TIMBER CONNECTIONS by Dr. Ad Leijten TU-Eindhoven The Netherlands Former PT-member

Brussels, 18-20 February 2008 Dissemination of information workshop 1 TIMBER CONNECTIONS by Dr. Ad Leijten TU-Eindhoven The Netherlands Former PT-member

Brussels, 18-20 February 2008 Dissemination of information workshop 2 We can not escape connections THE WEAKEST LINK

Brussels, 18-20 February 2008 Dissemination of information workshop 3 What kind of connections do we use?

Brussels, 18-20 February 2008 Dissemination of information workshop 4 Carpenter connections (not in Eurocode 5) National regulations apply

Brussels, 18-20 February 2008 Dissemination of information workshop 5 Carpenter connections compression forces

Brussels, 18-20 February 2008 Dissemination of information workshop 6 Glued connections (not in Eurocode 5) Structural Finger joints Glued in steel rods National regulations apply

Brussels, 18-20 February 2008 Dissemination of information workshop 7 What do we find in Eurocode 5 Section 8: Connections with metal fasteners Mechanical connections with Dowel type fasteners - Nails, staples, - punched metal plates, - screws, - dowels and bolts Connectors - Shear plates - Split-rings

Brussels, 18-20 February 2008 Dissemination of information workshop 8 Nails < 8 mm (EN14592) definition profiled nails Bolts > 8 mm Screws Dowels > 6 mm

Brussels, 18-20 February 2008 Dissemination of information workshop 9 Punched metal plate fasteners Split-ring connectors 64-104 mm diameter Bolts M12 to M20

Brussels, 18-20 February 2008 Dissemination of information workshop 10 Steel to timber

Brussels, 18-20 February 2008 Dissemination of information workshop 11 Eurocode 5 allows: Design by testing: EN 1075 tests punched metal plates connections EN 1380 tests nails,screws,dowels,bolted conn. EN 1381 - tests on stapled connections. EN 26891 Specimen density selection. EN 28970 Test procedure for connection tests. Design by calculation - Model provided in EN1995-1-1

Brussels, 18-20 February 2008 Dissemination of information workshop 12 Design by calculation - covers

a 3,t a 3,c EUROCODES Brussels, 18-20 February 2008 Dissemination of information workshop 13 Fastener spacing requirements Key: (1) Loaded end (2) Unloaded end (3) Loaded edge (4) Unloaded edge a) a 2 a 2 2 1 a 2 a 2 a 1 a 1 a1 a1 b) α a 4,t α α α a 4,c -90 α 90 90 α 270 0 α 180 180 α 360 (1) (2) (3) (4)

Brussels, 18-20 February 2008 Dissemination of information workshop 14 Fastener spacing requirements (bolts) Spacing and end/edge distances (see Figure 8.7) a 1 (parallel to grain) a 2 (perpendicular to grain) Angle 0 360 0 360 Minimum spacing or distance (4 + cos) d 4 d a 3,t (loaded end) a 3,c (unloaded end) a 4,t (loaded edge) a 4,c (unloaded edge) -90 90 90 < 150 150 < 210 210 270 0 180 180 360 max (7 d; 80 mm) max([1 + 6 sin ) d; 4d) 4 d max([1 + 6 sin ) d; 4d) max([2 + 2 sin ) d; 3d) 3 d

Brussels, 18-20 February 2008 Dissemination of information workshop 15 Spacing requirements determine timber dimensions Decrease member width? a 1 a 2

Brussels, 18-20 February 2008 Dissemination of information workshop 16 Spacing requirements determine timber dimensions k a1 a 1 Only for connectors Split-rings and shear plates a 2 k a1 a 1 k a2 a 2 k a2 a 2 2 2 0 ka1 1 ( ka1) + ( ka 2 ) 1 with 0 ka 2 1

Brussels, 18-20 February 2008 Dissemination of information workshop 17 Laterally loaded axially loaded The strength capacity model for laterally loaded dowel-type-fasteners (based on Johansen (1949) Background: -Structural Education Timber Program (STEP 1)(1994) -Timber Engineering; Thelandersson & Larsen (2003) ISBN 0-470-84469-8

Brussels, 18-20 February 2008 Dissemination of information workshop 18 single shear fasteners multiple shear

Brussels, 18-20 February 2008 Dissemination of information workshop 19 Double shear Embedment stresses

