Key Words Gusset plates, uniform force method, ductility, frame action, concentrically braced frames, hollow sections

Transcription

1 Heavy Brae Gusset Plate Connetions for Braed Steel Frames Author: Alistair Fussell Affiliation: Steel Constrution New Zealand In. Date: 20 th July 2011 Ref: CON1301 CONNECTIONS Key Words Gusset plates, uniform fore method, dutility, frame ation, onentrially raed frames, hollow setions Introdution Heavy rae gusset onnetions for the purposes of this paper are defined as gusset plates that onnet large tues (irular and square) or ukling restrained raes used in raed frames (onentri and eentri) for medium and high rise appliations. This inludes oth orner and midspan gusset plate onnetions (figure 1). While the guidane in this paper pertains prinipally to gusset plate onnetions, the priniples will also e appliale to UC rae/eam/ olumn onnetions. Referene is made to HERA Design and Constrution Bulletins (Clifton 1998, 2000, and 2001) that over this situation. Figure 1: A Conentrially Braed Frame Featuring Hollow Setion Braes and Mid-span and Corner Gusset Plate Connetions In onentrially raed frames the rae elements are the primary seismi resisting elements that dissipate energy y yielding or ukling. As a result of this and other differenes in the inelasti ehaviour of CBF ompared to EBF systems, gusset plate onnetions for CBF systems are sujet to greater dutility demand, oth in-plane and out of plane than for EBF systems. This paper egin with a onsideration of the seismi ehaviour of gusset onnetions ased on a review of past and urrent researh programmes into the seismi ehaviour of gusset plate onnetions for CBF systems. Various equilirium models inluding the Uniform Fore Method will also e disussed for alulating the design ations on the interfaes of gusset plate rae/eam/olumn onnetions. The Uniform Fore Method inludes a Dislaimer: SCNZ and the author(s) of this doument make no warrantee, guarantee or representation in onnetion with this doument and shall not e held liale or responsile for any loss or damage resulting from the use of this doument Steel Constrution New Zealand In

2 restrition that makes it hard for designers to always ahieve ompat gusset plate onnetions. A revised formulation to the traditional UFM proedure whih allows greater range of plate geometries is disussed. One of the requirements for seismi design to setion 12 of the Steel Strutures Standard (SNZ, 1997), is that the onnetions must possess adequate dutility to sustain the inelasti demands imposed on them during a design level earthquake event. Guidane is provided on how to ahieve dutile ehaviour of gusset plate rae/eam/olumn onnetions. The ehaviour of midspan gusset plates in a seismi event is different to orner gusset plate onnetions. These differenes will e disussed with reommendations given for designing midspan gusset plate onnetions. Seismi Behaviour of Gusset Plate Connetions Introdution The seismi performane of gusset plate onnetions has een the sujet of a long term North Amerian researh programme led y Charles Roeder. In partiular they have foused their attention on the performane of orner and midspan onnetions for onentrially raed frames (CBF) with raes effetive in ompression and tension. The reason for the interest in CBF systems is that the onnetions must e apale of withstanding yli rae ukling ehaviour. This out of plane ehaviour, see figure 2, imposes large rotation demands on the rae and the rae onnetions. Figure 2: A Gusset Plate Connetion Sujet to Out of Plane Rotation (Roeder and Lehman, 2008) Corner gusset plate ehaviour has een reasonaly well investigated, however there has een little attention given to midspan gusset plates (Roeder and Lehman 2008). The ehaviour of these onnetions are quite different to orner gussets as midspan gusset plates have less edge restraint and they are not sujet to additional ations indued y frame ation. The stress and strain distriution in gusset plates is very omplex. Therefore most design standards provide simple design and detailing rules to approximate this omplex ehaviour and to ensure adequate performane. Summary of Past Gusset Plate Researh Roeder and Lehman (2008) provide a useful summary of past researh into gusset plate onnetions. Prior to the reent North Amerian researh programme led y Charles Roeder, the experiments invarialy involved onnetion tests only and negleted frame ehaviour, and for the most part were monotoni resistant tests rather yli tests fousing on inelasti deformation apaity. A review of the results from past testing has onfirmed the following. Steel Constrution New Zealand In

