p I I I : C o n n e c t o r t ype B c o m b i n a t i o n s w i tness s e v e r e p in d e f o r m a t i o n s or f rac tu re 50 Figure 66: L o a d - d i s p l a c e m e n t c u r v e s of P B - c o m b i n a t i o n s 51 ircduo List of F igu res ix Figure 67: L o a d - d i s p l a c e m e n t c u r v e s of P A - c o m b i n a t i o n s 51 Figure 68: L S L : No d a m a g e o b s e r v e d at t he p in ho les 5 4 Figure 69: L S L : N o d e f o r m a t i o n s of t he c o n n e c t o r 54 Figure 70: L o a d - d i s p l a c e m e n t c u r v e s of T B - c o m b i n a t i o n s 55 Figure 7 1 : L o a d - d i s p l a c e m e n t c u r v e s of S - T B - c o m b i n a t i o n s 5 5 Figure 72: D G fir: Sp l i t t ing a l o n g rows of ho les 5 6 Figure 73: D G fir: S h e a r fa i lu re in the p l a n e of t he p ins 56 Figure 74: L o a d - d i s p l a c e m e n t c u r v e s of D B - c o m b i n a t i o n s 57 Figure 75: L o a d - d i s p l a c e m e n t c u r v e s of S - D B - c o m b i n a t i o n s 5 7 Figure 76: P S L : Sp l i t t ing a l o n g the rows of ho les 58 Figure 77: P S L : D e f o r m a t i o n of t he c o n n e c t o r p ins 5 8 Figure 78: L o a d - d i s p l a c e m e n t c u r v e s of P B - c o m b i n a t i o n s 59 Figure 79: L o a d - d i s p l a c e m e n t c u r v e s of S - P B - c o m b i n a t i o n s ) 5 9 Figure 80: L V L : Spl i t t ing a l o n g rows of ho les 6 0 Figure 8 1 : L V L : R ipped-o f f p ins - 6 0 Figure 82: L o a d - d i s p l a c e m e n t c u r v e s of M B - c o m b i n a t i o n s 6 0 Figure 83: L o a d - d i s p l a c e m e n t c u r v e s of S - M B - c o m b i n a t i o n s 61 Figure 84: D i f fe ren t a p p r o a c h e s to d e t e r m i n e t h e c o n n e c t i o n s t i f fness 63 Figure 85: C lass i f i ca t ion of m e m b e r s e t u p s a c c o r d i n g to tens i le p e r f o r m a n c e 69 Figure 86: C l o s e - u p of e n d g ra in : a) P S L ; b) L V L ; c) D o u g l a s Fir; d) L S L 7 0 Figure 87: C o n n e c t i o n m o d e l 71 / F igure 88: P lan of c o n n e c t i o n 7 3 Figure 89: Fa i lu re m o d e s a c c o r d i n g to E u r o p e a n Y ie ld M o d e l 7 4 Figure 90: D i f fe ren t tes t p r o c e d u r e s to d e t e r m i n e the e m b e d d i n g s t reng th of w o o d a n d w o o d - b a s e d ma te r i a l ; a ) , D IN E N 3 8 3 - 1 9 9 3 b) A S T M D 5 7 6 4 a - 1 9 9 7 80 Figure 9 1 : T h a l k i r c h e n Br idge , M u n i c h , G e r m a n y 8 3 Figure 92: S u p p o r t w i th n o d e a n d c o n n e c t e d b e a m s 8 3 Figure 93: N o d e in the t russ s y s t e m 83 Figure 94: C o n n e c t i o n s e t u p of I N D U O - c o n n e c t o r in 3 D - s p a c e t russ b e a m 84 Figure 95: Cu t t i ng t i m b e r m e m b e r s to r o u g h d i m e n s i o n s (Sl id ing T a b l e S a w ) 92 Figure 96: P lan ing of t i m b e r m e m b e r s to f ina l w i d t h a n d t h i c k n e s s (4 -S ided P laner ) 92 Figure 97: M a c h i n i n g of rows of ho les a n d V - g r o o v e by m e a n s of C N C - r o u t e r 92 Figure 98: M a c h i n i n g of c i rcu la r g r o o v e s ( C N C - r o u t e r ) 92 Figure 99: I N D U O - c o n n e c t o r ready to be e m b e d d e d in m a c h i n e d t i m b e r ha lves 93 Figure 100: C i rcu la r g r o o v e s to a c c o m m o d a t e S tee l S ide Pla tes 93 Figure 101 : A p p l y i n g P V A - c o n s t r u c t i o n g lue to bo th ins ide faces of the t i m b e r ha lves 93 Figure 102: Inser t ing the c o n n e c t o r in V - g r o o v e 93 Figure 103: J o i n i n g of bo th t i m b e r ha lves e n c l o s i n g the c o n n e c t o r 94 incduo List of F igu res x Figure 104: Inser t ing t h e c o m p o s i t e m e m b e r into hydrau l i c p ress ; P r e s s i n g t i m e : 30 m i n u t e s 94 Figure 105: A l te rna t i ve c o n n e c t i o n of t imber ha lves w i th regu lar w o o d - s c r e w s 6 x 1 0 94 Figure 106: S e t u p of s c r e w s 94 Figure 107: T a p e r i n g of t he tes t m e m b e r to s q u a r e d c r o s s - s e c t i o n 1 0 0 x 1 0 0 a n d 1 2 0 x 1 2 0 m m respec t i ve ly ( N C - s h a p e r ) 94 Figure 108: S p e c i m e n ready to be t e s t e d 95 Figure 109: S t a c k e d m e m b e r s of d i f fe rent c o n n e c t o r - m a t e r i a l s e t u p s be fo re tes t ing 9 5 Figure 110: S p e c i m e n c o n n e c t e d to upper m a c h i n e s u p p o r t 95 Figure 111 : S tee l S ide P la tes t rans fe r app l i ed load f r o m the lower m a c h i n e s u p p o r t to tes t m e m b e r 95 Figure 112: A s s e m b l y of S tee l S ide P la tes w i th 7 /8 - inch bo l ts 96 Figure 113: C l o s e - u p of s tee l r ings s l id ing into c i rcu la r g r o o v e 96 Figure 114: S tee l S ide P la tes a re p r e s s e d into g r o o v e s by m e a n s of regu lar c l a m p s 9 6 Figure 115: C o m p o n e n t s of lower c o u p l i n g : D i s tance p la tes a n d t w o 1-inch bol ts 96 Figure 116: L o w e r c o u p l i n g f a s t e n e d 97 Figure 117: S p e c i m e n c o n n e c t e d to tes t a p p a r a t u s , r e a d y to be t e s t e d 97 Figure 118: T e s t a p p a r a t u s cons i s t i ng of tes t m a c h i n e , con t ro l unit a n d PC 98 Figure 119: U n l o a d i n g of h e a v y tes t m e m b e r by m e a n s of a "mob i le ga l low" 98 Figure 120: G a l l o w in p l a c e to s u p p o r t t he tes t m e m b e r a f ter be ing u n c o u p l e d 98 Figure 121 : D i s a s s e m b l i n g of u p p e r a n d lower c o u p l i n g s 98 Figure 122: T e s t e d t ype A a n d B s p e c i m e n s , 1 2 0 x 1 2 0 m m , test ser ies 1 99 Figure 123: T e s t e d t ype A a n d B s p e c i m e n s , 1 0 0 x 1 0 0 m m , tes t ser ies 1 99 Figure 124: T e s t e d c o n n e c t o r - t y p e - B s p e c i m e n s , 1 0 0 x 1 0 0 m m , test ser ies 2 99 Figures125a-e: N o fa i lu re o b s e r v e d wi th T B - m e m b e r s e t u p s 100 Figures126a-b: Fa i lu re m o d e s o b s e r v e d w i th T A - m e m b e r s e t u p s 101 Figures 127a-f: Fa i lu re m o d e s o b s e r v e d w i th X B - m e m b e r s e t u p s 102 Figures 128a-c: Fa i lu re m o d e s o b s e r v e d w i th X A - m e m b e r s e t u p s 103 Figures 129a-g: Fa i lu re m o d e s o b s e r v e d wi th D B - m e m b e r s e t u p s 104 Figures 130a-c: Fa i lu re m o d e s o b s e r v e d w i th D A - m e m b e r s e t u p s 105 Figures 131a-g: Fa i lu re m o d e s o b s e r v e d wi th M B - m e m b e r se tups 106 Figures 132a-d: Fa i lu re m o d e s o b s e r v e d wi th M A - m e m b e r s e t u p s 107 Figures 133a-f: Fa i lu re m o d e s o b s e r v e d wi th P B - m e m b e r s e t u p s 108 Figures 134a-c: Fa i lu re m o d e s o b s e r v e d w i th P A - m e m b e r s e t u p s 109 inoduo List of T a b l e s xi List of Tables ( Table 1 : T e s t va r iab les 2 9 Table 2: Mo is tu re con ten t a n d dens i t y of t es ted s p e c i m e n s 4 0 Table 3: A n a l y z e d d a t a of tes t se r ies 1 41 Table 4: A n a l y z e d d a t a of tes t se r ies 2 5 2 Table 5: S ta t is t ics o n u l t imate load 62 Table 6: S ta t is t ics on d i s p l a c e m e n t at u l t imate load 62 Table 7: S ta t is t ics o n d i f fe rent 10 /40 - a n d 3 0 / 7 0 - c o n n e c t i o n s t i f fness 6 4 Table 8: 5 t h pe rcen t i le s t r e n g t h s of respec t i ve d is t r ibu t ions 66 Table 9: Charac te r i s t i c v a l u e s of m a x i m u m tens i le c a p a c i t y 67 Table 10: 5 t h pe rcen t i le of 10 /40-s t i f fness 6 7 Table 1 1 : 5 t h pe rcen t i le of 30 /70 -s t i f f ness 6 7 Table 12: E x a m p l e s h o w i n g a s t e p - b y - s t e p a p p r o a c h to d e t e r m i n e charac te r i s t i c v a l u e s fo r tens i le c a p a c i t y 81 Table 13: C o m p a r i s o n of charac te r i s t i c c o n n e c t i o n s t r e n g t h v a l u e s 81 Table 14: Ca lcu la t i on e x a m p l e o n s t i f fness a n d d i s p l a c e m e n t of I N D U O - c o n n e c t o r in 3 D - t r u s s b e a m 85 Table 15: R E L A N d a t a f i t t ing of s t reng th d a t a set 110 Table 16: R E L A N d a t a f i t t ing of s t i f fness d a t a set 110 Table 17: Ca l cu la t i on of charac te r i s t i c s t reng th v a l u e s ; P S L 111 Table 18: Ca l cu la t i on of charac te r i s t i c s t reng th v a l u e s ; D o u g l a s Fir 112 Table 19: Ca l cu la t i on of charac te r i s t i c s t reng th v a l u e s ; L V L 113 Table 20: Ca lcu la t i on of charac te r i s t i c s t reng th v a l u e s ; L S L 114 dedicated to my mother Anne inoduo A c k n o w l e d g e m e n t s xi i i Acknowledgements I would like to express my greatest thanks to my supervisors Dr. Helmut Prion and Dr. Frank Lam for their advice and guidance during my Master's program and this research. Their comments and constant support meant a great help to me and contributed very much to the successful completion of the project. I also would like to thank Robert Fiirst, Tom Wray, Robert Myronuk, George Lee and Emmanuel Sackey of the Department of Wood Science and Harald Schrempp of the Department of Civil Engineering for their great practical help in the preparation process of the test series. Their dedication helped to carry out the research in a very smooth and effective manner. I would like to thank especially Bruce Craig of TrusJoist, a Weyerhaeuser Business and Gordon White of Ainsworth Lumber Co. for providing the Microllam®LVL, Parallam®PSL, TimberStrand®LSL and Douglas-Fir plywood, respectively. Without their generous donations of material and their technical support, the research would have been a lot more difficult. For comprehensively supporting the research and supplying the INDUO-connectors, I would like to express my special gratitude to Paul Reichartz and his team of INDUO. jnoduo Introduction 1 1 Introduction Carpentry in Europe looks back on a very long tradition and history. Over centuries craftsman skills have been improved and passed on from one generation to the next. In the past a carpenter meant more than just manufacturing wooden structures; via their craftsmanship carpenters united the work of engineers, architects and contractors into one person. Until the end of the 18th century, being universal experts, they acted as general contractors. The industrial revolution changed the traditional construction habits. Especially the Central European countries witnessed a substantial shift from wood to steel, concrete and brick as major building materials. This development caused a severe depreciation of the carpenter's craftsmanship. Despite their knowledge and skills, carpentry's time and cost intensive manual-labor could not compete with the upcoming industrialized and engineered construction technology. Carpentry lost its dominating role both in the design process and the construction. Since then, carpenters gradually limited their field of work to the manufacturing of roof and truss structures. Until the 1980s, wood as construction material remained relatively dormant. In the last two decades, however, especially in Austria, Germany, and Switzerland, an increasing environmental consciousness changed people's attitude towards the 'established' and 'old' construction materials. Demanding a healthy and 'environmentally friendly' as well as a comfortable and cozy home, more and more willing homebuilders decided to use wood for the construction of their new houses. Furthermore, public authorities supported the use of wood for commercial as well as public projects. With the renaissance of wood as the most natural of all building materials, the old carpenter's skills were again in great demand. In addition, computer-controlled woodworking machinery enabled carpenters to manufacture complex and labor-intensive wooden structures at a competitive price level. Compared to North America, building your own house in Europe is a very costly endeavor. Contributing with their own labor force, many homebuilders