Brussels, 18-20 February 2008 Dissemination of information workshop 20 Starting point for strength model - Embedment strength Embedment = f emb f emb displacement embedment Idealised femb= fv

Brussels, 18-20 February 2008 Dissemination of information workshop 21 Determination of embedment strength EN 383 Eurocode 5 Design clause embedment strength Nails (not pre-drilled) Nails pre-drilled Bolts/dowels f h,k h,k -0,3 2 0,082 ρ k N/mm f = d = 0,082 (1-0,01 d ) ρ N/mm k 2

Brussels, 18-20 February 2008 Dissemination of information workshop 22 Parallel Perpendicular Taken from Sawata and Yasumura (2002). Background: Whale L. and Smith, I. The derivation of design clauses for nailed and bolted joint in Eurocode 5, In Proceedings of paper CIB-W18 paper 19-7-6/ Florenze 1986 Yasumura, M. and Sawata, K., Determination of embedment strength of wood for dowel-type-fasteners. In: Journal of Wood Science, nr. 48, 2002, Japan Wood Research Society, Inst. Of Wood Techn, Akita, Japan

Brussels, 18-20 February 2008 Dissemination of information workshop 23 Embedment test - parallel to grain Dowels Parallel All wood species 120,0 Embedment 100,0 80,0 60,0 40,0 20,0 0,0 200 400 600 800 1000 1200 Density 5mm data 7mm data 8mm data 12mm data 16mm data 20mm data 30mm data Background: Leijten, A.J.M. & Köhler, J.& A Jorissen, Review of Probability Data for Timber Connections with Dowel-Type Fasteners; In Proceedings of CIB-W18, paper 37-7-12, Edinburgh, UK, September 2004

Brussels, 18-20 February 2008 Dissemination of information workshop 24 Re-evaluation parallel to grain embedment results for future consideration in EC5? Nails (pre-drilled) Coniferous Bolts and dowels Coniferous Decideous f f f f f f h;0 h;90 h;0 h;90 h;0 h;90 = 0, 0104ρ = 0,046ρ = 0,097ρ =? = 0,087ρ = 0,106ρ d 1,35 0,27 d 1,48 0,42 d 1,07 0,25 d 1,09 0,25 d 1,13 0,46

Brussels, 18-20 February 2008 Dissemination of information workshop 25 Embedment - perpendicular to grain

Brussels, 18-20 February 2008 Dissemination of information workshop 26 Embedment at angle to grain Bolts/dowel: f h,0,k = 0,082 (1-0,01 d ) ρ k f h,,k = f h,0,k 2 2 k 90 sin α + cos α k 90 1,35 + 0,015 d for softwoods = 0,90 + 0,015 d for hardwoods

Brussels, 18-20 February 2008 Dissemination of information workshop 27 Coniferous: Embedment test - perpendicular to grain Embedding strength, based on average values embedding tests (European Whitewood (EW), European Redwood (ER), Canadian Spruce Pine Fir (CS)) fh;90 (Embedding strength) [N/mm 2 ] 40 35 30 25 20 15 10 Nails Bolts Nails ER Nails EW Nails CS Bolts ER Bolts EW Bolts CS Eurocode 5 Power (All) Power (Eurocode 5) Eurocode 5 5 0 0 5 10 15 20 Nail or Bolt diameter [mm] Needs adjustment in future EC5

Brussels, 18-20 February 2008 Dissemination of information workshop 28 Embedment rules Nails in: Plywood Hardboard Particleboard & OSB f h,k = 0,11ρ d h,k = 30 k f d t 0,3 0,6 f d t h,k = 65 0,3 0,7 0,1 Taken from DIN 1052

Brussels, 18-20 February 2008 Dissemination of information workshop 29 Mode I Mode II Mode III Some failure modes of single shear fasteners Some failure modes of double shear fasteners

Brussels, 18-20 February 2008 Dissemination of information workshop 30 Equations for every failure mode Fasteners in single shear F v,rk fh,1,kt1d (a) fh,2,kt2d (b) 2 2 fh,1,kt1d 2 t2 t2 3 t2 t2 β + 2β 1 + + + β β 1 + (c) 1 + β t1 t1 t1 t1 min h,1,k 1 4 β (2 + β) = f t d M y,rk Fax,Rk 1, 05 2 β (1 + β ) + β + 2 (d) 2 + β fh,1,k d t1 4 fh,1,kt2d 4 β (1 + 2 β) M 2 y,rk Fax,Rk 1, 05 2 β (1 + β ) + β + (e) 1 + 2 2 β fh,1,k d t2 4 2β Fax,Rk 1,15 2 M y,rk fh,1,kd + (f) 1 + β 4