3 a. Bukling Capaity of Gusset Plate Connetions The traditional approah to deriving the design apaity of a gusset plate in ompression has een to use the Whitmore width onept in onjuntion with an effetive length fator to allow the gusset plate to e treated as a olumn element. For orner gusset plates, Thornton has suggested an effetive length fator of k e = 0.65 is appliale in this situation. NZS 3404 (lause ()) speifies a similar value of 0.7. Comparison of previous test results has shown this method to e onservative (figure 3). Figure 3: Comparison of Measured Gusset Plate Bukling apaity to Predited y Thornton Model as a Funtion of Plate Slenderness Ratio (Roeder and Lehman, 2008). In-plane Staility Unsupported Compression Edge of Plate Design rules have also een proposed for the free length of gusset plates when there are ompressive stresses perpendiular to the free edge of the plate (Brown 1998, Astaneh-Asl, 1989). These were not found to provide reliale estimates of plate ompression apaity limited y edge ukling. An example of edge ukling is shown in figure 8. There was signifiant satter of results. The ratio of atual ultimate loads/predited loads varied from 0.2 to 7 (figure 4). They onluded that none of the methods were supported y the experimental results (Roeder et al., 2008). There is an edge slenderness hek given in the HERA New Zealand Strutural Steel Limit State Design Guide Volume 1 (Clifton, 1994). This is apparently ased on a equation from a British omposite ridge design standard for ridge truss gusset plate design. The author is not aware of any yli experimental testing to verify the use of the equation for seismi appliations. Figure 4: Auray of Edge Bukling Models as a Funtion of Gusset Plate Slenderness (Roeder and Lehman, 2008). Findings/ Reommendations of Roeder et. al Based on the researh to date, Roeder et. al (2008) has suggested the following: Steel Constrution New Zealand In

4 1. Brae end learane model (out of plane dutility for CBF systems: Braes effetive in ompression and tension) The proposed 8t p elliptial learane model provided similar or etter seismi performane while resulting in more ompat onnetions than the urrent 2t model (figure 5). The enefits of more ompat setions in terms of redued joint stiffness are disussed in the next point. Figure 5: Clearane Models for Brae Bukling Deformation a) 2t rule ) proposed 8t elliptial rule 2. In plane Gusset Plate Dutility Gusset plate onnetions must e designed for in-plane dutility. The reason for this requirement is that gusset plate onnetions, figure 7, are relatively rigid for in-plane ending. During severe seismi events, raed frames experiene large deformations that an result in inelasti onnetion demand and loal yielding of eams and olumns (Lehman et al. 2008). In this instane, appropriate apaity design is required to ensure the additional frame ations generated y these deformations, are onsidered in the design of the seondary elements, and the joints are appropriately designed and detailed to ensure dutility. This is more of an issue for onentrially raed frames where the frame deformations will tend to open and lose gusseted onnetions (figure 6a). In ontrast for eentrially raed frames the frame displaed shape when sujet to large inelasti demand will not ause suh large opening and losing ations (figure 6). Further evidene of this phenomenon was oserved y Lopez et. al. (2004). During testing of a full sale two storey frame with gusset plate onnetions, (figure 8a), researhers noted free edge ukling of a gusset plate during yli testing of a ukling restrained rae system (figure 8). The interesting aspet of this ehaviour was that the rae was in tension at the time. The frame interstorey drift when edge ukling ourred was 2.5%. The edge ukling was attriuted to losing of the joint due to frame ation (figure 7). (a) () Figure 6: Inelasti Deformed Frame Shapes a) Conentrially Braed Frame ) Eentrially Braed Frame Researh y Roeder (2008) has shown that the inelasti demand on gusset plate onnetion is suffiient to initiate raks in the orner of welded gusset plates due to the diagonal plate stresses indued y joint opening and losing at storey drifts of aout 2%, see figure 7. Steel Constrution New Zealand In