reduce construction costs by resorting to do-it-yourself (DIY) kits. For this reason, manufacturers of prefab houses or building components offer a variety of both hardware - structural components - and service - construction of the house up, to various degrees of completion. Depending on the skills of the homebuilder, all the finishing work can be done by DIY style. Due to good workability, wood and wood products are widely used in this area. inoduo Introduction 2 In the early 1990s, Paul Reichartz, a German businessman and consultant, came up with the idea to provide both carpenters and DIY-homebuilders with a simple and affordable state-of-the-art construction system, while meeting all performance and code requirements and satisfying customers' demands. After years of development and adjustment, in 1995 the INDUO-connection and construction system was introduced in the homebuilding market. INDUO® is a contemporary heavy-timber system. It consists of precisely prefabricated, easy-to-assemble wooden members of varying cross-section and length, connected to standardized steel nodes. These basic elements can be used to build up post-and-beam frames with varying configurations. These building elements are also well suited for highly loaded timber structures, such as 3-dimensional space trusses. Figure 1: Plan of connection Figures 2 & 3: Assembly of a single-family house constructed with the INDUO-system inoduo Introduction 3 Figure 4: 3-dimensional roof truss Figure 5: INDUO-system applied in 3-dimensional truss system Consisting of two timber halves and two special wood connectors, the primary framing members can be considered as composite beams. Before gluing the wooden halves together, they are specially machined to accommodate the cast steel connector. The connector element itself features a set of tapered spikes and has a female thread connection at both ends, which can be bolted to a variety of nodes or brackets. Located along the center-line of the member, the connector is designed to transfer axial and transversal loads. A common wrench is the only tool necessary for the assembly. Thus, both simple and complex structures can be erected fast and precisely. Figure 6: INDUO-connector and Figure 7: Embedded connector; beam ready to be timber halves before assembly pressed in hydraulic press inoduo Introduction 4 Since 1995 various research projects have focused on the manufacturing process of the joint with less emphasis on load carrying capacities (Fuhrer 1997). In a small test series Guldenpfennig (1996) proofloaded 12 INDUO-connections with Spruce gluelam up to 100kN. Only one of the specimens failed. In 2001, the original connector (type A) was superceded by an advanced version. The modified shape of the new connector (type B) allowed the capacity of the connection to be calculated according to the German code DIN 1052-1988 as a Tight-Fitting-Dowel joint (BlaB 2001). Tests to verify this design and calculation model were not conducted. Figure 8: INDUO-connector type A Figure 9: INDUO-connector type B In the past, the basic elements were made with solid softwood - mainly Spruce and Fir. Manufacturing the so-called "Quarter Logs" with exposed edge grain and pith forming the member corners allows the connector spikes to penetrate into the flat grain portion of the log. Visually more attractive, the Quarter Log can be made from small logs, possibly even peeler cores. Round timber Ripping into Rotation Milling Lamination Quarters Figure 10: Manufacture of INDUO-Quarter Logs inoduo In t roduc t ion 5 Being a high-performance connection method, the INDUO connector could be cost-effectively applied in heavy-timber construction with structural composite lumber. Since little is known about the connection's behavior in combination with such engineered wood products, composites like Laminated Strand Lumber (LSL), Parallel Strand Lumber (PSL) or Laminated Veneer Lumber (LVL) have so far not been considered as possible substitutes for solid wood. The intention of this Master research project was to gain more specific information on the joint's tension behavior with emphasis on ultimate strength and failure mode. As this connector is primarily suited for tension application, it was decided that testing of the INDUO-member deployed as beam elements, where transverse shear will be the dominant load, would not be done. 1.1 Research object ive Officially introduced in Germany, the INDUO®-system has been successfully marketed in many European countries. The system has been gradually improved to respond to various customer demands. INDUO's future goal is to enter the North American marketplace. Considering different construction standards and techniques as well as a different set of priorities, it is essential to adjust the system to North American demands and requirements. Concerning the adaptation of INDUO, the following issues are of substantial interest: • Investigation of connector type B's tensile behavior: - Mechanical properties (strength and stiffness) - Failure mode - Calibration of rational calculation model (BlaB 2001) • Influence of different member configurations on tensile behavior; Comparison of connector type A and type B with: - Beams built-up with different wood species (solid wood and wood composites) - Different methods of bonding the two member halves: screwed or glued - Members of different cross-sections inoduo In t roduc t ion 6 1.2 Scope A comprehensive test program, considering all the above-mentioned parameters and providing a sufficient sample size to create statistically significant data, was deemed to be beyond the scope of a Master thesis. Therefore, it was agreed to split the project into two separate test series, each focusing on different aspects. In series 1, beam members of all the possible combinations of material, connector type, member cross-section and lamination type were tested under tensile loading. Up to three specimen for each individual combination are meant to provide'general information and performance trends. Data assessment and experiences of series 1 then served as the basis for a detailed and more accurate investigation of one combination. With a sample size of up to 10, statistically more significant results for ultimate tensile strength were determined. This test series would then be used to calibrate an analytical model for calculating the tensile capacity of typical connections. inoduo Non-v i s ib le h e a v y t i m b e r jo in ts 7 2 Non-visible heavy-timber joints Comparing North American and European practices, the design approach of timber joints is significantly different. Since the 1970s, European architects and structural engineers have more and more tended to 'hide' or embed timber connections for esthetic and / or fire protection purposes, whereas contemporary North American heavy-timber design still prefers to expose the connection. Responding to the trend for non-visible joints, numerous new connection types have been developed in Europe. With improved mechanical properties and advanced performance, these innovative connections have been widely applied in various timber structures. Being embedded in a composite beam member, the INDUO-connector, which is the focus of research in this Master thesis, represents one of the above-mentioned non-visible timber joints. The following sections give an overview of various other non-visible connector systems. 2.1 Non-visible mechanical connector systems Most of the joints used in timber structures are mechanical connections. Being exposed, simple fasteners like nails, bolts, drift pins and lag screws, as well as advanced fastener types like shear plates, split rings, truss plates, sheet metal connectors and glulam rivets often present a problem with esthetic and fire protection demands. Embedded or hidden connectors, however, typically meet these requirements. Mechanical connectors, when proportioned carefully, can meet demands for high ductility which is important for equal load distribution and energy absorption. 2.1.1 The BSB-system (Tight-fitting dowel connection) Tight-fitting dowel connections consist of high-quality steel dowels and embedded steel plates. Driven into undersized pre-drilled holes, the dowels are kept in place by friction. In addition to being esthetically more pleasing, tight-fitting dowels are further distinguished from bolted connections by higher strength values and better failure performance. The dowel press fit prevents initial slip and guarantees a stiff connection as well as a more uniform load distribution. inoduo Non-v i s ib le h e a v y t i m b e r jo in ts 8 Figure 11: BSB-joints in heavy timber truss Figure 12: Scheme of BSB-connection Requiring highly accurate fabrication, Computer Numerically Controlled-equipment more and more substitutes the time-consuming and difficult manual labor of precisely drilling the dowel holes and machining the slots for the steel plate. Manufacture and assembly of the joint is mostly done in the shop under controlled conditions. Depending on the size, smaller components of the structure are prefabricated, then brought to the construction site and finally completed by connecting a limited number of joints. During fabrication, transport and assembly, moisture fluctuations of the wood must be strictly avoided. The BSB-connection is a highly optimized tight-fitting dowel connection that was developed in Switzerland (Mischler 2 0 0 0 ) . It is officially approved in many European countries and has been applied in many heavy-timber structures. Figure 13: BSB-system applied in a roof Figure 14: Footbridge constructed structure of a spa with BSB-connection irvoduo Non-v is ib le h e a v y t i m b e r jo in ts 9 2.1.2 The SpikeTec-system (Embedded nail plate) Invented by MERK1 and officially approved in Germany (DIBt2 2002), the SpikeTec-system is a further development of the truss-plate connection. It is mainly applied for large trusses and is typically used with composite structural lumber. The connection consists of the SpikeTec-connector - a steel plate with a double-sided set of spikes welded perpendicular to its surface - sandwiched between a pair of timber members. Figure 15: Close-up of SpikeTec-connection Figure 16: Nail plates are set in place The steel plate is 10mm thick; the spikes have a length of 50mm and a diameter of 5mm. LVL and glued-laminated timber are commonly used for the t r u s s members. In the manufacturing process, the nail plate is pressed into the flanging side members. To meet higher fire protection requirements, both timbers have to be countersunk by 5mm in the nail plate area. The wood thus completely encloses the connector. Optimized design enables the SpikeTec-connection to carry 50% more load parallel to grain than a conventionally bolted joint of the same size. Due to compact joint dimensions and reduced member cross-sections, the construction of large timber trusses has become competitive with structural steel. 1 M E R K H o l z b a u , A i c h a c h , G e r m a n y . L e a d i n g c o n t r a c t o r for h e a v y - t i m b e r c o n s t r u c t i o n s 2 D I B T : D e u s t c h e s Inst i tut fu r B a u t e c h n i k (German Institute for Construction Technology) irroduo Non-v is ib le h e a v y t i m b e r jo in ts 10 2.2 Non-visible composi te connector systems Unlike many mechanical connectors, composite joints with Epoxy, Resorcinol or non-shrink grout glued-in steel components, provide high stiffness and strength, low tolerances, easy fabrication, and for completely embedded connectors systems, good fire protection. The main disadvantages are brittle failure modes of the glued connection and deterioration of strength properties due to climate changes and poor quality of the glue bond. While glued wood-to-wood connections are common in traditional joinery, larger structures have typically relied on mechanical fasteners. This is mainly to facilitate construction and / or assembly on site. The advantage of high-performance glued incduo Non-v is ib le h e a v y t i m b e r jo in ts 11 connections can be combined with mechanical on-site connection methods by gluing metal connectors to the wood. Failure modes can then also be controlled by assuring a weak link in the steel element. 