Brussels, 18-20 February 2008 Dissemination of information workshop 31 Equations for every failure mode Fasteners in double shear F fh,1,kt1d (g) 0,5 fh,2,kt2d (h) f t d 4 β (2 + β ) M F = + + + + 2β Fax,Rk 1,15 2 M y,rk fh,1,kd + (k) 1+ β 4 h,1,k 1 y,rk ax,rk v,rk min1,05 2 β (1 β ) β (j) 2 2 β fh,1,kd t1 4

Brussels, 18-20 February 2008 Dissemination of information workshop 32 Some failure modes of single shear fasteners steel to wood connections thick steel plate t > d = fastener diameter thin steel plate t < 0,5d

Brussels, 18-20 February 2008 Dissemination of information workshop 33 Test results still higher than Johansen equations Cord effect: Only valid for Mode II and III Requires knowledge about withdrawal Theory for nails 15% extra Background: Kuipers, J. Van der Put, T.A.C.M., Betrachtungen zum Bruchmechanismus von Nagel verbindungen, In: Ingenieuholzbau in Forschung und Praxis, J. Ehlebeck and G. Steck, editors, Bruderverlag Karlsruhe 1982

Brussels, 18-20 February 2008 Dissemination of information workshop 34 Test results still higher than Johansen equations Cord effect: Fax/4 = withdrawal capacity/4 = estimated effect % are estimates Maximum Nails 15% Grooved nails 50% Screws 100% Bolts 25% Dowels 0%

Brussels, 18-20 February 2008 Dissemination of information workshop 35 Fasteners in double shear F fh,1,kt1d (g) 0,5 fh,2,kt2d (h) f t d 4 β (2 + β ) M F = + + + + 2β Fax,Rk 1,15 2 M y,rk fh,1,kd + (k) 1+ β 4 h,1,k 1 y,rk ax,rk v,rk min1,05 2 β (1 β ) β (j) 2 2 β fh,1,kd t1 4 Test results still higher than Johansen equations

Brussels, 18-20 February 2008 Dissemination of information workshop 36 Mode III theory Design value F F d 2 δ 1+ δ = s h, 1 kmod 0,9 = Fk = Fk = 0, 69 F y 1,3 m 2 M schroef Ms = yield bending moment steel fastener d Partial material factor of timber γm applied to Ms!! Better seperate: f k

Brussels, 18-20 February 2008 Dissemination of information workshop 37 Mode III Taking both γm - seperately F F F F d d d d 4δ = 1+ δ 4δ = 1+ δ k M y M e, k m, s e, k 1,1 4δ = 0,79 1+ δ = 0,79 F d d M schroef schroef e, k d f f h,1, k y m, h h,1, k 1,3 schroef k 0,9 f mod h,1, k factor = 0,79 F 0,69 F k k = 1,15 For Mode II factor is 1,05

Brussels, 18-20 February 2008 Dissemination of information workshop 38 Fasteners in double shear F fh,1,kt1d (g) 0,5 fh,2,kt2d (h) f t d 4 β (2 + β ) M F = + + + + 2β Fax,Rk 1,15 2 M y,rk fh,1,kd + (k) 1+ β 4 h,1,k 1 y,rk ax,rk v,rk min1,05 2 β (1 β ) β (j) 2 2 β fh,1,kd t1 4 Eurocode 5 equations

Brussels, 18-20 February 2008 Dissemination of information workshop 39 Yield moment of dowel type fasteners Large diameter bolts and dowels Small rotation at failure No full plastic yielding M y,rk = 0,3 f u,k d 2,6 R Rotation angle 20.0 18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 Background: Jorissen, A.J.M. Blass, H.J., the fastener yield strength in bending, In: Proceedings of CIB-W18 paper 31-7-6, 1998 0.0 8 10 12 14 16 18 20 22 24 26 28 30 diameter d

Brussels, 18-20 February 2008 Dissemination of information workshop 40 Failure of multiple of fasteners in a row Splitting Shear Tensile

