Proceedings of the METNET Seminar 2016 in Castellon by Häme University of Applied Sciences, HAMK

Virdi & Tenhunen (Editors)

Proceedings of the METNET Seminar 2016 in Castellon

METNET Annual Seminar in Castellon, Spain, on 11 – 12 October 2016

ELECTRONIC ISBN 978-951-784-785-8 (PDF) ISSN 1795-424X HAMKin e-julkaisuja 6/2016

Kuldeep Virdi and Lauri Tenhunen (Editors) HAMK

PRINTED ISBN 978-951-784-784-1 ISSN 1795-4231 HAMKin julkaisuja 4/2016

Editors: Kuldeep Virdi, City, University of London Lauri Tenhunen, Häme University of Applied Sciences (HAMK) Proceedings of the METNET Seminar 2016 in Castellon PRINTED ISBN 978-951-784-784-1 ISSN 1795-4231 HAMKin julkaisuja 4/2016 ELECTRONIC 978-951-784-785-8 (PDF) ISBN ISSN 1795-424X HAMKin e-julkaisuja 6/2016

Hämeenlinna, December 2016

INDEX Kuldeep Virdi & Lauri Tenhunen

PREFACE.................................................................................................................................. 4 PAPERS Ondrej Svoboda & Josef Machacek

STABILITY OF A STEEL TUBE ARCH SUPPORTING BARREL TYPE TEXTILE ROOF MEMBRANES........................................................................................................................... 7 Yvonne Ciupack and Hartmut Pasternak

BONDING TECHNOLOGY IN STEEL STRUCTURES

A SUMMARY OF RESEARCH ACTIVITIES AT CHAIR OF STEEL AND TIMBER STRUCTURES.............................. 19

Raimo Ruoppa, Kimmo Keltamäki, Rauno Toppila & Vili Kesti

RESEARCH OF ULTRA-HIGH-STRENGTH AND WEAR-RESISTANT STEELS USING ADVANCED TECHNIQUES...................................................................................................... 31 Kuldeep S. Virdi

STRENGTH OF CIRCULAR HOLLOW SECTION COLUMNS....................................................43 G.M. Gasii

THE FLAT DOUBLE-LAYER GRID-CABLE STEEL-CONCRETE COMPOSITE STRUCTURE...... 56 Jukka Laitinen & Tarja Meristö

APPLYING VISIONARY CONCEPT DESIGN TO ENERGY EFFICIENT RESIDENTIAL AREAS...63 Tarja Meristö & Jukka Laitinen

CLEANTECH BUSINESS OPPORTUNITIES IN THE FUTURE.................................................. 72 ABSTRACTS Markku Heinisuo, Marsel Garifullin, Kristo Mela, Teemu Tiainen & Timo Jokinen

MOMENT-ROTATION BEHAVIOUR OF WELDED TUBULAR HIGH STRENGTH STEEL JOINTS.......................................................................................................................80 Eduardo Bayo

STEEL JOINTS WITH BEAMS OF UNEQUAL DEPTH USING CRUCIFORM ELEMENTS FOR FRAME ANALYSIS.................................................................................................................84 Eduardo Bayo

CRUCIFORM FINITE ELEMENTS FOR THE MODELLING OF STEEL JOINTS IN FRAME ANALYSIS.............................................................................................................................. 86 Mirambell E., Real E. & Chacón R.

RECENT RESEARCH AND DESIGN DEVELOPMENTS IN STEEL AND COMPOSITE STRUCTURES AT UPC ...........................................................................................................87 M. Lozano, M.A. Serrano, C. López-Colina & F. López-Gayarre

CHARACTERIZATION OF RHS-IPE BEAM-COLUMN JOINTS ............................................... 89 Olli Ilveskoski

DIAPHRAGM STABILIZATION OF STEEL BUILDINGS IN FIRE ............................................90

Proceedings of the METNET Seminar 2016 in Castellon

PREFACE David Ricardo developed the theory of comparative advantage in 1817 (Ricardo, 1817). Ricardo stated that if two countries capable of producing two commodities do exchange the commodities in free market, then both countries would increase its consumption by exporting the goods for which it has a comparative advantage while importing goods from the other country. This assumes, however, that there exist differences in productivity between the two countries (Baumol & Binder, 2012). Since 1817, economists have attempted to generalize the theory of comparative advantage in broader settings. In the case of many countries and many commodities, the theoretical examination of comparative advantage requires quite a complex quantitative formulation (Deardorff, 2005). However, according to the new theory of international values, theory of comparative advantage can also be used to explain how global supply chains appear in general (Shiozawa, 2016). Partly intuitively, the participants of METNET collaboration have utilised the benefits of comparative advantage now over ten years, starting from 2006 in Berlin, by changing constantly expertise and knowledge related to metallic construction and industries in Europe (www.hamk.fi/metnet). During the last ten years, METNET-network has actively promoted best practices and enabled transfer of knowhow between companies, research and training organizations at the European level. This collaboration has brought about many positive steps both in theory and in practice. For example, the networking participants have accomplished successful international projects with defined development targets. The participants continue to plan new collaborative activities and projects. The METNET annual conferences and workshops have taken place in different countries and have been organised in cooperation with regional network members. The Department of Mechanical and Engineering Construction, Universitat Jaume I, Castellon, Spain, hosted the 11th METNET International Conference in October 2016. This Proceedings book presents the outcomes from the METNET Jaume seminar in 2016. The first papers presented at the seminar concerned Moment-Rotation behaviour of welded High Strength Steel tubular columns. The main connection studied was between rectangular hollow section columns and beams. The paper from the Czech Technical University, Prague, explored the interaction between a steel arch and the membranes spanning between the arches. It was concluded that the tension in the membranes helped to increase the buckling load of the arch significantly, compared with the buckling load of an arch without any membranes.

Proceedings of the METNET Seminar 2016 in Castellon

A paper from Lapland University in Finland covered issues with the performance against wearing of ultra-high strength steels. Use of modern measurement techniques was described. Modelling of the behaviour using finite element analysis showed satisfactory correlation with experiments. An innovative technique for joining steel elements using adhesives was the topic of the paper from Brandenburg Technical University, Cottbus, Germany. An extensive range of topics covered in the research were described. Certain recommendations on the usability of different adhesives were made. Specifically, limits on the thickness of the adhesive and the effect of temperature have been determined. The research helps to remove reservations over the use of adhesives in steel connections. The proceedings include short abstracts of some of the presentations made at the seminar. The wide range of research on steel and composite structures being done at the University of Catalonia, Barcelona, was described. The techniques used and the range of topics covered was truly extensive. The paper from University of Navarra described the need for a novel finite element, termed the cruciform element, to be used in non-linear frame analysis. The cruciform element has been implemented in analytical software. Very good correlation with more complex finite element modelling was demonstrated. A paper on the stability analysis of circular hollow section columns, based on finite difference method was the topic of a paper from City, University of London. Good correlation with a large series of previously published experiments was obtained. Two papers from the Laurea University of Applied Sciences in Finland covered a peep into the future. The first covered a framework for visionary concept designs for energy-efficient residential areas. The research, funded by European Regional Development programme, covered case studies from 3 regions in Finland. Another paper presented results from a survey in Finland on the current and future use of clean technologies. A list of key actors who are playing significant roles in developing solutions. Interestingly, the audience was used for collecting more data for the survey. One of the conclusions was that bureaucracy needs to be reduced at all levels, although how this was to be achieved was still an open question. Two presentations dealt with current research grant applications â&#x20AC;&#x201C; one dealing with sandwich panels and the other with connections between rectangular hollow sections and I-beams. Due to reasons of confidentiality, the papers have not been included in the proceedings. Kuldeep Virdi and Lauri Tenhunen

Proceedings of the METNET Seminar 2016 in Castellon

REFERENCES Baumol, William J. and Alan S. Binder (2012). Economics: Principles and Policy. Mason, USA. Deardorff, A. V. (2005). “How Robust is Comparative Advantage?” Review of International Economics. Volume 13 (5): 1004 – 1016. METNET (2016). METNET Ideas. www.hamk.fi/metnet Ricardo, David (1817) On the Principles of Political Economy and Taxation. London. Shiozawa, Y. (2016). The revival of classical theory of values. In Nobuharu Yokokawa et als. (Eds.) The Rejuvenation of Political Economy, May 2016, Oxon and New York: Routledge.

Proceedings of the METNET Seminar 2016 in Castellon

STABILITY OF A STEEL TUBE ARCH SUPPORTING BARREL TYPE TEXTILE ROOF MEMBRANES Ondrej Svoboda Czech Technical University in Prague Thakurova 7, Prague 6, Czech Republic Josef Machacek Czech Technical University in Prague Thakurova 7, Prague 6, Czech Republic

ABSTRACT Following extensive tests of a single (alone) built-in tubular steel arch and the identical arch supporting textile membrane roofing fastened to the arch, the corresponding numerical analyses are presented. The membranes made of the prestressed PVC coated polyester fabric Ferrari® Précontraint 702S were used as a currently standard and perfect material. The tests and results oriented to the loss of in-plane and out-of-plane stability of the arch are briefly described. SOFiSTiK software was employed to model orthotropic membrane and supporting steelwork behavior using geometrically nonlinear 3D analysis with imperfections (GNIA). The numerical results were validated using test data based on various membrane prestressing with excellent agreement. Finally results of vast parametric studies concerning stability of a central arch in five arches membrane assemblies are presented. Enormous savings in the tubular steel arches due to the membrane supporting effects are revealed, leading to omission of a stability check concerning the out-of-plane arch buckling.

INTRODUCTION Tensioned Fabric Structures (TSF) are becoming popular for their visual appearance, lightness and sufficient load-bearing capacity. Currently various membrane materials are available for both single layer double curved shapes (barrels, hypars, cones) and double/more layer pneumatic constructions (inflatable cushions, air supported pneumatic structures). The PVC coated polyester fabric (e.g. Précontraint FERRARI®) seems to be a rational choice for common single layer structures with the lifetime up to 20 years, acceptable cost and good joining possibilities (welding or sewing). Details on other materials as glass fabrics, expanded PTFE, polyethylene fabrics, ETFE, THV are described e.g. by Svoboda & Macháček (2015) or Macháček & Jermoljev (2016). Design of structures with TSF requires geometrically and materially nonlinear analysis with imperfections (GMNIA). In the analysis an appropriate input of the membrane material behaviour is required. The material is due

Proceedings of the METNET Seminar 2016 in Castellon

to its structure non-homogeneous, orthotropic (warp and fill directions) and non-linear. Based on experimental investigation, Kato et al. (1999) suggested a numerical model requiring 15 data parameters. The model proved an excellent agreement with tests but is rather too complicated for a practical use. Apart from a simplified approach, Gosling 2007 suggested “strain-strain-stress” approach using response surfaces linking strains to stresses through three dimensional representations. A non-linear material model with 5 parameters based on tests and depending on load ratios in warp and fill direction was proposed by Galliot and Luchsinger (2009). Model parameters are: warp and fill Young’s modules for the respective load ratio 1:1, the change in warp and fill Young’s modules (variation of the moduli on the whole range of load ratios) and the Poisson’s ratio. A simplified stress-strain model for coated plain-weave fabrics was developed by Pargana and Leitao (2015). The model consists of three nonlinear elements to model the yarns and an isotropic plate to model the coating. Nevertheless, in accord with recommendation of Tensinet Analysis & Materials working Group (Foster & Nollaert 2004) a simplified elastic approach may be employed using the simple plane stress theory. The supplied test data provides elastic modules for warp and fill directions and corresponding Poisson’s ratio, valid for anticlastic type of structures. Recently Uhlemann et al. (2015) analysed two approaches concerning simplified elastic constants for design of the membranes (Japanese MSAJ and acc. to European Tensinet & Materials working Group) and found the European approach more general and reasonable for PES/PVC material, but still with reservations. For novel unique structures the use of supporting components as slender as possible is necessary to follow the concept of a delicate, light and attractive structure with the membrane surface and to avoid any visual intervention of supporting steelworks into the membrane area. While membrane surface is exclusively tensioned, supporting construction is most often exposed to a compression and/or bending. This type of loading, in combination with slender elements, results into stability problems and design must be done with respect to these effects. Different situations occur, when membrane surface is joined with the supporting steel structure continually. In this case the membrane represents a spring support for the supporting structure, the critical (buckling) length of individual steel elements is changing, and both parts of the entire system cannot be investigated separately but as one complex structure using proper software package which allows integrated modelling and computing (e.g. EASY (technet), FORTEN (ixForten), SOFiSTiK, Rhino Membrane, NDN (membrane NDN)], etc.). Detailing of membrane structures and erection methods are well described by Seidel (2009), and the basic design in European Design Guide (2004). The paper deals with stabilizing effects of membranes to the respective supporting steelwork, based on numerical parametrical studies validated by tests. The tests relate to the structural model representing a concert stage structure with two supporting arches. The stability of the inner supporting arch is of the primary interest. Influence of various membranes prestressing

Proceedings of the METNET Seminar 2016 in Castellon

on the structure stabilization is analysed and compared with the test results. Parametrical study concerns nonlinear behaviour of tube arches with various geometries which support textile membranes.

LABORATORY TESTING Model Arrangement

Model of an Model outdoor covered stage (with an approx. reduction 1:10) was proposed of an outdoor covered stage (with an approx. reduction 1:10) was proposed using Formfinder Software. TheThesoftware enables manipulation with using Formfinder Software. software enables intuitiveintuitive manipulation with membrane shapes under required stress level in interactive way and export/import to membrane shapes under required stress level in interactive way and export/ import through to other programs throughfiles. DXF/DWG files. Supporting steelwork other programs DXF/DWG Supporting steelwork involves two involves(outer two supporting arches (outertubes), and inner CHS tubes), supporting arches and inner CHS bottom edgebottom steel edge wire ropes and steel wire ropes and membrane plates, see Figure 1. membrane plates, see Figure 1.

1: Setup Figure 1: Setup andFigure photo in laboratory.

and photo in laboratory.

® Précontraint The membrane is PVC iscoated polyester Ferrari ® The membrane PVC coated polyester fabric fabric Ferrari Précontraint 702S 702S with 2 2 opaque surface 830 g/m ). Dimensions of the of vertical inner tube with(weight opaque surface (weight 830 g/m ). Dimensions the vertical inner tubearch are LxH are and LxH =outer 4500x1200 tube arch has inclination of 60° in = 4500x1200arch [mm] tube [mm] archand hasouter inclination of 60° in respect to horizontal.

respect to horizontal.

The fabric ofThe Serge Ferrari group is made of polyester scrim coated both sides with fabric of Serge Ferrari group is made of polyester scrim coated both sides liquid PVC and PVDF for for joining. material is due to its with liquid PVCtopcoat, and PVDF weldable topcoat, weldable joining.The The material is due structure non-homogeneous, orthotropic orthotropic (warp and(warp fill directions) and and non-linear, with to its structure non-homogeneous, and fill directions) non-linear, with elastic constants presentedoffurther at chapter of numerical elastic constants presented further at chapter numerical modelling. Nevertheless, modelling. Nevertheless, the patented fabrication process ensures similar the patented fabrication process ensures similar elongation behaviour in both warp elongation behaviour in both warp and fill directions and minimum creep. and fill directions and minimum creep. Concerning material characteristics the biaxial Concerning material characteristics the biaxial test performed by Lab BLUM test performed by Lab BLUM Stuttgart Withwarp respect to these tests Stuttgart 2005 is available. With 2005 respectistoavailable. these tests both and fill both warp and fill braking loadings were considered as S ≈ 56 kN/m, while working ult braking loadings were considered as Sult ≈ 56 kN/m, while working loading S = S /5 ≈ 11.2 kN/m to exclude tearing and prestressing up to P = up to P = = Sult/5 loading Smax max ult ≈ 11.2 kN/m to exclude tearing and prestressing S /5 ≈ 2.24 kN/m. Nevertheless, prestressing in the case of the lab model max Nevertheless, prestressing in the case of the lab model was up Smax/5 ≈ 2.24 kN/m. was up to P ≤ 0.5 kN/m only. to P ≤ 0.5 kN/m only. The steel tubes from grade S355J0 were hot-formed in workshop. The investigated inner tube of ø 26.9x3.2 [mm] and outer one of ø 88.9x3.2 [mm] were welded to steel

Proceedings of the METNET Seminar 2016 in Castellon

The steel tubes from grade S355J0 were hot-formed in workshop. The investigated inner tube of ø 26.9x3.2 [mm] and outer one of ø 88.9x3.2 [mm] were welded to steel blocks to form a fixed-in frame acc. to Figure 1. Coupon tests of the inner steel tube resulted into average yield strength fy = 475 MPa, ultimate strength fu = 595 MPa and elongation 27.1 %. Modulus of elasticity was considered in accord with Eurocode 3 as E = 210 GPa. The membrane was fastened to the outer tube arch using riveted aluminium keder profile and to the inner tube using alternating pockets. The standard 7x7 wire peripheral rope with diameter of 6 mm inserted in a curved cuff fastened the membrane to corner plates and anchors. Deflections of the investigated inner arch were measured using transducers (electrical potentiometers) in vertical (V), transverse (H) and longitudinal (L) directions in positions as shown in Figure 2, together with loading points (P). In supports and midspan were located strain gauges denoted (T), always in four mutually perpendicular positions. The membrane cut resulted from EASY (technet) software, considering

The membrane cut resulted from EASY software, considering uniform uniform prestressing of roughly P = 0.5(technet) kN/m. During assembly the peripheral prestressing of roughly P = 0.5 kN/m. During assembly the peripheral ropes ropes were tightened and the membrane checked against wrinkling. Eightwere tightened andgauges the membrane checked against wrinkling. strain gauges strain were placed at various locations within Eight the membrane surfacewere placed for at rough various locations within the membrane surface for rough information information on prestressing values. The measuring before and after on prestressing values.resulted The measuring and after resulted into prestressing into averagebefore prestrain value of εprestressing y = 140 μm/m (i.e. P ≈ average0.09 prestrain value of ε y = 140 μm/m (i.e. P ≈ 0.09 kN/m) in perpendicular kN/m) in perpendicular direction with respect to the supporting arches directionand with to the arches and εxdirection = 450 μm/m P ≈ 0.30 εx =respect 450 μm/m (i.e. supporting P ≈ 0.30 kN/m) in parallel to the (i.e. arches. kN/m) in parallel direction to the arches.

Figure 2: Positions of loading (P), transducers (V, H, L) and strain gauges (T). Figure 2: Positions of loading (P), transducers (V, H, L) and strain gauges (T).

Test Results The investigation Test Results concerned exclusively the inner steel arch to find stabilizing effect of the membrane to its nonlinear behaviour. First the inner arch alone (without fastening the membrane)concerned was loaded and second, aftersteel the arch membrane assembly, the The investigation exclusively the inner to find stabilizing complete membrane structure. For loading calibrated pouches with steel pellets effect of the membrane to its nonlinear behaviour. First the inner arch alonewere used and suspended from seven given points P at the arch. The loadings (without fastening the membrane) was loaded and second, after the membranewere carefullyassembly, arrangedthe to complete simulate membrane uniform symmetrical loading andcalibrated asymmetrical loading structure. For loading pouches corresponding to pellets the firstwere in-plane buckling mode.from Maximum loadings forParch alone with steel used and suspended seven given points at the are given in Figure 3, for complete membrane structure in Figure 4. arch. The loadings were carefully arranged to simulate uniform symmetrical Distribution of maximum total load 80,24

81,36

81,09

80,50

81,02

81,14

81,56

Distribution of maximum total load

Total weight = 566,9 [kg]

100 kg]

71,01

69,82

71,27

Total weight = 241,96

Test Results of the METNET Seminar 2016 in Castellon The investigation concerned exclusively the inner Proceedings steel arch to find stabilizing effect11 of the membrane to its nonlinear behaviour. First the inner arch alone (without fastening the membrane) was loaded and second, after the membrane assembly, the complete membrane structure. For loading calibrated pouches with steel pellets were used and suspended from seven given points P at the arch. The loadings were loading to andsimulate asymmetrical loading corresponding to the first in-plane buckling loading carefully arranged uniform symmetrical loading and asymmetrical mode. Maximum loadings for arch alone are given in Figure 3, for complete corresponding to the first in-plane buckling mode. Maximum loadings for arch alone membrane structure in Figure 4. are given in Figure 3, for complete membrane structure in Figure 4. Distribution of maximum total load 80,24

81,36

81,09

80,50

81,02

81,14

81,56

Weight [kg]

Distribution of maximum total load

Total weight = 566,9 [kg]

100

60 40 20

71,01

69,82

71,27

Total weight = 241,96 [k ]

60 40

29,86

20 0

0 1

0,00

Loading points

Figure 3: Loading of the 3: innerLoading steel arch alone: - symmetrical, right - asymmetrical. Figure ofLeft the inner steel arch alone:

Left-symmetrical, right - asymmetrical.

Distribution of maximum total load 140

Total weight = 849,5 [kg]

120,49 122,45 122,03 120,20

Distribution of maximum total load

121,97 121,00 121,38

120

100

Weight [kg]

120

60 40 20

101,42

100,83

101,70

Total weight = 354,23 [k ]

80 60

50,28

40 20

0 1

Loading points

0 1

0,00

Loading points

Figure 4: Loading the arch joinedof withthe the membranes: Left - symmetrical, right membranes: - asymmetrical. Figure 4:ofLoading arch joined with the

Left-symmetrical, right – asymmetrical.

Individual loading steps amounted for roughly 1/10 of maximum loading, each followed steps by unloading. The tests terminated when deflections Individual loading amounted forwere roughly 1/10 ofabnormal maximum loading, each out-of-arch plane or in-arch-plane were reached. followed by unloading. The tests were terminated when abnormal deflections out-of-

arch plane or in-arch-plane were reached. Results for Symmetrical Loading

Results for Symmetrical Loading Deflections of the inner arch under increasing loading in vertical and

Deflections horizontal of the inner archare under loading in vertical and directions shownincreasing in Figure 5 (unloading is not included andhorizontal generally elastic). The arch without membrane buckled directions are shownwas in fully Figure 5 (unloading is not included and out-of-plane generally was fully at total approaching F0 = 5.5 kN, with associated vertical deflection elastic). The archloading without membrane buckled out-of-plane at total loading along all span down. On the other side the test with arch stabilized by thedown. On approaching F0 = 5.5 kN, with associated vertical deflection along all span membrane was terminated under total load of F M = 8.3 kN, showing very the other side the test with arch stabilized by the membrane was terminated under small and nearly linear increase of the mid span deflection. Stabilizing effect total load of ofFthe M = 8.3 kN, showing very small and nearly linear increase of the mid membrane is enormous. span deflection. Stabilizing effect of the membrane is enormous.

directions are shown in Figure 5 (unloading is not included and generally was fully The archloading without membrane buckled out-of-plane at total loading er archelastic). under increasing in vertical and horizontal approaching F = 5.5 kN, with associated vertical deflection along all span down. On 0 Figure 5 (unloading is not included and generally was fully Proceedings of the METNET Seminar 2016 in Castellon 12 other sidebuckled the testout-of-plane with arch stabilized the membrane was terminated under ithout the membrane at total by loading load ofvertical FM = 8.3 kN, showing N, withtotal associated deflection along allvery spansmall down.and On nearly linear increase of the mid with arch stabilized by the membrane wasofterminated underis enormous. span deflection. Stabilizing effect the membrane

N, showing very small and nearly linear increase of the mid ing effect of the membrane is enormous.

Figure 5: Symmetrical loadings:

Figure 5: Symmetrical loadings: Vertical deflections (left), transverse deflections (right).

Figure 5: Symmetrical loadings: Vertical deflections (left), transverse deflections (right). deflections (left), transverse deflections (right).

Results for Asymmetrical Loading Results for Asymmetrical Loading

cal Loading

Vertical and transverse horizontalhorizontal deflections underunder increasing loading Vertical and transverse deflections increasing loading are are shown in

horizontal deflections under increasing loading arearch shown in membrane (“0”) terminated shownof in the Figure 6. Testing of the without Figure 6. Testing arch without membrane (“0”) terminated under total loading arch without membrane (“0”) terminated under total loading under total loading F = 2.37 kN, giving maximal vertical deflection = 41.3 one  = 0 F0 = 2.37 kN, giving maximal vertical deflection  0 = = 41.3 mm and horizontal 0 ximal vertical deflection 41.3 mmone and horizontal 0 =horizontal 0 stabilized by membrane (“M”) mm  and = 3.5 mm. one The arch 3.5 mm. The arch stabilized by membrane (“M”) deflected much less, giving for the lized by membrane deflected (“M”) deflected much less, giving forloading the F M = F0 = 2.37 kN values much less, giving for the same = Fmm = and 2.37 values M stabilizing = 18.5 mm = 0.5 mm. The same loadingMF= M 0mm = 2.37 kN values 18.5 and kN =M 0.5 = 0.5 The = 18.5 mm. mm. The effectand of theM membrane stabilizing effect ofvertical the membrane concerning vertical ofdeflection resulted into concerning vertical deflection resulted into 45 %.Difference e membrane concerning deflection resulted intoreduction vertical deflections in positions and V3 (see Figure 2) gives V1 and V3 reduction of comparing 45 %.Difference comparing verticalV1deflections in positions = 68.3 mm and = 32.9 mm. (see Figure 2) gives 0 = 68.3 mm and M = 32.9 mm.

Measured displacements and stresses (not presented) prove expected Measured displacements and stresses (not presented) prove expected enormous enormous effect of the membrane on buckling load and strength of inner effect of thesupporting membrane on bucklingin load and strength arch, arch, particularly asymmetrical loadings. of Testinner valuessupporting are used particularly inforasymmetrical loadings. Test values are used for validation of numerical validation of numerical analyses. analyses.

FigureVertical 6: Asymmetrical loadings: Figure 6: Asymmetrical loadings: deflections (left, positive down), transverse deflections (right). Vertical deflections (left, positive down), transverse deflections (right). NUMERICAL ANALYSIS Numerical Model and Validation by Tests

Proceedings of the METNET Seminar 2016 in Castellon

NUMERICAL ANALYSIS Numerical Model and Validation by Tests SOFiSTiK software 2014 was used to perform GNIA (using N-R iteration) both for the inner arch alone and the complete membrane structure. Various meshing of the membrane was analysed (square sizes of 25, 50, 100 and 200 [mm]) giving nearly identical stress results (differences ≤ 0.2%), but with increase of computer time between largest and least mesh up to 107 times. Therefore, optimum mesh with 50 mm size was used throughout the analysis.