2.2.1 Glued-in steel rods In the past 30 years several researchers have investigated means of transferring high loads from wood members to steel rod elements. In the 1960s and '70s Scandinavian engineers (Riberhoit 1988) conducted initial research on inserting steel rods into predrilled and oversized holes filled with Epoxy or Resorcinol glue. Placed parallel to grain, this composite joint was the first successful application of gluing steel to timber. In Russia (Turkowskij 1991 3) ribbed steel bars were glued perpendicular to the grain in places where the Glulam was subjected to excessive bearing forces. Later, reinforcement bars were inserted at a 30° angle to reinforce the timber members for high shear stresses. In the late 1980s, extensive research was conducted at UBC (Madsen 1998), to develop a reliable glued-in steel rod connection. Madsen phrased guidelines to meet state-of-the-art performance requirements. They are: - High Strength - High Stiffness - Avoid brittle failure - Tolerate reverse loading - Loads transferred via specified path - Simplicity of design - Ease of manufacturing - Construction friendly - Attractive appearance - No field gluing - No field welding - Provide for corrosive environment (if needed) - Fire protection - MC of wood members less than 15% - Cost 3 R e s e a r c h h a d b e g u n a l ready in 1975 , h o w e v e r , r e m a i n e d u n k n o w n for the rest of t he w o r l d till 1 9 8 9 , b e c a u s e pub l i ca t ions w e r e in R u s s i a n . r irxxJuo Non-v is ib le h e a v y t i m b e r jo in ts 12 Investigating the performance of the wood to glued-in steel rod connection in general, in terms of strength, stiffness, different sizes and lengths of the rods, as well as the joint's behavior with rods perpendicular and at an angle to the grain, Madsen came up with a basic connection design suitable for various applications. Recess for plate Figure 17: Glued-in steel rod connection (Madsen) Anchor plates with pre-welded rods on one side are inserted in epoxy glue filled holes of the timber member. In the structure, these composite members are then connected to each other with bolts. Engaging a larger portion of the cross-section, the use of angular rods was found to increase load-bearing and shear capacities of the wood member. Figure 18: Applications for glued-in steel rod connections (Madsen) inoduo Non-v i s ib le h e a v y t i m b e r jo in ts 13 f Madsen Canada Buchanan Fairweather New Zealand A Turkovskij Russia Riberholt Denmark Figure 19: Research conducted on glued-in steel rod connections Many researchers worldwide have raised concerns about the sensitivity of the epoxy glue joint to climatic changes. The loss of strength and stability due to changing moisture contents and temperatures, as well as a required minimum temperature for the adhesive reaction, limits the use of epoxy for outside applications or structures with strict fire protection requirements. Kangas (2000) conducted fire resistance tests of epoxy glued-in V-form steel rod connections. To avoid premature failure of the joint, he found that all fire exposed steel parts have to be covered with rock wool and steel sheets. Due to severe loss of strength when heated above 50°C, epoxy glue was substituted by cement grout (Buchanan and Eiststetter 2000). While easy to handle, inexpensive and fire resistant, however, cement grout's poor adhesion to the timber represents a major problem, therefore requiring a mechanical bond. Reinforcement with pins and screws, driven into to the wood member before grouting, is one way to create a strong connection with good fire resistance. Figure 20: Sketch of test specimens (before grouting) (Buchanan) inoduo Non-v is ib le h e a v y t i m b e r jo in ts 14 The TiSCo®-system Building on the knowledge of more than 30 years of research experience in the field of glued-in steel connections, the German TiSCo®-system (Timber-Steel-Connector) represents a new type of composite connector. The tubular shape of the connector, providing a larger surface area to carry load and the use of an easy-to-handle, strong and temperature-tolerant vinyl-ester based compound mortar distinguishes this connector system from most other glued composite connections. The connector, which consists of a mild steel tube (125mm long, 48mm outer diameter and 3mm wall thickness) and a steel plate welded to one end, is inserted into the end grain of a wooden member. Featuring a threaded hole (M16), the steel plate acts as connector head, which can easily be connected to adjacent elements of the structure. To provide enhanced adhesion, the surface of the tube is sandblasted or ribbed. In addition, four longitudinal slots over most of the tube's length are meant to reduce residual stresses due to deformations of the wood. Figure 21: TiSCo-connector: Sandblasted and grooved version Figure 22: TiSCo-connector (Schreyer) inserted in end grain (Schreyer) The connection is manufactured in three steps: 1. Drill a circular hole, including a 20mm countersink to accommodate tube and head of the connector 2. Inject mortar into the hole and 3. Insert the connector with a twisting action, distributing the viscous mortar all over the glue splice. irxxJuo Non-visible heavy timber joints 15 After a short hardening time of approximately 10 minutes, the squeezed-out excess mortar can be removed; 1 hour later, loads can be applied. Figure 23: Manufacturing steps of the TiSCo-connection (Schreyer) Bathon and Schreyer (2000) investigated strength and stiffness properties of the TiSCo-connector. Under tensile loading, the connection fails abruptly with a withdrawal of the connector along the mortar surface. Under compression, after a failure of the mortar bond, wood compression causes an increase of loading capacity with ductile failure characteristics. Preliminary research showed that exposure to changing climates may result in a reduction of the tensile load capacity of the connection. Although, extensive research was conducted, TiSCo has never been tested and used in full-size timber structures, thus remaining a prototype. 2.2.2 The BVD®-system Successfully utilized in numerous heavy-timber structures, the BVD-system, developed and marketed by German engineer Peter Bertsche, provides high connection strength and stiffness properties in the longitudinal direction of the loaded member. The BVD-system consists of a cylinder-shaped main connector hanger with inside thread inserted into the end grain of a member, a large number of drift pins located perpendicular to the connector's longitudinal axis and a non-shrink cement grout that compounds wood and inoduo Non-v is ib le heavy t imber jo in ts 16 steel parts creating a composite joint. The recess for the main connector as well as the holes that accommodate the drift pins are predrilled and generously oversized to provide enough play for a uniform distribution of the cement grout poured into the voids between wood and steel parts. Figure 24a: Components of the BVD-system (Bertsche) Figure 24b: Different hanger sizes in various length configuration (Bertsche) The following set of pictures show the manufacture of the BVD-Joint: Figure 25: Manufacturing the BVD-connection (Bertsche) inoduo Non-v is ib le h e a v y t i m b e r jo in ts 17 The four steps of manufacturing and installing the BVD-joint are as follows: 1. Drill the pin holes. 2. Drill the large main connector hole in the end-grain of the member. 3. Insert connector hanger first, followed by the drift pins that interlock with the connector. Adjust and fix position of the connector. 4. Cover the surface of the member with plastic foil to protect the wood from being stained and pour cement grout into special feed openings. The grout is cured after 12 hours. Ready to be installed in the structure, however, the member is not to be fully loaded for another 12 hours. After 28 days the grout is completely cured, providing the maximum strength. The BVD-system has been widely applied in many heavy-timber structures all over the world. Over 250 major projects, including wooden bridges, large span roof structures and various custom timber constructions have been built using the high-strength BVD-connector. In addition, in the area of reconstruction and renovation of historic timber structures the system recently found a further field of application. fl Figure 26: Knee joint with BVD-connection (Bertsche) inoduo Non-v is ib le h e a v y t i m b e r jo in ts 18 For the largest timber structure ever been designed and constructed, the EXPO-Roof built for the world exposition in Hannover, Germany in 2000, the BVD-system was applied in one of the main joints that connect the tower columns with the foundation of the structure. Due to the size and the number of the column members, the BVD-joints were processed on a CN-controlled machining center. drift pins Bolts connected to main member Figure 27: Tower column with BVD-connection to foundation Figure 28: EXPO-Roof with towers supporting the roof structure inoduo M e t h o d s a n d Mater ia ls 19 3 Mater ia ls and M e t h o d s Chapter 3 describes the methodology of the research work on investigating the tensile performance of the INDUO-connector. As mentioned in the introductory chapter, limited knowledge on the strength and performance of both connector type A and type B and a set of various problems resulted in a subdivision of the practical research into two test series. Section 3.1 specifies the different wood materials used in the construction of the test members, including how these wood products are manufactured, and their material characteristics. Furthermore, section 3.1 introduces the different connector types. Section 3.2 describes in detail the scope of the research, the connection design, the fabrication of the test specimens and the set-up of the test apparatus. 3.1 Materials Microllam®LVL, Parallam®PSL, TimberStrand®LSL, Douglas-Fir plywood and Douglas-Fir lumber are the different wood materials used in this research project. 