Brussels, 18-20 February 2008 Dissemination of information workshop 41 Failure of multiple of fasteners in a row Caused by group effect Only for load component in grain direction Stress 13d Cumulatieve stress

Brussels, 18-20 February 2008 Dissemination of information workshop 42 Failure of multiple of fasteners in one row Caused by group effect effective number Nails: Empirical (Gehri) n = n Bolts & dowels: Fracture mechanics n ef ef = n k 0, 9 4 ef a 0 13d Spacing a a 1 14d a 1 = 10d a 1 = 7d a 1 = 4d Not predrilled a For intermediate spacings, linear interpolation of k ef is permitted Background: Double shear timber connections with dowel type fasteners, A.J.M. Jorissen, ISBN 90-407-1783-4, DUP Delft, 1998 1,0 0,85 0,7 - k ef Predrilled 1,0 0,85 0,7 0,5

Brussels, 18-20 February 2008 Dissemination of information workshop 43 Be careful with cheese connections

Brussels, 18-20 February 2008 Dissemination of information workshop 44 Steelplate in timber Chees connections Small dowels

Brussels, 18-20 February 2008 Dissemination of information workshop 45 Failure at fastener perimeter (Prof.Racher, Fr.) Block shear: Full penetration Plug shear Partial penetration Tensile or shear failure, which happens first? Literature: Johnson, H, Stehn, L, A Linear Fracture Mecanics Evaluation of Plug Shear Failure,In Proc of 8th world conf on Timber Engineering WCTE 2004, Finland

Brussels, 18-20 February 2008 Dissemination of information workshop 46 F bs,rk 1,5 A = max 0,7 A net,t net,v f t,o,k f v,k Tensile failure Shear fracture Correlation parameters 2 v,1 v,2 v,3 v,4 t,1 t,2 1 Fasteners keep straight Not correct in EC5 see previous sheet v,5 v,6 v,7 v,8 t ef

Brussels, 18-20 February 2008 Dissemination of information workshop 47 Amendment A1: Contains: New design rules for: compressive strength perpendiculer to grain Present rules unsafe 3 1 l l ef

Brussels, 18-20 February 2008 Dissemination of information workshop 48 Amendment A1 Addition Axially loaded screws Traditional screws Diameter thread=smooth shank Not very effective not hardened low bending moment >8mm requires predrilled holes

Brussels, 18-20 February 2008 Dissemination of information workshop 49 Amendment A1 Addition Axially loaded screws Very effective hardened self tapping high axial stiffness Background: Blaß, HJ; Bejtka, I: Self-tapping screws as reinforcements in connections with dowel-type fasteners. In: Proceedings. CIB-W18 Meeting, Karlsruhe, Germany 2005. Paper 38-7-4 Blass H.J. Joints with dowel-type-fasteners, In: Timber Engineering Thelanderson and Larsen, editors, Wiley & Sons (2003):

Brussels, 18-20 February 2008 Dissemination of information workshop 50

Brussels, 18-20 February 2008 Dissemination of information workshop 51 Some examples Leg or coach screw

Brussels, 18-20 February 2008 Dissemination of information workshop 52 Some examples Spax-S Heco Topix/Fix Rapid Komprex

Brussels, 18-20 February 2008 Dissemination of information workshop 53 Tecfi Woodpecker BMF Torx SFS WT

Brussels, 18-20 February 2008 Dissemination of information workshop 54 Check: - Withdrawal failure - Tear-off failure of the head - Pull through of the head - Tensile failure of the screw - Torsional capacity - Group effect (neff number of effective fasteners)

Brussels, 18-20 February 2008 Dissemination of information workshop 55 -Withdrawal: Code proposals in EU countries (1) Eurocode 5 = (1) R 1,5 ( d l ) 0,8 ax, a, k = π ef 2 3 3,6 10 ρk 2 sin a + 1,5 cos a (2) R ax; k = 2 k 2 6 60 10 ρ 2 sin a + cos 4 3 a d l ef (3) (4) ax, k ( 1,5 + 0, d ) ( l d ) ρ R 6 R = nom hec nom ( π d l ) 3 3,0 10 ρk 2 sin a + 1,5 cos 0,8 ax, k = 1,7 1 ef 2 a