Arch Alone Analysis The tube arch was introduced as ø 26.9x3.2 [mm], with built-in supports, material S355J0 and Young’s modulus E = 210 GPa. Initial geometry as measured was introduced into analysis (theoretical radius R = 2709 mm with out‑of-plane (transverse) deflection at the crown w0 = 17 mm). The elastic buckling analysis of the perfect arch under symmetrical loading in accordance with Figure 3 (left) resulted into total critical loads TLcr,1 = 6.7 kN reduction of (out-of-plane 45 %.Difference comparing deflections in positions V1 single wave buckling),vertical TLcr,2 = 15.4 kN (out-of-plane two waves buckling), 18.7 kN etc. (see Figure 2) gives TL 0cr,3 = =68.3 mm(in-plane and two 32.9 buckling), mm. M =waves

and V3

GNIA transverseand deflection curves symmetrical prove loadingexpected are shownenormous Measured displacements stresses (notforpresented) in Figure 7 left, indicating the first out-of-plane bifurcation earlier but in effect of the membrane on buckling load and strength of inner supporting arch, accordance with test, roughly at 4.9 kN. Vertical deflections for asymmetrical particularly inloading asymmetrical valuesseeare used for validation of numerical correspondloadings. also well toTest test results, Figure 7 right. analyses.

6: Asymmetrical loadings: Figure 7: Arch alone, testFigure and GNIA results. Left: transverse deflections for symmetrical loading at midspan. Right: verticaldeflections deflections for asymmetrical Vertical (left, loading. positive down), transverse deflections (right). NUMERICAL ANALYSIS Numerical Model and Validation by Tests SOFiSTiK software 2014 was used to perform GNIA (using N-R iteration) both for the inner arch alone and the complete membrane structure. Various meshing of the membrane was analysed (square sizes of 25, 50, 100 and 200 [mm]) giving nearly

Proceedings of the METNET Seminar 2016 in Castellon

Full Membrane Structure Analysis The membrane FERRARI® 702 was considered with linear elastic orthotropic behaviour taking account for influence of load ratios according to Galliot and Luchsinger 2009. The necessary five parameters were taken as: warp and fill Young’s moduli for 1:1 load ratio Ew1:1 = 635.3 kN/m and Ewf1:1 = 661.9 kN/m, the change in warp and fill Young’s moduli ΔEw = 295 kN/m and ΔEf = 168 kN/m, the Poisson’s ratio ν = 0.196. The warp direction was considered as perpendicular to the arch plane and conversion to stresses due to approximate thickness of the membrane as 0.7 mm. Initial geometrical imperfections of the arch were measured just before the first loading and were found to be 5.0 mm in horizontal direction at the top of the arch and +/-10.0 mm in vertical direction at the quarters of the arch. Initial shape of the membrane surface was generated automatically by the AutoCAD as a surface with minimal area. The shape was exported into SOFiSTiK software. For each level of the membrane prestress the corresponding value of the perimeter cable pretension was received from formula:

SS = r ⋅ SM (1)

where SS r SM

is a tension force in a perimeter cable [kN], radius of the cable curvature [m], prestress in the membrane perpendicular to the cable [kN/m].

The equilibrium state and final unloaded shape of the structure was found under initial SOFiSTiK software calculations. The results of GNIA under various pretension of the membrane P [kN/m] is shown in Figure 8. The level of pretension is the crucial parameter for the arch stability and resistance capacity. Higher pretension in membrane logically means higher stability and smaller deflection of the investigated arch. It should be recalled, that the real pretension during the testing was influenced by the alternating “pockets” connection of the membrane to the inner arch and due to this fact rather nonuniform and slacked. As mentioned, the randomly measured values during the test at several localities (which however can’t be considered as a reliable ones), were between 0.09 and 0.3 [kN/m]. This is reflected in numerical analysis with pretension close to zero (0.2 kN/m) which provides results very near to the test ones. Therefore, the received GNIA results with the corresponding prestress 0.2 kN/m justify use of the model for following parametric studies.

Proceedings of the METNET Seminar 2016 in Castellon

the test ones. Therefore, the received GNIA results with the corresponding prestress 0.2 kN/m justify use of the model for following parametric studies. the test ones. Therefore, the received GNIA results with the corresponding prestress 0.2 kN/m justify use of the model for following parametric studies.

Figure 8: Symmetrical loading of the membrane model: Comparison the test and GNIA vertical deflections Figure 8:8:Symmetrical loading of theofmembrane model: Figure Symmetrical loading of the membrane model: under various pretension. Comparison of the test and GNIA vertical deflections under various pretension.

Comparison of the test and GNIA vertical deflections under various pretension.

Parametrical Parametrical Study Parametrical StudyStudy The principal goal of the study to determine the stabilizing effect of common TheThe principal goal ofofthe study was to determine the stabilizing stabilizing effect principal goal the study was to was determine the effect of of common common textile membranes on buckling and nonlinear behaviour of slender textile membranesononbuckling bucklingand and nonlinear nonlinear behaviour supporting textile membranes behaviourofofslender slendersteel steel supporting steel supporting arches. Both in-plane and out-of-plane arch buckling was arches. Bothin-plane in-planeand and out-of-plane out-of-plane arch arch buckling was studied forforvarious arches. Both buckling was studied various studied for various geometries of arches and membranes in a parametrical geometries of arches and membranes in a parametrical study. Inner arches in the 5 5 geometries of arches membranes in a parametrical study. Innerarranged arches in the study. Innerand arches in the 5 arch assembly with practical dimensions arch assembly with practical dimensions arranged in a barrel type structure in arch assembly with practical dimensions in9 awere barrel type structure in in a barrel type structure in accordingarranged with Figure investigated. To according withFigure Figure were investigated. evaluate of of lateral according with 9 9were To evaluate theinfluence influence lateral evaluate the influence of investigated. lateral boundaryTo conditions, thethe outer (edge) arches boundary conditions, theouter outer (edge)byarches arches either (supported a a were either the flexible (supported a truss)were or continuously fixed in all three by by boundary conditions, (edge) were eitherflexible flexible (supported truss) or continuously fixed in all three directions. directions. fixed in all three directions. truss) or continuously

5xB

Figure 9: Five arch assembly and outer lateral flexible support. Figure 9: Five assembly andassembly outer lateral flexible support. Figure 9: arch Five arch and outer lateral flexible support.

The study covers arch spans of L = 6 ÷ 20 [m], rise H = L/10 ÷ L/2 and spacing of the arch spans ofwas L = 6considered ÷ 20 [m], rise = L/10 ÷direction L/2 and spacing five archesThe B study = 3 ÷covers 10 [m]. Loading inHvertical (alternatively Theinstudy covers arch spans of L = 6 ÷ 20 [m], rise H = L/10 ÷ L/2 and spacing of the of the five arches B = 3 10 [m]. Loading was considered in vertical direction 2÷ only, together with various radial direction) as 1 kN/m simulating snow loading in radial direction) as 1 kN/m2 simulating snow loading only, fiveprestressing arches B(alternatively =of3 the ÷ 10 [m]. Loading was considered in vertical direction (alternatively membranes, ranging from 4 to 7 kN/m in both directions. The arch 2 togetheras with ofsnow the membranes, ranging 4 to 7 various in radial direction) 1 various kN/m simulating loading only, together with dimensions were designed for prestressing resulting internal forces according to from Eurocode 3 from kN/m in both directions. The arch dimensions were designed for resulting prestressing the membranes, from 80÷90 4 to 7 kN/m in both directions. steel gradeofS355 in an iterative ranging way, to reach % of their design capacity.The arch internal forces according to Eurocode 3 from steel grade S355 in an iterative dimensions were designed for resulting internal forces according to Eurocode 3 from way, to reach 80÷90 % of their design capacity. First the buckling of the arch alone built-in supports to find insteel grade S355 in an iterative way, with to reach 80÷90 % ofwas theirinvestigated design capacity.

plane and out-of-plane buckling loads under considered loading (Figure 10).

First the buckling of the arch alone with built-in supports was investigated to find inplane and out-of-plane buckling loads under considered loading (Figure 10).

Proceedings of the METNET Seminar 2016 in Castellon

First the buckling of the arch alone with built-in supports was investigated to find in-plane and out-of-plane buckling loads under considered loading (Figure 10).

Arch alone: in-plane buckling (left), out-of-plane(left), bucklingout-of-plane (right). FigureFigure 10:10: Arch alone: in-plane buckling buckling Figure 10: Arch alone: in-plane out-of-plane buckling (right). (right).

Second the all five arch assembly (including lateral support) was analysed to

in-plane out-of-plane bucklinglateral loads for the mid-arch under loading Second the thefind all five five arch archand assembly (including support) was to Second all assembly support) was analysed analysed to find findininincluding various prestressing of the membranes (Figure 11). plane and out-of-plane buckling loads for the mid-arch under loading including plane and out-of-plane buckling loads for the mid-arch under loading including variousprestressing prestressing of of the the membranes membranes (Figure (Figure 11). various 11).

Figure11: 11:Arch Arch with membranes: membranes: mid mid arch arch in-plane in-plane buckling (left), Figure (left), mid mid arch archout-ofout-ofFigure 11:with Arch with membranes: mid arch buckling (left), midbuckling planein-plane buckling (right). arch out-of-plane buckling (right). plane buckling (right). Fortube example arch ø [mm], 133x5 [mm], grade S355, with 12 m, m, H H= For example archtube ø 133x5 steel steel grade S355, with L L==12 = 3.6 m For example3.6tube arch 133x5 [mm], steel grade S355, with L = 12 m,mesh H = 3.6 m m and B = ø3 m and uniformly loaded vertically under 3 kN/m with and B = 3 m and uniformly loaded vertically under 3 kN/m with mesh size 250 mm, and B = 3 msize and uniformly loaded vertically under kN/m with meshbuckling size 250 mm, mm, gives following buckling loads: 3arch alone in-plane gives following 250 buckling loads: arch alone in-plane buckling N cr.y = 189 kN, out-ofgives following buckling loads: arch alone in-plane buckling N = 189 kN, out-ofcr.y N = 189 kN, out-of-plane buckling N = 70 kN; arch with membranes 2 cr.yN cr.z under vertical loading 1kN/m membranes and plane buckling cr.z = 70 kN; arch with = 70loading kN; arch with2 and membranes 1kN/m plane buckling Ncr.z under vertical 1kN/m prestressedunder by P ≈vertical 5.0 kN/mloading gives Ncr.y = 2 and prestressed by P ≈ 5.0 kN/m gives Ncr.y = 346 kN and Ncr.z = 542 kN. Similar relations kN≈and kN.NSimilar relations were obtained therelations kN and Ncr.z = 542 throughout kN. Similar prestressed346 by P 5.0 NkN/m gives cr.z = 542 cr.y = 346 were obtained throughout the parametrical study, indicating that the membrane parametrical study,the indicating that the membrane support to that the stability of were obtained throughout parametrical study, indicating the membrane support to the stability of arch is substantial. arch is substantial. support to the stability of arch is substantial.

Conclusions Conclusions The main conclusions are: The main conclusions are: effect of textile membranes on both in-plane and out-of1. Tests confirmed decisive 1. Tests confirmed decisive effectand of textile membranes on both in-plane and out-ofplane supporting arch stability strength. plane supporting arch stability and strength. 2. GNIA with linear elastic orthotropic behaviour of the textile membrane and using 2. GNIA with linear elastic orthotropic behaviour of the textile membrane and using SOFiSTiK software proved to be adequate for numerical modelling. to be adequate for numerical modelling. influence the 3. SOFiSTiK The real software value ofproved the membrane prestressing substantially 3. The real value of theof membrane prestressing substantially the deflections and strength the supporting arch structure (see Figure influence 8).

Proceedings of the METNET Seminar 2016 in Castellon

CONCLUSIONS The main conclusions are: 1.

Tests confirmed decisive effect of textile membranes on both in-plane and out-of-plane supporting arch stability and strength. 2. GNIA with linear elastic orthotropic behaviour of the textile membrane and using SOFiSTiK software proved to be adequate for numerical modelling. 3. The real value of the membrane prestressing substantially influence the deflections and strength of the supporting arch structure (see Figure 8). 4. Parametric studies of large barrel membrane structures supported by a row of steel arches resulted into huge savings in a practical design of supporting steel arches due to enormous increase of both in-plane and particularly out-of-plane buckling loads in comparison to the ones of an buckling loads inincomparison toNthe ones of an arch alone (e.g. in the shown cas arch alone (e.g. the shown case cr.z increased 7.7 times), mentioned increased 7.7 times), mentioned Ncr.zalready by Svoboda and Macháček 2016. already by Svoboda and Macháček 2016. oads in comparison to the ones an arch alone (e.g.study in the shown 5. 5. Within thetheof scope parametric study for case the alone tubethe arch Within scope ofofthethe parametric for the tube arch in- alone the in-plan ased 7.7 times), mentioned already by Svoboda and Macháček 2016. plane buckling of the arch is not decisive (N (Ncr.in 2÷4 NNcr.out ), while buckling of the arch is not decisive ), while for the arch wit cr.in  2÷4 cr.out scope of the parametric study for the tube arch alone the in-plane for the arch with membranes the out-of-plane arch buckling is not decimembranes the out-of-plane arch buckling is not decisive (Ncr.out  2÷3 Ncr.in). sive (N 2÷3 NN ). ), while for the arch with f the arch is not decisive (Ncr.out cr.in cr.in  2÷4 cr.out

es the out-of-plane arch buckling is not decisive (Ncr.out  2÷3 Ncr.in). ACKNOWLEDGEMENT

DGEMENT

ACKNOWLEDGEMENT

The support of the grant GACR No. 105/13/25781S is gratefully acknowledged. The support of the grant GACR No. 105/13/25781S is gratefully acknowledged.

f the grant GACR No. 105/13/25781S is gratefully acknowledged. REFERENCES

REFERENCESSoftware GmbH, Wien. http://www.formfinder.at/main/software/. Formfinder Foster, B. Software Mollaert, M. (2004): European Design Guide for Tensile Surfac Formfinder GmbH, Wien. http://www.formfinder.at/main/software. oftware GmbH, Wien. http://www.formfinder.at/main/software/. Structures, TensiNet, 354 p. Mollaert, M. (2004): European Design Guide forA Tensile Surface Galliot, Luchsinger R.H. (2009): simple model describing Foster, C., B. Mollaert, M. (2004): European Design Guide for Tensile Surfacethe non-linear biaxia es, TensiNet, 354 p. tensileTensiNet, behaviour Structures, 354 p.of PVC/coated polyester fabrics for use in finite elemen uchsinger R.H. (2009):analysis. A simpleComposite model describing the Vol. non-linear Structures, 90, No.biaxial 4, pp. 438-447. behaviour of PVC/coated polyester fabrics for use in finite element Galliot, C., R.H. (2009): A simple model describing the non-linear Galliot, C.,Luchsinger Luchsinger, R.H. (2009): Non-linear properties of PVC-coated fabric Composite Structures, Vol. 90, No. 4, pp. 438-447. biaxial tensile behaviour of PVC/coated polyester fabrics for use in finite eleused in tensairity structures. Proc. ICCM17, Edinburgh, 10 p. ment analysis. Composite Structures, of Vol.PVC-coated 90, No. 4, pp. 438 – 447. uchsinger, R.H. Gosling, (2009): Non-linear properties P. (2007): Basic philosophy and calling fabrics notice. Tensinet analysis & Materia ensairity structures. Proc. ICCM17, Edinburgh, 10 p. working group, Tensinews (www.tensinet.com), No. 13, pp. 12-15. Galliot, and C., Luchsinger, R.H. (2009): Non-linear properties of PVC-coated fab007): Basic philosophy calling notice. Tensinet analysis & Material ixForten 4000. http://www.ixforten.com/ rics used in tensairity structures. Proc. ICCM17, Edinburgh, 10 p. group, Tensinews (www.tensinet.com), 13, pp. Kato, S., Yoshino, T., No. Minami, H.12-15. (1999): Formulation of constitutive equations fo . http://www.ixforten.com/ fabric membranes based on of fabric lattice model. Eng. Structures Gosling, P. (2007): Basic philosophy andconcept calling notice. Tensinet analysis & shino, T., Minami, Material H. Vol. (1999): Formulation of constitutive equations forpp. 12 – 15. working group, Tensinews (www.tensinet.com), No. 13, No. 8, pp. 691-708. embranes basedLaboratorium on concept of fabricStuttgart lattice model. Structures, BLUM (2005):Eng. Report on biaxial tests Précontraint 702 4000. http://www.ixforten.com/ 8, pp. 691-708. ixForten http://arc-ae.net/ARC1010-Pisco/. BLUM StuttgartMachacek, (2005): Report on biaxial tests Précontraint 702.in interaction with non-metall J., Jermoljev, D. (2016): Steel structures -ae.net/ARC1010-Pisco/. membranes, J. Civil Engineering and Management do , Jermoljev, D. (2016): Steel structures in interaction with non-metallic 10.3846/13923730.2015.1128482. nes, J. Civil Engineering and Management doi: membrane NDN software. http://www.ndnsoftware.com/. 13923730.2015.1128482. Pargana, J.B., Leitao, W.M.A. (2015): A simplified stress-strain model for coate DN software. http://www.ndnsoftware.com/. plain-weave fabrics used in tensioned fabric structures. Eng. Structures, Vo

Proceedings of the METNET Seminar 2016 in Castellon

Kato, S., Yoshino, T., Minami, H. (1999): Formulation of constitutive equations for fabric membranes based on concept of fabric lattice model. Eng. Structures, Vol. No. 8, pp. 691 – 708. Laboratorium BLUM Stuttgart (2005): Report on biaxial tests Précontraint 702. http://arc-ae.net/ARC1010-Pisco/. Machacek, J., Jermoljev, D. (2016): Steel structures in interaction with non-metallic membranes, J. Civil Engineering and Management doi: 10.3846/13923730.2015.1128482. membrane NDN software. http://www.ndnsoftware.com/. Pargana, J.B., Leitao, W.M.A. (2015): A simplified stress-strain model for coated plain-weave fabrics used in tensioned fabric structures. Eng. Structures, Vol. 84, pp. 439 – 450. Rhino Membrane. http://www.membranes24.com/rhino-membrane.html. Seidel, M. (2009): Tensile Surface Structures - A Practical Guide to Cable and Membrane Construction, John Wiley & Sons, 240 p. SOFiSTiK 2014. http://www.sofistik.de/. Svoboda, O., Macháček, J. (2015): Behaviour of steel arch stabilized by a textile membrane, 2nd Int. Conf. on Innovative Materials, Structures and Technologies, IOP Conf. Series: Materials Science and Engineering 96, pp. 1 – 8, http:// iopscience.iop.org/article/10.1088/1757-899X/96/1/012055/pdf. Svoboda, O., Macháček, J. (2016): Tubular steel arch stabilized by textile membranes, J. Advances in Technology Innovation (AITI), Vol. 1, No. 2, pp. 50 – 52, http://ojs.imeti.org/index.php/AITI/article/download/180/315 technet gmbh Berlin-Stuttgart. http://www.technet-gmbh.com. Uhlemann, J., Stranghöner, N., Saxe, K. (2015): Stiffness parameters for architectural fabrics: An analysis of two determination procedures. Structural Engineering International, Vol. 25, No. 1, pp. 9 – 19.

Proceedings of the METNET Seminar 2016 in Castellon

BONDING TECHNOLOGY IN STEEL STRUCTURES A SUMMARY OF RESEARCH ACTIVITIES AT CHAIR OF STEEL AND TIMBER STRUCTURES Yvonne Ciupack and Hartmut Pasternak Chair of Steel and Timber Structures, Brandenburg University of Technology, Cottbus, Germany

ABSTRACT Bonding technology is an innovative alternative to typical joining methods in steel structures. For more than 10 years studies have taken place at the the Chair of Steel and Timber Structures on the potential of bonded steel joints within the scope of various research projects, which are presented in this paper. Two new application examples for faรงade construction are currently being investigated. For these structures constructive solutions, mechanical models, design rules and testing procedures for cyclic loading were developed. The efficiency of refurbishment measures made with bonded CFRP laminates is a second research field of the Chair. The future work will focus on long-term effects and cyclic loading of bonded steel joints.

INTRODUCTION The joining technology plays an essential role in steel structures. It influences the costs of manufacturing and erection as well as the quality of execution. The requirements on joining techniques affect the static and dynamic carrying capacity, the safety and reliability, the geometric precision as well as aesthetics and durability. The application of traditional joining methods is limited regarding the cross section reduction through drilled holes, the residual stresses through welds and the fatigue strength through notch effects together with aesthetic reasons. As an alternative to welding and bolts, the bonding technology provides considerable advantages for this requirement profile. Bonded joints are successfully used in other industries, e.g. automotive and aviation. Despite many technical developments and improved long-term resistance of adhesives, there is no noteworthy example of application of adhesive bonding in structural engineering. The Chair of Steel and Timber Structures investigates the innovative bonding technology within the scope of different research projects, some of which are presented in this paper.

Proceedings of the METNET Seminar 2016 in Castellon

DEVELOPMENT OF NEW APPLICATIONS OF BONDING TECHNOLOGY In a first research project (Dilger et al. 2008) two different application examples of bonded steel joints in steel façades were developed. With these constructions a research unit was created, which still functions and has been expanded during recent years. Examples of a bonded façade connection and years. of will a bonded façade and façade reinforcement façadeExamples reinforcement be presented in theconnection following.

recent will be presented in the following.

Bonded façade connection Bonded façade connection Trapezoidal façades are used especially for industry halls. Commonly, the façades are used especially for industry halls. Commonly, the connectionTrapezoidal of a trapezoidal sheet with the support structure is realized with self connection of a trapezoidal sheet with the support structure is realized with drilling screws. Thus, the cross section of structural elements is reduced in the self drilling screws. Thus, the cross section of structural elements is reduced region of in thethejoint, leadswhich to aleads decreased loadload carrying regionwhich of the joint, to a decreased carryingcapacity capacity and stress concentration. Thisconcentration. detail means degradation, particularly for for thethefatigue limit and stress This detail means degradation, particularly limit strength ofThis the construction. This unsatisfactory situation can be strength offatigue the construction. unsatisfactory situation can be prevented by using prevented by using bonding technology. By bonding the trapezoidal sheets to bonding technology. By bonding the trapezoidal sheets to the support structure with the support structure with the assistance of modified connection profiles, the the assistance of modified connection profiles, the cross section reduction can be cross section reduction can be avoided. avoided. Such an alternative joining method does not only provide an advantage

Such an alternative does onlythe provide anofadvantage regarding thejoining carryingmethod capacity, but alsonot improves aesthetics the façade regarding the carrying but also improves the dents aesthetics of the façade view by view capacity, by avoiding visible fastener heads. Errors, or scratches, which can avoiding visible fastener heads. Errors, dents or scratches, occur during the mounting, are avoided. Besides, an essentialwhich benefitcan is theoccur during preservation of the selfBesides, cleaning effect of the façade through of the mounting, are avoided. an essential benefit is the theabsence preservation of the screws at the outer shell. self cleaning effect of the façade through the absence of screws at the outer shell. It is necessary to abide by the rules for bonding-suitable construction, in

It is necessary to abide by the rules for bonding-suitable construction, in order to order to successfully establish this innovative façade joint. The connection successfully establish this the innovative is designed so is designed so that complete façade bonding joint. processThe can connection be realized under that the complete bonding realized under definedprofile conditions in a defined conditions in aprocess laboratory.can Thus,be a specially shaped connection laboratory.is Thus, a specially shaped connection profilefaçade is required. design required. Different design variants for the bonded connectionDifferent are in Figure 1(a). variants forshown the bonded façade connection are shown in Figure 1(a). cover

seal L-profile

T-profile

-profile

outside isolation glass

beam screw self-drilling

inside reinforcement bondline

Figure Figure 1: (a) Structures for bonded façade connections with different shape of 1: (a) Structures for bonded façade connections with different shape of connection profile; connection (b) Bonded façade reinforcementprofile; (b) Bonded façade reinforcement Bonded façade reinforcement The modern city panorama is defined by glass façades and slender structural

Proceedings of the METNET Seminar 2016 in Castellon

Bonded façade reinforcement The modern city panorama is defined by glass façades and slender structural components. Architects and builders demand high side view transparency, as well as segmented and wide opened façades. Therefore, transom-mullion façades are often realized. With this construction, the wind loads are transferred from the outer shell to the mullion profiles, which induce the forces on to the transom profiles at certain points. This way, the transom profiles are stressed by bending moments and transfer the loads to the foundations. To ensure the required side view transparency, small transom profiles need to be used, which can be arranged with high distances in-between. Thus, a reduction of the cross section dimensions, as well as reduction of profile stiffness and carrying capacity is required simultaneously. For hollow profiles, one possible solution can be found by using sections with thicker profile walls. In contrast, this kind of element is produced with a cold forming process, which is limited to certain sheet thicknesses. The bonding technology provides an innovative and simple alternative. With the application of an inner reinforcement, as shown in Figure 1(b), the transom stiffness can be increased without changing the outer dimensions. This new compound cross section shows an increased carrying capacity compared with typical façade profiles and allows the retention of common mounting methods on site. The bonding process is conducted by means of a pneumatic method in a laboratory. The pre-assembled profile will be connected with the mullion elements on site by use of dowel type fasteners. Such kind of façade structure provides high transom distances with slender profiles, so that the construction meets the requirement of side view transparency. The principle of a hollow façade profile with bonded inner reinforcement is comparable to the carrying behaviour of glued laminated timber. That means, a composite section is created to increase carrying capacity and stiffness. In contrast to timber structures, the stiffness of the compound steel elements is so high, that a proof of the bondline could be decisive. This fact should be considered in a design model.

DESIGN RULES FOR BONDED STEEL JOINTS For the definition of design rules, according to the standards, mechanical models are essential for the description of the bondline behaviour. In the scope of different research activities (Dilger et al. 2008, Meinz 2010), two design models for the mentioned application examples were developed. The fundamental ideas of these resistance models will be presented in the following sections.

Proceedings of the METNET Seminar 2016 in Castellon

Design of bonded façade connection For the design of bonded façade connections an analytical model is used, based on the weakest-link-theory. The essential mode of failure can be identified by comparing the failure modes of the joint components. These are the connection profile, the bondline and the trapezoidal sheet. The resistance of the whole joint can be expressed as the smallest resistance of its components. Based on the theory of elastically supported slabs, Meinz (Meinz 2010) developed a design approach for the prediction of the bondline resistance. Therefore, the joint is divided into partial models for every supporting direction. A superposition of these partial models at the end of the design procedure provides the three-dimensional system and the complex stress distribution in the bondline. The principle of the design concept is shown in Figure 2.