3.1.1 Microllam®LVL Laminated Veneer Lumber is an engineered wood product created by layering dried, graded and adhesive-coated wood veneers into blocks of material. Rotary-peeled on a lathe, the veneer is typically produced in thicknesses of 2.5, 3.2, and 4.2 mm. The adhesive used in Microllam®LVL is phenol formaldehyde, continuously applied to the veneer sheets by passing under a glue-curtain. Layered with the grain running in the lengthwise direction and specifically located in the veneer block to assure optimized strength properties, the laminations are cured in a heated press, fabricating a continuous billet. After exiting the press, the billet is sawn to standard dimensions, either 610 mm or 1220 mm wide and 19 mm to 89 mm thick and is finally stored to cool down. Dependent on customer orders, stocked L V L is ripped and cut to the required length in a separate line. irxxluo M e t h o d s and Mater ia ls 20 Veneer End Cutting ,-"~..a Ultrasonic - " ^ i Grading Shipping Grading ^jliyiicrollarti Figure 29: Manufacture of LVL (TrusJoist) Figure 30: Close-up of LVL With a consistent moisture content, LVL is virtually free from warping and splitting and can be easily worked using conventional woodworking tools. Compared to common lumber products, due to defect removal and dispersal, LVL as a solid, highly predictable and uniform wood product offers higher reliability and lower variability. For the research project LVL of grade 1.9E was used. One important benefit of LVL is that the veneering and gluing process creates large timbers from underutilized species of small trees. Besides the most common species Douglas Fir, Southern Pine and Spruce, in North America, Aspen and Yellow Poplar are increasingly being used. irroduo Methods and Materials 21 3.1.2 ParallanTPSL Representing a more recent development of structural composite lumber, Parallel Strand Lumber is another lengthwise oriented structural wood product, created by layering dried, adhesive-coated veneer strands parallel into blocks of material. Similar to the manufacture of LVL, the veneer is typically produced from Douglas Fir, Southern Pine or Yellow Poplar and either rotary-peeled to a veneer ribbon of 2.5 and 3.2mm thickness at the plant or purchased and delivered to the plant. The adhesive used in Parallam®PSL is resorcinol or phenol formaldehyde with a small admixture of wax to avoid moisture absorption of the composite. In the manufacturing process veneer is clipped to strands between 12.5 to 25mm width and up to 2.4m length. In a sorting machine strands shorter than 300mm are removed. After being coated with adhesive in an immersion bath, the strands are dried and then passed through a distribution system, where density and strength of the finished product is set by controlling the mass flow. Being layered and aligned approximately parallel to the product axis, the strands are gathered in a conveyor hutch to form a continuous billet of required mass per length. The strand mat is slowly fed into the press, which applies pressure for densification and cures the adhesive using microwave energy. Figure 31: Manufacture of PSL (TrusJoist) inoduo M e t h o d s a n d Mate r ia l s 22 After cooling, finished billets of up to 279 x 483 mm in cross-section are ripped, cut and sanded to required dimensions. For handling reasons, the billet is cut to lengths of up to 22m. Because it is a continuous process any length is theoretically possible. Finally, the end grain of each finished member is treated with a sealant to avoid moisture absorption. Figure 32: Close-up of PSL Figure 33: PSL applied in heavy-timber structure (Forestry building, UBC) Besides its good workability and high strength, the unique and appealing parallel grain structure of PSL satisfies esthetic demands and is often left exposed as a design element. Independent of the wood species of the strands, PSL is generally provided in a 2.0E grade. In terms of recovery and efficiency, the manufacture of Parallam®PSL utilizes 64 percent of a log, whereas traditional sawmilling processes convert only 40 percent of a log into lumber. irvoduo M e t h o d s a n d Mater ia ls 23 3.1.3 TimberStrancTLSL Laminated Strand Lumber is one of the latest developments in engineered wood product technology. It is also a strand-based product with fiber orientation slightly more random than PSL. In the manufacturing process, around 75 percent of a log of low-density hardwood species such as Yellow Poplar, Aspen and Cucumbertree is utilized. The adhesive used in TimberStrand®LSL is polymeric diphenylmethane diisocyanate (MDI) with a small admixture of wax to avoid moisture absorption of the composite. LSL is created by cutting the log into fine strands of 0.75 to 1.3mm thickness, roughly 25mm width and up to 300mm length. After drying and removing short pieces, the strands are conveyed to blenders where they are coated with adhesive and wax. Aligning the strands approximately parallel to the product axis, a 2.4 m wide continuous mat of specified mass is formed and cut to the appropriate pressing length. In the press the mat is densified and cured with injected steam, creating a composite with minimal density variation throughout its thickness. After exiting the press and cooling, the LSL billets with a rough size of up to 140mm thickness, 2.4m width and 11m length are sanded, ripped and cut to final dimensions, for structural material ranging from 32mm to approximately 100mm thickness. Strander Figure 34: Manufacture of LSL (TrusJoist) ircduo Methods and Materials 24 Although LSL is manufactured as panels, it is mainly used as linear elements, such as rimboards and studs. In terms of manufacturing, LSL is similar to Oriented Strand Board (OSB), except that OSB is conventionally hot pressed and LSL strands are longer and more-or-less parallel aligned, thus enhancing bending and axial strengths in the main direction. LSL is available in 1.3E, 1.5E and 1.7E grades. 1.5E grade was used for the research project. 3.1.4 Douglas-Fir plywood Plywood and LVL were originally developed in the 1930s for the manufacture of wooden airplane propellers and other high-strength aircraft parts. Since the 1950s as a substitute for solid wood sheathing, particularly in North America, plywood rapidly advanced to a highly deployed construction material. Although losing much of its market share to OSB, it remains one of the most important engineered wood products. The fabrication of plywood is similar to LVL, except that the grain direction of sequential veneer sheets is alternated and the layer set-up is symmetrical to the centerline. Dependent on structural or non-structural, exterior or interior application, different plywood grades are available. Typically Douglas-Fir or other softwood veneers and waterproof formaldehyde adhesives are used to build up the panel. In the manufacture, plywood panels are produced to sizes of up to 6 x 12m, then ripped and cut to standard dimensions of 1.2 x 2.4m; thicknesses range from 12.5mm to 38mm. inoduo M e t h o d s a n d Mate r ia l s 25 Veneer lathe Venoer cutter Dryer •Veneer sorting for inner and outer plies Glue application Panel layup Assembled panels placed between heated platens in press _^L/~^>C E d9 e t r i m m i r , g . patching, r - sanding and grading Bundling and shipping Figure 36: Manufacture of plywood (CWC) For the research project plywood of 50 and 60mm thickness was needed. Since standard plywood is limited a maximum thickness of 25.4mm, a set of one 12.5mm and two 19mm as well as a pair of two 25.4mm thick panels respectively were glued-laminated in the shop to build up the required panel dimension. Providing grade A on both faces and grade C1 for all inner plies, the 12.5mm, 19mm and 25.4mm thick panels consisted of 5, 7 and 9 cross-plies, respectively. Figure 37: Close-up of thick Douglas-Fir plywood incduo M e t h o d s and Mate r ia l s 26 3.1.5 Douglas-Fir lumber Contrary to Europe, where timber is individually cut to customer demands, in North America sawmills mostly produce standard dimensions. In addition, dimension lumber sizes are expressed in nominal imperial units. A nominal 2x4 lumber member (pronounced 'two-by-four') for example has a cross-sectional area of 38 x 89mm. For the test specimen manufacture nominal 3x6 Douglas-Fir lumber (64 x 140mm) was directly purchased from a sawmill, since building material 180kN. LSL Code Max Load [kN] cracking noises Failure conments TA1-100 124.1 >60kN ductile bearing and splitting perp. to strands TA2-100 112.7 > 50kN ductile bearing and splitting perp. to strands TA1-120 132.7 > 80kN ductile bearing and splitting perp. to strands TA2-120 130.9 > 70kN ductile bearing and splitting perp. to strands TB1-100 177.6 >120kN brittle failure of glueline TB2-100 > 180.0 >145kN no failure of glueline TB1-120 > 180.0 no no TB2-120 > 180.0 no no TB3-120 > 180.0 >160kN no Table 3: Analyzed data of test series 1 inoduo Results 42 X-LVL Code Max Load [kN] cracking noises failure comments XA1-100 104.2 > 45kN ductile bearing and splitting perp. to grain XA2-100 108.2 > 60kN ductile bearing and splitting perp. to grain XA1-120 127.2 > 50kN ductile bearing and splitting perp. to grain XA2-120 119.1 > 40kN ductile bearing and splitting perp. to grain XB1-100 174.2 >125kN brittle net section failure at last pair of pins XB2-100 163.8 >130kN brittle net section failure at last pair of pins -XB1-120 > 180.0 no no XB2-120 > 180.0 >140kN no XB3-120 > 180.0 no no DG Fir Code Max Load [kN] cracking noises failuro comments DA1-100 99.4 > 55kN rather ductile bearing and splitting DA2-100 82.0 > 50kN rather ductile bearing, splitting and group tear-out DA1-120 96.7 >70kN rather ductile bearing and splitting DA2-120 95.4 > 60kN rather ductile bearing and splitting DB1-100 165.2 > 90kN brittle violent failure, row shear-out along rows of holes DB2-100 165.8 >105kN brittle violent failure, splitting, deformed pins DB1-120 > 180.0 >140kN no close to failure, heavy crack, noises DB2-120 > 180.0 >120kN no close to failure, heavy crack, noises DB3-120 155.5 >90kN brittle splitting along rows of holes LVL Code Max Load [kNl cracking noises ! failure comments MA1-100 81.3 >45kN brittle bearing, splitting and group tear-out MA2-100 72.0 >35kN rather ductile bearing, splitting and group tear-out MA1-120 ' 76.0 >30kN rather ductile bearing and splitting MA2-120 81.9 > 50kN rather ductile bearing and splitting MB1-100 139.9 >70kN brittle splitting along rows of holes MB2-100 169.7 >100kN brittle splitting with group tear out, deformed pins MB1-120 145.0 >105kN brittle splitting along hole line, deformed pins MB2-12Q 167.6 >120kN brittle splitting along hole line, ripped-off pins MB3-120 163.9 > 110kN brittle splitting and group tear-out, deformed pins Table 3 (continued): Analyzed data of test series 1 inoduo Resu l ts 43 PSL Code Max Load [kN] cracking noises failure comments PA1-100 90.0 >40kN brittle bearing, splitting and group tear-out PA2-100 98.9 >50kN brittle bearing, splitting and group tear-out PA1-120 100.5 > 60kN rather ductile bearing, splitting and group tear-out PA2-120 87.0 > 4 5 k N rather ductile bearing, splitting and group tear-out PB1-100 143.6 >100kN brittle splitting, group tear-out, deformed pins PB2-100 166.1 >110kN brittle splitting, deformed pins PB1-120 171.3 >105kN rather ductile splitting along rows of holes, deformed off pins PB2-120 157.5 > 90kN brittle splitting along rows of holes, ripped-off pins PB3-120 151.0 >110kN rather ductile splitting along rows of holes, deformed off pins Table 3 (continued): Analyzed data of test series 1 4.1.1 Group I (TimberStrancfLSL & Douglas-Fir plywood) Figure 52: Group I: No failure for connector type B Figure 53: Group I: Severe bearing; a typical failure combinations mode for connector type A combinations In terms of strength and stiffness, combinations of connector type B and LSL or X-LVL outperformed all other specimens of the test series. None of the larger 120x120mm cross-sections could be failed at the maximum test load of 180kN. Damage to the fiber structure or deformation of the connector was not observed (Figure 52). While loading the test specimen, cracking noises indicating failure propagation, were rarely observed. Some of the smaller type B dimensions (100x100mm), however, failed. While both had a similar stiffness and high ultimate loads, LSL specimens typically failed due to a poorly inoduo Resu l t s 44 fabricated glue bond, whereas some X-LVL members experienced a sudden net section failure of the wood around the last pair of pins. In both cases failure was not governed by the connection itself. All connector type A combinations failed before reaching the 180kN test limit. After linear-elastic behavior up to an average of 95kN, the connection showed a comparatively ductile performance, before most of the specimens finally failed in a brittle manner caused by splitting perpendicular to the strands and the veneer layers, respectively. Investigating the failure area, the wood structure of the spike holes was significantly damaged due to excessive bearing (Figure 53). Displacements of more than 150% of the spike diameter were measured. The average ultimate load determined for LSL specimen was 125kN and 115kN for X-LVL. Interesting data was collected