5 Figure 7: Cross Diagonal Stresses in Welded Gusset Connetions (Roeder and Lehman, 2008) They found if the gusset plate welds were sized for the tensile apaity of the plate, the raks remained stale, allowing the joint to ontinue to dependaly sustain inelasti demand even up to greater interstorey drifts. This raking was noted regardless of the gusset plate design. Sine this was a deformation indued demand, there was little enefit in inreasing plate size or thikness as this only attrated larger stresses. In fat this is a good reason to make the joint as ompat as possile. This requirement to size the gusset welds for the plate tensile apaity is onsistent with the guidane given in the ommentary to NZS (a) () Figure 8: Two Storey Experimentally Tested BRB Frame a) Frame Elevation ) Bukling of Free Edge of Gusset Plate (Lopez et. al., 2004) Steel Constrution New Zealand In

6 3. Midspan Gusset Plates (a) Photograph of midspan gusset ) Comparison of measured ompression to orner gusset data and design models Figure 9: Midspan Gusset Plate Bukling Oserved in 3 storey Full Sale Test (Roeder, 2008) With only one edge restrained, midspan gusset plates have signifiantly less resistant to end rotation and hene the proaility of gusset plate ukling is inreased. The rotational restraint provided to the plate y the eam and the floor system is therefore important to the ukling ehaviour of the plate. They suggest a more onservative approah is required to the design of midspan gusset plates (figure 9). They reommend the effetive length fator k e e inreased to 1.4. Furthermore, vertial stiffeners should e installed and a parrallel 8t p rae end learane model utilised, see figure 10. Figure 10: Proposed 8t Parallel Clearane Model for Midspan Gusset Plate Connetions (Roeder et. al, 2008) 4. Strength Hierarhy Roeder et. al. (2008) proposed a strength hierarhy for CBF rae and gusset plate onnetions to promote dutile ehaviour. This involved suppressing undesirale failure modes suh as plate frature olt and weld frature in favour of more dutile limit states as plate yielding. Analysis of Gusset Plate Connetions Introdution The first step in designing orner gusset plate onnetions is to estalish the design ations. Over the years a numer of simple methods, all satisfying stati equilirium, have een used for estalishing plate support fores suh as the Uniform Fore Method, Truss Analogy, Parallel Fore Method and the KISS method. Of these, the method adopted in the AISC Manual of Steel Constrution (2005) for determining fores on the gusset interfaes is the Uniform Fore Method. Of the methods onsidered at the time, this was found to provide the est means for the safe and eonomi design of diagonal rae onnetions (Muir 2008). The Uniform Fore Method is ased on the lower ound theorem. This theorem states: If a distriution of fores within a struture (or onnetion, whih is a loalised struture) an e found, whih is in equilirium with the external loads and whih satisfies the limit states, than the externally applied loads is less than or at most equal to the load that would ause onnetion failure. In other words, as long as suffiient dutility is present and all Steel Constrution New Zealand In

7 appliale limit states are satisfied, design an proeed ased on any aritrary distriution of fores, as long as the distriution satisfies equilirium. Desription of the Uniform Fore Method The goal of the Uniform Fore Method is to otain a statially admissile fore distriution whih would produe no moment at the onnetion interfaes and would e appliale to a wide range of geometries (Muir 2008). In the asene of moments, these onnetions may e designed for shear and normal fores only. For the fore distriution show in the free ody diagrams, figure 11, to remain free of moments on the onnetion interfae, the following expression must e satisfied. tan e tan e (eqn. 1) One and have een determined, the fatored axial and shear fores at the gusset interfaes an e omputed using the following equations. H V P r e P r H e r P V P r where: r 1 2 e 2 2 e The asis for the uniform fore method is that the fores at the gusset to olumn interfaes, 2 2 H V, and gusset to eam interfaes, respetively, see figure H V, are required to pass through working points at (o, e ) and (e, o) Steel Constrution New Zealand In