Model of bonded facade connection y

A Partial model for x-direction ΔFA

z B

B Partial model for z-direction A

ΔFB 2

Figure 2: Model for the bonded façade connection

Design of bonded façade reinforcement The concept for the design of bonded façade reinforcement is structured, so that the Design of bonded façade reinforcement composite section is formed by the partial cross section of the reinforcement profile and the first is the test of a quasi-rigid compound. Therefore, Thehollow conceptprofile. for the The design of step bonded façade reinforcement is structured, a dimensionless shear-effect-value is introduced and deduced so that the composite section is formed by the partial cross section offrom the the differential reinforcement and the hollow The first step the test ofcompound. a equation of beamprofile deflection and jointprofile. displacement for is semi-rigid If this quasi-rigid Therefore, aofdimensionless shear-effect-value test has been compound. met, a comparison the bondline strength withis the determined introduced and To deduced from the beam deflection stresses follows. describe thedifferential complexequation stress ofdistribution in the bondline a and joint displacement for semi-rigid compound. If this test has been met, procedure is included which is based on the theory of elastically supported slabs. a comparison of the bondline strength with the determined stresses follows. This procedure is equivalent to the method for bonded façade connections. To take To describe the complex stress distribution in the bondline a procedure is into account the concentrated load introduced at the joints between included which is based on the theory of elastically supported slabs. Thisthe transom and mullion profiles, effective lengths are determined based on numerical studies for procedure is equivalent to the method for bonded façade connections. To different connection details of typical façade joints. take into account the concentrated load introduced at the joints between the transom and mullion profiles, effective lengths are determined based on studies for different connection details of typical façade joints. Modelnumerical calibration

In a further project (Pasternak 2012), the presented models were calibrated aiming at the development of a general procedure of concept calibration for bonded steel joints. Basis for the calibration process is a description model for the prediction of the resistance which contains stochastic and deterministic parameters. A comparison of

Design of bonded façade reinforcement The concept for the design of bonded façade reinforcement structured, so that Proceedings ofisthe METNET Seminar 2016the in Castellon composite section is formed by the partial cross section of the reinforcement profile and the hollow profile. The first step is the test of a quasi-rigid compound. Therefore, a dimensionless shear-effect-value is introduced and deduced from the differential equation of beam deflection and joint displacement for semi-rigid compound. If this test has been met, a comparison of the bondline strength with the determined stresses follows. To describe the complex stress distribution in the bondline a procedure is included which is based on the theory of elastically supported slabs. Model calibration This procedure is equivalent to the method for bonded façade connections. To take into account the concentrated load introduced at the joints between the transom and In a further project (Pasternak 2012), the presented models were calibrated mullion profiles, effective lengths are determined based on numerical studies for aiming connection at the development of a general procedure of concept calibration for different details of typical façade joints.

bonded steel joints. Basis for the calibration process is a description model for

the prediction of the resistance which contains stochastic and deterministic Model calibration Inparameters. a further project (Pasternakof2012), the presented models wereiscalibrated aiming A comparison test result with model results the essence of atthe themethod. development of a general procedure of concept calibration for bonded steel joints. Basis for the calibration process is a description model for the prediction of the resistance which contains stochastic and deterministic parameters. A comparison of To result determine characteristic of a bondline based on test with model results is thematerial essence properties of the method. Eurocode, so-called butt joint tests (DIN EN 15870) and shear lap tests (DIN

ENdetermine 14869) characteristic were carriedmaterial out. The resultsoffor tensionbased strength (σk), shear To properties a bondline on Eurocode, socalled butt (τjoint tests (DIN EN 15870) and shear shearmodulus lap tests (G (DIN ENsummarized 14869) were strength modulus (Ek) and k), Young’s k) are carried in Tableout. 1. The results for tension strength (σk), shear strength (τk), Young’s modulus (Ek) and shear modulus (Gk) are summarized in Table 1. Table 1: Characteristic material properties of adhesives [N/mm²] Table 1: Characteristic material properties of adhesives [N/mm²] Adhesive basis

k

k

MS polymer

1.31

1.32

5.64

0.54

Epoxy

38.7

29.3

3063

308.9

In the next step, investigations focused on specimen component tests for the application examples. The test setup needed ensure that a failure oftests the bondline In the next step, investigations focused ontospecimen component for the can be observed. A general the warranty of a cohesive failure,ofwhich application examples. Therequirement test setup isneeded to ensure that a failure the can be met by specific surface pre-treatments, e.g. blasting, of the steel components

bondline can be observed. A general requirement is the warranty of a cohesive failure, which can be met by specific surface pre-treatments, e.g. blasting, of the steel components being connected. Figure 3 shows the test setup and some being connected. Figure 3 shows example. the test setup and some results for the first results for the first application

application example.

2500

Force [N]

2000 1500 1000 500 0

10 15 20 Deformation [mm]

Figure 3: (a) Test setup; (b) Test results

The essential task of the bonded façade connection in the mounted state is the The essential of support the bonded façade of connection in the state is transfer of wind loads task to the structure the façade. Themounted critical wind suction transfer ofin wind the support structure of the façade. The critical loads arethe considered the loads tests,tobecause bondlines react more sensitive to tension wind loads are considered in the was tests, modelled because bondlines react corrugated more loads than tosuction compression. The specimen to be one sensitive to tension loads than to compression. The specimen was modelled segment of a 500 mm long trapezoidal sheet. For the bonding procedure a silane be one corrugated segment of aused, 500 mm long trapezoidal sheet. For the (see modified toadhesive (MS polymer) was because it behaves elastically Table 1) and thus a deformability of the joint was ensured. The load introduction occurred perpendicular to the joint with a testing rate of 2 mm/min to create a quasistatic stress state in the bondline. For all specimen tests a cohesive failure of the bondline could be achieved and a strongly non-linear behaviour was observed.

10 15 20 Deformation [mm]

Figure 3: (a) Test setup; (b) Test results

The essential task of the bonded façade connection in the mounted state is the transfer of wind loads to the support structure of the façade. The critical wind suction loads are considered in the tests, because bondlines react more sensitive to tension procedure a silane modified adhesive (MS polymer) was used, loads bonding than to compression. The specimen was modelled to be one corrugated because it behaves elastically (see Table 1) andFor thusthe a deformability of the joint segment of a 500 mm long trapezoidal sheet. bonding procedure a silane was ensured. The load introduction occurred perpendicular to the joint with a (see modified adhesive (MS polymer) was used, because it behaves elastically testing mm/min to create a quasi-static stateThe in the bondline. Table 1) and rate thusofa2 deformability of the joint was stress ensured. load introduction For all specimen tests a cohesive failure of the bondline could be achieved occurred perpendicular to the joint with a testing rate of 2 mm/min to createand a quasia strongly non-linear behaviourFor wasall observed. static stress state in the bondline. specimen tests a cohesive failure of the bondline could be achieved and a strongly non-linear behaviour was observed. a

F/2 3 w

2 w

F/2 0

2 w

3 w

w 1

F/2 90 90

136.5

136.5 273

136.5

136.5 273 1000

250 200

136.5

136.5 273

Force [kN]

Proceedings of the METNET Seminar 2016 in Castellon

150 100 50

F/2 90 90

0,0

2,0

4,0

6,0

8,0

Deformation in axis 0 [mm]

Figure 4: (a) Four point bending test; (b) Test results.

To investigate the reinforced façade profile, four point bending tests on a 1 m hollow investigate the reinforced façadewere profile, four out point(Figure bending a1 sectionTowith inner steel reinforcement carried 4).tests Theoncompound m the hollow section with inner steelwas reinforcement carried out (Figure between components being joined made by anwere epoxy adhesive. Through the 4). The of compound the components joined was compound made by ancan be high stiffness this kindbetween of adhesive (see Table being 1) a quasi-rigid epoxy adhesive. Through the high of state this kind adhesive assumed. The real loading situation in thestiffness mounted was of modelled in (see tests with a load Table introduction into thecompound specimens connections. A cohesive failure of 1) a quasi-rigid canatbebolted assumed. The real loading situation the bondline due to restoring at in the profile be registered in the mounted state was forces modelled tests with ends a loadcould introduction into the for all tests. Aspecimens significantatimprovement of profile stiffnessfailure could of bethe achieved compared bolted connections. A cohesive bondline due to to a reference façade transom reinforcement (dashedfor line Figure 4). restoring forces at thewithout profile ends could be registered all in tests. A significant improvement of profile stiffness could be achieved compared to a reference façade transom without reinforcement (dashed line in Figure 4).

Based on Eurocode (DIN 1990) the model calibration was realized in Based on Eurocode (DIN 1990) the model calibration was realized in a final l calibration was realized in athe final step of The results are partial safety factors for the design of t research. step of the research. The results are partial safety factors for the design of the y factors for the design of the mentioned application examples of bonding technology. For the bonded façade mentioned application examples of bonding technology. For the bonded façade gy. For the bondedconnection, façade connection, a factor of  = 2.2 was found and for the façade reinforcement, M = 1.6. a factor of M = 2.2 was found and for the façade reinforcement, ade reinforcement, M = 1.6. = 1.6. Conversion factors The objective of implementing conversion factors is to enhance the Conversion n factors is to enhance thefactors design with themodel result that the resistance can be modified with the following equ modified with the following equation. Rk The objective ofR implementing conversion factors is to enhance the design  i d model with the result M that the resistance can be modified with the following equation. The conversion factors i take into account the different

influences on nt the different influences onresistance. the value ofBecause the carrying and deformation behaviour of a deformation behaviour R ofk asignificantly bondline isaffected by the environmental temperature and the bondl = ⋅ηi R d l temperature and the bondline γ M thethickness, conversion factors t and m are introduced to the presented troduced to the presented models. possible Itto is describe the bondline resistance as a function of the spe nce as a function of the specific influence value, based on small scale specimen tests with various temperatures ts with various temperatures thicknesses. and bondlineThe results of these experiments are considered in relatio ents are considered in relationvalues, to reference which are generated from tests at normal condition (e.g. Table

factor factor of of M=M 2.2 = 2.2 was was found found and and forfor thethe façade façade reinforcement, reinforcement, M=M 1.6 =1 factor of M = 2.2 was found and for the façade reinforcement, M = 1.6. Conversion factors Conversion Conversion factors factors Conversion factors Proceedings of the METNET Seminar 2016 in Castellonmodel 25 The objective of implementing conversion factors is toconversion enhance the design The Theobjective objectiveof ofimplementing implementing conversion factors factors is isto toenhance enhance thet The objective of implementing conversion factors is to enhance the design model with the result that the resistance can be modified with the following equation. with with thethe result result that that thethe resistance resistance can can bebe modified modified with with thethe following following e with can be modified with the following equation. Rk the result that the resistance R R R  d

i R k  MR   d M i

k k    M M i i

R   dR d

iThe take into account theidifferent influences onthe the valueinfluences ofinfluence The conversion factors The take into into account account the different different conversion conversion factors factors i take conversionthe factors takeand into the the different influences onathe i take intoaccount account different influences on the value of The The conversion factors resistance. Because carrying deformation behaviour of bondline is resistance. resistance. Because Because the thedeformation carrying carryingand anddeformation deformation behaviour behaviour of o value of resistance. Because the carrying and behaviour of a resistance. Because the carrying and deformation behaviour ofthickness, a bondline is significantly affected bysignificantly the environmental temperature and the bondline significantly affected affected by by the the environmental environmental temperature temperature and and the the bon bo bondline affected is significantly affected by the environmental temperature significantly by the environmental temperature and the and bondline Itthickness, the conversion factors  and  are introduced to the presented models. is t m thetheconversion conversion factors factors t and andmare are are introduced introduced thepresented presen the bondline thickness, the conversion factors introduced to to tothe t and m thetothe conversion factors  possible and mto are introduced toofresistance the models. It is possible describe bondline resistance as athe function the presented specific influence presentedthe models. It ist to describe bondline asas a a possible possible to describe describe the thebondline bondline resistance resistance as a function functionof ofthe thesp describe the bondline resistance as a scale function of the specific influence value, possible based ontosmall specimen tests with various temperatures and bondline function of thescale specific influence value, based on small specimen tests value, value, based based on on small small scale scale specimen specimen tests tests with with various various temperature temperatu value, based ontemperatures small scaleexperiments specimen tests with various and bondline with various and bondline thicknesses. The results of these thicknesses. The results of these are intemperatures relation toconsidered reference thicknesses. thicknesses. The The results results of considered of these these experiments experiments are are considered in in relati rel thicknesses. The results of these experiments are considered in relation to reference experiments are considered in relation to reference values, which are generated values, which are generated from tests at normal condition (e.g. Table 1). Figure 5 values, values, which which are are generated generated from from tests tests at at normal normal condition condition (e.g. (e.g. Tab T from tests atfor normal condition (e.g. Table 1).at Figure 5 conversion presents the results values, which are generated from tests normal condition (e.g.for Table 1). Figure 5 presents the results the teststhe and the determined factors at different presents presents the results results for for the the tests tests and and the the determined determined conversion conversion fac fa the tests and the determined conversion factors at different temperatures for presents results for the tests and the determined conversion factors at different temperatures forthe the MS polymer-based adhesive. A polymer specific behaviour, a specif temperatures temperatures for for the the MS MS polymer-based polymer-based adhesive. adhesive. A A polymer polymer spe the MS polymer-based A polymer adhesive. specific behaviour, a reduction temperatures for thestrength MSadhesive. polymer-based A polymer specific behaviour, a reduction of the bondline with increasing temperature, can be observed. reduction ofincreasing of thethebondline bondline strength strength with with increasing increasing temperature,can c of the bondline reduction strength with temperature, can be observed. Due temperature, reduction of the bondline strength with increasing temperature, can be observed. Due to thetofact that the conversion factors are referred to a defined limit state and Due Due to to thethe fact fact that that the the conversion conversion factors factors are are referred referred to to aa defined define the fact that the conversion factors are referred to a defined limit state and Due thetofact that the conversion factors are referred to a ofdefined limit state and thereby to thereby atorequired target reliability Eurocode, the determination of these factors a required target of Eurocode, the determination these thereby thereby toreliability to a required aofrequired target target reliability reliability of of Eurocode, Eurocode, thethe determination determinatio thereby a based required target reliability of resistance Eurocode, the determination of these factors is based onfactors thetodesign values of on the resistance (see figure 5). is onis the design values ofdesign the figure 5). is based based on the the design values values of(see of the the resistance resistance (see (see figure figure 5).5). is based on the design values of the resistance (see figure 5). Butt joint test Butt joint test

Lap shear test Lap shear test

Strength [N/mm²] Strength [N/mm²]

Strength [N/mm²]

Strength [N/mm²] Strength [N/mm²]

Strength [N/mm²]

Lap shear test Butt joint test Butt joint test

Lap shear test 3,0 3,0 1,8 1,8 1,8 3,0 1,6 1,8 2,5 1,6 1,6 2,5 2,5 1,4 1,6 1,4 1,4 2,5 2,0 1,4 2,0 2,0 1,2 1,2 1,2 2,0 1,0 1,2 1,0 1,0 1,5 1,5 1,5 0,8 t = 1,02 1,0 t = 1,02 0,8 0,8t = 1,02 1,5 1,0 t = 1,18 0,6 t = 1,18 0,8 t = 1,02 1,0 1,0t = 1,18 0,6 0,6 t = 0,95 t = 0,95 0,4 1,0 t = 1,18 0,6t = 0,95 t = 0,86 0,4 0,4 t = 0,86 t = 0,86 0,5 t = 0,95 0,5 0,5 t = 0,31 t = 0,33 0,2 0,4 t = 0,33 t = 0,3 t = 0,86 t = 0,31 t = 0,31 0,2 0,2 0,5 t = 0,33 0,0 0,0 t = 0,31 0,0 0,2 0,0 0,0 0,0 ‐400 ‐20 040 40 20 40 60 80 100 ‐40 ‐40 ‐40 ‐200,0 0 20 40 60 80 100 ‐40 ‐40 0,020 ‐20 ‐200 020 2040 4060 6080 80 100 ‐20 ‐20 020 60 60 80 80 100100 ‐40 ‐20 0 20 40 60 80 100 ‐40 ‐20 0 20 40 60 80 100 Temperature [°C] Temperature [°C] Temperature [°C] Temperature [°C] Temperature [°C] Temperature [°C] Temperature [°C] Temperature [°C] Mean value of test results Design value Mean Mean value value of test of test results results Design Design value value Mean value of test results Design value 3,0

Figure 5: Conversion factors for a MS polymer adhesive Figure 5: Conversion factors for a MS polymer adhesive

Figure Figure 5: 5: Conversion Conversion factors factors forfor a MS a MS polymer polymer adhesive adhesi Figure 5: Conversion factors for a MS polymer adhesive

LONG-TERM BEHAVIOUR OF OF BONDED STEEL JOINTS LONG-TERM BEHAVIOUR BONDED STEEL JOINTS LONG-TERM LONG-TERM BEHAVIOUR BEHAVIOUR OFOF BONDED BONDED STEEL STEEL JOINTS JOINTS LONG-TERM BEHAVIOUR OF BONDED STEEL JOINTS The presented investigations focused on quasi-static load situations and short-term The presented investigations focused on quasi-static load situations and The The presented presented investigations investigations focused focused onon quasi-static quasi-static load load situations situation The presented investigations focused onthat quasi-static load situations short-term It is that known, however, the bondline strength can and loading. It short-term is known,loading. however, the bondline strength can be reduced by longloading. loading.It Itis isknown, known,however, however,that thatthethebondline bondline strength strength can canbebere be reduced bycyclic long-term influences cyclic loading. That is why ongoing loading. It and is known, however, thatandis the bondline strength can be reduced by longterm influences loading. That why ongoing research projects at the term term influences influences and andcyclic cyclic loading. loading. That Thatis iswhy whyongoing ongoing research researc research projects at the Chair aim at taking into account such effects. term influences and cyclic loading. That is why ongoing research projects at the Chair aim at taking into Chair account such effects. Chair aim aim at at taking taking into into account account such such effects. effects. Chair aim at taking into account such effects. Cyclic loading The objective of an ongoing research project is to carry out cyclic test procedures for the application of the bonded façade connection, taking into account long-term effects (Pasternak et al. 2016). In a first step, representative load functions for the test sequences were deduced from actual measured wind speed and air temperature data from various weather stations. The main

Proceedings of the METNET Seminar 2016 in Castellon

Cyclic loading The objective of an ongoing research project is to carry out cyclic test procedures for of the project was presented at the Metnet Seminar (Ciupack theprocedure application of the bonded façade connection, taking into2015 account long-term et al.(Pasternak 2015). effects et al. 2016). In a first step, representative load functions for the test sequences were deduced from actual measured wind speed and air temperature data from tovarious stations. The main procedure theloading, project was In order describeweather the general behaviour of bonded joints underof cyclic presented at the Metnet 2015 (Ciupackand et al. some investigations onSeminar small scale specimens the2015). façade connection were carried out under constant amplitudes. The stress ratio was chosen with a In value order of to 0.1, describe general behaviour of bondedtensile joints loading under cyclic loading, whichthe means a varying but permanent for the some investigations onheating small of scale specimens the façade connection bondline. To prevent the adhesive due and to mechanical load, the tests were carried under amplitudes. stress was chosen with and a value of were out carried outconstant with a frequency of 5 The Hz. If 2·106 ratio load cycles are reached 0.1,nowhich means a varying but permanent tensile loading for the bondline. To damage is detected, the value of the load is assigned within the fatigue prevent heating of the adhesive due to mechanical load, the tests were carried out strength. The results of these examinations for a polyurethane adhesive are with a frequency of 5 Hz. If 2·106 load cycles are reached and no damage is shown in Figure 6. detected, the value of the load is assigned within the fatigue strength. The results of these examinations for a polyurethane adhesive are shown in Figure 6. All specimens failed with a cohesive failure of the bondline. The notch is calculated with a value the buttThe jointnotch tests sensitivity and of All sensitivity specimens(k) failed with a cohesive failureofof6.19 thefor bondline. (k) 6.74 for the façade connection. The threshold of the load, which the joint canfaçade is calculated with a value of 6.19 for the butt joint tests and of 6.74 for the withstandThe permanently value of 3.3 kN k), is evaluated connection. threshold(fatigue of the limit load, Swhich the joint with can awithstand permanently for the small scale specimen and of 1.3 kN for the specimen component. (fatigue limit Sk), is evaluated with a value of 3.3 kN for the small scale specimen and of 1.3 kN for the specimen component. 102

Maximum load [kN]

Adhesive basis: Substrat: Steel Load: Axial, R = 0.1

Small scale specimen: Sk = 3.3 Specimen component: Sk = 1.3

Acrylate sinusoidal

f = 5 Hz Small scale specimen

101

f = 5 Hz Specimen component f = 1 Hz Small scale specimen Without fracture 100 103

104

105 Cycles [-]

106

107

Sk Fatigue strength [kN]

Figure 6: Comparison of test results of cyclic tests

Additional investigations showed a dependency of the bondline strength on the testing frequency (circular symbols 6) for of small specimens. Additional investigations showedina Figure dependency the scale bondline strengthThe on lower thethe frequency, the smaller the fatigue limit strength. Based on these experiments it testing frequency (circular symbols in Figure 6) for small scale specimens. canThe be lower concluded that the real wind load sequences (low frequency and small the frequency, the smaller the fatigue limit strength. Based on amplitudes) mean higher fatigue damagethat thanthe observed in load the standard these experiments it can be concluded real wind sequencesexperiments. (low Thefrequency future work the amplitudes) Chair will focus the fatigue possibilities of modelling the actual and of small meanon higher damage than observed amplitude collective and frequencies as well as methods to reduce testing in the standard experiments. The future work of the Chair will focus on thetime of cyclic tests. possibilities of modelling the actual amplitude collective and frequencies as well as methods to reduce testing time of cyclic tests. Creep behaviour If the trapezoidal façade is used as structural or stabilizing construction element, the bonded connection is loaded by wind loads and dead loads simultaneously. Such

Creep behaviour

If the trapezoidal façade is used as structural or stabilizing construction element, the bonded connection is loaded by wind loads and dead loads

Proceedings of the METNET Seminar 2016 in Castellon

simultaneously. Such long-term impacts lead to creep effects in the bondline. To study the general creep behaviour of bondlines, shear creep tests at overlapped steel plates bonded with a structural epoxy adhesive were carried (Sahellie et al. 2015). The testing time totalled 110 days and the testing long-termout impacts lead to creep effects in the bondline. To study the general creep condition was air temperature of 20 °C ± 2 °C and relative humidity of 65% ± behaviour4%. ofThe bondlines, shear creep tests at overlapped steel plates bonded with a objective was to estimate the bondline strength for a defined lifetime. structuralTherefore, epoxy adhesive were carried outmaterials (Sahellie et al. 2015). Theand testing time two model for viscoelastic were used, Findley’s totalled 110 days and the testing condition was air temperature of 20 °C ± 2 °C and Burger’s models.

relative humidity of 65% ± 4%. The objective was to estimate the bondline strength for a defined lifetime. Therefore, twoshear model forbehaviour viscoelastic materials were used, The test results showed that the creep is dependent on the Findley's magnitude and Burger's of themodels. applied stress. Moreover, it was found that Findley’s model fits better to the data for shorter lifetime periods in general. However, it

found showed that longerthat lifetime adhesives should be predicted Burger’s on the The testwas results the ofshear creep behaviour is by dependent model. Findley’s model, whichMoreover, indicated huge relative errors, is not useful formodel fits magnitude of the applied stress. it was found that Findley’s longshorter lifetimeslifetime due to the unlimited retardation spectrumit of thatfound that better to predicting the data for periods in general. However, was model (Feng et al. 2005). longer lifetime of adhesives should be predicted by Burger's model. Findley's model, which indicated huge relative errors, is not useful for predicting long lifetimes due to the unlimited retardation spectrum of that model (Feng et al. 2005).

REINFORCEMENT OF STEEL STRUCTURE WITH BONDED CFRP LAMINATES

REINFORCEMENT OF STEEL STRUCTURE WITH BONDED CFRP LAMINATES General investigation General An investigation innovative alternative to typical reinforcement measurements of steel An innovative alternative to or typical reinforcement measurements ofcreated steel structure, structure, e.g. welding screwing of additional steel sheets, could be by bonding of CFRPof(carbon fibre reinforced polymer) laminates to the by steel e.g. welding or screwing additional steel sheets, could be created bonding of members. In the scope of anpolymer) initial research projectto(Pasternak al. 2015) CFRP (carbon fibre reinforced laminates the steelet members. In the and efficiency steel-CFRP-composites studied. and The efficiency scope of applicability an initial research projectof(Pasternak et al. 2015) were applicability main objectives werewere the testing based and analytical demonstration the of steel-CFRP-composites studied. The main objectives were theoftesting based potential, as well as the development of recommendations for utilization of and analytical demonstration of the potential, as well as the development of bonded CFRP reinforcements in steel structure. recommendations for utilization of bonded CFRP reinforcements in steel structure. Different influences of parameters, e.g. bondline thickness, joint length and surface Different influences of parameters, e.g. bondline thickness, joint length and pre-treatment of components being connected, were studied at double lap shear surface pre-treatment of components being connected, were studied at double joints. Moreover, agingMoreover, tests were on shearonlap specimens, which were lap shear joints. agingperformed tests were performed shear lap specimens, outsourced at various temperatures and climate conditions. The corrosion which were outsourced at various temperatures and climate conditions. Theresistance was determined means was of salt spray tests. corrosionby resistance determined by means of salt spray tests. b)

60 60 50

100

110

Steel adherend CFRP laminate

CFRP laminate

a) Steel adherend Bondline

400 100

120

160

120

100

Figure 7: (a) Double steel-CFRP laminate-assemblies; (b) dumbbell specimen Figure 7: (a) Double steel-CFRP laminate-assemblies; (b) dumbbell specimen

Proceedings of the METNET Seminar 2016 in Castellon

These investigations were carried out using tensile shear tests on bonded double steel-CFRP laminate-assemblies based on DIN EN 1465 (see Figure 7a). It was shown that the carrying capacity of the specimens can be increased by increasing the joint length up to 100 mm. From this specific joint length, no essential improvement of bondline strength was possible. For all tests an adhesive failure at the laminates surface was observed, which can be explained by the characteristic of the employed adhesives, commonly used for the bonding of CFRP laminates to concrete. Thus, the adhesive is designed to have a lower strength than connected steel members. Taking into account the aging effects, the specimens were tested for climate change conditions based on DIN EN ISO 9142 with a single dependency (temperature) and double dependency (temperature and humidity). Depending on the kind of adhesive, climate degradation change conditions based on DIN EN ISO were 9142observed. with a single dependency effects, but also post-curing effects,

(temperature) and double dependency (temperature and humidity). Depending on the kindThe of adhesive, degradationwere effects, but also effects,specimens were observed. described experiments repeated for 8 post-curing mm thick dumbbell The described experiments were repeated for 8 mmtothick dumbbell influences, specimens (see (see Figure 7b) loaded in tension. Supplementary the mentioned Figure the 7b) arrangement loaded in tension. Supplementary mentionedsingle-sided influences, the of the CFRP reinforcement to wasthe investigated, arrangement of the CFRP reinforcement was investigated, andofdoubleand double-sided. Some results are presented in Figure single-sided 8. The efficiency sided. Some results are presentedis in Figure 8.diagram, The efficiency of this this reinforcing measurement shown in the which shows thatreinforcing the measurement is of shown in the diagram, which is shows that the resistance the double reinforced specimen 38% higher thanresistance that of the of the double reinforced specimen is 38% higher than that of the reference specimen. reference specimen. 300 250 Force [kN]

200 150

Without CFRP Single-sided CFRP laminate Double-sided CFRP laminate

100 50 0

Figure 8: Diagram CFK

30 Deformation [mm]

Figure 8: Diagram CFK

Refurbishment of steel members stressed by fatigue An ongoing research project (Ummenhofer et al. 2016) deals with the refurbishment of steelbonded members CFRP stressed by fatigue of steelRefurbishment structure with laminates. The main idea is to substitute common refurbishment procedures, e.g. welded steel sheets, with this innovative An ongoing research project (Ummenhofer et al. 2016) of deals with welds the rehabilitation method. Further, the possibility of refurbishment cracked will refurbishment of steel structure with bonded CFRP laminates. The main be investigated. The efficiency of bonded CFRP laminates or sheets for the idea is to common refurbishment e.g. welded steel reduction and thesubstitute avoidance of crack growth will procedures, also be studied. In this manner, sheets, with this innovative rehabilitation method. Further, the possibility different application dependent retrofitting techniques will be developed and of loaded refurbishment of cracked welds will investigated. The efficiency cyclically specimens will be tested. Thebefocus is on long-term effects,ofe.g. the bonded CFRP or sheets for thethe reduction the avoidance of crack The loss of pre-stress in laminates CFRP sheets through creepand behaviour of a bondline. growthofwill also be studied. this manner, is different application dependent development a suitable testingIn methodology planned for the characterizing of retrofitting willbehaviour be developed and cyclically loaded specimens the strength and techniques deformation of CFRP reinforced steel construction, will be tested. The focus is on long-term effects, e.g. the loss of pre-stress in aiming to establish this new refurbishment procedure. CFRP sheets through the creep behaviour of a bondline. The development of a

CONCLUSIONS AND FUTURE WORK Bonding technology is more than just an alternative joining method for steel structures, which can be shown by various applications and research projects in

Proceedings of the METNET Seminar 2016 in Castellon

suitable testing methodology is planned for the characterizing of the strength and deformation behaviour of CFRP reinforced steel construction, aiming to establish this new refurbishment procedure.