from the screw-laminated members S-TA1-100 and S-TB1-100. Compared to their glued-laminated counterparts, the screw-bonded specimens seemed to perform similarly or even better. Providing above average stiffness, S-TB-100 was not failed. The type A specimen presented the highest ultimate load of all TA combinations and remained on a high level before failing rapidly in splitting and shear (Figure 55). incduo Resu l t s 45 Figure 55: Load-displacement curves of TA-combinations incduo Resu l t s 46 Displacement [mm] Figure 56: Load-displacement curves of XB-combinations Figure 57: Load-displacement curves of XA-combinations irroduo Resu l ts 47 4.1.2 Group II (Douglas Fir & Microllam®LVL) Figure 58: Group II: Splitting along the rows of Figure 59: Group II: Bearing and relatively large holes; a typical failure mechanism of connector displacements due to shear failure of connector type B combinations type A combinations Two major trends were observed in this group: • Combinations with connector type A showed relatively ductile failure characteristics after reaching the maximum load with a comparatively low average ultimate load of 77kN for LVL and 93kN for Douglas Fir. • Unlike type A specimens, connector type B combinations provided higher stiffness and ultimate strength, but typically failed in a very brittle and abrupt manner, providing an average ultimate load of 157kN for LVL and 169kN for Douglas Fir. In general, severe splitting and shear failure along the rows of holes caused the sudden failure. In addition, Type A specimens sustained significant bearing and group tear-out. The pins of failed type B connectors were deformed and showed cracks at the pin base (Figure 131f, page 107). For type A connectors no evidence of damage to the connector was observed. Although two out of the ten type B specimens (DB1-120 and DB2-120) survived, severe cracking noises just before reaching the limit load and a delayed failure of DB2-120 15 seconds after the loading had stopped, indicated that 180kN is near the ultimate load. incduo Resu l t s 48 3 4 5 Displacement [mm] Figure 60: Load-displacement curves of DB-combinations 180 160 140 120 100 I 80 TJ BJ 3 60 40 20 0 3 4 5 Displacement [mm] ™^-DA1-100 -©-DA2-100 -H-DA1-120 - & - D A 2 - 1 2 0 /if V Al/ ( )v r i i i i i i i Figure 61: Load-displacement curves of DA-combinations inoduo Resu l t s 49 180 , Displacement [mm] Figure 62: Load-displacement curves of MB-combinations 180 160 140 120 100 Displacement [mm] -O -MA1-100 -e— MA2-100 -S -MA1-120 -A-MA2-120 Figure 63: Load-displacement curves of MA-combinations incduo Resu l ts 50 4.1.3 Group III (Parallam® PSL) Figure 64: Group III: Bearing and tear-out of connector Figure 65: Group III: Connector type B type A combinations combinations with severe pin deformations or fracture As evident from the inconsistent performance for both type A and type B combinations, the connection's performance was governed by the local interaction of pins or spikes and the strand structure and was thus relatively random. The cavities between the strands seemed to act like pre-occurred damage to the wood structure. Therefore, an above average accumulation of these cavities in the area of a pin or spike hole led to premature crack propagation, resulting in splitting and a brittle failure mechanism. Investigating the connection region of type B members, it was found that specimens with a more interwoven strand layout performed in a more ductile manner, whereas members with strictly parallel alignment had failed in a brittle way. The advantageous interlocking of strands seemed to have less effect on type A connections. In general, type A specimens failed due to severe bearing followed by shear-off along the rows of holes. Approaching the ultimate load level, often entire strand segments were torn-out (Figure 64). With heavy deformations and partially ripped-off pins, connection type B's typical failure mechanism was splitting along the rows of holes, dependent on aforementioned parameters resulting in very brittle or more ductile behavior. The average ultimate loads determined for types A and B specimens were 94kN and 157kN, respectively. inoduo Resu l ts 51 180 Disp lacement [mm] Figure 66: Load-displacement curves of PB-combinations 3 4 5 Disp lacement [mm] Figure 67: Load-displacement curves of PA-combinations inoduo Resu l t s 52 4.2 Test series II As mentioned earlier, test series 2 consisted of a larger number of specimens with selected attributes. This provided a more representative statistical data base for those connection types that proved to be the most promising for practical applications. The following combinations were tested: S-MB-100 MB-100 S-PB-100 PB-100 S-TB-100 TB-100 S-DB-100 DB-100 Due to a more detailed data set developed for each setup, the specimens were no longer grouped according to observed behavior, but evaluated in their individual connector and material categories. LSL Code cmckmg noises |KN| failure comments (TB1-100) 177.6 >120kN brittle failure due to bad glueline (TB2-100) 180.0 >145kN no failure of glueline, no connection failure TB3-100 > 180.0 no no TB4-100 > 180.0 >170kN no TB5-100 > 180.0 no no TB6-100 > 180.0 >170kN no TB7-100 > 180.0 >160kN no TB8-100 > 180.0 no no TB9-100 > 180.0 no no TB10-100 > 180.0 >170kN no (S-TB1-100) > 180.0 no no S-TB2-100 > 180.0 no no S-TB3-100 > 180.0 no no S-TB4-100 > 180.0 >160kN no S-TB5-100 > 180.0 >170kN no Table 4: Analyzed data of test series 2 incduo Resu l t s 53 DG Fir Code Max Load [kN] crack ing noises fai lure comments ( D B 1 - 1 0 0 ) 165.2 > 1 2 0 k N bri t t le v io lent fai l . , row shear -ou t a long rows of ho les ( DB2-100 ) - 1 6 5 . 8 > 1 1 0 k N brit t le v io lent fa i lure, spl i t t ing, d e f o r m e d p ins DB3-100 177.6 > 1 4 0 k N brit t le sp l i t t ing, g r o u p tear -ou t DB4-100 > 180.0 > 1 3 0 k N no h e a v y c r a c k i n g no ises , c lose to fa i lure DB5-100 143.5 > 9 5 k N bri t t le ear ly c r a c k i n g no ises , spl i t t ing DB6-100 130.9 > 1 1 0 k N bri t t le ab rup t fa i lure , spl i t t ing DB7-100 151.4 > 1 2 0 k N bri t t le ab rup t fa i lure, spl i t t ing DB8-100 > 180.0 > 1 5 0 k N no h e a v y c r a c k i n g no ises DB9-100 > 180.0 no no DB10-100 > 180.0 > 1 7 0 k N no S-DB1-100 171.3 > 1 2 5 k N bri t t le ab rup t fai l . , sp l i t t ing, par t ia l net sec t ion fai l . S-DB2-100 173.3 > 1 3 0 k N bri t t le ab rup t fa i lure, spl i t t ing, r ipped-of f pin S-DB3-100 > 180.0 > 1 2 0 k N no h e a v y c rack , no ises , c lose to fa i lure S-DB4-100 > 180.0 n o no S-DB5-100 170.5 > 9 0 k N brit t le v io lent fa i lure, spl i t t ing PSL Code Max Load [kN] c rack ing noises fai lure ' comments ( PB1-100) 143.6 > 1 0 0 k N brit t le sp l i t t ing, g r o u p tear -out , d e f o r m e d p ins ( PB2-100) 166.1 > 1 1 0 k N bri t t le sp l i t t ing, d e f o r m e d p ins PB3-100 141.7 > 1 3 0 k N brit t le ab rup t fa i lure , spl i t t ing, PB4-100 135.9 > 1 2 0 k N bri t t le h e a v y c r a c k i n g no ises , v io lent fa i lure, sp l i t t ing, r ipped-of f pin PB5-100 157.7 > 1 1 0 k N bri t t le sp l i t t ing, g r o u p tear-out PB6-100 152.9 > 1 1 0 k N bri t t le v io len t fa i lure , spl i t t ing PB7-100 166.1 > 1 5 0 k N brit t le sp l i t t ing, s t rong ly d e f o r m , p ins PB8-100 171.5 > 1 0 0 k N bri t t le ab rup t fa i lure, spl i t t ing PB9-100 145.5 > 1 3 0 k N brit t le ab rup t fa i lure, sp l i t t ing, d e f o r m , p ins, f i ssures at p in b a s e PB10-100 151.9 > 1 1 0 k N brit t le v io len t fa i lure , spl i t t ing S-PB1-100 160.5 > 9 0 k N brit t le sp l i t t ing, g r o u p tear -out , d e f o r m , p ins & f i ssu res at p in base S-PB2-100 166.4 > 1 2 0 k N brit t le ab rup t fa i lure, spl i t t ing S-PB3-100 156.8 > 1 0 0 k N bri t t le v io lent fa i lure, sp l i t t ing, row of p ins r ipped-of f S-PB4-100 171.2 > 1 3 0 k N bri t t le v io lent fa i lure , sp l i t t ing, g roup tear -out S-PB5-100 > 180.0 > 0 k N no h e a v y c rack , no ises , c lose to fa i lure Table 4 (continued): Analyzed data of test series 2 incduo Resu l t s 54 LVL Code Max Load [kN] cracking noises [ ! ! ! — ! ~ ! • — failure comments ( MB1-100 ) 139.9 >70kN brittle splitting along rows of holes ( MB2-100 ) 169.7 >100kN brittle splitting, group tear-out, deformed pins MB3-100 142.8 > 80kN brittle abrupt failure, splitting, deformed pins MB4-100 157.3 >140kN brittle abrupt failure, splitting, 3 ripped-off pins, fissures at pin base MB5-100 146.9 >120kN brittle abrupt failure, splitting, partial row shear-out MB6-100 147.6 >130kN brittle abrupt failure, splitting, deformed pins MB7-100 139.7 >105kN brittle splitting, deformed pins MB8-100 139.3 >100kN brittle "slower" failure, splitting, deform, pins MB9-100 147.6 >120kN brittle abrupt failure, splitting, ripped-off row of pins MB10-100 168.9 >120kN brittle violent failure, splitting, ripped-off row of pins S-MB1-100 157.1 >110kN brittle abrupt failure, splitting S-MB2-100 148.9 >100kN brittle splitting, 2 ripped-off pins S-MB3-100 168.9 >120kN brittle splitting, deformed pins S-MB4-100 175.6 >70kN brittle violent failure, splitting, group tear-out S-MB5-100 147.2 >80kN brittle splitting, group tear-out Table 4 (continued): Analyzed data of test series 2 4.2.1 Performance 4.2.1.1 TimberStrantfLSL r « * Figure 68: LSL: No damage observed at the pin holes Figure 69: LSL: No deformations of the connector Similar to the observations made in test series 1 , in terms of ultimate strength, LSL outperformed all other materials. None of the 13 specimens were failed. At the limit load of 180kN the testing was stopped and the members were unloaded. For 6 specimens however, cracking noises were noticed around 160 to 170kN, indicating the beginning of inoduo Resu l t s 55 failure development. For the rest of the sample set no cracking noises or signs of distress were observed. Examining the connection area of the test members after cutting them open, it was found that neither the wood structure nor the connectors were damaged or deformed. 3 4 5 Displacement [mm] Figure 70; Load-displacement curves of TB-combinations 180 3 4 5 Displacement [mm] -0-S-TB1-1OO -O-S-TB2-100 - B - S - T B 3 - 1 0 0 A S-TB4-100 - *~S-TB5-100 Figure 7 1 : Load-displacement curves of S-TB-combinations inoduo Resu l t s 56 As listed in figures 70 and 71, it was evident that the alternative screw-bond of the timber halves did not have an apparent influence on the stiffness performance of the connection. Further evaluation of the test data (Sections 4.2.2 and 8.2) confirmed this. 4.2.1.2 Douglas Fir Figure 72: DG fir: Splitting along Figure 73: DG fir: Shear failure in the plane of the pins rows of holes Even though six of the 13 test specimens survived the 180kN load limit, for almost all Douglas Fir members heavy cracking noises were observed. Failure typically occurred very abruptly. Splitting along the rows of holes and shear-off along the plane of the pins caused extremely violent and brittle failures. Comparing and analyzing screw-bonded and glued-laminated specimens, it was found that the alternative screw-lamination did not have a significant impact on the failure itself. Without exception, both types of lamination presented very similar failure mechanisms with failure in the plane of the pins. Investigating the failure areas of the screw-bonded specimens, no evidence was found that the screws influenced or contributed to the overall tensile strength and performance of the connection. incduo Resu l ts 57 Displacement [mm] Figure 75; Load-displacement curves of S-DB-combinations inoduo Resu l t s 58 Figure 76: PSL: Splitting along the Figure 77: PSL: Deformation of the connector pins rows of holes Except for one specimen, member combinations with PSL failed due to splitting along the rows of holes before reaching the 180kN test limit. While