8 Figure 11: Uniform Fore Method Equilirium Model Beam/Brae/Column Connetion (Clifton, 1998) This aritrary requirement to pass the gusset to olumn interfae fore through a working point at (o, e ) an result in oddly shaped or disproportionally large gusset plates. Muir (2008) has suggested a slight modifiation to the Uniform Fore Method that removes this requirement for the gusset to olumn interfae ations to pass through the point (0, e ). It therefore, still provides statially admissile solutions with no moments at the gusset to olumn, gusset to eam and eam to olumn interfaes, while providing solutions for a wider range of plate geometries. The original Uniform Fore Method an still e applied to gusset geometries that do not omply with equation 1. However, if this approah is adopted, moments will e introdued to either one or oth of gusset plate interfaes to ensure equilirium. This ase is desried and illustrated y way of worked example in AISC (2005). Speial ase 1: Modified Working Point There are ost and performane advantages in detailing ompat gusset plate onnetions (Roeder and Lehman 2008). The size of gusset plate an, depending on the onnetion geometry, e redued y allowing eentri working points (figure 12). This noding eentriity will introdue additional ending ations in either or oth of the eam or olumn elements. The eentriity, e, of the line of the rae with respet to the working point of the frame entre lines an e omputed as follows (Tamoli 2010): e e ysin e xos The total moment due to noding eentriity is: M noding * P.e Steel Constrution New Zealand In

9 Figure 12: Non-onentri Working Points (modified Tamoli, 2010) To determine the gusset interfae ations and the frame moments, the following requirements will apply: 1. Moment equilirium aout the eam/ olumn working point must e estalished. 2. There must e no moments at the gusset plate interfaes or the eam/ olumn onnetion. The eam moment is: * M He V The olumn moment aout the frame entreline working point (point A) is: * M e Ve H For equilirium aout working point A: M A H e H e V V e P.e 0 and M M P.e The steps involved in estalishing the distriution of onnetion design ations assoiated with eentri noding are: 1. For the seleted working point alulate e and P*e. 2. Use an appropriate equilirium method to estalish the gusset interfae fores (V,H, V,H ). In this instane y is used in lieu of e and x is used in lieu of e (see figure 12). 3. Calulate olumn and eam moments M and M respetively. 4. Chek equilirium has een estalished. Steel Constrution New Zealand In

10 A range of working point loations an e used for the onnetion shown in figure 12. The use of a node point at loation eliminates any transfer of the horizontal omponent rae fore to the olumn. The distriution of gusset plate interfae fores an e determined y using e =0 in the original or modified Uniform Fore Method. The olumn will e sujet to a moment equal to the vertial omponent of the rae fore ating on the olumn fae. If the olumn length and stiffness are the same aove and elow the floor level, the olumn moment will e shared equally aove and elow the onnetion. i.e. M ol where: M ol V e 2 P os.e The olletor eam will not e sujet to any moment as the working point is loated at the eam entreline. The AISC Steel Constrution Manual (2005) inludes an example for the working point at loation C. When this approah is adopted the gusset interfaes are only sujet to shear fores only. In this instane the gusset interfae design ations are omputed with e and e set to zero. Moments equal to V e and V e are required to re-estalish onnetion equilirium. The distriution of the olumn moment will e the same as for the example with the working point loated at point B. The eam and olumn elements must e designed for the additional flexural ations. Some issues to onsider in identifying a preferred working point loation are: 1. Column sizes are generally governed y axial loading. They an therefore often aommodate additional moments due to noding eentriity without requiring a hange in memer size. 2. It is desirale to minimise tension fores on endplate onnetions, partiularly the gusset to olumn onnetion as this will result in smaller flange tension fores that have to e aommodated y the olumn. 3. Large interfae ations (shear, tension or ompression), impose large shear, tension or earing demands on the wes of the eams or olumns that may govern the sizing of these elements. Speial ase 2: Minimise eam shear reation If the eam fore R (figure 11) is large and ats in the same diretion as V *, it may e desirale to redistriute some or all of it to the olumn onnetion (Clifton 2000). In this ase the gusset plate ats as haunh to transfer the vertial V from the eam to the olumn interfae onnetion. This adjustment of vertial shear fore disrupts equilirium (Muir 2008). If it is preferale to avoid moments at the gusset to olumn and eam to olumn interfaes, an additional moment must e introdued at the gusset to eam interfae. The magnitude of this moment is: M H e V V Note this moment onnetion is easier to ahieve if the gusset plate is welded diretly to the eam rather than an endplate detail. Column Moments Column moments are present even for onentri onnetions. The magnitude of this moment is given y: M,noding max Ve, Ve H e In addition there will e moment due to the eentri eam shear fore equal to: M,e,shear R. e Steel Constrution New Zealand In