CONCLUSIONS AND FUTURE WORK Bonding technology is more than just an alternative joining method for steel structures, which can be shown by various applications and research projects in different fields. Despite many advantages of bonded joints, the civil engineering and especially steel construction community is reluctant to embrace this innovative technique. Often it is justified by doubts about the durability and a lack of experience. Interpreting such doubts as open questions and challenges creates the possibility and the potential for exceptional innovations. The general interest in developing standards as a basis for analysis and design of adhesive joints in steel is expected to rise with a growing number of functional and numerical applications. The future work for the presented application examples for façade structures focuses on experimental investigations under variable amplitudes and the creation of a scientific based guideline for the design of bonded steel joints. For the introduction of alternative innovative bonded steel joints, the effort is limited to the development of engineering-models, planning and construction design. General technical approvals or individual approvals can be achieved more easily, thus sustainably increasing the innovative capacity of small and medium-sized enterprises.

REFERENCES Dilger K., Pasternak H., et al. (2008), IGF Project No. 169 ZBG, Neue Konstruktionen durch Einsatz von Klebverbindungen im Stahlbau, Forschungsbericht für die Praxis P654, Forschungsvereinigung Stahlanwendung e. V., FOSTA, Verlags- und Vertriebsgesellschaft, Düsseldorf, Germany. Meinz, J. (2010), Kleben im Stahlbau – Betrachtungen zum Trag- und Verformungsverhalten und zum Nachweis geklebter Trapezprofilanschlüsse und verstärkter Hohlprofile in Pfosten-Riegel-Fassaden, Dissertation, Brandenburgische Technische Universität Cottbus, Germany. Pasternak H., Dilger K., et al. (2012), IGF Project No. 16494 BG, Entwicklung eines Eurocode-basierten Bemessungskonzepts für Klebverbindungen im Stahlbau (in Anlehnung an DIN 1990), Germany. DIN EN 15870 (2009), Adhesives – Determination of tensile strength of butt joints, German version EN 15870:2009.

Proceedings of the METNET Seminar 2016 in Castellon

DIN EN 14869 (2004), Structural adhesives - Determination of shear behaviour of structural bonds - Part 2: Thick adherends shear test; German version EN 14869-2:2004. DIN EN 1990 (2010), Eurocode 0: Basis of structural design; German version EN 1990:2002 + A1:2005 + A1:2005/AC:2010. Pasternak H., Dilger K., et al. (2016), IGF Project No. 18161 BG, Untersuchungen zum Tragverhalten und der Lebensdauer von Klebverbindungen im Stahlbau unter zyklischer Belastung, ongoing research project, Germany, 01.04.2014 – 31.12.2016. Ciupack Y. et al. (2015), Adhesive bonded steel structures under cyclic loading, Virdi, K.; Tenhunen, L. (Editors), Proceedings of the METNET Seminar 2015 Budapest, Häme University of Applied Science, 2015. Sahellie S., Pasternak, H. (2015), Expectancy of the lifetime of bonded steel joints due to long-term shear loading, Archives of civil and mechanical engineering 15, pp. 1061 – 1069. Feng C.-W., Keong C.-W., Hsueh Y.-P., Wang Y.-Y., Sue, H.-J. (2005), Modeling of long-term creep behavior of structural epoxy adhesives, International Journal of Adhesion & Adhesives 25, pp. 427 – 436. Pasternak H., Feldmann M., Ummenhofer Th. et al. (2015), IGF Project No. 17700BG, Systematische Untersuchungen zur Verstärkung von Stahlkonstruktionen mit kohlefaserverstärkten Kunststoffen (CFK) - STAKOK, DVS, Germany. DIN EN 1465 (2009), Adhesives - Determination of tensile lap-shear strength of bonded assemblies; German version EN 1465:2009. DIN EN ISO 9142 (2004), Adhesives - Guide to the selection of standard laboratory ageing conditions for testing bonded joints; German version EN ISO 9142:2003. Ummenhofer Th., Pasternak H., Feldmann M. (2016), IGF Project No. 19032 BG, Einsatz von geklebten Kohlestoff-Faserverbundwerkstoffen zur Sanierung ermüdungsgeschädigter Stahlkonstruktionen (FASS), ongoing research project, Germany, 01.02.2016 – 31.07.2018.

Proceedings of the METNET Seminar 2016 in Castellon

RESEARCH OF ULTRA-HIGH-STRENGTH AND WEARRESISTANT STEELS USING ADVANCED TECHNIQUES Raimo Ruoppa, Kimmo Keltamäki, Rauno Toppila Lapland University of Applied Sciences, Finland Vili Kesti SSAB, Finland

ABSTRACT The usability of ultra-high-strength and wear-resistant steels has been researched for several years at Lapland University of Applied Sciences. Bendability tests have been carried out by using a commercial press brake and, recently, a new “ultimate bending machine” was built up to allow steels with higher strength and thickness to be tested. Research techniques have been developed in cooperation with SSAB and the University of Oulu. The techniques, as well as ultra-high-strength and wear-resistant steels, are reviewed in this paper. Welding and wearing have also been researched for many years in cooperation with SSAB. The most common welding technique is mechanized MIG/MAG welding in method tests but shielded metal arc welding is also used for repair welding research. The quality and mechanical properties of welds are assessed using destructive testing results. Wear-resistance tests are specialized field tests conducted under reallife conditions in the Outokumpu Chrome Oy Kemi mine, performed in cooperation with Tapojärvi Oy. The usual test targets of the wear-resistant steels are cutting edges and feed hopper plates. The wear of the wear-resistant plates was measured with the ATOS optical measurement system from GOM GMBH at Lapland University of Applied Sciences.

INTRODUCTION Lapland University of Applied Sciences has an “Arctic Steel and Mining” (ASM) research team. The main focus of its work has been in the usability of ultra-high-strength and wear-resistant steels, as well as stainless steels. The team has participated together with the University of Oulu, SSAB Europe, and some other organizations in a research cluster which has recently published various papers at conferences and in journals. This paper mainly focuses on reviewing the techniques which have been utilized within these projects, as well as the introduction of ultra-high-strength steels. Detailed results of the projects can be read in the papers listed in the references below.

Proceedings of the METNET Seminar 2016 in Castellon

ULTRA-HIGH-STRENGTH AND WEAR-RESISTANT STEELS Ultra-high-strength steels (UHSS) are usually considered to be steels with a yield strength of more than 550 N/mm2 and an ultimate tensile strength of more than 700 N/mm2. That is at least about 1.5 â&#x20AC;&#x201C; 2.5 times higher compared to regular structural steels. Ultra-high-strength steels are suitable for fabrication but their higher strength and lower ductility make their processing more challenging and it becomes essential to follow instructions carefully. The high strength of UHS steels can be utilized in lightweight structures, which influences the costs and increases the lifetime of the devices. They are The high suitable strengthforofapplications UHS steelswhere, can be in weight lightweight structures, which forutilized example, is critical. Reducing the influences the costsweight and also increases the lifetime of the devices. They are suitable for applications reduces the consumption of steel, which also influences the carbon where, for example, weight is critical. Reducing the weight also reduces the consumption dioxide emissions during the production of the steel. Typical applications of of steel, which also are, influences the carbon dioxide emissions duringthe the production of the UHS steels for example, the booms and frames of cranes, frames of steel. Typical ofand UHS steels are,see forFigure example, the booms and frames of trucksapplications and their beds, parts of cages; 1.

cranes, the frames of trucks and their beds, and parts of cages; see Figure 1.

Figure 1.steels Typical Figure 1. Typical applications of UHS [SSAB] applications of UHS steels Wear-resistant steels are typically used in applications such as the scoops and lip plates of Wear-resistant steels are typically used such trucks as the scoops and earth movers, mining machines, the parts of in theapplications cement mixer and batching plants lip plates of earth movers, mining machines, the parts of the cement mixer that are subject to wear, agricultural and wood-handling machinery, bed structures, trucks and batching plants that are subject to wear, agricultural and woodfeeders and funnels, and the edges of different crushers; see Figure 2. handling machinery, bed structures, feeders and funnels, and the edges of different crushers; see Figure 2.

Figure 1. Typical applications of UHS steels Wear-resistant steels are typically used in applications such as the scoops and lip plates of 33 earth movers, mining machines, the parts of the cement mixer trucks and batching plants that are subject to wear, agricultural and wood-handling machinery, bed structures, feeders and funnels, and the edges of different crushers; see Figure 2. Proceedings of the METNET Seminar 2016 in Castellon

Figure 2. Typical applications of wear resistant steels [SSAB]

Figure 2. Typical applications of wear resistant steels

In the example below a demonstrationisis provided provided by of a of life-a life-cycle In the example shownshown below a demonstration bymeans means approach of howfrom a change from steel standard steel toinUHSS in vehicles approachcycle of how a change standard to UHSS vehicles reduces carbon reduces carbon dioxide emissions. If, for example, 1.3 millionsteel tons of dioxide emissions. If, for example, 1.3 million tons of standard is standard replaced with one steel replaced oneproduction million tons UHS steel in the production of cars: million tons ofisUHS steelwith in the of of cars: 1. upgrading when upgrading to a high-strength steel, the application its per- even 1. when to a high-strength steel, the application retains retains its performance formance even though steelinisweight used. savings This results in weight savings and though less steel is used. This less results for the steel application for less the steel means that less steelfewer needsresources to be produced. means that steelapplication needs to beand produced. Additionally, are needed; Additionally, fewer resources are needed; 2. as 2. muchasasmuch 90% as of 90% the reduced environmental impact can be related the use of the reduced environmental impact can beto related to phase of lighterthe vehicles, through reduced fuel consumption; use phase of lighter vehicles, through reduced fuel consumption; 3. from a life-cycle perspective, this case shows the substantial savings 3. from a life-cycle perspective, this case shows the substantial savings that can be can be achieved throughsteels. using high-strength steels. achievedthat through using high-strength

Figure 3. Example of the reduction of the carbon dioxide emissions through the use of UHS steels [SSAB] Figure 3. Example of the reduction of the carbon dioxide emissions through the use of UHS steels A. When 300,000 tonnes less steel needs to be produced, the CO₂ emissions from upstream suppliers will decrease by 200,000 tonnes since less energy and raw material are needed. B. The reduction in steel produced by 300,000 tonnes results in 500,000 tonnes fewer CO₂

Proceedings of the METNET Seminar 2016 in Castellon

A. When 300,000 tonnes less steel needs to be produced, the CO₂ emissions from upstream suppliers will decrease by 200,000 tonnes since less energy and raw material are needed. B. The reduction in steel produced by 300,000 tonnes results in 500,000 tonnes fewer CO₂ emissions from SSAB’s steel production. C. When the current European vehicle fleet is upgraded, CO₂ emissions will decrease by 7.3m tonnes. D. The total CO₂ savings arising from this hypothetical case are around 8m tonnes.

BENDABILITY Manufacturing components for UHS steel applications almost invariably require bending, which is the most used, the best, and in many cases the only choice for forming, especially when the thickness of the sheet is increased. In modern high-strength steel structures (for example new generation boom and bed structures), bending is increasingly used and it often replaces, for example, welding. This leads in many cases to better fatigue durability of the structure at the same time as a reduction of the production costs. Increasing the strength, however, makes bending more challenging and it

very important to obtain information about bendability the Increasing becomes the strength, however, makes bending more challenging and itand becomes very factors have anabout influence on the process. aim have is to an have important to obtainwhich information bendability and the While factorsthe which influence a process as isefficient trouble-free as as possible, the on the process. Whilethat theisaim to haveand a process that is efficientfollowing and trouble-free as instructions be emphasized. makes it very perform possible, following the must instructions must beThat emphasized. Thatimportant makes it tovery important to full-scale bendabilitytests testswhich which are carried machines perform full-scale bendability carriedout outwith withreal realbending bending machines and samples large enough large to correspond to actual parts. Onparts. the basis ofbasis the tests, and samples enough to correspond to actual On the of the correct instructionstests, can correct be provided to customers. instructions can be provided to customers.

In the bending, steel plate isplate bentiswith, for example, a hydraulic press In the abending, a steel bent with, for example, a hydraulic pressbrake brake using a toolset containing a punch and die. Theandprocess, usually usually considered as three-point using a toolset containing a punch die. The process, considered as bending, isthree-point illustrated in Figure is4.illustrated in Figure 4. bending,

Figure 4. Bending a steelplate plate in in a V-die: a) puncha) in lower position; b) punch position; released Figure 4. Bending of a ofsteel a V-die: punch in lower b) punch released

In Figure 4, spring-back may also be noticed. This phenomenon typically takes place in three-point bending. After the punch is released, the angle of the plate is increased as a

Proceedings of the METNET Seminar 2016 in Castellon

In Figure 4, spring-back may also be noticed. This phenomenon typically takes place in three-point bending. After the punch is released, the angle of the plate is increased as a result of the elastic deformation which is, in addition to plastic deformation, present in the steel. The spring-back may be measured using a video camera. Two images are sampled from the video, one from the lower end of the punch and another after the release of the punch; see Figures 4a and b. The spring-back angle is measured from the difference between the angles detected from the images. In general, greater the strength of the steel, the higher the spring-back, and knowledge of the magnitude of the springback is valuable for foundries manufacturing steel products. When the strength of steel increase, its formability usually decrease. Figure 4 also illustrates the bending radius, which, in practice, corresponds to the radius of the punch. When the bending radius decreases, the strain on the outer surface increases and the probability of failure becomes higher. Therefore, the bending radius has to be large enough to avoid failures. In bendability tests, the minimum bending radius (Rmin) of the material, i.e. the smallest radius which is able to be used without failures, is usually defined. The result of the test is evaluated by means of visual inspection. Samples which have been bent using various bending radii are shown in Figure 5aâ&#x20AC;&#x2030;â&#x20AC;&#x201C;â&#x20AC;&#x2030;c. As a result, various degrees of failure have been developed on the surface. For the quality of the bend, certain requirements have been laid down and on the basis of these requirements, the result of the test will either be accepted or not.

a) No failures;b) b) and and c) failures appeared in the steel as a result of toosteel small a bending Figure 5. a)Figure No5.failures; c) failures appeared in the as a radius result of too small a bending radius

After the minimum bending radius has been reached, a sufficient number

of repetitions is carried thereached, accepted a bending radius. Henceofthe After the minimum bending radiusout hasusing been sufficient number repetitions is durability the material is able to be determined and durability the steel manufacturer carried out using the of accepted bending radius. Hence the of the material is able is able to guarantee a minimum bending radius, threeatimes the to be determined and the steel manufacturer is ablefortoexample, guarantee minimum bending thickness of the sheet (Rmin=3xt). The value usually depends on the grade and radius, for example, three times the thickness of the sheet (Rmin=3xt). The value usually of the steel and typically varies between 2 and 6 with UHS steels. depends on thickness the grade and thickness of the steel and typically varies between 2 and 6 with UHS steels. In addition to basic bendability testing, detailed investigations of the samples have also been carried out. With the GOM ARGUS 3D forming analysis In addition to basicsamples bendability testing, detailed investigations of more the samples system, have been studied with the aim of getting detailed have also been carriedinformation out. With the GOM ARGUS 3Dphenomena forming analysis samples about the metallurgical involvedsystem, in bending [Arola have been studied withet al. the2015, aimRuoppa of getting moreIn detailed information metallurgical et al. 2014]. this technique, a grid is about marked the on the phenomena surface involved in bending [Arola et al. On 2015, Ruoppa al. 2014]. this of the sample before bending. the basis of theet change in the In grid, a technique,

a grid is marked on the surface of the sample before bending. On the basis of the change in the grid, a computer system calculates the strain distribution on the surface. The system is illustrated in Figure 6a and an example of a strain map is shown in Figure 6b. As a result, information about the processes during failure and the factors affecting them is

radius, for example, three times the thickness of the sheet (Rmin=3xt). The value usually depends on the grade and thickness of the steel and typically varies between 2 and 6 with steels. of the METNET Seminar 2016 in Castellon 36UHS Proceedings In addition to basic bendability testing, detailed investigations of the samples have also been carried out. With the GOM ARGUS 3D forming analysis system, samples have been studied with the aim of getting more detailed information about the metallurgical phenomena involved in bending [Arola et al. 2015, Ruoppa et al. 2014]. In this technique, computer calculates thesample strain distribution on the surface. system a grid is marked on system the surface of the before bending. On theThe basis of the change is illustrated in Figure 6a and an example of a strain map is shown in Figure in the grid, a computer system calculates the strain distribution on the surface. The system 6b. As a result, information about the processes during failure and the factors is illustrated in Figure 6a and an example of a strain map is shown in Figure 6b. As a affecting them is gained, which helps the steel producer to develop products result, information about the processes during failure and the factors affecting them is and customer instructions. gained, which helps the steel producer to develop products and customer instructions.

Figure 6. a) Measurement of of the strain the GOM ARGUSwith 3D forming b) example3D forming Figure 6. a) Measurement the distribution strain with distribution the analysis GOMsystem; ARGUS of strain mapping in bent sample analysis system; b) example of strain mapping in bent sample

A more detailed study study of the samples hasalso also performed by microscopic A more detailed of the samples has beenbeen performed by microscopic examinations. Figure 7 shows microscopic imageimage of theofcross-section of of a bend. From examinations. Figure 7a shows a microscopic the cross-section a the image, bend. scratches surface, as well the reduction the thickness, can be Frombeneath the image,the scratches beneath theas surface, as well asofthe reduction detected. Hardness may can alsobebe measured frommay thealso cross-section of the thickness, detected. Hardness be measured and from by the means of minimum hardness (redand line); the neutral axis has been located inthe different cross-section by means of minimum hardness (red line); neutralcases axis [Kesti et al. 2014 & has 2015]. neutral axis refers the line zero deformation. By beenThe located in different casesto [Kesti et al.which 2014 &represents 2015]. The neutral axis refers to theaxis line which zero deformation. By means neutralof products means of the neutral the sorepresents called k-factor, which is utilized in of thethedesign axis the so called k-factor, which is utilized in the design of products which are manufactured by bending, is defined. If the k-factor is known, material loss which areand manufactured by after-treatment bending, is defined. If the k-factor is known, material loss and the demand for are reduced.

the demand for after-treatment are reduced.

Figure 7. Figure Microscopic ofcross-section the cross-section of the 7. Microscopic image image of the of the bend. The neutral axisbend. is markedThe by redneutral points basedaxis on theis marked by red points based on the hardness measurements hardness measurements Bending a steel plate by means of a hydraulic press needs a force with a magnitude that depends on the strength and dimensions of the material being bent, as well as the dimensions of the tools being used. When bending tests are carried out, the force is usually measured. On the basis of the measurements, models for the prediction of force

Proceedings of the METNET Seminar 2016 in Castellon

Figure 7. Microscopic image of the cross-section of the bend. The neutral axis is marked by red points based on the hardness measurements Bending a steel plate by means of a hydraulic press needs a force with a magnitude that depends on the strength and dimensions of the material being

Bending abent, steel of a hydraulic needsWhen a force with tests a magnitude that as plate well asby themeans dimensions of the toolspress being used. bending are depends carried on theout, strength and dimensions of the material being bent, as well as the the force is usually measured. On the basis of the measurements, dimensions of the tools being used. When bending tests are carried out, the force is models for the prediction of force have also been developed [Ruoppa et al. usually measured. On the basis of the measurements, models for the prediction of force 2014] which may be utilized in planning product manufacturing. As the have alsostrength been developed [Ruoppa al. 2014] be utilized planning and thickness of theetsteels being which tested may are increasing allinthe time, product manufacturing. As theforstrength thickness of therising. steelsBeside beingthe tested are220increasing all the demand strongerand machines has been regular the time, ton the press demand for stronger machines has been rising. Beside the regular brake which has been in use, an idea for a new “ultimate bending 220-ton press brake which hasproposed been in[Toppila use, anetidea for and a new “ultimate machine” was machine” was al. 2011] it was built as bending well. proposed [Toppila et al. 2011] and it was built as well.

Figure 8. Figure a) Hydroforming machine b) tools bending built in the machine 8. a) Hydroforming machine and b)and bending built in thetools machine Figure 8a shows a hydroforming machine which is located at the Tornio campus. The Figure 8a shows a hydroforming machine which is located at the Tornio maximum force of the main cylinder is 3000 tons, which is adequate for heavy-duty campus. The maximum force of the main cylinder is 3000 tons, which is bending. Using the machine for bending tests was planned in cooperation with SSAB and adequate for heavy-duty bending. Using the machine for bending tests was the special tools shown in Figure 8b were designed and constructed. From now on it will planned in cooperation with SSAB and the special tools shown in Figure 8b be possible to bend even 60-mm-thick UHS steel plates and the tests have already been were designed and constructed. From now on it will be possible to bend even started. The results have been encouraging and the research will be continued. 60-mm-thick UHS steel plates and the tests have already been started. The results have been encouraging and the research will be continued.

WELDING

WELDING The cooperation between Ruukki and Kemi-Tornio University of Applied Science (currently SSAB and Lapland University of Applied Sciences) within welding research has continued for several years. The first research projects were focused on different cutting methods, welding wires, UHS and welding, and the repair welding of wear-resistant steels. Advanced techniques were used for the investigation of the quality and mechanical properties of the welds. In the majority of the welding tests, the SFS-EN ISO 15614-1 standard is applied. This standard “Specification and qualification of welding procedures for metallic materials. Welding procedure test. Part 1: Arc and gas welding of

The cooperation between Ruukki and Kemi-Tornio University of Applied Science (currently Proceedings of theLapland METNET Seminar 2016 in Castellon SSAB and University of Applied Sciences) within welding research has continued for several years. The first research projects were focused on different cutting methods, welding wires, UHS and welding, and the repair welding of wear-resistant steels. Advanced techniques were used for the investigation of the quality and mechanical properties of the welds. In thesteels majority welding tests, the ISO 15614-1 is applied. and of arcthe welding of nickel andSFS-EN nickel alloys” includesstandard all the tests which This standard “Specification and qualification of welding procedures properties for metallicofmaterials. are necessary for ascertaining the quality and mechanical the Welding procedure test. Part 1: Arc and gas welding of steels and arc welding of nickel welds. The most common investigation cases reported are pWPS (Preliminary and nickel alloys” includes all the tests which are necessary for ascertaining the quality Welding Procedure Specification) in most some common cases an official qualification to and mechanical properties of the welds.but The investigation cases reported WPS (Welding Procedure has also been In order are pWPS (Preliminary WeldingSpecification) Procedure Specification) butcarried in someout. cases an official to ensure reliability the test results, the ASM Steelcarried and Mining) qualification to the WPS (Weldingof Procedure Specification) has(Arctic also been out. In order R&D the group followsofthe ISO/IEC 17025(Arctic laboratory “General to ensure reliability theSFS-EN test results, the ASM Steel standard and Mining) R&D group follows the SFS-EN 17025 laboratory standard “General requirements for the requirements forISO/IEC the competence of testing and calibration laboratories”. Some competence of testing and calibrationbeing laboratories”. Someare of shown the investigation techniques of the investigation techniques that are used in Figure 9. being that are used are shown in Figure 9.

Figure 9. Tensile test, impact test, and test hardness test Figure 9. Tensile test, impact test, and hardness Lapland University of Applied Sciences has also worked in cooperation with the Czech Lapland University of Applied has also worked cooperation Technical University in Prague. TheSciences first research project wasinfocused on thewith use and the Czech Technical University in Prague. The of first was evaluation of advanced techniques for the investigation theresearch behaviourproject of high-strength on the usepresented and evaluation of advanced techniques for the investigation steel.focused The techniques included optical strain measurement in tensile tests, a crackofpropagation study, the idea of steel. using aThe hydroforming for included bending tests. the behaviour of and high-strength techniquesmachine presented Welded connections of common-grade steels been investigated andand verified optical strain measurement in tensile tests,have a crack propagation study, the for several detailed rules for both the for welding technology process and structural ideayears; of using a hydroforming machine bending tests. Welded connections engineering design are therefore clearly given in design standards and other prescriptions. of common-grade steels have been investigated and verified for several On the other hand, new or improved steel processing techniques allow structural steel years; for exceeding both the welding technology processThe andmain structural grades with detailed a tensile rules strength 1300 MPa to be produced. reason why therefore clearly in design standards and new engineering high-strengthdesign steels are cannot be used in thegiven engineering praxis is a lack of other knowledge other hand, new or improved processingare techniques aboutprescriptions. the behaviourOn of the such steels in structures. These steel characteristics controlled not steel grades with a tensile strengthbut exceeding MPato the only allow by thestructural mechanical properties of the base material also with1300 regard technological process and procedures and quantities, such as, for instance, heat to be produced. The its main reason why new high-strength steels cannot beinput, used in the engineering praxis is a lack of knowledge about the behaviour of such steels in structures. These characteristics are controlled not only by the mechanical properties of the base material but also with regard to the technological process and its procedures and quantities, such as, for instance, heat input, grade, the diameter and type of welding consumable used, the welding method, number of passes (single or multilayer welds), preheating level, cooling rate, geometry of the weld, and others. [Sefcikova et al. 2015].