loading the member, cracking noises indicated failure propagation, leading to a sudden but less violent failure. For most of the connectors strong deformations and fissures at the base of the pins were found; some pins were ripped-off. Similar to Douglas Fir specimens, glued-laminated and screw-bonded members performed similarly. incduo Resu l t s 59 - - 0 - S - P B 1 - 1 O O - e - S - P B 2 - 1 0 0 - B - S - P B 3 - 1 0 0 A S - P B 4 - 1 0 0 • - > r ~ S - P B 5 - 1 0 0 0 1 2 3 4 5 6 7 8 Displacement [mm] Figure 7 9 ; Load-displacement curves of S-PB-combinations irvoduo Resu l ts 60 4.2.7.4 MicrollarrfLVL Figure 80: LVL: Splitting along rows of holes Figure 81: LVL: Ripped-off pins Before reaching the limit load, all LVL specimens mostly failed due to splitting along the rows of holes. Similar to the observations made for Douglas Fir, the LVL combinations failed in an extremely violent and brittle manner. Typically, the connector pins were strongly deformed and partly ripped-off. Glued-laminated and screw-bonded members showed similar behavior. Figure 82: Load-displacement curves of MB-combinations inoduo Resu l ts 61 Figure 83: Load-displacement curves of S-MB-combinations inoduo Resu l ts 62 4.2.2 Strength and Stiffness Since glued-laminated and screw-bonded test specimens performed similarly under tensile loading, it was decided to merge both sets of samples. Therefore by providing a larger sample size of 15 specimens, statistically more significant values for the characteristic strength and stiffness could be calculated. Tables 5 and 6 display information on average ultimate strength and displacement for connector type B combinations. For specimens that were not failed, it was assumed that their ultimate load equated 180kN. Ultimate Load [kN] LSL DG Fir PSL LVL min value > 180.00 130.88 135.91 139.30 max value > 180.00 > 180.00 > 180.00 175.62 mean > 180.00 168.63 157.85 153.16 std dev 15.27 12.62 12.30 COV 9.1% 8.0% 8.0% Table 5: Statistics on ultimate load Displacement at ultimate load [mm] LSL DG Fir PSL LVL min value 0.91 1.12 0.6 0.72 max value 2.77 2.45 2.41 2.4 mean 1.57 1.71 1.43 1.49 std dev 0.50 0.43 0.56 0.57 COV 32.0% 25.1% 39.2% 38.3% Table 6: Statistics on displacement at ultimate load To determine the stiffness S of a connection, typically equation 4.1 is chosen to calculate the specific stiffness properties. S= 1 0 h ) (4.1) ^ 4 0 - ' ' l O where: incduo Resu l t s 63 S = St i f fness of the jo in t [ N / m m ] F 4 0 = S t reng th p roper ty at 4 0 % of the u l t imate load [N] F 1 0 = S t reng th p roper ty at 1 0 % of the u l t imate load [N] d 4 0 = D i s p l a c e m e n t at 4 0 % of t h e u l t imate load [ m m ] d 4 0 = D i s p l a c e m e n t at 1 0 % of the u l t imate load [ m m ] Due to inconsistent stiffness performance at the beginning of the loading process, it was found that the 10%-40% approach does not precisely represent the actual stiffness of the connection. To provide more accurate values, stiffness was determined by using the 30%- and 70%-ultimate load points. 180 Figure 84: Different approaches to determine the connection stiffness Figure 84 shows typical load-displacement curves of the test series, presenting a "softer" (DB5-100) as well as a "stiffer" behavior (PB8-100) at the beginning of the loading process. Comparing the dashed and solid straight lines indicating different stiffness, it is evident, that the 30/70-approach creates more realistic results than the 10/40-method. incduo Resu l t s 64 For this reason the stiffness of the connection is determined as follows: S = t l l ) " ^ 3 0 (4.2) w h e r e : S = S t i f fness of t he jo int [ N / m m ] F7)> = S t r e n g t h p roper ty at 7 0 % of the u l t imate load [N] F 3 0 = S t r e n g t h p rope r t y at 3 0 % of the u l t imate load [N] d 7 0 = D i s p l a c e m e n t at 7 0 % of the u l t imate load [ m m ] d 3 0 = D i s p l a c e m e n t at 3 0 % of the u l t imate load [ m m ] Statistics on stiffness values calculated using both approaches are presented in table 7, showing that the 30/70 method generally results in a significantly lower variability of values and, except for Douglas Fir, a smaller average stiffness. S t i f f n e s s [N /mm] S - T B / T B S-DB / DB 10/40 30/70 10/40 30/70 min value 79,412 91 ,139 43 ,626 86,747 max value 360,000 " 2 0 0 , 0 0 0 300 ,000 3 1 8 , 7 9 7 ' mean 166,686 135.223 119,752 170.557 std dev 82,695 35 ,740 68,109 70,471 COV 4 9 . 6 % 2 6 . 4 % 5 6 . 9 % 4 1 . 3 % S-PB / PB S - M B / M B 10/40 30/70 10/40 30/70 min value 53,581 96 ,454 39 ,678 62,105 max value 2,293,556 265,9:19 439 ,058 281,529 mean 351,768 •«> 154,228 186,075 ' ' 140,235 std dev 592,150 52 ,960 145,808 v 61 ,473 COV 1 6 8 . 3 % 3 4 . 3 % 7 8 . 4 % 4 3 . 8 % Table 7: Statistics on different 10/40- and 30/70-connection stiffness inoduo Resu l t s 65 Average ultimate strength and stiffness are elementary mechanical properties used to develop a basic understanding of the fasteners behavior. To model the characteristic connection strength, however, typically the lower 5 t h percentiles of the ultimate strengths have to be determined. Looking for the most accurate distribution to generate the 5 t h percentile values, a subroutine of the RELAN6 software was used to fit Normal, Lognormal, 2P- and 3P-Weibull distributions to the ultimate strength data. The software calculated an overall data fitting error for each data set and developed a distribution function that fits a curve to all data points. Applying these fitted functions, the following formulas were used to calculate the 5 t h percentile values of the respective distributions: Normal Distribution: Xp =/J,-k(T (4.3) w h e r e : x P = S t r e n g t h p rope r t y at t he 5t h pe rcen t i l e [N] u = M e a n v a l u e [N] k = Fac to r re la ted to percent i le P, level of c o n f i d e n c e a n d s a m p l e s ize (k = 1.645) a = S t a n d a r d dev ia t ion [N] Lognormal Distribution: A",, = / • < T ' ' J / ' 1 (4-4) w h e r e : x P = S t r e n g t h p r o p e r t y at t he 5t h pe rcen t i l e [N ] Pm = Log m e a n v a l u e a i n = Log s t a n d a r d dev ia t ion zp = S t a n d a r d n o r m a l n u m b e r (z sco re ) for a g i ven percent i le (z P i 0 0 5 = 1.645) a n d (4.5) // l n = l n / / - 0 . 5 - a l n : (4.6) 6 R E L A N : REL iab i l t y AA/alysis s o f t w a r e d e v e l o p e d in the D e p a r t m e n t of Civi l E n g i n e e r i n g at U B C incduo Resu l t s 66 w h e r e : Pin = Log m e a n va lue ai n = Log s t a n d a r d dev ia t ion u = M e a n v a l u e [N] a = S t a n d a r d dev ia t ion [N] Weibull Distribution: jt, = A- 0 + wi{- ln( l - />)]*" ' <4-7> w h e r e : X p = S t reng th p rope r t y at t he 5 t h pe rcen t i le [N] x 0 = Loca t i on p a r a m e t e r ( x 0 = 0 fo r a 2 P - W e i b u l l d is t r ibu t ion) m = Sca le p a r a m e t e r k = S h a p e p a r a m e t e r P = Percen t i l e v a l u e Analyzing the data generated by RELAN, it was found that 3P-Weibull functions provided the best fit on the lower tail of the ultimate strength data set. Table 15 in section 8.2 presents detailed information on the different fitting errors and distribution functions computed by RELAN. With these fitted functions, the following 5 t h percentiles were calculated: 5th percenti le strength [kN] Normal Lognorm. 2P-Weib. 3P-Weib. LSL (180.00) DG Fir 141.98 144.75 138.24 138.24 PSL 135.75 137.98 132.98 135.64 LVL 133.19 133.81 131.07 137.48 Table 8: 5th percentile strengths of respective distributions inoduo Resu l t s 67 Accounting for the short term duration of loading (test specimens were typically failed in 5 to 7 minutes), the 5 t h percentile results were multiplied with a factor KDOi to generate the characteristic values of maximum tensile capacity (Table 9). w h e r e : x P = S t r e n g t h p roper ty at t he 5TH pe rcen t i l e [N] KDOL = Fac to r to a c c o u n t fo r shor t t e r m load ing ( K D O L = 0.8) T k = Charac te r i s t i c v a l u e fo r the m a x i m u m tens i le c a p a c i t y [N] Characterist ic values of maximum tensile capacity [kN] Normal Lognorm. 2P-Weib. 3P-Weib. LSL (144.00) DG Fir 113.58 115.80 110.59 410.59 PSL 108.60 110.39 106.38 108.51 LVL 106.55 107.05 104.85 * : 109.98 Table 9: Characteristic values of maximum tensile capacity The RELAN data fitting subroutine was also used to compute the 5 t h percentile values of the connection stiffness. Similar to the results of the 5 t h percentile of the connection strength, 3P-Weibull distributions provided the most accurate data fit. The following tables present the complete set of results for all distributions and the 10/40- as well as the 30/70-method to determine the individual connection stiffness. 10/40 Normal Lognorm. 2P-Weib. 3P-Weib. LSL 54,615 62,809 51,169 74,200 DG Fir 35,270 41,227 34,229 45,758 PSL 41,946 41,710 37,687 73.608 LVL 25,140 30,231 18,292 38,402 Table 10: 5 percentile of 10/40-stiffness 5th percenti le st i f fness [N/mm] J30770 Nolmal. Lognorm. 2P-Weib. ;3P-Weib. LSL 77,658 82,623 74,769 „85,913 DG Fir 71,800 78,982 68,820 88,012 PSL 84,626 86,900 80,860 101,784 LVL 52,888 60,378 52,027 63,571 Table 11: 5 percentile of 30/70-stiffness incduo D i s c u s s i o n 68 5. D i s c u s s i o n 5.1 Evaluat ion of test results Appraising the findings of both tests series, with respect to member material and connector type, two major conclusions were reached: 1. Material: Superior tensile strength and failure-free performance of type-B test members significantly distinguishes LSL from LVL, PSL and Douglas-Fir lumber. Assuming the 180kN upper bound limit as the capacity of the unfailed specimens, the latter three generally had 30% weaker characteristic strength properties accompanied by brittle failure modes under tensile loading of the joint. In combination with connector type A, LSL furthermore showed an advantageously ductile failure behavior, providing the highest average ultimate load value of all type A member setups. Conclusion: LSL outperforms LVL, PSL and Douglas-Fir lumber. 2. Connector: Connector type B combinations presented high characteristic tensile strength values and failed typically in splitting along the rows of holes with bending, and in some cases rupture of the pins, whereas type A member setups were primarily engaged in bearing and group tear-out of the wood, eventually failing in tension perpendicular to the strands or the veneer layers; Douglas Fir members typically failed in splitting of the wood. In addition, type A combinations presented a 60% lower average ultimate tensile strength. Conclusion: Connector type B is stronger than type A, but causes abrupt and very brittle failures at high ultimate load levels. Figure 85 shows all possible material/connector setups, indicating combinations with weak (white), stronger (light grey) and most beneficial (dark grey) tensile strength properties. Due to the promising performance of LSL-type A combinations in test series 1, this connection setup is specially indicated (light grey dot). inoduo D i s c u s s i o n 69 LSL LVL PSL DG Fir Type A Type B Figure 85: Classification of member setups according to tensile performance The axial resistance of a dowel-type connection with multiple fasteners is primarily dependent on the dimension, the strength, the number, the spacing and the edge distances of the fasteners, as well as the mechanical properties and characteristics of the member material. In the case of the INDUO-connector, except for the loaded edge distance (set to 10d), all fastener related parameters are defined by the connector itself, leaving only the cross-section and the material of the member as variable factors. Thus, the tensile strength of the connection is directly dependent on the quality and strength of the member material. Every wood or wood-based material features man-made and / or natural characteristics