11 The moment due to eentri shear loading should e distriuted to the olumn in aordane with NZS 3404 lause When omining the moments due to these effets, aount should e taken of their sign. This moment should e added to the other apaity design derived olumn ations. Design of Gusset Plate Connetions Introdution The geometry of gusset plate onnetions is ditated y the layout of the rae and olletor eam elements, the fastener and weld arrangement and if the rae an ukle, suh as ours in some onentrially raed frames, minimum rae learane requirements. A ommonly used onnetion for heavy raed frames is to slot a hollow setion rae symmetrially over a gusset plate. The rae is often held in position temporarily on site with an eretion olt and site welded in plae. A olted endplate onnets the rae and olletor eam to the olumn, see figure 13. Figure 13: Gusseted Hollow Setion Brae/Beam/Column Connetion The design of the gusset plate, endplate, welds and olts for this onnetion will e overed in the following setions. Another form of heavy rae onnetion whih is similar ut involves a UC rae is shown in figure 14 The detailed design of this type of onnetion is disussed in HERA Design and Constrution Bulletins DCB no. 47 and 56. The methodology is illustrated y way of worked example in DCB no. 61. These Design and Constrution Bulletins are availale free of harge from the HERA wesite, To referene the douments, lik on the Strutural Systems division ta on the HERA home page. Steel Constrution New Zealand In

12 Figure 14: UC Brae/Beam/Column Connetion (Clifton 2000) Designing Gusset Plate Connetions for Dutility Aording to NZS 3404 (SNZ, 1997), onnetions must exhiit dependale strength and dutility in order to maintain the integrity of the strutural system throughout the expeted range of seismi indued deformations determined in aordane with this Standard and the Loadings Standard (SNZ. 2004). The priniple means for ahieving dutility is to promote dutile mehanisms and suppress undesirale modes of onnetion ehaviour. For example, for the onnetion shown in figure 13, the undesirale modes of ehaviour are: 1. Weld failure (gusset plate to endplate or gusset plate to eam) 2. Bolt frature (shear or tension) 3. Column flange tension yielding 4. Column we rippling adjaent to the olletor eam ottom flange 5. Blok shear failure The desirale modes of ehaviour for the same onnetion are: 6. Endplate yielding in tension 7. Enlarging of olt holes The following approah is suggested to ahieve the desired dutile onnetion ehaviour. 1. Design the gusset and endplate for the design ations determined using an appropriate equilirium model suh as the uniform fore method. 2. Size the gusset interfae welds to arry the tensile design apaity of the gusset plate. 3. Suppress mode 3 ehaviour (olt tensile failure) when sizing the endplate. This will e disussed further in setion Ensure the ply material in the olt onnetions does not exeed the maximum value given y the following equation t p,max 0.9 d f where: t p is the endplate thikness d f is the endplate olt diameter This equation is taken from Clifton (2005). It is intended to ensure there is loal oversizing of the olt holes efore the olt experienes shear failure. Steel Constrution New Zealand In