Brussels, 18-20 February 2008 Dissemination of information workshop 56 Model uncertainty evaluation of test results Eurocode 5 25 (1) (4) (2) (3) 20 Frequentie.. 15 10 5 0-1 -0,8-0,6-0,4-0,2 0 0,2 0,4 0,6 0,8 1 ln(model/meetpunt) Current EC5 design rule unsafe

Brussels, 18-20 February 2008 Dissemination of information workshop 57 New proposal in Amendment A1: Requirement Screws as defined in EN 14592 6 mm d 12 mm R 0, 9 0, 8 0, 52 d.l ef ρk ax,a,k = nef 2 2 sin a + 1, 2 cos a All other screws: R ax,a,k = n ef sin 2 f ax.d.l ef a + 1, 2 cos 2 ρ a ρ k a 0, 8 Parameters determined by tests EN 14592 (EN 1382)

Brussels, 18-20 February 2008 Dissemination of information workshop 58 - Head tear off - Axial screw strength: Determine by tests: EN 14592 R - Torsional capacity EN 14592 = A ax f ax

Brussels, 18-20 February 2008 Dissemination of information workshop 59 Head pull through EN 14592 Test standard EN 1383 Eurocode 5: R ( ) ( ) ) 1 2 1 2 dk - ds 3, 0 fc,90 k R k = π 2 2, = A ax f ax Zulassung Germany 9.1-235: N k = 16,0 d k 2 In the absents of information Clause 8.5.2. bolt washers

Brussels, 18-20 February 2008 Dissemination of information workshop 60 short elastisch behaviour Large non-elastic traject

Brussels, 18-20 February 2008 Dissemination of information workshop 61 Model uncertainty design rule 9.1-235andn Eurocode 5 for screws with washers 9 Eurocode 5 Zulassung 8 7 6 Frequentie 5 4 3 2 1 0-1 -0,5 0 0,5 1 1,5 ln (model/meetwaarde)

Brussels, 18-20 February 2008 Dissemination of information workshop 62 - Group (tear out) effect: Due to a lack of background information Based on test by Gehri (empirical): n ef = n 0,9

Brussels, 18-20 February 2008 Dissemination of information workshop 63 Splitting by perpendicular to grain forces F Ed α b h e h F v,ed,1 F v,ed,2 b/2 b/2 b a) b) Design clause 8.1.4 (3) is formulated as a maximum shear force criterion on either side of the connection Background: Leijten A.J.M. & Vander Put T.A.C.M, Evaluation of Perpendicular to Grain Failure of Beams caused by Concentrated Loads of Joints, In: Proceedings of CIB-W18, paper 33-7-7, Delft, 2000.

Brussels, 18-20 February 2008 Dissemination of information workshop 64 Splitting by perpendicular to grain forces Design clause 8.1.4 (3) is formulated as a maximum shear force criterion on either side of the connection Fracture mechanics background 14 is calibration parameter w = (political factor) F 90,Rk = 14b w 0,35 wpl max 100 for punched metal plate fasteners w = 1 1 for all other fasteners h e he 1 h

Brussels, 18-20 February 2008 Dissemination of information workshop 65 Splitting by perpendicular to grain forces Some empirical models consider Fastener spacing F k k s r 18a = k skr 6. 5+ 2 h 1 = max 1. 4a 0. 7 + h n = 2 n h 1 i= 1 hi 90,k 2 ( t ef h ) 0. 8 f t,90,k

Brussels, 18-20 February 2008 Dissemination of information workshop 66 Splitting by perpendicular to grain forces Fracture mechanical model Consider energy balance after crack appearance

Brussels, 18-20 February 2008 Dissemination of information workshop 67 Splitting by perpendicular to grain forces UK-Test results versus design proposals Strength [N] 50000 45000 40000 35000 30000 25000 20000 15000 10000 5000 0 0 0,2 0,4 0,6 0,8 1 he/h Test Results Eurocode 5 design Punched metal plate fasteners

Brussels, 18-20 February 2008 Dissemination of information workshop 68 Comparison between models Eurocode 5

Brussels, 18-20 February 2008 Dissemination of information workshop 69 Splitting by perpendicular to grain forces Assumed governing failure mechanism is shear Not by tensile stresses perpendicular to grain For loaded edges >0,7 h no splitting h e h Simply supported max F Cantilever beam max F/2 F h h e F/2

Brussels, 18-20 February 2008 Dissemination of information workshop 70 Frequently heard: Not found in Eurocode We don t know everything Research in progress Future changes and additions are expected