that influence the material and connection strength. When the applied load exceeds the capacity of the wood joint, failure typically initiates at the weakest spot of the connection. Solid wood with its non-homogeneous structure contains various weakening characteristics and 'natural defects' such as knots, checks and varying density (growth-rings, late-early wood) that present such weak points. To create a more uniform and less heterogeneous wood-based material, engineered wood products were developed, eliminating major wood defects and evenly distributing minor weakening characteristics over the entire volume of the member. In the manufacturing process, however, where the original fiber structure is partially destroyed, the recreated wood product shows man-made defects that likewise present undesired characteristics. A close look at the end grain of PSL (Figure 86a) reveals relatively large voids embedded in the strand structure (white circles). In addition, PSL strands and LVL plies feature little surface cracks (white box) that derive from the peeling and drying process during the veneer manufacture as incduo Discuss ion 70 well as from bending while forming and pressing the strand mat into a billet (PSL manufacture). Figure 86: Close-up of end grain: a) PSL; b) LVL; c) Douglas Fir; d) LSL LSL's advantageous connection strength properties originate from the high density (0.68g/cm 3 7 ) and uniformity of the material and its interwoven strand structure. In comparison to PSL, the end grain of LSL does not show any visible cavities (Figure 86d) or initial damage to strands, because the thin and flexible LSL strands overlap and bend without creating hollow spaces and surface cracks, thus reducing potential weak spots in the composite structure. Providing a more cross-layered strand orientation than PSL, the interwoven fiber structure of LSL seems to act like inner reinforcement of the mainly parallel aligned composite. This material property is considered to be responsible for the fact that, in 7 Density value provided by manufacturer TrusJoist irroduo D i s c u s s i o n 71 contrast to Douglas Fir, LVL and PSL, none of the LSL specimens failed in wood splitting along the rows of holes. In conclusion, LSL's performance is based on high material density and uniformity, accounting for a high embedding strength and the unique strand structure that reduces splitting of the material. 5.2 Compar ison of character ist ic s t rength values For the most common fasteners used in contemporary wood construction, timber codes provide the necessary information to calculate and dimension all structural components related to the joint. For connection techniques that have not been introduced into the code, reliable design information has to be supplied by the manufacturer of the fastener. This data is typically generated in specific test series. 5.2.1 Connection Model Due to the costly procedure for an official approval and certification of the INDUO-connector type A, in 2000 it was decided to modify the fastener shape so that it can be modeled and calculated as a "tight-fitting dowel connection with inside steel plate" according to DIN1052-1988 (BlafB 2001). Connector body ( = inside steel plate) Pin (= tight-fitting dowel) Wood (= side member) Figure 87: Connection model With this connection model, the specific design strength for the tensile capacity of the INDUO-connector can be calculated according to any timber code, provided that the respective code includes dowel-type fasteners. The characteristic tensile strengths inoduo D i s c u s s i o n 72 derived from the results of test series 2 can then be compared and verified with corresponding values generated from code design strengths. Using the aforementioned connection model, in section 5.2.2 these design strength values will be determined according to the new German DIN1052-2000 (Draft), the European EC5, the Canadian CSA 086.1 and the US-American ASCE 16-95 timber code. All four codes are based on a Limit States Design philosophy, but vary in detail due to different safety approaches. With a step-by-step approximation of the different code results, the characteristic tensile strength for each design approach will be generated and compared with the data developed from the test series. 5.2.2 Determination of code design values To create a uniform and comparable set of results, values for member dimensions, material properties, service conditions and duration of loading are defined as follows: Connection Model: Dowel-type connection with inside steel plate and wooden side members. Number of dowel-type fasteners: 12 Number of shear planes per fastener: 2 Components of connection: Side members: Wood or wood product: Douglas Fir, No1. & better grade LSL, 1.5E grade LVL, 1.9E grade PSL, 2.0E grade Moisture content: MC < 19% Cross-sectional area: 33 x 100mm Main member: Inside steel plate, spherical cast iron of EN-GIS-500-7 grade defined in EN1563 Characteristic ultimate tensile strength: f u k = 500 N/mm2 Characteristic yield strength: f y k = 320 N/mm2 Cross-sectional area: 14 x 100mm incduo D i s c u s s i o n 73 Fastener: Tight-fitting dowel 8 / bolt 9, spherical cast iron of EN-GIS-500-7 grade, defined in EN1563 Dimension of fastener: d = 8mm, I = 80mm Service condition and duration of loading: Duration of loading: Medium or standard term loading (1 week - 6 months) Service conditions: Temperature: 20° centigrade Relative humidity of the surrounding air: 65%, exceeding 85% only for a few weeks of the year Connection is not exposed to any corrosives Treatment: The wooden side members are not impregnated with any strength reducing chemicals Spacing and distances of fasteners: Parallel to grain direction: a-i = 40.4mm • 5.05d Perpendicular to grain direction: a 2 = 25.0mm • 3.13d Loaded edge: a 3 = 80.0mm • 10.0d Unloaded edge: a 4 = 37.5mm • 4.67d „ 1 3 r- - j CO r o (0 ^ CO o o o o o 33 0 — 33.* 30 0 - 40 4 30 0 • Ll a 3 = 1 0 d 40 4 1-D a , = 5 . 0 5 d • 40 4 • 10 0 10 0 Figure 88: Plan of connection according to DIN1052-2000 and EC5 according to CSA 086.1 and A S C E 16-95 inoduo D i s c u s s i o n 74 Failure modes according to the European Yield Model: EC5, CSA 086.1 and ASCE 16-95 use the European Yield Model (EYM) to describe typical failure modes occurring in dowel-type connections. Based on these failure models the codes provide equations to generate the nominal lateral strength resistances per shear plane and fastener. The failure modes are defined as follows: Failure mode I: B e a r i n g - d o m i n a t e d y ie ld of t he w o o d f ibers in c o n t a c t w i th the f a s t e n e r Failure mode II: F a s t e n e r y ie ld in b e n d i n g at o n e p last ic h inge po in t per s h e a r p lane a n d b e a r i n g - d o m i n a t e d y ie ld of t he w o o d f i be rs in c o n t a c t w i th f a s t e n e r Failure mode III: B e a r i n g - d o m i n a t e d y ie ld of t he m a i n m e m b e r in c o n t a c t w i th the f a s t e n e r (Not c o m p a t i b l e w i th ac tua l p e r f o r m a n c e of I N D U O - c o n n e c t o r ) Failure mode IV: F a s t e n e r y ie ld in b e n d i n g at t w o p last ic h inge po in ts per s h e a r p lane w i th l im i ted loca l i zed c r u s h i n g of w o o d f i be rs near the s h e a r p lanes (Not c o m p a t i b l e w i t h ac tua l p e r f o r m a n c e of I N D U O - c o n n e c t o r ) Figure 89: Failure modes according to European Yield Model Since the failure modes III and IV do not represent the characteristic performance of the INDUO-connector and in addition neither of these failure types was observed in the test series, only mode I and II will be considered in the design calculations. Each of the following sections (5.2.2.1 to 5.2.2.5) presents for the design procedure and the equations used to calculate the axial strength for dowel-type connections according to the respective timber code. 0 | O l O l o I oi IO o o inoduo D i s c u s s i o n 7 5 5.2.2.1 Dim052-2000 (Draft) R, = R. (5.1) w, , • / ! , • / / . (5.2) <-R=j2-fi M^-I),.c' (5.3) provided that (5.4) nun M1 o n \U)d M , =0 .26- /" . d: (5.5) (5.6) A, =0.082 (I-0.01 d) /v£, (5.7) (5.8) w h e r e : F d R d R k kmod Ym n e f n n r n r n s d My, k fh,k Veq Ii fu,k Pk = D e s i g n f o r c e [N] = D e s i g n v a l u e of t he load-car ry ing c a p a c i t y of t he c o n n e c t i o n [N] = Charac te r i s t i c l oad-ca r ry ing capac i t y pe r s h e a r p lane a n d f a s t e n e r [N] = Fac to r a c c o u n t i n g fo r the ef fect of load d u r a t i o n a n d m o i s t u r e con ten t ( k m o d =0 .8 ) = Par t ia l f ac to r for s tee l in t i m b e r c o n n e c t i o n s (Ym = 1.1) = Ef fec t ive n u m b e r of f a s t e n e r s in a row = N u m b e r of f a s t e n e r s in a row = E f fec t i ve n u m b e r of f a s t e n e r s in a row = N u m b e r of r o w s = N u m b e r of s h e a r p lanes = F a s t e n e r s p a c i n g in gra in d i rec t ion [ m m ] = F a s t e n e r d i a m e t e r [ m m ] = Charac te r i s t i c f a s t e n e r y ie ld m o m e n t [ N m m ] = C h a r a c t e r i s t i c e m b e d d i n g s t reng th of t he w o o d [ N / m m 2 ] = R e q u i r e d m i n i m u m t h i c k n e s s of t he w o o d s ide m e m b e r ( t r e q s I, = 3 3 m m ) = E m b e d d i n g leng th of f as tene r in the w o o d s ide m e m b e r [ m m ] = Charac te r i s t i c u l t imate tens i le s t reng th of t he f a s t e n e r [ N / m m 2 ] = C h a r a c t e r i s t i c dens i t y of t he w o o d [ k g / m 3 ] = R e d u c t i o n fac to r a c c o u n t i n g for a , < a ^ r e q = 7 d incduo Discuss ion 7 6 5.2.2.2 EC5 (Eurocode 5) Fd. PH-nt-n, n, • J, (5.16) t T h e lesser p u of m o d e 1 a n d 2: Failure mode I: P.t = o . s - / , • ! (5.19) w h e r e : P r = F a c t o r e d la tera l s t r e n g t h of a bo l ted c o n n e c t i o n [N] = R e s i s t a n c e fac to r (cp - 0.7) Pu = PU(KD- KSF- KJ) p u = La tera l s t reng th res i s tance fo r load ing in g ra in d i rec t ion [N] K T = F i re - re ta rdan t t r e a t m e n t fac to r (KT = 1.0) K S F = Serv i ce cond i t i on f ac to r (KSF = 1.0) K D = L o a d d u r a t i o n fac to r (K T = 1.0) n s = N u m b e r of s h e a r p l a n e s n r = N u m b e r of f a s t e n e r r o w s n F = N u m b e r of f a s t e n e r s in a row JF . = JQ-JL'JR JL JR JG d s f i G f 2 = Fac to r fo r l o a d e d e n d d i s t a n c e (JL= 1-0) = Fac to r fo r n u m b e r of rows ( J L = 0.8) = Fac to r fo r t w o to m a x i m u m 12 f a s t e n e r s in a row = E m b e d d i n g leng th of f a s t e n e r in the w o o d s ide m e m b e r [ m m ] = F a s t e n e r d i a m e t e r [ m m ] = F a s t e n e r s p a c i n g in the row [ m m ] = E m b e d d i n g s t r e n g t h of the w o o d [ N / m m 2 ] (f, =63-G-(1-0.01d)) = M e a n o v e n - d r y d e n s i t y = E m b e d d i n g s t r e n g t h of the ins ide s tee l m e m b e r [ N / m m 2 ] (set to f 2 = 10,000 N / m m 2 ~ inf in i te e m b e d d i n g s t reng th ) = Y ie ld s t r e n g t h of t h e s tee l f a s t e n e r [ N / m m 2 ] incduo D i s c u s s i o n 78 5.2.2.4 ASCE 16-95 7. <, A Z' / (5.20) Z'=Z n, C. C\, C, C\ (5.21) t * T h e lesser Z of m o d e 1 a n d 2: C. = m t Failure mode I: Z = 1.06 z A Z u Z ' z n F n r n Cg C M C T C A R E A E S E M A s Am R e s i s t a n c e fac to r c o n n e c t i o n s (• Lognorm. Normal - Loqnorm. 2P-Weib. 3P-Woib. LSL (180100) LSL (144.00) DG Fir 141.98 144.75 138.24 138.24 DG Fir 113.58 115.80 110.59 110.59 PSL 135.75 137.98 132.98 135164 PSL 108.60 110.39 106.38 108.51 LVL 133.19 133.81 131.07 137S48 LVL 106.55 107.05 104.85 109)98 Table 15: RELAN data fitting of strength data set RELAN Data Fitting (Error of Distribution Fit) WHO Normal Loqnorm 2P-Woib. 3P-Weib LSL 0.5306 0.2042 0.3935 0.1346 DG Fir 0.3923 0.0656 0.2408 0.0547 PSL 1.8531 1.1627 1.5708 '• >\ 0.6728'. , .