Proceedings of the METNET Seminar 2016 in Castellon

grade, the diameter and type of welding consumable used, the welding method, number of Arctic welding has been developed within TEKESofproject passesResearch (single orinto multilayer welds), preheating level, cooling rate,the geometry the weld, and others. [Sefcikova et al. 2015]. WELDARC 2015 – 2017 (Improvement of productivity and quality of welding on special steels in Arctic conditions). The aim of this project is to reach a Research welding has been withinand thetoTEKES WELDARC new into levelArctic of know-how related to developed Arctic welding developproject collaboration 2015-2017 (Improvement of productivity and quality of welding on special steels in Arctic between steel producers and downstream enterprises in Lapland as well as conditions). The aim of this project is to reach a new level of know-how related to Arctic research and educational institutes. The eco-efficiency of the welding of ultrawelding and to develop collaboration between steel producers and downstream high-strength wear-resistant is developed; the occupational safety of enterprises in Laplandand as well as researchsteels and educational institutes. The eco-efficiency of operations is enhanced by developing more precise control for those steels the welding of ultra-high-strength and wear-resistant steels is developed; the occupational are susceptible to coldbycracking; themore monitoring the conditions of thethat safety that of operations is enhanced developing precise of control for those steels are susceptible coldup cracking; of the conditions and of the welding is built up welding istobuilt (Figurethe 10);monitoring the enterprise networking problem-solving (Figurecapability 10); the enterprise networking capability the Arcticofwelding for the Arctic weldingand areproblem-solving improved; these are a fewfor examples the are improved; these are a few examples of the practical applications developed from the practical applications developed from the project results [Keltamäki 2016]. project results [Keltamäki 2016].

FigureFigure 10. Monitoring welding conditions (Rovaniemi Arctic Power laboratory) 10. Monitoring of of welding conditions (Rovaniemi Arctic Power laboratory)

WEARING

WEARING Wearing has also been under research for several years at Lapland University of Applied Sciences. The best places to evaluate wearing conditions are in mines. Finnish mines also beenand under research several years Lapland research. UniversityThe such Wearing as Kittilä,has Sodankylä, Kemi have for many targets forat wearing of Applied The the bestnearest places one to evaluate wearing Outokumpu Kemi Sciences. mine, being to the ASM R&Dconditions group, is are the in most commonly used one, and conditions in that Sodankylä, mine are also the hardest mines. Finnish mines such as Kittilä, and Kemi havewhen manyconsidering targets wearing. Therefore, it is possible to compare and the laboratory tests. Many for wearing research. The Outokumpu Kemi real-life mine, being nearest one to the application-oriented laboratory wear test devices for such purposes have been developed ASM R&D group, is the most commonly used one, and conditions in that mine by the Tampere Wear Center located in southern Finland. are also the hardest when considering wearing. Therefore, it is possible to compare real-life and laboratory tests. Many application-oriented laboratory wear test devices for such purposes have been developed by the Tampere Wear Center located in southern Finland.

Proceedings of the METNET Seminar 2016 in Castellon

Kemi mine, being the nearest one to the ASM R&D group, is the most commonly used one, and conditions in that mine are also the hardest when considering wearing. Therefore, it is possible to compare real-life and laboratory tests. Many application-oriented laboratory wear test devices for such purposes have developed by the Tampere Wear Center located Kemi mine, being the nearest one been to the ASM R&D group, is the most commonly used one, in southern Finland. in that mine are also the hardest when considering wearing. Therefore, it is and conditions possible to compare real-life and laboratory tests. Many application-oriented laboratory wear

Field of the ofbeen a loading machine and the wear a feed test devices for suchcutting purposes developed by the Tampere Wearplates Center located Field tests oftests the cutting edge ofedge ahave loading machine and the wear plates of a of feed hopper in southern Finland. hopper carried with the aim of evaluating the application-oriented were carried outwere with the aim out of evaluating the application-oriented test methods and to get test methods and to get more detailed information for the numerical modellingwere more detailed the modelling the plates wear.of The Field testsinformation of the cuttingfor edge of anumerical loading machine and theofwear a feedplates hopper of the wear. plates were constructed using and the Hardox. special wear-resistant constructed using theThe special wear-resistant steels Raex The wearing were carried out with the aim of evaluating the application-oriented test methods and to getof the side plates was measured with for the ATOS optical measurement system from GOM GMBH steels Raex and Hardox. The wearing of the side plates measured with the more detailed information the numerical modelling of thewas wear. The plates were constructed usingmeasurement special Sciences. wear-resistant steels Raex andmeasured Hardox. The wearing of the at Lapland University ofthe Applied plates were using ATOS before ATOS optical systemThe from GOM GMBH at Lapland University sideand plates waswearing. measured with the ATOS optical measurement system in from GOM GMBH assembly after Some ATOS measurements are shown Figures 11 and 12. of Applied Sciences. The plates were measured using ATOS before assembly at Lapland University Applied Sciences. The plates were measured using ATOS before The weights of the platesof were also measured before after wearing and so12. it was and after wearing. Some ATOS measurements areand shown 11 and and assembly and after wearing. Some ATOS measurements are shownininFigures Figures 11 12. possibleThe to weights report the total material loss exactly. Hardness tests and microscopic analysis thethe plates were also also measured before before and after wearing so it was The weightsof of plates were measured and after and wearing and are alsopossible included report in thethe normal procedure. total material exactly. Hardness tests andHardness microscopictests analysis so it wasto possible to report theloss total material loss exactly. and are also included in the normal procedure. microscopic analysis are also included in the normal procedure.

Figure 11. Wearing of cutting edge

FigureWearing 11. Wearingofofcutting cutting edge Figure 11. edge

Figure 12. Wearing of side plate

Proceedings of the METNET Seminar 2016 in Castellon

ACKNOWLEDGEMENTS The authors would like to acknowledge the financial support of MeripohjolaHanke and TEKES – the Finnish Funding Agency for Technology and Innovation – for their support for the METNET Network.

REFERENCES Arola, A-M., Kesti, V., Ruoppa, R. (2015). The Effect of Punch Radius on the Deformation of Ultra-High Strength Steel in Bending. Proceedings of the 16th International Conference on Sheet Metal, pp 139 – 146, Erlangen, Germany, 16 – 18 March, 2015. Joutsenvaara, J. (2013). Forming limit curve of high strength steel with the help of digital image correlation. Proceedings of the METNET Seminar 2013, Luleå, Sweden, 22 – 23 October 2013. Joutsenvaara, J. (2012). Investigation of strain localization of high strength steel with the help of digital image correlation. Proceedings of the METNET Seminar 2012, Izmir, Turkey, 10 – 11 October 2012, HAMK UAS, ISBN 975951-784-592-2, pp.19 – 30. Keltamäki, K. (2016). WELDARC, The Lapland University of Applied Sciences explores the improvement of productivity and quality of welding on special steels in Arctic conditions. Accepted for publishing in Pan European Networks: Science & Technology 20 September 2016. Kesti, V., Kaijalainen, A.J., Mourujärvi, J., Ruoppa, R. (2015). Bendability and microstructure of Optim® 700 MC Plus. Nordic Steel Construction Conference, Tampere, Finland, 23 – 25 September 2015. Kesti, V., Kaijalainen, A., Väisänen, A., Järvenpää, A., Määttä, A., Arola, A-M., Mäntyjärvi, K., Ruoppa, R. (2014). Bendability and microstructure of direct quenched Optim 960QC. Materials Science Forum, Trans Tech Publications, Switzerland Vols. 783 – 786, pp 818 – 824. Ruoppa, R., Toppila, R., Kesti, V., Arola, A-M. (2014). Bendability tests for ultra-high-strength steels with optical strain analysis and prediction of bending force. Proceedings of the METNET Seminar 2014, Moscow, Russia, 21 – 22 October 2014, HAMK UAS, ISBN 975-951-784-693-6, pp. 68 – 78. Ruoppa, R., Toppila, R., Sipola, J., Kesti, V. (2012), Bending properties of some ultra-high-strength steels. Proceedings of the METNET Seminar 2012, Izmir, Turkey, 10 – 11 October 2012. HAMK UAS, ISBN 975-951-784-592-2, pp.10 – 18.

Proceedings of the METNET Seminar 2016 in Castellon

Sefcikova, K., Brtnik, T., Dolejs, J., Keltamäki, K., Toppila, R. (2015). Mechanical Properties of Heat Affected Zone of High Strength Steels. IOP Conference Series: Material Science and Engineering. 2nd International Conference on Innovative Materials, Structures and Technologies, Riga Technical University in Riga, Latvia, 30 September-2 October 2015. Siltanen, J., Kesti, V., Ruoppa, R. (2014). Longitudinal bendability of laser welded special steels in a butt joint configuration, Proceedings of International Congress on Applications of Lasers & Electro-Optics, October 19 – 23, 2014, San Diego, USA. SSAB website (2016). URI: http://www.ssab.fi/ssab-konserni/kestava-kehi tys#!di=discover401DE95158784C5FA3329B8231ADD611, Cited 15 August 2016. SSAB imagebank (2016). URI: http://imagebank.ssab.com/SSAB/#1480314741475_0, Cited 15 August 2016. Toppila, R., Dolejs, J., Kauppi, T., Brtnik, T., Joutsenvaara, J., Vaara, P., Ruoppa, R. (2011). Investigation of behavior of HSS using advanced techniques. METNET Seminar 2011 Aarhus, Denmark, 12 – 13 October 2011. HAMK UAS, 978-951-784-556-4, pp 24 – 36.

Proceedings of the METNET Seminar 2016 in Castellon

STRENGTH OF CIRCULAR HOLLOW SECTION COLUMNS Kuldeep S. Virdi Emeritus Professor of Structural Engineering Department of Civil Engineering City, University of London, UK

INTRODUCTION Circular Hollow Sections offer considerable benefits when used as compression members. The geometry of the section offers the maximum possible value of radius of gyration, resulting in superior performance against instability. As advances in materials technology result in higher grades of steel, it becomes imperative to reassess the instability behaviour of structural members made from higher grades of steel. Eurocode 3 (EN 1993-1-1:2003) permits the use of steel grades up to S460. Recent research has targeting use of circular hollow sections made of aluminium (Zhou and Young 2006) and further as concrete filled steel tubes (Han et al. 2007). Although nonlinear finite element method can be used to determine the ultimate failure load of such columns, the modelling process takes considerable time and effort and for accurate modelling using solid elements often requires considerable computer time per run even on modern day computers. This paper describes a fast and yet accurate method for calculating the strength of eccentrically loaded circular hollow section columns. The method is based on the use of finite differences. A range of applications of the finite difference approach has previously been described by the author (Virdi 2006).

ADVANCED CALCULATION MODEL The method of analysis presented here simulates the process of physically testing a slender column. A small axial load is applied and the deflected shape of the column is calculated. The load is then increased and the process repeated, until for a given applied load no equilibrium deflected shape is obtainable. The highest load, for which a deflected shape can be obtained, is taken as the failure load of column. An interpretation of this is that the stiffness of the column at the maximum applied load has vanished. Appropriate nonlinear stress-strain characteristics of steel are used. Initial imperfections in the form of lack of straightness can be specified. The method described here is for an axial load acting at an eccentricity.

Proceedings of the METNET Seminar 2016 in Castellon

STEEL STRESS-STRAIN CHARACTERISTICS It is customary to idealise the stress-strain characteristic of steel as elastic plastic or a series of straight lines. Strain hardening can be considered by adopting an elastic-elastic bi-linear relationship, where the post-yield elastic modulus is adopted as a very small fraction (1/10000 or similar) of the Young’s modulus of steel. Current version of Eurocode 3 (EN 1993-1-1:2003) permits characteristics fromof tests when performing Eurocode the 3 use (ENof stress-strain 1993-1-1:2003) permitsobtained the use stress-strain characteristics non-linear structural analysis. obtained from tests when performing non-linear structural analysis.

STRUCTURAL MODEL STRUCTURAL MODEL The analysis described in thisinpaper is for eccentrically loadedconcrete concrete The analysis described this paper is for eccentrically loaded filledfilled tubular columns. tubular The column may different of loading at theattwo columns. Thehave column may haveeccentricity different eccentricity of loading the ends. The two ends. The model is shown in Figure 1. model is shown in Figure 1. P

P exA

x Figure 1 - Structural model

exB

Figure 1 - Structural model

The cross-section is described by the outer radius Ro and the inner radius The cross-section is described by the outer radius Ro and the inner radius Ri as shown Ri as shown in Figure 2. The effective line of compression joins the two in Figure 2. The effective line of compression joins the two end eccentricity positions end eccentricity positions of the applied thrust. The governing equation of of the applied thrust.at aThe governing equation ofcan equilibrium at a point distance from equilibrium point distance from the origin be stated thus: the origin can be stated thus: M = P (e x + u ) (2) M  P (e x  u ) (2)

MM is the where, at where, the point consideration, x is is theinternal internal moment, moment, e xeis thethe effective at theunder point under consideration, effective and u is the u is the deflection . deflection . eccentricity, and eccentricity, Equation (2) re-written can be re-written Equation (2) can be as: as: u  M / P  ex u = M / P − e (3) x

(3)

Since equilibrium to be satisfied the deflected state thecolumn, column, the internal Since equilibrium needs toneeds be satisfied in theindeflected state ofofthe u .this the internal stress-resultants upon the displacements, Hence, this u . Hence, equation may be stress-resultants depend upon thedepend displacements, equation may be written in symbolic form as: written in symbolic form as: (4) u  U (u ) (4) u = U (u )

where U iswhere a complex function, not easily expressed closedform. form. U is a complex function, not easily expressedin in a a closed A solution of the above equation requires calculation of internal stress resultants taking into account the column curvature, cross section shape and material stress-strain characteristics. This is achieved by establishing the moment-thrust-curvature

Proceedings of the METNET Seminar 2016 in Castellon

A solution of the above equation requires calculation of internal stress resultants taking into account the column curvature, cross section shape and material stress-strain characteristics. This is achieved by establishing the moment-thrust-curvature relations. The solution also requires calculation of the deflected shape to be consistent with curvatures that cause the internal strains and stresses at all points along the length of the column.

MOMENT-THRUST-CURVATURE RELATIONS The moment-thrust-curvature relations are a complex function due to the generality of the material stress-strain characteristics of steel. An analytical formulation is, in general, not feasible. Recourse is taken to numerical general, not feasible. Recourse is taken numerical It is assumed methods. It is assumed that to upon flexure, a methods. plane cross-section remains that plane upon flexure, a plane cross-section plane ascorollary shown in Figure 2. This as shown in Figure 2.remains This results in the that the strains varyresults linearly general, not feasible. is linearly takenThe towithin numerical methods. is assumed that in the corollary that the Recourse strains vary the cross-section. The neutral axis within the cross-section. neutral axis may well beItinside or outside the upon flexure, a plane remainsalso plane as shown inalso Figure 2. This resultsof may well be inside or cross-section outside the The diagram shows the stresscross-section. Thecross-section. diagram shows the stress-strain characteristics in the characteristics corollary that the strains linearly within the cross-section. The neutral axis strain ofrelative steel relative to theaxis. neutral axis. steel tovary the neutral

may well be inside or outside the cross-section. The diagram also shows the stressstrain characteristics to the axis. as: The of twosteel stress resultants canneutral be The two stress resultants canrelative be expressed as:expressed The two stress resultants can be expressed P   dAas:

(5)

(5)  and (6) M x dA (6) and P  dA (5) M   x dA (6) and x hasequation, the originx athas the centroid the cross-section. In the above equation, In the above the origin atof the centroid of the cross-section. A

AA A

x hasthe the origin atresultants the centroid of the cross-section. In above equation, Forthe purposes of For evaluating numerically, the cross-section is purposes of stress evaluating the stress resultants numerically, the crossdivided into a number n isofdivided slices into parallel to thenneutral axis and the integrations section a number of slices parallel to the neutral axisare and For purposes of the evaluating stress the atcross-section is replaced by summations. Thethe strain at a resultants point withinnumerically, the cross-section obtained integrations are replaced by summations. The strain aispoint withinby: the divided into a number n of slices parallel cross-section is obtained by:to the neutral axis and the integrations are replaced by summations. The strain (7)  at  xa point within the cross-section is obtained by:   from x the neutral where, x is the distance of the point axis and  is the (7) curvature(7) at the cross-section. The depth of the neutral axis from the top of the cross-section is where, xbyisdthe distance of the point from the neutral axis and  is the curvature at denoted n. where, x is the distance of the point from the neutral axis and φ is the the cross-section.curvature The depth of cross-section. the neutral axis topneutral of theaxis cross-section at the The from depththe of the from the topisof denoted by d n . the Ro cross-section is denoted by d . n

Ro Ri

 si

si

 si



si

Neutral Axis

Strains

Neutral Axis Steel Stress

dn dn

replaced by summations. The strain at a point within the cross-section is obtained by: replaced by summations. The strain at a point within the cross-section is obtained by: (7)   x (7)   x Proceedings of the METNET Seminar 2016 in Castellon 46 where, x is the distance of the point from the neutral axis and  is the curvature at where, x is the distance of the point from the neutral axis and  is the curvature at the cross-section. The depth of the neutral axis from the top of the cross-section is the cross-section. The depth of the neutral axis from the top of the cross-section is denoted by dn . denoted by d n . Ro Ri

Ro Ri  si

 si



si xi

si



Neutral Axis Neutral Axis Strains

Steel Stress Steel Stress

Strains

Figure 2 - Discretization of the cross-section for moment-thrust-curvature Figure Discretization of theforcross-section for moment-thrust-curvature Figure 2 2 - -Discretization of the cross-section moment-thrust-curvature calculations calculations calculations Equations (5) and (6) become:

Equations (5) and (6) become: Equations (5) and (6) become:

n (8) P  si nAsi (8)

P   A M    A x (9) M    A x i 1

i 1

si n

i 1

(9)

(8) (9)

The stress at the of slice for from steel,the Asicentroid is the area of the thethe distance tothe the mid-height steel zonewithin the σ slice, andcentre xi is si is evaluated Thestress atcentre the centre of the slice steel,∆ area The stress at the of the slice forfor steel, AsiAis the area of the si is evaluated si is si is evaluated

of the slice. of the steel zone within the slice, and xi is the distance from the centroid to the mid-height of the slice.

Figure 3 shows an enlarged view of a typical slice, bounded between the lines 1-1 and Figure 3 shows an enlarged view of a typical slice, bounded between the lines 2-2. 1-1 and 2-2.

 

 1

1



2

2 RI





Figure 3 - View of inner and outer zones within a slice.

The angle  1 can be obtained for a distance e from the top as follows:

1  cos1{( Ro  e) / Ro }

(10)

R IwithinCa slice. R O Figure 3 - View of inner and outer zones R Oand outer zones within a slice. R I3 - View C of inner Figure R R C Figure I3 - View of inner Oand Proceedings outer zones within a slice. of the METNET Seminar 2016 in Castellon 47 Figure 3 View of inner and outer zones within a slice. e  1 can be obtained for a distance e from the top as follows: The angle3 - 1View can of beinner obtained for a distance e from the top as follows: Figure and outer zones within a slice. 1 The angle obtained for a distance e from the top as  1 can   cos {( Rbe (10)follows: o  e) / Ro } 1 The angle  1 can be 1obtained for a distance e from the top as follows: 1  cos {( Ro  e) / Ro } (10) The angle  1 can be obtained for a distance e from the top as follows: 1 angles cosa distance {( Ro  ee) from / for Ro }the (10) angle for obtained as follows: α 1 can 1 for equations can be The written thebe other required thistopstrip. For 1   cos {( R  e ) / R } (10) Similar equations can be written for the other angles required for this strip. For 1 o o , 1   cos {( R  e ) / R } (10) (10) example, 1 Similar equations can beo writteno for the other angles required for this strip. For 1 Similar equations can be written for angles required for this strip. For example,  2  cos {( Ri can e be the )written / Rother i } for1the other angles required for(11) Similar equations this strip. Similar equations For canexample, be written for the  2 other  cos angles {( Ri  erequired   ) / Ri } for this strip. For example, (11)  1 example,  2and  cos2-2{(can Ri be e  shown  ) / Ri }to be given (11) a of the whole slice between the lines 1-1 1 (11)   cos {( R  e   ) / R } (11) The area of the 2whole slicei between the lines 1-1 and 2-2 can be shown to be given i 1   cos {( R  e   ) / R } (11) to be given by: 2 i i The area of the whole slice between the lines 1-1 and 2-2 can be shown The area of the whole slice between the lines 1-1 and 2-2 can be shown to be 2 2 The area the lines 1-1 and 2-2 can be shown (12) to be given by: W of Ro the ( 2 whole  sin  2slice sin between given by: 2 )  R2o (1  sin 1 sin 1 ) 2 The area of the whole sliceWbetween shown by: (12)  Ro ( 2 the  sinlines  2 sin1-1  2 )and  Ro2-2 (1 can  sinbe 1 sin 1 ) to be given 2 2 by: (12) (12) W  Ro within ( 2  sin  2 same sin  2 ) slice  Ro ( 1  sin 1 sin 1 ) , the area of the hollow zone the can beobtained by 2 2 (12) W  R (   sin  sin  )  R (   sin  sin  ) Likewise, the area of the hollow zone within the same slice can be obtained by o R 2and using 2 2 o 1 angles: 1 1 appropriate g the radius Ro withLikewise, 2 i 2 thearea of the hollow zone within the same slice can be obtained by (12) W  R (   sin sin  )  R (   sin  sin  ) with hollow using appropriate angles: replacing the Ro 2the o 2the radius 2 oR i and 1 1 Likewise, area of zone the angles: same slice can be obtained by replacing the radius Ro with Ri1 and usingwithin appropriate Likewise, the area of the hollow zone within the same slice can be obtained by 2 replacing the radius R 2 o with R i and using appropriate angles: (13) H R  2the sin hollow (  1  sin  1 sinthe  1 ) same i ( area 2  sinof 2 )  R2i zone 2 Likewise, the within slice can be obtained by using appropriate angles: eplacing the radius Ro withHRi Rand 13) (13) i (  2  sin  2 sin  2 )  Ri (  1  sin  1 sin  1 ) using appropriate angles: eplacing the radius Ro with Ri and 2 2 (13) H  R (   sin  sin  )  R (   sin  sin  ) i is 2simply:2 the slice 2 1 1 1 a of the steel zone The within the slice isi simply: 2 area of the steel zone within 2 (13) H  area Ri ( of sin  2 )  R sinslice  1 sinis 1simply: ) The steel within the 2 the 2 sin zone i (1  2 2 H  R (   sin  sin  )  R (   sin  sin  ) i 2 2 2 i 1 1 1 The area of the slice is simply: (14)(14) (13) S steel W  Hzone within the (14) S  W  H The area of the steel zone within the slice is simply: It would be appreciated that, using these equations, accurate evaluation of The area of the steel zone slice is simply: (14) S  W  Hevaluation of integrals be appreciated that, usingwithin thesethe equations, accurate integrals in Equations (5) and (6) can be obtained very rapidly. (14) S  W  H It would be appreciated that, using these equations, accurate evaluation of integrals ons (5) and (6) can be obtained very rapidly. (14) of integrals S W H in Equations (5) and (6) can be obtained rapidly. accurate evaluation ETERMINATION OFbe THE POSITION OF NEUTRAL AXIS It would appreciated that, using thesevery equations, t would be appreciated that, using equations, accurate in Equations (5) and (6)these canPOSITION be obtained very rapidly. DETERMINATION OF THE OF NEUTRAL AXISevaluation of integrals he criterion used to establish the position of neutral axis at a given point of is integrals that the be appreciated that,be using thesevery equations, evaluation nt would Equations (5) and (6) can obtained rapidly. accurate The criterion to establish the position of neutral axis at ashould given point is alue of the force stress-resultant obtained from the above procedure match n Equations (5) and (6) can beused obtained very rapidly. that the value theends force stress-resultant obtained from the above procedure e externally applied force ܲ atofthe of the column. should match the externally applied force at the ends of the column.

he procedure adopted here to calculate the depth of neutral axis ݀ n to the neutral The procedure adopted here to calculate the depth of neutral axis n to the xis is a variation ofneutral Newton’s of of iteration theofequation: axis ismethod a variation Newton’susing method iteration using the equation:

d r 1  d r  P ' (d r ) / P (d r )

(15)(15)

where, d r is a trial value of the depth of neutral axis for iteration r , P ( d r ) here, d r is a trial value of the depth of neutral axis for iteration r , P(d r ) is the value is the value of the stress resultant calculated for the current trial value of d r , ' ' the stress resultant for the rate current trial value , P ( d ) is the current d P calculated (d r ) is the current of change in P ( dof , and is the improved ) d r r r +1 r value of the depth of neutral axis from the current iteration. Except for te of change in P(d r ) , and d' r 1 is the improved value of the depth of neutral the axis first trial,

P (d ) can

be obtained numerically from the current value of

r om the current iteration. Except for the first trial, P ' ( d r ) can be obtained the value obtained in the previous iteration Experience P(d r ) and P (d r −1 ) .numerically indicates the method converges extremely rapidly even when parts om the current value of P(dthat and the value obtained in the previous iteration ) Pof( dthe r r 1 ) cross-section have significant non-linear stresses. Experience indicates that the method converges extremely rapidly even when parts the cross-section have significant non-linear stresses.

ETERMINATION OF EQUILIBRIUM DEFLECTED SHAPE

o determine the deflected shape of the column under an applied load, an influence

rom therate current iteration. Except for the trial,improved P ' ( d r ) can be obtained numerically of change in P is the value of the depth of neutral axis (d r ) , and d r 1first ' rom thefrom current ofiteration. the value obtained in the iteration P(d r ) andExcept P ( dnumerically r 1 ) the value current for the first trial, P previous ( d r ) can be obtained Proceedings of the METNET Seminar 2016 in Castellon 48 the Experience indicates theofmethod converges extremely even when parts P ( d ) from currentthat value the value obtainedrapidly in the previous iteration P(d r ) and r 1 of the cross-section have significant non-linear stresses. . Experience indicates that the method converges extremely rapidly even when parts of the cross-section have significant non-linear stresses.