13 5. To protet the olumn flange from tension yielding, the tensile design apaity of the endplate must e less than that of the olumn flange. It is expensive to add tension stiffeners, therefore the endplate should only e the minimum thikness required to transfer the design ations. This requirement will typially e satisfied if the flange is thiker than the endplate. Calulations are required to onfirm this strength hierarhy has een ahieved. 6. Strategies for preventing rippling of the olumn we adjaent to the olletor eam ottom flange inlude: i. Stop the flexile endplate 30mm short of the inside fae of the eam ottom flange (DCB no. 47, Clifton (1998)), see figure 15 ii. Assume frame ation generates a flexural moment in to the olletor eam equal to its setion apaity M s. Chek the olumn we for a onentrated ompression load equal to the ompression apaity of the ottom flange ie. N* = A f f y. The asis for this approah is disussed in Fussell (2010a). Figure 15: Beam/Brae/Column Connetion with Endplate Stopped Short of Beam Bottom flange (Clifton, 1998) Connetion Design Proedure 1. Initial sizing of onnetion In DCB no. 56 Clifton (2000), guidane is given on initial seletion of onnetion omponents inluding olt numers and endplate sizes. Consideration should e given to introduing onnetion eentriity to produe smaller gusset plates, see figure Gusset Plate Design i. Determine gusset plate interfae design ations using an appropriate equilirium model. ii. The gusset plate shall e sized to satisfy the following limit states. a) Blok Shear ) Whitmore Setion Yielding ) Plate Bukling Note the following effetive length fators are appliale for plate ukling: a. Corner gusset k e = 0.7. Midspan gusset k e = 1.4 Vertial stiffeners similar to figure 10 should e used for midspan gusset plates. d) Plate Comined Stresses Steel Constrution New Zealand In

14 The ritial loations for the stress hek are at the gusset plate interfaes. e) Gusset Unsupported Edge Speifi design guidane on limit state a) to e) is provided in Fussell (2010) or the HERA New Zealand Strutural Steel Limit State Design Guide Volume 1 (Clifton 1994). Note the lok shear model in Clifton (1994) has een revised, see Fussell (2010). iii. Size Welds a) Brae to Gusset Weld This should e sized for the apaity design derived rae load. ) Gusset to Column and Gusset to Beam Weld These welds shall e sized for the tensile apaity in kn/mm of the plate. 3. Brae to Colletor Beam Endplate The tension apaity of the end plate must e determined using a yield line proedure. An appropriate proedure is given in step 1A, setion 2.8 of SCI puliation no. 207/95, Joints in Steel Constrution: Moment Connetions (1995). The tension apaity is derived y onsidering three modes of failure for an equivalent tee stu, figure 16. Figure 16: Endplate modes of failure in tension (SCI, 1995) The first two modes involve full or partial flange yielding. Mode 3 ehaviour whih involves olt failure alone is not permitted. The olt apaity N tf must e greater than the tension apaity for modes 1 and 2. Steel Constrution New Zealand In

15 Additional olt prying fores must e aounted for when heking olt apaities. Rather than impliitly alulating the prying fore it reommended the olt fore is inreased y 40%. This is onsistent with the oservation that prying fores an vary from 0 40% depending on the plate geometry (SCI, 1995). The olts should therefore e heked for omined shear tension where: * V * V n * H * 1.4H n where: V* is the gusset interfae shear fore derived using an appropriate equilirium model. H* is the gusset interfae tension fore derived using an appropriate equilirium model. To ensure olts do not failure in a shear mode, the plate thikness t p should e limited as follows: t 0.9d p,max Where d f = diameter of fastener f 4. Column Tension Capaity Chek The olumn flange tension apaity alulation proeeds in a similar manner to the rae to olletor eam endplate. As previously disussed the olumn tension apaity should exeed that of the rae to olletor eam endplate. As a rule of thum the olumn flange thikness should exeed that of the endplate to satisfy the requirement. Speifi alulations are required to hek this. 5. Column We Crippling If the rae to olletor eam to olumn endplate is not left short of the inside fae of the eam ottom flange, the olumn should e heked using the we rippling provisions of NZS3404 (lause 5.13) for a design onentrated load equal to: * N A f where A f f y f y is the ottom flange area. is the yield stress of the olletor eam. 6. Colletor Beam We The olletor eam we should e heked for loal yielding adjaent to the gusset plate. 7. Column We The we of the olumn should e heked for loal yielding adjaent to olted onnetion. Steel Constrution New Zealand In