-LVL 2.1072 0.6898 1.0657 0r48331 Parameters of Fitted Distribution 10/40 « S « i i N o r r n a l Lognorm. 2P-Weib 3P-Weib. mean std. dev. mean std. dev. loe m k loe m k LSL 152,205 59,325 166,861 91,702 0 179,855 2.3629 68,005 103:262 1.0557 DG Fir 109,185 44,933 121,628 74,088 0 129,734 2.2292 391332 86,376 1.1431 PSL 149,511 65,389 211,144 197,957 0 192,074 1.8238 73,562 93 711 . 0.3895 LVL 121,317 58,466 182,918 193,393 0 182,318 1.2918 34.438 1381028 0.8366 5th percentile stiffness [N/mm] Characteristic value of maximum stiffness (10/40) 10140 Normal Lognorm. 2P-Wiib. 3P-Weib. mmmom. Normal Lognormal 2P-Weibull 3P-Weibull LSL 54,615 62,809 51,169 74.200 LSL 43,692 50,248 40,935 59,360 DG Fir 35,270 41,227 34,229 451758 DG Fir 28,216 32,981 27,383 36,606 PSL 41,946 41,710 37,687 731608 PSL 33,557 33,368 30,149 58,886 LVL 25,140 30,231 18,292 381402 LVL 20,112 24,185 14,634 30,721 RELAN Data Fitting (Error of Distribution Fit) 30/70 Normal Lognorm. 2P-Woib. 3P-Weib. LSL 0.1106 0.0791 0.1229 0.0666"' • DG Fir 0.2077 0.0050 0.1789 0.0228 PSL 0.2076 0.1145 0.2321 0 0493 LVL 0.1732 0.0267 0.1251 -. 0.0255 Parameters of Fitted Distribution 30/70 Normal Lognorm. 2P-Wcib. 3P-Weib. mean std. dev. mean std. dev. loe m k loe m k LSL 133,047 33,671 134,782 37,716 0 145,792 4.4479 74.887 67,382 1.6409 DG Fir 163,499 55,744 171,960 75,428 0 185,950 2.9882 79,608 > • 99,783 1.2004 PSL 148,250 38,677 152,587 48,879 0 162,911 4.2402 99,326 ' , * 55,783 0.9514 LVL 134,873 49,839 142,501 68,877 0 154,646 2.7265 52J263 98,391 1.3729 5th percentile stiffness [N/mm] 30/70 Normal* Bllbqnorm.' 2P?Weib. : 3P-Weib.«-LSL 77,658 82,623 74,769 85.913 DG Fir 71,800 78,982 68,820 88;012 PSL 84,626 86,900 80,860 101,784 LVL 52,888 60,378 52,027 6j>B71 Table 16: RELAN data fitting of stiffness data set inoduo Appendices 111 Metric units Imperial units (inch, lbs, psi) PSL, 2.0E l)psii|n strength (factored resistance) • No group rcriuctiun factor No rRs i s t . ince .•' modification factor Adjustment of embedding strength Clididctersitii; vdlue for tonsils capacity Charact. tensile strength of wood (N/mm2): F t 15.5 2,250 DIN 1052-2000 74,378 105,551 116,106 *> 116,106 MOE of wood (N/mm2): E 13,790 2,000,000 ECS ' 89,323 1111,019 Equivalent spec, gravity p S G 0.50 0.50 — * i lu.u jy of wood (a/cm") CSA 086.1 23,528 54,332 77,617 94,795 r . . . . — !H,/SS Embedding length main member (mm): L2 14.0 0.551 ASCE 16-95 69,257 73,400 112,923 Embedding length side L1 33.0 1.299 * member (mm): Pin diameter (mm): d 8.0 0.315 Spacing axial (mm): a1 40.4 1.591 Spacing perp.(mm): a2 25.0 0.984 End distance loaded edqe (mm): a3 80.0 3.150 Edge distance (mm) a4 37.5 1.476 DIN 1052-2000 CSA 086.1 Factor accounting for DOLand MC: k mod 0.8 Resistance factor: PHI 0.7 Partial factor for steel in timber connections: gamma m 1.1 Embedding strength wood (N/mm2): f 1 29.0 Charact. ultim. tensile strength of steel dowel (N/mm2): f u,k 500 Specific gravity / mean oven-dry density (g/cnf1): G 0.5 Charact. yield moment of steel dowel (Nmm): M y,k 35.6B9 Embedding strength steel main member ( - infinite) (N/mm2): f 2 10000 Charact. embedding strength wood (N/mm1): fh,D,k 37.72 Bolt yield strength (N/mm2): f y 320 Charact. embedding strength mod (N/mm2): f h,k 32.04 Number of shear planes: n s 2 Modification factor spacing / emb. strength: k a 0.85 Number of fasteners: n F 12 Factor for effective n: n ef 4.23 Factor for 2 to 12 fasteners in a row: J G 0.54 Factor for laoded end distance: J L 1.00 Charact. Load-carrying capacity per shear plane, per fastener (N): R k . : 6,047 Factor for number of rows: J R 0.80 J G * J L * J R = J F 0.43 ECS 11 = 28.4 11 = 36.97 Factor accounting for DOL and MC: k mod 0.8 Failure Mode I Lateral resistance per shear plane, per fastener (N): p U 6,121 8,155 Partial factor for wood and wood composites: gamma m,w 1.3 Failure Mode II Lateral resistance per shear plane, per fastener (N): p U 3,234 3,950 Partial factor for steel in timber connections: gamma m 1.1 KD = KT= KST= 1 p U = P U Charact. ultim. tensile strength of steel dowel (N/mm2): fu.k 500 f 1 = 38.61 A Charact. yield moment of steel dowel (Nmm): M y,k 34,133 t r • Charact. embedding strength wood (N/mm2): fh,0,k 37.72 Charact. embedding strength mod (N/mm2): fh.k 32.04 ASCE 16-95 Modification factor spacing / emb. strength: k a 0.85 Resistance Factor Connection: PHI z 0.65 Factor for effective n: n ef = n 6 Time effect factor: lamda 0.80 Dowel bending yield strength (psi): F yb 60,000 Failure Mode 1 Design value of the load-carrying capacity per shear plane, per fastener (N): R d 5,205 L m (in): L 2 0.55 Failure Mode II Design value of the load-carrying capacity per shear plane, per fastener (N): R d 722 L s (in): L 1 1.30 Failure Mode II Lateral resistance per shear plane, per fastener (gamma m =1) (N): R d 4,585 Embedding strength steel main member ( - infinite) (psi): F em 1,450,000 Embedding strength wood side member (psi): F es 5,600 L m / L s = R t 0.42 F em / F es = R e 258.93 Imperial units Metric units (N) Failure mode I: Nominal lateral design value for a single fastener (lbs): Z l 3,804 16,921 Failure mode II: Nominal lateral design value for a single fastener (lbs): Z l l 2,644 11,763 Factor failure mode II: k 3 0.56 MC factor: C m 1 Temperature factor: C t 1 Group action factor C g 0.944 MOE steel (psi): E m 30,458,000 MOE wood (psi): E s 2,000,000 R EA R E A 0.155 X-section main member (in2): A m 2.170 X-section side members (in2): A s 5.115 spacing (in): s 1.590 Slip modulus in dowel-type wood-to-steel connections: gamma 47,725 u 1.004 m 0.912 Number of fasteners in a row: n 6 Number of fast, rows: n r 2 Tot. number of fast.: n F 12 Table 17: Calculation of characteristic strength values; PSL inoduo Appendices 112 Imperial units {inch, lbs, Metric units psi) Douglas Fir ' '• No.1 & better Design slienylh ^(factored ; resistance)' No group reduction factor ("Kg" = 1) No lesistanrc1 modification f 5.998 7.808 Partial factor for steel in timber connections: gamma m 1.1 Failure Mode II Lateral resistance per shear plane, per fastener (N): p U 3,189 illiiililtiitii! 3,831 Charact. ultim. tensile strength of steel dowel (N/mm2): fu.k 500 KD = KT = KST = 1 p U = P U Charact. yield moment of steel dowel (Nmm): M y.k 34,133 f 1 = 36.97 Charact. embedding strength wood (N/mm2): fh,0,k 36.97 Charact. embedding strength mod (N/mm2): fh.k 31.40 ASCE 16-95 Modification factor spacing/ emb. strength: k a 0.85 Resistance Factor Connection: PHI z 0.65 Factor for effective n: n ef = n 6 Time effect factor: lamda 0.80 Dowel bending yield strenqth (psi): F yb 60,000 Failure Mode I Design value of the load-carrying capacity per shear plane, per fastener (N): Rd 5,101 L m (in): L 2 0.55 Failure Mode II Design value of the load-carrying capacity per shear plane, per fastener (N): R d . C- 3,672 L s (in): L1 1.30 Failure Mode II Lateral resistance per shear plane, per fastener (gamma m =1) (N): R d 4,521 Embedding strength steel main member (~ infinite) (psi): F em 1,450,000 Embedding strength wood side member (psi): F es 5,500 L m / L s = Rt 0.42 F em / F es = R e 263.64 Imperial units Metric units (Nj Failure mode I: Nominal lateral design value for a single fastener (lbs): Z l 3.736 16,619 Failure mode II: Nominal lateral design value for a single fastener (lbs): Z l l 2,609 ' 11,604 Factor failure mode II: k 3 0.56 MC factor: C m 1 Temperature factor: Ct 1 Group action factor: C g 0.926 MOE steel (psi): E m 30,458,000 MOE wood (psi): E s 1,522,896 R E A R EA 0.118 X-section main member (in2): A m 2.170 X-section side members (in2): A s 5.115 spacinq (in): s 1.590 Slip modulus in dowel-type wood-to-steel connections: gamma 47,725 u 1.005 m 0.901 Number of fasteners in a row: n 6 Number of fast, rows: n r 2 Tot. number of fast.: n F 12 Table 18: Calculation of characteristic strength values; Douglas Fir incduo Appendices 113 Imperial units (inch, lbs, Metric units psi) LVL, ' at Design strenqth (factored resistance) No group ; No lesistance / Adjustment ol reduction factor : modification factor ' embedding ("Kg" = 1) ("PHI"=1) strength Charactersitic value fur tensile capacity Charact. tensile strength of wood (N/mm2): F t 12.4 1,805 DIN 1052-2000 74,378 105,551 116,106 116,106 MOE of wood (N/mm2): E 13,100 1,900,000 ECS 89,323 - * • 110,039 110,039 Equivalent spec, gravity of wood (q/cm2) p SG 0.50 0.50 CSA 086.1 23,528 54,332 77,617 94,795 94,795 Embedding length main member (mm): L2 14.0 0.551 ASCE1635 69,034 73,400 112,923 112.923 Embedding length side member (mm): L1 33.0 1.299 Pin diameter (mm): d 8.0 0.315 Spacing axial (mm): a1 40.4 1.591 Spacing perp.(mm): a2 25.0 0.984 End distance loaded edqe (mm): a3 80.0 3.150 Edge distance (mm) a4 37.5 1.476 DIN 1052-2000 CSA 086.1 Factor accounting for OOL and MC: k mod 0.8 Resistance factor: PHI 0.7 Partial factor for steel in timber connections: gamma m 1.1 -Charact. ultim. tensile strength of steel dowel (N/mm2): f u,k 500 Embedding strength wood (N/mm2): f 1 29.0 Charact. yield moment of steel dowel (Nmm): M y.k 35,669 Specific gravity./ mean oven-dry density (g/cm2): G 0.5 Charact. embedding strength wood (N/mm2): fh,0,k 37.72 Embedding strength steel main member ( - infinite) (N/mm2): f 2 10000 Charact. embedding strength mod (N/mm2): f h,k 32.04 Bolt yield strength (N/mm2): f y 320 Modification factor spacing / emb. strength: k a 0.85 Number of shear planes: n s 2 Factor for effective n: n ef 4.23 Number of fasteners: n F 12 Factor for 2 to 12 fasteners in a row: J G 0.54 Charact. Load-carrying capacity per shear plane, per fastener (N): R k 6,047 Factor for laoded end distance: J L 1.00 ECS Factor for number of rows: J R 0.80 J G * J L * J R = J F 0.43 Factor accounting for DOL and MC: k mod 0.8 11 = 28.4 f1 = 36.97 Partial factor for wood and wood composites: gamma m,w 1.3 Failure Mode I Lateral resistance per shear plane, per fastener (N): p U '6,121 8.155 Partial factor for steel in timber connections: gamma m 1.1 Failure Mode II Lateral resistance per shear plane, per fastener (N): p U 3,234 3.950 Charact. ultim. tensile strength of steel dowel (N/mm2): f u,k 500 KD = KT = KST = 1 p U= P u Charact. yield moment of steel dowel (Nmm): M y,k 34,133 f 1 = 38.61 Charact. embedding strength wood (N/mm2): fh.O.k 37.72 Charact. embedding strength mod (N/mm2): f h,k 32.04 ASCE 1635 Modification factor spacing / emb. strenqth: k a 0.85 Resistance Factor Connection: PHI z 0.65 Factor for effective n: n ef = n 6 Time effect factor: lamda 0.80 Dowel bending yield strength (psi): F yb 60,000 Failure Mode I Design value of the load-carrying capacity per shear plane, per fastener (N): R d 5,205 L m (in): L 2 0.55 Failure Mode II Design value of the load-carrying capacity per shear plane, per fastener (N): R d 3,722 L s (in): L 1 . 1 30 Failure Mode II Lateral resistance per shear plane, per fastener (gamma m =1) (N): R d 4,585 Embedding strength steel main member (~ infinite) (psi): F em 1,450,000 Embedding strength wood side member (psi): F es 5,600 -L m / L s = R t 0.42 F em / F es = Re 258.93 Imperial units Metric units (N) Failure mode I: Nominal lateral design value for a single fastener (lbs): Z l 3,804 16.921 Failure mode II: Nominal lateral design value for a single fastener (lbs): Z l l 2,644 Factor failure mode II: k 3 0.56 MC factor: C m 1 Temperature factor: Ct 1 Group action factor: C g 0.941 MOE steel (psi): E m 30,458,000 MOE wood (psi): E s 1,900,000 R EA R EA 0.147 X-section main member (in2): A m 2.170 X-section side members (in2): A s 5.115 spacing (in): s 1.590 Slip modulus in dowel-type wood-to-steel connections: gamma 47,725 u 1.004 m 0.910 Number of fasteners in a row: n 6 Number of fast, rows: n r 2 Tot. number of fast.: n F 12 Table 19: Calculation of characteristic strength values; LVL inoduo Appendices 114 Metric units Imperial units (inch, lbs, psil LSL, ! oL n