DETERMINATION OF EQUILIBRIUM DEFLECTED SHAPE

DETERMINATION OF EQUILIBRIUM DEFLECTED SHAPE DETERMINATION EQUILIBRIUM DEFLECTED SHAPE load, an influence To determine the deflected shape OF of the column under an applied oefficient method is used, based on finite difference formulae for curvatures. The To determine deflected shape ofshape the of column under an applied load, Tothe determine the deflected theofcolumn an applied load,an an influence olumn length is divided into a number of segments equal under step length, although it coefficient method is used, based on finite difference formulae for curvatures. The influence coefficient method is used, based on finite difference formulae s also possible to formulate a solution based on unequal step size. Because of the column length is divided into a number of segments of equal step length, although it for curvatures. The column length is divided into a number of segments of peed with which aequal solution is obtained by this method, greater accuracy can be step length, although it is also possible to formulate a solution based is also possible to formulate a solution based on unequal step size. Because of the achieved simply by on using a larger number of segments than obtained trying to unequal step size. Because of equal the by speed with whichrather a solution speed with which a solution is obtained this method, greaterisaccuracy can be adaptively vary the step size. by this accuracy can of be equal achievedsegments simply by using achieved simply bymethod, using greater a larger number rathera larger than trying to number of equal segments rather than trying to adaptively vary the step size. adaptively vary the step size. P

2 3

s-1 s

s+1

n-1

s-1 s u s+1 n-1 B u1 u A 1 2 3 2u us+1 n-1 u u 3 u s un-1 x 1 u s-1 2u u u u 3 x s-1 s s+1

Figure 4 - Finite difference discretization of the column length Figure 4 - Finite difference discretization of the column length

Figure 4 - Finite difference discretization of the column length

Equation for each of displacements, the unknown displacements, u n ,of Writing Equation (4)Writing for each of the(4)unknown u1 , u 2 , …u1 ,uun ,2 ,a…set a set of simultaneous equations is obtained, represented below in symbolic imultaneous is obtained, represented below in symbolic form. Writingequations Equation form. (4) for each of the unknown displacements, u1 , u 2 , … u n , a set of given by: simultaneous equations is obtained, represented below in symbolic form. u  U (u ) (16)(16)

1 given by: (16) ur 1  vector uur U[(uI})Kat] iteration U r ) r , an improved (17) (ur number Starting vector with anassumed r r , an improved ur at iteration{unumber solution is Startinggiven withby: an assumed solution is given by: matrix isuar Jacobian Ur,r )an improved ur unit urr at  [iteration I  K]K1(number (17) solution Starting with assumed the and matrix, defined by: is In the equationan above I  is vector 1  u ur 1  ur  [ I  K ] 1 (ur  U r ) (17) (17) In the equation above I  is the unit matrix K  is a Jacobian matrix, defined by: U i and In the equation above [I ]Kis  the unit matrix and [K ] is a Jacobian matrix, (18) ij In the equation above defined by: I  is the unit matrix u j and K  is a Jacobian matrix, defined by: U i K  (18) ij (18) U In practice, when using the finite difference a triK ij  u ji formulation, matrix K  is reduced to (18) u j difference formulation, matrix [K ] is practice, whenbelow. using the finite diagonal form,Inas explained reduced to a the tri-diagonal form, as explained below. matrix K  is reduced to a triIn practice, when using finite difference formulation, In practice, when the to finite difference formulation, matrixisKapproximated tri is reduced tobya the When limiting the analysis small displacements, curvature diagonal form, as using explained below. When limiting the analysis to small displacements, curvature is approximated s second derivative. In finite difference form, at point , the curvature is: diagonal form,byas explained below. the second derivative. In finite difference form, at point s , the curvature is: When limiting the analysis to small displacements, curvature is approximated by the secondlimiting derivative. In finite to difference point ,uthe is: 2 u s s When the analysis small curvature is approximated by the d 2displacements, u form,u sat (19) 1  s 1 curvature  = = (19) s 2 s , the curvature is: 2 form, at point second derivative. In finite difference h dz2 u  2 u s  u s 1 d u s = = s 1 (19) 22 2 u u slength. d u 1  2hu s It s 1 be shown easily that: can where, h is the step size along  s = thedzcolumn = (19) h2 dz 2 ds the column ds length. 1It dcan s be shown 1 easily that: where, h is the step size along = =  = 2 (20) can be shown easily that: where, h is the step size along dus 1the column dus 1 length. 2It du h s

When When limitinglimiting the analysis to small curvature is approximated by theby the to small curvature is When limiting the analysis analysis to displacements, small displacements, displacements, curvature is approximated approximated by the the When limiting the analysis to small displacements, curvature is approximated s econdsecond derivative. In finite difference form, at point , the curvature is: ss ,, the second derivative. finite form, at is: WhenIn limiting the analysis to small displacements, curvature is approximated by by tt derivative. In finite difference difference form, at point point the curvature curvature is: s second derivative. In finite difference form, at point , the curvature is: s , Seminar second derivative. In finite difference Proceedings form, atof point the curvature the METNET 2016 in Castellonis: 49 2 2 u  2 u  u 2 u  2 u  u d u d us 1 u ss s11  2s2u1ss  u ss 11 s u 1  s =  ss 2= (19) (19) = =d u2 = = s21 d 2 2us (19) 1  2 u s  u s 1 s 2u = u ss  2 d h dz 1  2 u s  u s 1 h dz  = (1 dz 2  ss = dzh 22 = (1 hh 2 dz h is the column length.length. It can be shown easily easily that: that: here,where, h is step size along the It be is the thesize stepalong size the along the column column length. It can can be shown shown easily that: where, h step h is the step size along the column length. It can shown where, h where, is the step size along the column length. It can be shown easily bethat: shown easily easily that: that: where, h is the step size along the column length. It can be ds d 1 d   1 d d 1 d    1 dsss =s =dsss  = s 1 =dsss = 1 (20) (20) (20) = ds dd shss 2 = h222 11 ddss = 11 dus 1 du (2 s 2 dusss 111 dus 1 du dusss d111 s2 du= 2 du du ss = = h 2 du = h 22 (2 du du s  1 s  1 s dus 1 dus 1 2 du s h dU s dU s ss ddss dU s  dU dU dU s ddU dU dU2  2 2  = sss = = s = lso, Also, s dU   sss2 d Also,  (21) dU dU  (21) 2  (21) = = (21) Also,  2 s s s 2  dU s  dU dU d  du d  du d  h 2 du d  du d h = (2 Also, s s = s  dss  s h s  2  s s s  s s du d  du  s 2  = (2 Also,   ss ss du ss = s dd dd duss ss du duss ss  hh 2  dU s dU dU s dU dss d  dU s 1dU dU dU sss 11 imilarly, =dU sss = = dsss = (22) (22) ss dU Similarly, Similarly, dU dd dU 2s Similarly, = = (22) s 22ss = (22) dU dU dU ss 11 du d  du d  h d  du d h Similarly, (2 s s 1= s s s 1 du s s  1 s  1 s s du d  du d  h 1 1 1 1= s Similarly, = (2 2 ss  ss du ss  s dd dd du ss 11 ss du duss 11 ss hh 2 dU s dU dU s dU d d dU s 1dU 1 ss 1 nd, and, =dU ss = = dss = (23) (23) and, = dUsss dU = dUdU (23) d dU  and, (23) 1 2s 22s s s dU dU d dU  1 dus 1 du d  du d  h 2 d  du d  h and, = = (2 s s s s s s  1 s duss 11 dss duss 11= dss h = and, (2 2 du dd dd duss 11 ss du duss 11 ss hh 2 dU s dU dU dU ss 1 dU s 11 dU dU dU sss follows that, that,It follows that, = sss =s dU = dU (24) (24) s = It follows  dU 1 (24) s s dU dU dU ss 1 dU du s 1 du du du 2 du 2 It follows that, = (2 s  s 2 du du sss 111 s 1 dusss 111 dussss = It follows that, = = (2  2 du du du s  1 s  1 s du s 1 dus 1 2 dus s  1to rom Equation (19),From it(19), can be deduced that, for all that, t not to ,tossss or From it be that, for all ttallnot equal −111,,s,:sss or Equation (19), can be deduced forequal t not equal or sss :::  From Equation Equation (19), it can can be itdeduced deduced that, for all not equal to or From Equation (19), it can be deduced that, for all t not equal  11 ,, ss or From Equation (19), it can be deduced that, for all t not equal to to ss  or ss :: ds d =dss0 = = 0 0 d ss = 0 d dut du dutt = 0 du t 1 du s t not equal to , or1 his leads to the result that for all t not equal to This leads to the result that for all t 1 1 ::  1 ,,s ss 1or or: ss  This leads to the result that for all t not equal to sss  This leads leads to to the the result result that that for for all all tt not 1 1 ::  11 ,, ss or not equal equal to to ss  or ss  This This leads dU to the result that for all t not equal to s − 1 , s or s + 1 : dU s = dU0ss = = 0 0 dU dU ss = 0 dut du dutt = 0 du dutt only , there are terms terms centred n other words, for each row in matrix   K , there only three centred In other words, for each row in matrix   K arethree only non-zero three non-zero non-zero terms centred In other words, for each row in matrix K  , there are , there are only three non-zero In other words, for each row in matrix   K In other words, for each row in matrix K  , there are only three non-zero terms terms centr centr

[ ]

In other words, for each row in matrix K , there are only three non-zero terms centred over the diagonal. Obviously, in the first and last rows of the matrix, there will only be two, and not three, elements. Further, Equation (24) indicates that terms on either side of the diagonal have simple relationship with the diagonal terms. It should be noted that elements in one row do not have any simple relation with elements in any other row. Thus, the coefficient matrix generated by the above procedure is not symmetric. To define the elements of the diagonal, at each point along the length, integrations have to be performed twice, once for the current value of the curvature, defined by Equation (19) and once by giving a small change to the deflection u s , say of magnitude γ . The values of moment stress-resultant obtained in the two integrations lead respectively to the functions U s1 and U s 2 giving approximately:

other row. Thus, the coefficient matrix generated by the above procedure is not symmetric. To define the elements of the diagonal, at each point along the length, integrations of the METNET Seminar 2016 in Castellon 50 TotoProceedings have be performed twice,of once for the current value the curvature, defined by define the elements the diagonal, at each pointofalong the length, integrations Equation (19) and once by giving a small change to the deflection u , say of magnitude have to be performed twice, once for the current value of thes curvature, defined by (19) and once by stress-resultant giving a small change to theindeflection u s , say of magnitude The values of moment obtained the two integrations lead  . Equation values of moment stress-resultant obtained in the two integrations lead  . The to U s1 and U s 2 giving approximately: respectively the functions approximately: respectively to the functions U s 2UsU dU s U s1 and 2 giving s1 = (25) du dU (25) U  s 2  U s1 s s = (25)  dus From this result, the other two non-zero elements in the row s of matrix K  can be defined Equation (24). It two would beelements appreciated that From using thisFrom result, other two non-zero the row smatrix ofabove matrix K  can be this the result, the other non-zero elements ininthe row s ofthe [K ] scheme can be defined using Equation Itthe would be appreciated that the above represents very efficient way of forming equations required the defined ausing Equation (24). It (24). would benonlinear appreciated that the aboveforscheme scheme represents a very efficient way of forming the nonlinear equations solution. represents a very efficient way of forming the nonlinear equations required for the required for the solution. solution. DETERMINATION OF ULTIMATE LOAD DETERMINATION OF ULTIMATE LOAD DETERMINATION OF ULTIMATE LOAD As stated earlier, the procedure for calculating the ultimate load commences with applying a small axialearlier, load andprocedure establishing the equilibrium deflected shape using thewith Asearlier, stated the forcalculating calculating the ultimate load load commences As stated the procedure for the ultimate commences method described above. The applied load is then increased and the procedure is the applying smalland axialestablishing load and establishing the equilibrium deflected applying with a small axiala load the equilibrium deflected shape using shape using the method described above. The applied load is then increased repeated, an equilibrium shape is notload obtainable in a pre-defined of is methoduntil described above. The applied is then increased and thenumber procedure and thestage, procedure is calculations repeated, until can an equilibrium shape is not obtainable in iterations. At this the revert to the last solution obtained and repeated, until an equilibrium shape is not obtainable in a pre-defined number of a pre-defined of iterations. At this stage, calculations can revert reduce the increment innumber the the load. In this manner, thethe ultimate load can beobtained obtainedand iterations. At this stage, calculations can revert to the last solution to the last solution obtained and reduce the increment in the load. In this for any desired level of accuracy. reduce the increment in the load. this manner, the ultimate load can be obtained manner, the ultimate load can beInobtained for any desired level of accuracy. for any desired level of accuracy. Load

Load

No convergence Ultimate load No convergence Ultimate Typical Equilibrium point load Typical Equilibrium point

Deflection Deflection

Figure 5 – Ultimate load calculation procedure Figure 5 – UltimateFigure load calculation 5 –procedure Ultimate load calculation procedure A computer program labelled TUBCOLS has been developed using the above method. To A establish validity of thelabelled method, comparisons with testsusing reported in literature Athe computer program TUBCOLS beendeveloped developed thethe above computer program labelled TUBCOLS hashas been using above method. are To now described. method. establish the validity the method, comparisons tests in literature establish the To validity of the method,of comparisons with tests with reported reported in literature are now described. are now described. COMPARISON WITH TESTS REPORTED BY GANG SHI et al. COMPARISON WITH TESTS REPORTED BY GANG SHI ET AL. COMPARISON WITH TESTS REPORTED BY GANG SHI et al. Gang Shi et al. (2014) conducted tests on 24 circular hollow section columns. The columns were loaded with equal eccentricity at both ends. The observations included lateral deflections. Key parameters of the columns are given in Table 1.

Proceedings of the METNET Seminar 2016 in Castellon

It should be pointed out that the length given is the distance between centres of rotation as listed for each test specimen in Ref [5]. The eccentricity values listed were stated to be derived from the longitudinal strain values measured around the periphery of the column which were converted to an applied bending moment, which divided by the axial load was adopted as the eccentricity. Table 1 - Common properties

Test Specimen

Outer Diameter Mm

Tube Thickness mm

Effective Length mm

Out of Straightness mm

Loading Eccentricity mm

D420-20-1

274

6.01

2362

-0.60

10.20

D420-20-2

275

6.01

2361

-0.40

-2.77

D420-20-3

274

5.87

2362

0.60

2.71

D420-30-1

273

5.85

3305

-0.30

1.84

D420-30-2

273

5.95

3307

-0.75

-2.75

D420-30-3

273

5.91

3306

-2.25

11.72

D420-40-1

275

5.89

4248

1.60

5.88

D420-40-2

276

5.83

4249

0.85

-3.98

D420-40-3

271

5.91

4248

-1.05

6.57

D420-50-1

274

5.87

5193

2.95

0.10

D420-50-2

273

5.88

5192

-1.10

7.90

D420-50-3

273

5.88

5193

-0.20

2.85

D420-60-1

273

5.85

6134

0.40

-1.33

D420-60-2

273

5.95

6136

0.80

2.02

D420-60-3

272

5.91

6137

1.20

1.44

D420-20-4

274

6.01

2362

0.70

-29.33

D420-20-5

274

6.01

2362

1.00

6.79

D420-20-6

274

5.87

2362

-2.50

13.56

D420-40-4

273

5.94

4248

1.15

2.26

D420-40-5

276

6.11

4249

0.25

1.47

D420-40-6

269

6.15

4248

-1.20

0.15

D420-60-4

272

5.99

6134

0.55

4.30

D420-60-5

275

6.00

6134

-0.75

-2.55

D420-60-6

272

5.81

6135

-1.75

2.95

In the calculations presented here, steel is assigned a multi-linear stress strain characteristic shown in Figure 6 is used, using the experimentally recorded parameters reported in (Gang Shi et al. 2014). These are listed below in Table 2.

Proceedings of the METNET Seminar 2016 in Castellon

Stress

f u f y



Stress

f u

f y



 



 

p py p u uy Strain Strain Strain





Strain

Figure 6 – Idealised stress-strain Figure 6Idealised – characteristic Idealised stress-strain characteristic Figure 6 – Figure 6 –stress-strain Idealised Figure 6 –stress-strain Idealised characteristic stress-strain characteristic characteristic

2 –Table Material Properties Material Properties TableTable 2 – Table Material 22 ––Properties Material Table 2 Properties – Material Properties fy

εu εy fu fy fu fu p εp fy MPa fy fy fuMPa εyy ε y εp εyu εp εu MPa MPa MPa 2179 MPa 21195 D420-20-1 MPa 438 MPaMPa564MPa 151329 D420-20-1438 21195 438 438564 564 438 2179 151329 D420-20-1 D420-20-1 D420-20-1 564 2179 2179 21195 21195 151329 D420-20-2 438 564 2179 564 21195 2179 151329 D420-20-2 438 564 2179 21195 151329 D420-20-2 D420-20-2 438 D420-20-2 564 2179 2179 21195 21195 151329 D420-20-3 449 438564588 438 2269 564 20574 2179 146414 D420-20-3449 588 449 2269 20574 146414 D420-20-3 D420-20-3 D420-20-3 588 2269 2269 20574 20574 146414 D420-30-1 450449 449588571 2321 588 24600 2269 145107 D420-30-2 457 576 2406 24735 153722 D420-30-1 D420-30-1 450 D420-30-1 450 571 450 571 2321 571 2321 24600 2321 24600 D420-30-1 450 571 2321 24600 145107145107 D420-30-3 451 575 2336 24642 141022 D420-30-2D420-30-2 D420-30-2 576 2406 576 2406 24735 2406 24735 153722 D420-30-2457 457 457576 576 457 2406 24735 153722 D420-40-1 445 564 2202 25782 155329 D420-30-3D420-30-3 451 D420-30-3 451 575 451 575 2336 575 2336 24642 2336 24642 D420-30-3 451 575 2336 24642 141022141022 D420-40-2 464 586 2273 22078 152648 D420-40-1D420-40-1 445 D420-40-1 445564 564 445 564 2202 564 2202 25782 2202 25782 155329 D420-40-1 445 2202 2578225568 155329141368 D420-40-3 450 570 2305 D420-40-2D420-40-2 464 D420-40-2 464586 464 586 2273 586 2273 22078 2273 22078 152648 D420-40-2 586 2273 2207824600 152648145107 D420-50-1 450464 571 2321 D420-40-3D420-40-3 450 D420-40-3 450570 450 570 2305 570 2305 25568 2305 25568 141368 D420-50-2 477450 577 2387 D420-40-3 570 2305 2556826743 141368144093 D420-50-1 D420-50-1 450 D420-50-1 450571577 450 571 2321 571 2321 24600 2321 24600 145107 D420-50-3 477450 2387 D420-50-1 571 2321 2460026743 145107144093 D420-50-2 D420-50-2 477 D420-50-2 477 577 477 577 2387 577 2387 26743 2387 26743 144093 D420-60-1 450 571 2321 24600 145107 D420-50-2477 477 577 2387 26743 144093 D420-50-3 D420-50-3 D420-50-3 577 2387 2387 26743 26743 144093 D420-60-2 457 477577576 477 2406 577 24735 2387 153722 D420-50-3 477 577 2387 26743 144093 D420-60-1 D420-60-1 450 D420-60-1 571 2321 2321 24600 24600 145107 D420-60-3 451 450571575 450 2336 571 24642 2321 141022 D420-60-1457 571 457 2321 24600 145107 D420-60-2 D420-60-2 D420-60-2 576 2406 2406 24735 24735 153722 D420-20-4 473450 457576561 2440 576 33155 2406 127424 D420-20-5 473457 451575561 2440 575 33155 2336 127424 D420-60-3 D420-60-3 D420-60-3 575 2336 2336 24642 24642 141022 D420-60-2451 576 451 2406 24735 153722 D420-20-6 473451 473561561 2440 561 33155 2440 127424 D420-20-4 D420-20-4 D420-20-4 561 2440 2440 33155 33155 127424 D420-60-3473 575 473 2336 24642 141022 D420-40-4 475 473561561 473 2400 561 37378 2440 135251 D420-20-5 D420-20-5 473 D420-20-5 561 2440 2440 33155 33155 D420-20-4 473 561 2440 33155 127424127424 D420-40-5 475 563 2526 30690 127424 D420-20-6D420-20-6 473 D420-20-6 473561 561 473 561 2440 561 2440 33155 2440 33155 127424 D420-20-5 473 2440 3315531398 127424119578 D420-40-6 469 558 2393 D420-40-4D420-40-4 475 D420-40-4 475561 475 561 2400 561 2400 37378 2400 37378 135251 D420-20-6 561 2440 3315533155 127424127424 D420-60-4 473473 2440 D420-40-5 D420-40-5 475 D420-40-5 475563561 475 563 2526 563 2526 30690 2526 30690 127424 D420-60-5 473475 561 2440 D420-40-4 561 2400 3737833155 135251127424 D420-40-6D420-40-6 469 D420-40-6 469558 469 558 2393 558 2393 31398 2393 31398 119578 D420-60-6 473475 561 2440 33155 127424 D420-40-5473 2526 30690 D420-60-4D420-60-4 D420-60-4 473561 563 473 561 2440 561 2440 33155 127424 2440 33155 127424 D420-40-6473 469 473561 558 473 2393 3139833155 119578 D420-60-5D420-60-5 D420-60-5 561 2440 561 2440 2440 33155 127424 D420-60-4 473 561 2440 33155 127424 D420-60-6D420-60-6 473 D420-60-6 473561 473 561 2440 561 2440 33155 2440 33155 127424 D420-60-5

473

561

2440

33155

127424

D420-60-6

473

561

2440

33155

127424

εp εu 21195 151329 21195 151329 20574 146414 24600 145107 24735 153722 24642 141022 25782 155329 22078 152648 25568 141368 24600 145107 26743 144093 26743 144093 24600 145107 24735 153722 24642 141022 33155 127424 33155 127424 33155 127424 37378 135251 30690 127424 31398 119578 33155 127424 33155 127424 33155 127424

Proceedings of the METNET Seminar 2016 in Castellon

Figure 7 â&#x20AC;&#x201C; Load deflection plot for column D420-50-2

Figure 7 shows a typical plot of applied load and calculated deflections, in this

Figure 7 shows a typical plot of applied load and calculated deflections, in this case case for Specimen D420-20-1. The reducing stiffness of the column can be for Specimen D420-20-1. of the column can be readily readily visualised, The givingreducing confidence stiffness in the calculated value of the ultimate visualised, giving load. confidence in the calculated value of the ultimate load. 3 lists the computed test failure ratio.The The mean mean value of Table 3 listsTable the computed and testand failure loadsloads and and theirtheir ratio. value ofisthe ratio obtained is 0.940 with a standard of 7.4%. Clearly, the ratio obtained 0.940 with a standard deviation of deviation 7.4%. Clearly, for this set, the for this set,the the test correlation the test and computed correlation between and between computed failure loads is failure good loads and isis on the good and is on the conservative side. conservative side. Table 3 - Comparison between computed and test failure loads

Table 3 - Comparison between computed and test failure loads Column D420-20-1 D420-20-2 D420-20-3 D420-30-1 D420-30-2 D420-30-3 D420-40-1 D420-40-2 D420-40-3 D420-50-1 D420-50-2 D420-50-3 D420-60-1 D420-60-2

Test Column D420-20-1 D420-20-2 D420-20-3 D420-30-1 D420-30-2 D420-30-3 D420-40-1 D420-40-2

Test Strength, Pt

Strength, kN Pt kN 2090 2090 2173 2173 2178 2178 2075 2075 2075 2075 2094 2094 2027 2027 2100 2100 2026 2038 1997 2012 1957 1974

Computed Strength, Pc

Computed kN Strength, Pc kN 1973.3 1973.3 2137.8 2137.8 2129.0 2129.0 2147.9 2147.9 2148.4 2148.4 1918.4 1918.4 1904.9 1904.9 2126.7 2126.7 1941.4 1997.2 1866.4 2090.2 1924.6 1800.9

Ratio Pc / Pt

Ratio 0.944 Pc / Pt 0.984 0.978 1.035 1.035 0.916 0.940 1.013

0.944 0.984 0.978 1.035 1.035 0.916 0.940 1.013 0.958 0.980 0.935 1.039 0.983 0.912

Proceedings of the METNET Seminar 2016 in Castellon

D420-40-3

2026

1941.4

0.958

D420-50-1

2038

1997.2

0.980

D420-50-2

1997

1866.4

0.935

D420-50-3

2012

2090.2

1.039

D420-60-1

1957

1924.6

0.983

D420-60-2

1974

1800.9

0.912

D420-60-3

1960

1795.4

0.916

D420-20-4

2033

1761.1

0.866

D420-20-5

2375

2172.0

0.915

D420-20-6

2100

2042.1

0.972

D420-40-4

2356

2177.9

0.924

D420-40-5

2449

2347.7

0.959

D420-40-6

2306

2319.3

1.006

D420-60-4

2277

1762.7

0.774

D420-60-5

2329

1862.8

0.800

D420-60-6

2285

1762.9

0.772 Mean

0.940

Standard Deviation

0.074

Gang Shi et al. (2014) calculated the strength of their test columns using the finite element method. They obtained a mean value of 0.96 with a standard deviation of 5.98%. They commented that last three columns were made of galvanised steel which gave much higher test results compared with the calculated values based on the measured material properties. They attributed this to the fact that these columns were galvanised. The galvanisation process appears to have modified the material properties. Ignoring the three results, the mean value of the ratio between the calculated load and the experimental failure load was found to be 0.962. The standard deviation of the 21 results is 4.62%. These numbers may be considered excellent correlation with experiments. The numerical results presented here, based on the finite element method give accuracy which compares very well against the finite element method.

Proceedings of the METNET Seminar 2016 in Castellon

CONCLUSION The paper describes a method for the ultimate load calculation of columns made of circular hollow sections. The method, based on the finite difference method is shown to give good results when compared with 24 tests reported by Gang Shi et al. (2014). The overall accuracy of the method is very similar to that obtained by the finite element method. The method presented here requires just a few lines of input to define the geometry, material properties, imperfections and loading, making it suitable for large scale parametric studies.

REFERENCES EN 1993-1-1 (2003). CEN. Eurocode 3 - Design of steel structures - Part 1-1: General rules and rules for buildings. EN 1993-1-1:2003. European Committee for Standardisation, Brussels. Zhou J-H and Young B (2006). Aluminium alloy circular hollow section beamcolumns. Thin-Walled Structures, 44, 131–140. Han L-H, Yao G-H and Tao Z (2007). Behavior of concrete-filled steel tubular members subjected to combined loading. Thin-Walled Structures, 45, 600– 619. Virdi K S (2006). Finite difference method for nonlinear analysis of structures. Journal of Constructional Steel Research, 62, 1210–1218. Shi G, Jiang X, Zhou W, Chan T-M and Zhang Y (2014). Experimental study on column buckling of 420 MPa high strength steel welded circular tubes. Journal of Constructional Steel Research, 100, 71–81.

Proceedings of the METNET Seminar 2016 in Castellon

THE FLAT DOUBLE-LAYER GRID-CABLE STEELCONCRETE COMPOSITE STRUCTURE G.M. Gasii Ph.D., Associate Professor Poltava National Technical Yuri Kondratyuk University, Poltava, Ukraine Faculty of Civil Engineering Department of Structures from a Metal, Wood and Plastics

ABSTRACT The paper studies constructive concept of the flat double-layer grid-cable steelconcrete composite structure. The flat double-layer grid-cable steel-concrete composite structure is the new kind of roof system for long-span buildings. The composite structure has been developed in Poltava National Technical Yuri Kondratyuk University on department of structures from metal, wood and plastics. The feature of the constructive concept is in the bottom chord that consists of flexible bars and top chord that is made from a concrete slab. The flexible bars are made from segments of steel cables that have special details at the ends through which the segments are connected to each other. The bottom chord is not complex and its manufacturing and assembly does not take much time. Due to the use of steel cables, the total weight of the structure decreases, saving materials and, as a result, the total cost of construction also decreases. The paper describes the results of the numerical investigation of load bearing capacity of a modular steel-concrete composite spatial unit.