16 Referenes Astaneh-Asl, A, Simple Methods for Design of Steel Gusset Plates, Proeedings ASCE Strutures Conferene, San Franiso, CA Brown, V.L.S., Staility of Gusseted Connetions in Steel Strutures, A thesis sumitted in partial fulfillment of Dotor of Philosophy in Civil Engineering, University of Delaware, 1988 Clifton, G.C., New Zealand Strutural Steelwork Limit State Design Guides Volume 1, Report R4-80, NZ Heavy Engineering Researh Assoiation, Manukau City, 1994 Clifton, G.C., Design Guidane for Beam/Brae/Column Gusset Connetions, HERA Steel Design and Constrution Bulletin, no. 47, 1998 Clifton, G.C., Design Conepts for Brae/Beam/Column Connetions in Braed Steel Frame Seismi-Resisting Systems, HERA Steel Design and Constrution Bulletin, no. 56, 2000 Clifton, G.C., Design Example no. 61.1, Design of a Brae/Beam/Column Connetion in a Braed Steel Frame, HERA Steel Design and Constrution Bulletin, no. 56, 2001 Clifton C., Semi-Rigid Joints for Moment-Resisting Steel Framed Seismi-Resisting Systems, PhD Thesis, Department of Civil and Environmental Engineering, University of Aukland, 2005 (Also availale as a HERA doument) Fussell, A.J., Design of Bukling Restrained Brae Systems, Steel Strutures Seminar Series, Steel Constrution New Zealand, Manukau City, Marh 2010a Fussell, A.J., Light Brae Cleat Connetions for Braed Steel Frames, Steel Strutures Seminar Series, Steel Constrution New Zealand, Manukau City, Otoer 2010 Lehman, D.E., Roeder, C.W., Herman, D., Johnson, S., and Kotulka, B., Improved Seismi Performane of Gusset Plate Connetions, ASCE, Journal of Strutural Engineering, Vol.134, No. 6, Reston, VA, 2008 Lopez, W.A., Gowie, D.S., Lauk, T.W. and Saunders, M., Strutural Design and Experimental Verifiation of a Bukling Restrained Braed Frame System, Engineering Journal, 4 th Quarter, Amerian Institute of Steel Constrution, 2004 Muir, L.S., Designing Compat Gussets with the Uniform Fore Method, Engineering Journal, Furst Avarter, 2008, Amerian Institute of Steel Constrution, Chiago, Illinois, 2008 Roeder, C.W. and Lehman, D.E., Performane and Behaviour of Gusset Plate Connetions, Amerian Institute of Steel Constrution, North Amerian Conferene, Phoenix, Arizona, 2008 SCI, Joints in Steel Constrution: Moment Connetions, P207, Steel Constrution Institute, Asot, U.K., 1995 SNZ, Steel Strutures Standard (Inorporating Amendments 1 and 2), NZS 3404:1997, Standards New Zealand, Wellington, 2007 SNZ, Strutural Design Ations Part 5: Earthquake Ations New Zealand, , Standards New Zealand, Wellington, 2004 Tamoli, A.R. (editor), Handook of Strutural Steel Connetion Design and Detail, 2 nd edition, MGraw Hill, New York, 2010 Steel Constrution New Zealand In