INTRODUCTION AND PROBLEM STATEMENT In the current development of scientific and technological achievements and increasing social needs of the population, there is a need to find more effective structural systems, including roof systems. The main requirements that apply to the structural systems of buildings, in addition to reliability and load capacity, is an architectural expression, aesthetics, ergonomics and good indicators of energy efficiency. The important aspect of designing and finding a constructive concept for new designs is the use of reliable and advanced materials. Steel and concrete with various modern fillings belong to the materials that satisfy these requirements. The effectiveness of the developed designs also depend on how materials are used and on the behavior of constructions. It means that materials should be

Proceedings of the METNET Seminar 2016 in Castellon

under appropriate forces, which they are resisting well for example steel is used in tension or compression elements but concrete is used as compressed elements. Considering this fact, it has been decided to combine steel bars and concrete slabs in a unified spatial design and further its research to a broad implementation in practice of construction. This is a promising direction of development of building structures.

ANALYSIS OF RECENT SOURCES OF RESEARCH AND PUBLICATIONS The analysis has showed that traditionally the most known large-span spatial designs are made entirely of steel members, including flat double-layer grid [1]. However, there are examples of steel combined systems [2]. In addition, steel-concrete composite constructions are used widely in a variety of designs [3, 4]. These constructions are used in many sectors of industrial and civil construction [5]. Steel-concrete composite constructions are used as roof systems, slab system, columns, different plate structures and bearing elements of buildings and structures [6].

HIGHLIGHT UNSOLVED PARTS OF THE GENERAL PROBLEM Based on the analysis of previous studies and taking into account the advantages steel-concrete composite constructions the issue of development of the structural concept of the new spatial structure systems with its use has not been fully investigated.

PROBLEM FORMULATION The aim of the study is to identify the most promising and effective designs using the theoretical studies of the current state of construction and spatial steel and concrete composite structures. Given the physical and mechanical properties of materials and properties of structural elements is to offer and develop the new type of space structure systems.

THE MAIN MATERIAL AND RESULTS The development of the construction industry is accompanied by the introduction of effective materials. The use of these materials allows structure systems with the necessary strength characteristics and technical-economic indicators, but together with their development there is a need for the development of new geometric shapes and structural concept. Grid systems are the best known among long span and spatial structures. These systems have a great ability and flexibility to the form finding that is evidenced by many original shapes in the world [7].

Proceedings of the METNET Seminar 2016 in Castellon

Flat double-layer grids are allocated more frequently from the general class of the grid systems than others. That is why, for the development of the new type of space structure systems has been taken the flat double-layer grids as a basis. The original concept of structural system has been developed using world experience of shaping large-span constructions. This is about the flat double-layer grid-cable steel-concrete composite structure (Figure 1). This is the new kind of a spatial construction, the essence of which lies in appropriate use of materials and their properties [8].

Figure 1. Segment of The flat grid-cable steel-concrete composite structure: 1 – a top chord; 2 – a Figure 1. Segment of double-layer The flat double-layer grid-cable steel-concrete composite diagonal; 3 – a bottom chord; 4 – a nodal boltedstructure: connections 1 – a top chord; 2 – a diagonal; 3 – a bottom chord; 4 – a nodal bolted connections

The flat double-layer grid-cable steel-concrete composite structure consists

modular steel-concrete composite spatial units that are connected to each Theof flat double-layer grid-cable steel-concrete composite structure consists of other through the nodal bolted spatial connections (Figure in a complete design. modular steel-concrete composite units that are 2) connected to each other through the nodal bolted connections (Figure 2) in a complete design.

Elements of construction are made with automated processes, especially this

Elements are made withparts automated processes,the especially appliesoftoconstruction the production of nodal and processing ropes. this applies to the production of nodal parts and processing the ropes.

modular steel-concrete composite spatial unit is made segments of TheThe modular steel-concrete composite spatial unit is made from from segments of steel steel tubes and reinforced concrete slab. tubes and reinforced concrete slab. TheThe main feature of the reinforced concrete slab is is the structural concept. Slab is main feature of the reinforced concrete slab the structural concept. Slab made from from cement-sand mortarmortar and steel wovenwoven nets. nets. The The slab slab has has a size of is made cement-sand and steel a size 1500×1500 mm in mm plan in and depth 60 mm. slab isThe reinforced woven with steel of 1500×1500 plan andof depth ofThe 60 mm. slab is with reinforced nets and steel bars (Figure 3).

woven steel nets and steel bars (Figure 3).

The nets for reinforcement are made from thin steel wire that had a diameter of reinforcement thinmm. steelThe wiredistance that hadbetween a diameter 1.2The mm.nets Thefor steel nets have aare cellmade size from of 20×20 the of 1.2 mm. The nets a cellissize of 20×20 distance that between woven steel nets is steel 5 mm. Thishave distance provided withmm. steelThe disc-gaskets were placed between nets. Steel disc-gaskets are placed at every 200 mm. Steel nets and steel disc-gaskets are connected together with a cement-sand mortar in a single structure. This way of reinforcement of the slab allows reducing a weight

Proceedings of the METNET Seminar 2016 in Castellon

the woven steel nets is 5 mm. This distance is provided with steel disc-gaskets that were placed between nets. Steel disc-gaskets are placed at every 200 mm. Steel nets and steel disc-gaskets are connected together with a cement-sand mortar in a single structure. This way of reinforcement of the slab allows reducing a weight and cost of materials in comparison with conventional reinforced concrete structures with the same bearing capacity [8].

andand costcost of materials in comparison withwith conventional reinforced concrete structures of materials in comparison conventional reinforced concrete structures withwith thethe same bearing capacity [8].[8]. same bearing capacity

Figure 2. The nodal bolted connection: 1 –The a connecting detail; 2 connection: – a nut; 3 – a connecting plate; 4 – a diagonal; Figure 2. nodal bolted Figure 2. The nodal bolted connection: 5 – 1a bottom chord; 6 – a detail; boltdetail; –1a–connecting 2 –2a–nut; 3 –3a–connecting plate; 4 –4a–diagonal; 5 –5a– a a connecting a nut; a connecting plate; a diagonal; bottom chord; 6 –6a–bolt bottom chord; a bolt

Figure 3. A3.Part of steel reinforcing mesh: Figure A Part of steel reinforcing mesh: 1 – a wire d=15 mm; 2 – a steel disc-gasket t=5 mm; A=25 mm

Figure 3. A Part reinforcing mesh:21 –– aa wire d=15 mm; 2 – a steel t=5 mm; A=25 mm 1 –of asteel wire d=15 mm; steel disc-gasket t=5 disc-gasket mm; A=25 mm TheThe average modulus of elasticity for for different parts of the slabslab hashas been used for for average modulus elasticity of of the been used The average modulus ofofelasticity for different differentparts parts the slab has been used modeling its behavior. modeling its behavior.

for modeling its behavior.

Proceedings of the METNET Seminar 2016 in Castellon

Analysis of the stress-strain state of the construction has been investigated with the Analysis of the stress-strain state of the construction has been investigated FE method. For this had been defined physical and mechanical properties of Analysis theFEstress-strain state thebeen construction has beenand investigated with the with of the method. For thisofhad defined physical mechanical materials via experimental testing (Figure 4). The boundary conditions in the FE method. this had via been defined physical and mechanical properties of propertiesFor of materials experimental testing (Figure 4). The boundary numerical modeling are shown in Figure 5. The von Mises stress contour shown in materials via experimental testing (Figure 4). The boundary conditions conditions in the numerical modeling are shown in Figure 5. The von Mises in the Figure 6 is the result of the solving of numerical model under the estimated load. numerical are shown in Figure 5. The von stressofcontour shown in stress modeling contour shown in Figure 6 is the result of Mises the solving numerical Figure 6 is under the result of the solving model the estimated load.of numerical model under the estimated load.

a) a)

b)b)

c)c)

Figure 4. The stress-strain curve:Figure a) of concrete; b) of steel reinforcement; c)curve: of steel Figure 4. The The stress-strain 4. stress-strain curve:

concrete;b) b) of of steel steel reinforcement; a)a)ofofconcrete; reinforcement;c)c)ofofsteel steel

Figure 5. The boundary conditionals: The boundary conditionals: L – the length, 1.5 m; t – the thing, 0.06 m; q – the uniformly distributed L –Figure the5.length, 1.5 m; t – the thing, 0.06 m; q – the uniformly distributed load that load that equals snow load for Ukraine 5. The boundary conditionals: Figure equals snow load for Ukraine

L – the length, 1.5 m; t – the thing, 0.06 m; q – the uniformly distributed load that equals snow load for Ukraine

Proceedings of the METNET Seminar 2016 in Castellon

Figure The von stress Figure1 6. 6. The von1; Mises Mises stress contours: Figure 6. The von Mises stress contours: – the region 2 – the region 2 contours: 1 1– – the the region region 1; 1; 2 2– – the the region region 2 2

Figure 7. The vonFigure Mises stressThe contours at the destruction loadcontours at the destruction load Figure 7. 7. The von von Mises Mises stress stress contours at the destruction load Figure a failure (destruction) model of connection of developed Figure 7 77 shows shows model of nodal nodal connection of the the of developed Figure shows aa failure failure(destruction) (destruction) model of nodal connection the construction. The contour of solid von Mises stresses (see Figure 6) shows that construction. The contour of solid von Mises stresses (see Figure 6) shows that the the developed construction. The contour of solid von Mises stresses (see Figure maximum stress that has appeared is less than the limit stress of steel. The factor of maximum stress that has appeared is less than the limit stress of steel. The factor of 6) shows that the maximum stress thatmakes has appeared istoless than the limit safety of steel nodal details is 1.3. That it possible optimize the shape of safety of steel nodal details is 1.3. That makes it possible to optimize the shape of stress of steel. The factor ofstresses safety ofare steel nodal details is 2. 1.3. That makes it the the steel steel details. details. The The largest largest stresses are in in the the regions1 regions1 and and 2. possible to optimize the shape of the steel details. The largest stresses are in the regions1 and 2.

The destruction of steel details (see Figure 2) came in the regions 1 and 2 when a load had exceeded the estimated load 1.3 times.

Proceedings of the METNET Seminar 2016 in Castellon

DISCUSSION AND CONCLUSIONS The study deals with a new kind of spatial construction that consist of modular units. From the numerical investigation is obtained the stress-strain state of the flexible bottom chord of the flat double-layer grid-cable steel-concrete composite structure. The von Mises stress contours have been obtained with FE method. The stress contours gave an opportunity to identify the locations of the maximum stresses and strains. The factor of safety was obtained using information about the maximum stress and strains. Research confirms the effectiveness of structural concept of the construction.

REFERENCES [1] Chilton J. Space grid structures. Boston: Architectural Press, 2007,180 p. [2] Ivanyk I., Vybranets Y. and Ivanyk Y. Research of composite combined prestressed construction. Acta Scientiarum Polonorum, 2014. no. 13(2), pp. 81– 88. [3] Gasii G.M. Technological and design features of flat-rod elements with usage of composite reinforced concrete. Metallurgical and Mining Industry, 2014, no. 4, pp. 23–25. [4] Стороженко Л.І. and Гасій Г.М. Нові композитні матеріали кріплення гірничої виробки. Науковий вісник Національного гірничого університету, 2015, no 4, pp. 28–34. [5] Storozhenko L.I., Gasii G.M. and Gapchenko S.A. The new composite and space grid cable-stayed construction. Academic journal. Industrial Machine Building, Civil Engineering. PoltNTU. Poltava, 2014, no. 1, pp. 91–96. [6] Nizhnik O.V Construction of steel and concrete composite girderless floor. Building construction. NIISK. Kyiv, 2013, no. 78(1), pp. 144–149. [7] Storozhenko L.I. and Gasii G.M. Experimental research of strain-stress state of ferrocement slabs of composite reinforced concrete structure elements. Metallurgical and Mining Industry, 2014, no. 6, pp. 40–42. [8] Стороженко Л.І., Гасій Г.М., Гапченко С.А. Просторові сталезалізобетонні структурно-вантові покриття: Монографія, Полтава: ТОВ «АСМІ», 2015, 218 с.

Proceedings of the METNET Seminar 2016 in Castellon

APPLYING VISIONARY CONCEPT DESIGN TO ENERGY EFFICIENT RESIDENTIAL AREAS Jukka Laitinen (jukka.laitinen@laurea.fi) Tarja Meristรถ (tarja.meristo@laurea.fi) FuturesLabs CoFi Laurea University of Applied Sciences

INTRODUCTION In the EU 40% of the energy consumption and 36% of CO2 emissions are related to buildings (European Commission). Therefore, the energy efficiency of the buildings, and more broadly residential areas, has an important role when looking for solutions to the problems caused by climate change. In this paper we will focus on the energy efficiency of residential areas. By definition, a residential area is a physical and functional entity containing building blocks and also public and commercial services, such as grocery shops, day nurseries, schools, parks and recreation areas within a walking distance. The case areas in this paper consist of three residential areas in Finland which are in different phases of their life cycle. The framework for this paper consists of future research combined to participatory design. The timeframe is reaching up to the next 20 years. As a methodological tool we apply visionary concept design (Kokkonen et al. 2005; Meristรถ et al. 2009). The data collection for future scenarios forming the basis for the visionary concept design process took place in a futurology study course held for 25 MBA students in Laurea University of Applied Sciences during spring 2016. The students, working in small groups, collected data and created alternative scenarios and visionary concepts for all three residential areas. This aim of this paper is to introduce the methodological process and its outcomes. The results will open new views of the future when developing energy efficient residential areas from different perspectives including logistics, housing, energy, place and space, demographics, services, construction and safety & security. The results include also trends collected as background information for the scenarios. As a conclusion, a set of scenarios with visionary concepts will be presented.

Proceedings of the METNET Seminar 2016 in Castellon

RESEARCH DESIGN The focus of this paper is in three research questions: 1. 2. 3.

What are the drivers formulating the future alternatives for energy efficient residential areas? How can thematic scenario alternatives support the participatory design? What are the concrete visionary concepts supporting energy efficiency in the case areas?

The methods applied in this study are scenario approach and visionary concept design. Scenario working is a method within the field of futures research (Bell 1997, Masini 1993). Scenario working includes mapping alternative futures, identifying factors and development paths leading to different future outcomes. The action scenario approach (MeristĂś 1989) incorporates also the evaluation of the significance of the scenarios for the user. Finally, based on the evaluation necessary actions are suggested. Scenarios are descriptions of different futures. Besides including the description of the competitive environment with factors like politics, economy, society, technology and environment the approach also incorporates the process of development. Scenarios are different from forecasts, as scenarios are usually not measured by their probability of occurrence (Schwartz 1996). Scenarios are not exact descriptions of the future; they are rather verbal descriptions of both qualitative and quantitative nature. Our framework is based on a multiple scenario approach i.e. at least two alternative scenarios are constructed. Furthermore, each scenario leads to various possible choices of strategies and alternative visionary concepts based on need in alternative futures (Kokkonen et al. 2005). The method for creating visionary future product concepts consists of five main steps (Figure 1). The first step is the identification of change factors, which forms basis for the second step i.e. scenario building. The third step is the identification of product needs in each scenario. The fourth step is the actual generation of future product concepts based on the market need identified in each scenario. The fifth and last step of the method is the timing of R&D â&#x20AC;&#x201C;activities and operations. This step also includes other considerations concerning the contribution the visionary concepts might have to the companyâ&#x20AC;&#x2122;s business planning or strategy (Kokkonen et al. 2005). External change factors build up the framework for the scenarios and also for the visionary concepts. However, the factors are interrelated with the actors, and that is why pure objective knowledge about factors is not enough. Different actors have different interests, aims and needs. Therefore, it is important to take this actor-view into consideration when creating scenarios for the visionary concepts. It is preferable that the same group of people who are generating future concepts are also building the scenarios for the basis of the concepts phase. When people are involved into the process from the

Proceedings of the METNET Seminar 2016 in Castellon

very beginning, they have time to assimilate the content of the scenarios. This boosts group members to create more and better ideas in the phase of concept generation (Leppimäki et. al. 2008). There are several ways to present and illustrate the visionary concepts. However, the concept description should include at least the main features of the concept, estimations on its market potential and technical feasibility. These are the main dimension of the concept by which also a picture of the future operational environment is depicted. Further, illustrations, sketches, animations or other visual presentation material are excellent ways to communicate the concepts and related ideas (Leppimäki et. al. 2008).

Figure 1. The five main steps to create visionary concepts

Figure 1. The five main steps to create visionary concepts (adapted from Kokkonen et. al. 2005). (adapted from Kokkonen et. al. 2005).

The time perspective of the visionary concepts is long. Therefore, they are

The time perspective of the of visionary concepts is long. Therefore, they areoftargeted at the targeted at the markets the future, realized with the technology the future markets of the future, realized with the technology of the future and guided by societal norms and guided by societal norms and legislation of the future. Consequently, and legislation of the future. Consequently, some details and features of the concept are some details and features of the concept are difficult to describe and are more difficult to describe and are more or less a matter of imagination. Several benefits arise from less time a matter imagination. Severalfeature benefits from the longTo time range theorlong rangeofwhich is an essential in arise visionary concepts. begin with, which is an essential feature in visionary concepts. To begin with, visionary visionary concepts enable systematic examination of alternative future developments concepts enable systematic examination of alternative futureindevelopments because future scenarios are illustrations of the operational environment the future. It also takes into account driving forces (e.g. changes values, technological breakthroughs because future the scenarios are illustrations ofinthe operational environment in and new markets) as well asinto market potential, uncertainty and (e.g. challenges related to future the future. It also takes account the driving forces changes in values, in technological alternative scenarios. Moreover, and visionary product concept design and breakthroughs new concepts markets)enable as well as market potential, R&D for the future, over next product generation visualizes the future as products which are uncertainty and challenges related to future in alternative scenarios. corresponding to the market needs (Leppimäki et. al. 2008).

Moreover, visionary concepts enable product concept design and R&D for the future, over next product generation visualizes the future as products which are corresponding to the market needs (Leppimäki et. al. 2008).

CASE AREA DESCRIPTIONS

This study is based on the ongoing ELLI project financed by European Regional Development Fund ERDF. In the project there are three case residential areas which all are located in Southern Finland: Engelinranta in Hämeenlinna, Askonalue in Lahti and Peltosaari in Riihimäki. Engelinranta in Hämeenlinna is located close to the city center by the lakeside. The size of

Proceedings of the METNET Seminar 2016 in Castellon

CASE AREA DESCRIPTIONS This study is based on the ongoing ELLI project financed by European Regional Development Fund ERDF. In the project there are three case residential areas which all are located in Southern Finland: Engelinranta in Hämeenlinna, Askonalue in Lahti and Peltosaari in Riihimäki. Engelinranta in Hämeenlinna is located close to the city center by the lakeside. The size of the area is 47 hectares of which 23 hectares is land area and 24 hectares is water area. It has been planned to build apartments for about 2600 inhabitants. Additionally, there will be spaces for the different business activities. In the construction of Engelinranta, it is possible to follow the principles of the sustainable development because the area leans on existing community technical networks, services, street network, public transport and refreshment network. Askonalue has a central location close to center of Lahti with good transportation connections. Askonalue is an old industrial area and its size is more than 30 hectares. A historical industry milieu and existing service and actor network of the area are a strong foundation for the regional development which is carried out in cooperation with the town inhabitants and other actors. Peltosaari residential area in Riihimäki has been built in the immediate vicinity of the city center during 1970s and the 1980s. The strength of Peltosaari is its location being close to the city center and close to nature. The biggest problems are connected to the condition of the apartments and real estates, energy efficiency and to the social problems. The town of Riihimäki has committed itself strongly to develop Peltosaari to become an attractive and ecological residential area. In ELLI project these areas will be analysed in order to find new business opportunities for cleantech field. In practice, the areas are still in planning phase.

DATA AND PROCESS The data collection for future scenarios forming the basis for the visionary concept design process took place in a futurology study course held for 25 MBA students in Laurea University of Applied Sciences during spring 2016. The process consisted of six steps which were as follows: 1. 2.

Describing a residential area by core competence tree (adapted from Hamel & Prahalad 1996). Collecting future data (megatrends, wild cards, weak signals) for scenario building. Structuring collected data into market / technology / society perspectives according to scenario filter model (Meristo et al. 2009).

Proceedings of the METNET Seminar 2016 in Castellon

Selecting drivers from market / technology / society perspectives and formulating alternative fourfold tables based on those drivers. Choosing one fourfold table for more specific processing and drafting alternative future scenarios by considering assumptions and consequences in each block. Analysing a residential area by SWOT-analysis in alternative scenarios, i.e. recognizing strengths, weaknesses, opportunities and threats in each scenario. Developing proactive and defensive action alternatives in each scenario, i.e. what to do to avoid threats or vice versa, how to exploit opportunities. Creating visionary concepts responding the challenges recognized in each scenario. Those concepts were related to products, services, business models as well as other action models concerning the development of case areas towards sustainability.

RESULTS In this paper the results focus on driving forces behind the scenarios and visionary concepts based on the scenarios. As explained in the previous chapter, the student groups collected future data of megatrends, wild cards and weak signals. At first they applied PESTE approach for data collection, i.e. including political, economic, social, technological and ecological factors. Then, they structured the data into market / technology / society perspectives and chose drivers for the scenarios. The summary of the main drivers are introduced in Table 1. It is remarkable that drivers covered all the dimensions - i.e. market, technology and society – which ensured varied viewpoints about the future even though the general focus was in energy efficient residential areas. Table 1. Summary of the main drivers behind the scenarios (collected from Student Reports 2016). Market

Technology

Society

•

• •

• • •

• • • •

Ownership form of apartments Citizen driven business Sustainable economy Sufficiency of space Building types

• • •

Robotics Speed of adoption of new technology Extent of technological exploitation Humans vs. technology Renewable vs non-renewable energy sources

• • •

Services in residential areas Communality Socioeconomic situation of consumers Basic values Consumer habits Reputation of residential area

Based on the drivers each group created alternative scenarios. After that, visionary concepts based on alternative thematic scenarios were constructed, e.g. from space & place perspective the area with skyscrapers based on legotype system was imagined, including the monorail transportation network above the roofs. On the other hand, a group focusing on living theme produced

Proceedings of the METNET Seminar 2016 in Castellon

a concept â&#x20AC;&#x153;together we are strongerâ&#x20AC;? by using sharing economy concept at the level of residential area. Construction theme focused mostly on concepts utilizing robotics as stuntmen in dangerous tasks. Groups with logistics and services themes had smart concepts in the web for different purposes, including safety & security services (Student Reports 2016). As an illustrative example, we present the scenarios and visionary concepts from the group focusing on the energy theme. The main drivers behind the scenarios were energy sources (vertical axis) and consumer habits (horizontal axis). As a result there are four alternative scenarios which are introduced in Figure 2.

Figure 2. Scenario examples based on group work with energy focus Figure 2. Scenario examples based on group work with energy focus (Student Reports 2016, Lajunen et. al. 2016). (Student Reports 2016, Lajunen et. al. 2016). Scenarios 1. Ad hoc world and Scenario 4. Visionary world have the focus in renewable energy where as in Scenario 2. Laissez faire world and Scenario 3. Balance of horror the emphasis is on non-renewable energy. Additionally, consumer habits cause variety to the scenarios, depending on whether consumers are responsible or irresponsible. Visionary concepts were created for each scenario (Figure 3).

Proceedings of the METNET Seminar 2016 in Castellon

Figure 2. Scenario examples based on group work with energy focus (Student Reports 2016, Lajunen et. al. 2016).

Figure 3. Visionary concepts related to the energy theme (adapted from Student Reports, Lajunen et. al. 2016).

CONCLUSION Even if the focus is in energy efficient residential area the drivers shaping the future can be very diverse. In this study the driving forces were divided according to the scenario filter model including market, technology and society dimensions. The market drivers include factors such as ownership form of apartments, citizenâ&#x20AC;&#x2122;s role, sustainability in business, sufficiency of space and type of buildings. On the other hand, many of the drivers are related to technology. An essential driver regarding the future is the role between renewable and non-renewable energy forms. Other technology-related drivers consist e.g. of robotics, extent of technological exploitation, the speed of implementing new technological solutions and role between humans and technology. Additionally, society related drivers are notable including e.g. services provided in residential areas, communality, basic values and life styles. This study was carried out by student groups and each group had their own thematic viewpoint. Thematic groups dealt with logistics, housing, energy, place and space, demographics, services, construction and safety & security. Alternative scenarios from each thematic perspective were constructed. The scenarios formed in each group shared a language and basis for concept design

Proceedings of the METNET Seminar 2016 in Castellon

work. Timeframe to the future for the next 20 years helped participants to get feet from the ground and think also unthinkable solutions and concepts. Some of the concepts may be suitable only for one certain scenario whereas some concepts may suit several or all scenarios. For example, shared facilities are a must in Scenario 3. Balance of horror, but they work well also in other scenarios. On the other hand, startup hubs in Scenario 4 require more sophisticated conditions to work. The advantage of visionary concept design is that it enables systematic examination of alternative future developments. Future scenarios are illustrations of the operational environment in the future, which also takes into account the driving forces from different perspectives. Thematic focus of each group opened also new insights into residential areas and their energy efficiency, even more broadly, to their sustainable development in the future.

REFERENCES Bell, W. (1997) Foundations of Futures Studies I: History, Purposes, Knowledge. New Brunswick, NJ: Transaction Publishers, 1997. European Commission (2016) https://ec.europa.eu/energy/en/topics/energy-efficiency/buildings Hamel, G. & Prahalad, C.K. (1996). Competing for the Future. Harvard Business School Press. Kokkonen, V., Kuuva, M., Leppimäki, S,, Lähteinen, V., Meristö, T., Piira, S., Sääskilahti, M. (2005). Visioiva tuotekonseptointi - työkalu tutkimus- ja kehitystoiminnan ohjaamiseen. (Visionary concept design – a tool for steering R&D activities). Technology Industry Association in Finland. (In Finnish). Leppimäki, S., Laitinen, J., Meristö, T., Tuohimaa, H. (2008) Visionary Concept: Combining Scenario Methodology With Concept Development. In Wagner, C. (ed.) Seeing the Future Through New Eyes. World Future Society. Masini, E. (1993) Why Futures Studies? Grey Seal, London. Meristö, T. (1989) Not Forecasts but multiple scenarios when coping with uncertainties in the competitive environment. European Journal of Operatios Research Vol 38, pp. 350-357 ,1989. Meristö T., Leppimäki S., Laitinen J. & Tuohimaa H. (2009) Scenario Filter Model: The framework to generate scenarios and select visionary concepts for development. ISPIM conference proceedings.

Proceedings of the METNET Seminar 2016 in Castellon

Schwartz, P. (1996) The Art of the Long View: Planning for the Future in an Uncertain World New York: Currency Doubleday. Student Reports (2016) for Futurology study module. Laurea University of Applied Sciences, Finland (unpublished student reports). •

Färm, M., Salminen, S. & Aro, J. (2016) Demographics theme.

•

Heikkilä, L., Savolainen, S. & Väisänen, M. (2016) Space & place theme.

•

Ilveskoski, A-M., Palotie, P. & Sburatura, M. (2016) Logistics theme.

•

Javanainen, K., Mikkola-Ahokas, A. & Nurmi, T. (2016) Security & safety theme.

•

Lager, N., Hirsso, J. & Kleimola, M. (2016) Construction theme.

•

Lajunen, H., Rautemaa, M., Sainio, J. & Velling, E. (2016) Energy theme.

•

Luostarinen, K., Länsivire, H. & Norppa-Röpelinen, P. (2016) Living theme.

•

Rantanen, K., Rinkinen, N. & Rostedt, P. (2016) Service theme.

Proceedings of the METNET Seminar 2016 in Castellon

CLEANTECH BUSINESS OPPORTUNITIES IN THE FUTURE Tarja Meristö (tarja.meristo@laurea.fi) Jukka Laitinen (jukka.laitinen@laurea.fi) FuturesLabs CoFi Laurea University of Applied Sciences

ABSTRACT In this paper we will focus on cleantech business opportunities (Meristö & Laitinen 2014) which the future will provide for the companies as well as on the skills and capabilities required to exploit those opportunities. The research questions are as follows: 1. 2. 3.

What are the main drivers for the future concerning energy-efficient and sustainable solutions? What are the skills and competences needed to meet the future challenges in this field? What are the barriers and / or enablers when developing cleantech business for the future?

The data collected for this research paper consist of web-survey replies received from industry leaders, governmental decision makers, NGO representatives and individual experts from this field. The main drivers vary from international environmental agreements to consumer market behavior, depending on how strictly the legislation, but also the purchasing power of consumers will guide the development. Skills concerning digitalization and automation, but also marketing and communication are important among skills concerning new materials and renewable energy development work. Taboos in this field are discussed, too.

INTRODUCTION Environmentally friendly solutions in different business sectors have been sought during the last few decades. Energy sector and the process industry have been forerunners in developing eco-efficient production processes from the 1960´s. For example, pulp and paper industry has faced environment, health and safety issues very early in 1970´s in the consumer market. In the late 1980´s, and even more during 1990´s and in the new millennium, climate change discussion took a big role in everyday media. All businesses

Proceedings of the METNET Seminar 2016 in Castellon

and industrial sectors had to meet the challenges concerning corporate social responsibility (Meristö 2013). Today the new economy (Erber & Hageman 2005) has been replaced by circular economy (Pearce & Turner 1989) and all the industries are looking for new opportunities from sustainable solutions (Kettunen 1998), in processes and products and services. Value chain throughout the whole life-cycle includes many different opportunities for the growth of cleantech business. In the beginning, the input of raw materials and energy is the key enabler (or barrier) to cleantech business. The more renewables are used in the energy side, the cleaner the business is in the long run. Also, the more renewables or recycled materials are used, the greater will be the opportunities for the cleantech defined business. Logistics and transportation along the value chain constitute a remarkable factor when defining the sustainability rate of the businesses or communities. The sustainable community, by definition, has at least a chance for logistics with no emissions, i.e. safe routes for walking and bicycling or electric car network. Also, the construction materials and multi-perspective energy solutions based on renewables and bio-energy will form the basis for green and clean business opportunities in the construction sector and in community planning, too. (Tuohimaa et. al. 2011; Kettunen et. al. 2015)

RESEARCH DESIGN The framework for this study consists of futures research methodology (Masini 1993) including mega- and mini-trend analysis (Vanston & Vanston 2011) and scenario approach (Meristö et.al. 2012) concerning especially the future of residential areas and their opportunities for cleantech business. Another part of the framework includes the skills and competences (Meristö et. al. 2015) required to exploit these opportunities not only at company level but at national level, too. The paper focuses on cleantech business opportunities which the future will provide for the companies as well as on the skills and capabilities required to exploit those opportunities. The research questions are as follows: 1. 2. 3.

The data collected for this work comprises mostly on replies from a websurvey run in May-September 2016 by us in the project ELLI financed by European Regional Development Fund ERDF. The survey has been sent to the actors in the field of sustainable business and energy-efficient solutions in South Finland in different regions, namely Lahti, Riihimäki, Hämeenlinna and Lohja region with their surroundings.

Proceedings of the METNET Seminar 2016 in Castellon

RESULTS The survey was sent to the actors from the cleantech cluster in many roles: companies from different industries, government organizations at local, regional and national level, research institutes and educational organizations as well as NGOs and experts and active citizens from this field. Altogether, 50 replies were received until the 20 September 2016. From the respondents about one third came from different industries (36%) and almost another third from research and education (30%), 14% came from decision making side, 10 % from active citizens and 8% were experts. Some of the respondents represented themselves in several different roles. None of the replies were from EU level nor from NGOs in the field of environment, inhabitants´ or lobbying interests. Two thirds from the replies were male, one third female. Most answers came from the Lahti region (33%) and from Western Uusimaa region (29%) and the rest were from Hämeenlinna (13%) and Riihimäki (4%) region or other places in Finland (21%). The survey included six different research areas of questions: 1) Level of cleantech competences in Finland, 2) Special expertise / skills in cleantech in that region, 3) Key actors in energy-efficient residential regions, 4) Taboos, 5) Future Energy Solutions in Residential areas, 6) Key Trends concerning Future Energy Solutions. Also background information from the replier and open feedback were asked. The results in this paper are divided into three parts, by answering the research questions as follows. First, the main drivers for the future concerning energy-efficient and sustainable solutions will vary from today to the future according to the next diagram, where the average values of the survey results are presented (Figure 1). The future will differ from the present concerning almost all the factors asked in the survey: Time frame of the decision–making from short term will move towards long term, the market of the Finnish cleantech industry will focus more on the global instead of the local market, role of subventions concerning energy efficiency will decrease in the future, willingness to have residential areas without cars will increase in the future, energy savings will be easily realized by automation in the future, the focus of energy sources will move from non-renewables to renewables, the climate change issues will worry in the future not only experts but even ordinary citizens everywhere and finally, the role of circular economy will be a remarkable industry instead of a present boom phenomenon and also the skills and competences related to the cleantech in Finland will be broader than today. Only the norms of the international environmental agreements appear to be almost the same in the future. In the same way, the willingness to pay more for green solutions does not change a lot in the future from that which exists today. Lifestyle in Finland will continue both in urban environments and in the countryside as well, although at global level the urbanization is a strong megatrend.

Figure 1: Main drivers for the future and the directions of the change needed.

The future will differ from the present concerning almost all the factors asked in the survey: Time frame of the decisionâ&#x20AC;&#x201C;making from short term will move towards long term, the market of the Finnish cleantech industry will focus more on the global instead of the local market,

Figure 1: Main drivers for the future and the directions of the change needed.

as follows. First, the main drivers for the future concerning energy-efficient and sustainable solutions will vary from today to the future according to the next diagram, where the average values of the survey results are presented (Figure 1).

Proceedings of the METNET Seminar 2016 in Castellon

but even ordinary citizens everywhere and finally, the role of circular economy will be a remarkable industry instead of a present boom phenomenon and also the skills and competences related the cleantech in Finland will be broader than today. Only the norms Proceedings of the METNETto Seminar 2016 in Castellon of the international environmental agreements appear to be almost the same in the future. In the same way, the willingness to pay more for green solutions does not change a lot in the future from that which exists today. Lifestyle in Finland will continue both in urban environments and in the countryside as well, although at global level the urbanization is a strong megatrend. Second result from the study, concerning the skills and competences needed to meet thethe future challenges in this in the future, is as to follows Second result from study, concerning thefield skillstoday and and competences needed meet the (Figure 2). future challenges in this field today and in the future, is as follows (Figure 2).

Figure 2:Skills Skills competences in clean tech industry Figure 2: and and competences in clean tech industry needed in the future. needed in the future. The scale to estimate the level of the skills and competences in the cleantech is from 1 to 4: The scale to estimate the level of the skills and competences in the cleantech 1 = bad, 2= satisfactory, 3= good and 4= excellent. None of the listed skills and competences is from 1Â toÂ 4: 1 = bad, 2= satisfactory, 3= good and 4= excellent. None of reached the level good (number 3) in this survey. Energy technology skills, water technology the listed skills and competences reached the level good (number 3) in this skills, waste technology skills and digital skills and competences as well as competences survey. Energy technology skills, water technology skills, waste technology concerning natural resources and materials were estimated approximately to the level 2.5 skills and digital skills and competences as well as competences concerning (almost good). The weakest evaluation points went to the RDI skills, network skills, customer natural resources and materials were estimated approximately to the level 2.5 understanding and business skills, all getting values under the satisfactory level 2. (almost good). The weakest evaluation points went to the RDI skills, network Legislation skills reached barely satisfactory level. The estimations concerning the future skills, customer understanding and business skills, all getting values under statement did increase the evaluation points throughout the whole list of competences. The the satisfactory level 2. Legislation skills reached barely satisfactory level. highest ranking, level 3.5 belongs to the waste technology skills and competences and close The estimations concerning the future statement did increase the evaluation points throughout the whole list of competences. The highest ranking, level 3.5 belongs to the waste technology skills and competences and close to that were ranked also regarding energy technology skills and water technology skills. In the future, none of the skills were evaluated under the level 2 (satisfactory), but under the level 3 (good) in the future still are the competences having the lowest score today, namely skills concerning e.g. customer understanding, business RDI , networking and legislative issues. Outside the given list one of the replies mentioned the branding skills, still being very low concerning the Finnish cleantech industry. Third, the barriers and enablers when developing cleantech business for the future were found among the taboos recognized in this survey, but also from

the future, none of the skills were evaluated under the level 2 (satisfactory), but under the level 3 (good) in the future still are the competences having the lowest score today, namely skills concerning e.g. customer understanding, business RDIof, the networking and legislative Proceedings METNET Seminar 2016 in Castellon issues. Outside the given list one of the replies mentioned the branding skills, still being very low concerning the Finnish cleantech industry. Third, the barriers and enablers when developing cleantech business for the future were found among the taboos recognized in this survey, but also from the key actors estimated the key estimated in this field. The most powerful actors are decisionin this field. Theactors most powerful actors are decision-makers at EU and national level, but also at EU and national level, also regional well as result regionalmakers decision-makers as well as thebut companies fromdecision-makers this field (Figureas3). This from is this 3).and Thisjust result means the focus means the thatcompanies still the focus onfield the (Figure enablers after that,that thestill business will have is on the enablers and just after that, the business will have opportunities to opportunities to develop. develop. Big

Quite big

Small

Very small

(Value: 4)

(Value: 3)

(Value: 2)

(Value: 1)

Cannot answer (Value: 0)

1. Companies (avg: 3,00) 2. Decision makers (municipal) (avg: 3,08) 3. Decision makers (national) (avg: 3,33) 4. Decision makers (EU) (avg: 3,33) 5. Education & research (avg: 2,28) 6. Environmental organisations (avg: 2,19) 7. Residentsâ&#x20AC;&#x2122; associations (avg: 2,37) 8. Lobbying organisations (avg: 2,22) 9. Experts (avg: 2,66) 10. Active citizens (avg: 2,52)

Figure 3: Key thepower power concerning the future regions. of energy-efficient regions. Figure 3: actors Key actorsin in the gamegame concerning the future of energy-efficient The barriers and enablers related to the taboos were also recognized in this The barriers and enablers related to the taboos weretoalso thismarket survey. The survey. The subvention driven policy seems be arecognized barrier forinthe subvention driven policy seems to be a barrier for the market driven growth, but also on the driven growth, but also on the industry side there is a lack of holistic view industrytoside there is a lack of holistic view to the branch and holistic solutions are still rare; the branch and holistic solutions are still rare; instead of the systematic, instead multidisciplinary of the systematic, co-operation multidisciplinary co-operation in thenetworks international networks in the international many of the many of the companies and entrepreneurs entrepreneurs inin this this field fieldwork workalone, alone,farfarfrom fromthe themarket market and companies and customers. Also, the long-term life-cycle analysis in the eyes of the sustainability missing. and customers. Also, the long-term life-cycle analysis in the eyes of isthe Perhapssustainability active citizens together with research actors and other experts could raise is missing. Perhaps active citizens together with research actors these issues more in the table the future, whenissues planning and residential and other expertsincould raise these more inbuilding the tablenew in the future, areas. Environment, health and safety (EHS) are related to each other and customers with high when planning and building new residential areas. Environment, health and awareness of the risks are in the future probably willing to pay to reduce these risks, safety (EHS) are related to each other and customers with high awareness of too. At the risks are in the future probably willing to pay to reduce these risks, too. At the moment, active citizens in this field still can get a sign of a green person without a touch of the reality, was one of the barriers mentioned in the survey replies.

CONCLUSIONS According to the survey results, no one estimated the significance of nonrenewable energy sources very high in the future and almost all of the replies ranked it low or even less significant, when developing residential areas (Figure 4). On the other hand, solar power and geothermal energy were evaluated to

the moment, active citizens in this field still can get a sign of a green person without a touch of the reality, was one of the barriers mentioned in the survey replies.

Proceedings of the METNET Seminar 2016 in Castellon CONCLUSIONS

According to the survey results, no one estimated the significance of non-renewable energy sources very high in the future and almost all of the replies ranked it low or even less significant, when developing residential areas (Figure 4). On the other hand, solar power and geothermal energy were evaluated to be very significant or at least with a significant significant at least with a significant score in the received energy palette score inbe thevery energy palette or in the future. Bio-energy based on waste a few higher in the future. Bio-energy based on waste received a few higher than these points than bio-energy based on the wood. The figure as follows will points summarize bio-energy based on the wood. The figure as follows will summarize these estimates concerning different energy sources and their importance in the future after next estimates concerning different energy sources and their importance in the 20 years. future after next 20 years.

Very significant (Value: 4)

Significant (Value: 3)

Little

Not

Cannot

significant

answer

(Value: 2)

(Value: 1)

(Value: 0)

1. Solar energy (avg: 3,00) 2. Geothermal heat (avg: 3,29) 3. Wind energy (avg: 2,30) 4. Bioenergy (wood) (avg: 2,56) 5. Bioenergy (waste) (avg: 2,96) 6. Non-renewal energy (avg: 1,85) 7. Nuclear energy (avg: 2,46)

Figure 4: The significance of the alternative energy solutions of residential areas in the Figure 4: The significance of the alternative energy solutions of residential areas in the future. future. The view holistic viewlong-run, in the long-run, natural resources and economic The holistic in the includingincluding natural resources and economic calculations at calculations at global level will open the eyes for new opportunities in the global level will open the eyes for new opportunities in the cleantech sector. Its impact on cleantech sector. Its impact on the globe and its state of the art will be the globe and its state of the art will be dramatic, if developing business with an attitude of dramatic, if developing business with an attitude of saving the world. Also, the saving the world. Also, the skills and competences have to be updated, and that will require competences have to bepolicy updated, and Also, that will require changes in have changesskills in theand education and government as well. the networking activities the education and government policy as well. Also, the networking activities to be developed to the everyday life level, without handicap in attitudes in multicultural world. have to be developed to the everyday life level, without handicap in attitudes in multicultural world. REFERENCES

Erber, G. & Hagemann, H. (2005) The New Economy in a Growth Crisis. In Huebner. K. REFERENCES (ed.) The New Economy in Transatlantic Perspective. Taylor and Francis Inc., New York, USA. Erber, G. & Hagemann, H. (2005) The New Economy in a Growth Crisis. In Huebner. K. (ed.) The New Economy in Transatlantic Perspective. Taylor and Francis Inc., New York, USA. Kettunen, J. (1998) Can Finland Still Live from the Forest? Finnish Forest Cluster to 2020, pp 302-339, in a Book Green Kingdom, eds. Reunala, A, Tikkanen, I, Åsvik, E. Otava Ltd.. Kettunen, J., Meristö, T. & Laitinen, J. (2015) . Bioenergy Scenarios 2033 – a Platform for the Innovation. In Proceedings of The XXVI ISPIM Innovation Conference held in Budapest (Hungary) on 14 June to 17 June 2015. Editors:

Proceedings of the METNET Seminar 2016 in Castellon

Huizingh Eelko, Torkkeli Marko, Conn Steffen, and Bitran Iain. LUT Scientific and Expertise Publications 40. Tutkimusraportit – Research Reports. Masini, E. (1993) Why Futures Studies? Grey Seal, London. Meristö, T. (2013) Yritysten planetaarinen vastuu ja sen kehittyminen yrityskentässä. Futura 2/2013, 48-51. (Planetary Corporate Responsibility and its Development in the Field.) Meristö, T. & Laitinen, J. (2015) Future Skill Profiles – a Key to Renew Marine Industry. In Proceedings of The XXVI ISPIM Innovation Conference held in Budapest (Hungary) on 14 June to 17 June 2015. Editors: Huizingh Eelko, Torkkeli Marko, Conn Steffen, and Bitran Iain. LUT Scientific and Expertise Publications 40. Tutkimusraportit – Research Reports. Meristö, T. & Laitinen, J. (2014) Kestävän liiketoiminnan mahdollisuudet Länsi-Uudellamaalla – nykytila ja tulevaisuuden näkymiä. (Opportunities for Sustainable Business in Western Uusimaa County – a Present Situation and Alternative Future Views). Novago, Lohja (in Finnish). Meristö, T., Laitinen, J. & Tuohimaa, H. (2012) Scenario Filter Model as an Innovation Catalyst. The Proceedings of the 5th ISPIM Innovation Symposium, Seoul, Korea - 9-12 December 2012. Pearce, D & Turner, R. K. (1989). Economics of Natural Resources and the Environment. Johns Hopkins University Press. Tuohimaa, H., Meristö, T., Kettunen, J. & Laitinen, J. (2011) Sustainable Community Scenarios – A Challenge for Innovation Management in the Public Sector. Proceedings of the XXII ISPIM Conference, 12-15 June 2011, Hamburg, Germany. Vanston, J. & Vanston, C. (2011) Minitrends: How Innovators & Entrepreneurs Discover & Profit From Business & Technology Trends. TFI Texas.

Proceedings of the METNET Seminar 2016 in Castellon

MOMENT-ROTATION BEHAVIOUR OF WELDED TUBULAR HIGH STRENGTH STEEL JOINTS Markku Heinisuo, Marsel Garifullin, Kristo Mela, Teemu Tiainen, Timo Jokinen Tampere University of Technology, Korkeakoulunkatu 10, FI-33720, Tampere, Finland The field of applications of tubular structures covers a large range of structures, such as bridges, lattice masts, buildings and large span structures. Implementing high strength steel (HSS) to tubular joints allows reducing the consumption of metal (Johansson 2003), however it implies increased material and manufacturing costs, which must be considered when economical designs are pursued. In cost and emission optimization of HSS structures an essential task is the evaluation of the feasibility of the structure. In cost and emission analysis one important issue is welding duration and energy consumption using different welding technologies with different base materials, weld sizes and welding positions. Any new design methods have to be validated with experiments. The overall research of tubular joints was mainly adapted by Wardenier (1982) who first proposed the design formulas based on the classical yield line theory. The experiments for welded tubular moment joints have been presented in (Brockenbrough 1972), (Korol et al. 1977), (Kanatani et al. 1980), (Yura et al. 1981), (Ono et al. 1991), (Ishida et al. 1993), (Davies et al. 1996), (Owen et al. 2001), (Christitsas et al. 2007), (Fadden et al. 2015), (Chen et al. 2015), and (Tong et al. 2015). Tests for stainless steel SHS welded joints are presented in (Feng and Young 2006). However, experiments of welded HSS tubular moment joints cannot be found in published literature. The design of joints mean checks of resistance, stiffness and ductility. The moment resistance of a joint is the most important quantity in design. The rotational stiffness is needed when the global structural analysis model based on beam elements is made. The rotational stiffness has a great effect when cost optimal solutions are searched both in sway frames (Anderson et al. 1997), (Grierson and Xu 1993), (Simoes 1996), (Weynand et al. 1998), (Haapio et al. 2011) and in non-sway frames, as well (Bzdawka 2012). The rotational stiffness has an influence also on the buckling length of the members (Boel 2010), (Snijder et al. 2011) and (Heinisuo and Haakana 2015). It has been found theoretically (Heinisuo et al. 2016), that the weld size has a large effect on the initial rotational stiffness of welded tubular moment joints. Very few investigations have been conducted to explore the moment-rotation behaviour of welded joints made of HSS. This research involves 17 tests performed to observe the moment-rotation relationship and ductility of the joints. The experimental data are compared with the results of comprehensive finite element modelling (FEM) and manual calculations according to the Eurocode.

Proceedings of the METNET Seminar 2016 in Castellon

MAIN RESULT The method for calculation of the moment resistance proposed in (EN 19931-8, 2005) allows getting rather safe results, the average Mj,Rd / Mu,exp ratio is 0.61 (Fig. 1). Applying the correlation coefficient related to HSS makes the (M*j,Rd /M 0.50). Thisofshows there u,exp ratio shows thatresults there very is noconservative need to reduce the moment resistance HSSthat joints as prescribed in is no need to reduce the moment resistance of HSS joints as prescribed in Eurocodes.. Eurocodes..

The full paper will present the results including the following: • Moment-rotation relationship of the joints; • The effect of size on the initial rotational stiffness; The full paper will present theweld results including the following: • The ductility of the joints.

  

Moment-rotation relationship of the joints; of the tests study the effectstiffness; of the weld size on moment of TheAnother effect ofaim weld size on was the to initial rotational resistance, rotational stiffness and ductility of the joint. The ductility of the joints.

Another aim of the tests was to study the effect of the weld size on moment of resistance, REFERENCES rotational stiffness and ductility of the joint. Anderson, D., Hines, E.L., Arthur, S.J. and Eiap, E.L. 1997. Application of artificial neural networks to the prediction of minor axis steel connections. ComREFERENCES puters & Structures 63(4): 685–692.

Anderson, D., Hines, E.L., Arthur, S.J. and Eiap, E.L. 1997. Application of artificial neural Boel,toH. 2010. BucklingofLength of Hollow Section Members in Lattice networks the prediction minorFactors axis steel connections. Computers & Structures 63(4): Girders. Ms Sci thesis, Eindhoven: Eindhoven University of Technology. 685–692. Boel, H. 2010. Buckling Length Factors of Hollow Section Members in Lattice Girders. Ms Sci R.L. 1972. Strength of Square-Tube Connections under Comthesis,Brockenbrough, Eindhoven: Eindhoven University of Technology. bined Loads. JournalStrength of the Structural Division 98(12): 2753–2768. Brockenbrough, R.L. 1972. of Square-Tube Connections under Combined Loads. Journal of the Structural Division 98(12): 2753–2768. Bzdawka, K. 2012. Optimization Office Building Frame with Semi-Rigid Bzdawka, K. 2012. Optimization of Office of Building Frame with Semi-Rigid Joints in Normal and Joints in Normal and Fire Conditions,. PhD Thesis, Tampere University of 1038. Fire Conditions,. PhD Thesis, Tampere University of Technology, Publication Technology, Publication 1038. Chen, Y., Feng, R. and Wang, J. 2015. Behaviour of bird-beak square hollow section X-joints under in-plane bending. Thin-Walled Structures 86: 94–107. Chen, Y., Feng, R. and Wang, J. 2015. Behaviour of bird-beak square hollow Christitsas, A.D., Pachoumis, D.T., Kalfas, C.N. and Galoussis, E.G. 2007. FEM analysis of section X-joints under in-plane bending. Thin-Walled Structures 86: 94–107. conventional and square bird-beak SHS joint subject to in-plane bending moment — experimental study. Journal of Constructional Steel Research 63(10): 1361–1372. Christitsas, A.D., Pachoumis, D.T., Kalfas, C.N. and Galoussis, E.G. 2007. FEM Davies, G., Owen, J.S. and Kelly, R. 1996. Bird beak T-joints in square hollow sections: a finite element investigation. In Proceedings of the International Offshore and Polar Engineering Conference. Int Soc of Offshore and Polar Engineerns (ISOPE), pp. 22–27. Fadden, M., Wei, D. and McCormick, J. 2015. Cyclic Testing of Welded HSS-to-HSS Moment Connections for Seismic Applications. Journal of Structural Engineering 141(2):

Proceedings of the METNET Seminar 2016 in Castellon

analysis of conventional and square bird-beak SHS joint subject to in-plane bending moment — experimental study. Journal of Constructional Steel Research 63(10): 1361–1372. Davies, G., Owen, J.S. and Kelly, R. 1996. Bird beak T-joints in square hollow sections: a finite element investigation. In Proceedings of the International Offshore and Polar Engineering Conference. Int Soc of Offshore and Polar Engineerns (ISOPE), pp. 22–27. Fadden, M., Wei, D. and McCormick, J. 2015. Cyclic Testing of Welded HSSto-HSS Moment Connections for Seismic Applications. Journal of Structural Engineering 141(2): 04014109. Feng, R. and Young, B. 2006. Tests of stainless steel RHS X-joints. Welding in the World 50(SPEC. ISS.): 250–257. Grierson, D.E. and Xu, L. 1993. Design optimization of steel frameworks accounting for semi-rigid connections. In G. I. N. Rozvany, ed. Optimization of Large Structural Systems, Volume II. Dordrecht: Kluwer Academic Publisher, pp. 873–881. Haapio, J., Jokinen, T. and Heinisuo, M. 2011. Cost simulations of steel frames with semi-rigid joints using product model techniques. In L. Dunai, M. Ivanyi, K. Jarmai, N. Kovács, & L. G. Vigh, eds. Proceedings of EUROSTEEL 2011. Brussels: ECCS, pp. 2151–2156. Heinisuo, M., Garifullin, M., Jokinen, T., Tiainen, T. and Mela, K. 2016. Surrogate modeling for rotational stiffness of welded tubular Y-joints. Proceedings of The Eighth International Workshop on Connection in Steel Structures (Connections VIII). Heinisuo, M. and Haakana, Ä. 2015. Buckling of members of welded tubular truss. Nordic Steel Construction Conference 2015. Ishida, K., Ono, T. and Iwata, M. 1993. Ultimate strength formula for joints of new truss system using rectangular hollow sections. In Proceedings of 5th International Symposium on Tubular Structures. pp. 511–518. Johansson, B. 2003. Use and design of high strength steel structures. Revue de Metallurgie. Cahiers D’Informations Techniques 100(11): 1115–1123+vi–ix. Kanatani, H., Fujiwara, K., Tabuchi, M. and Kamba, T. 1980. Bending tests on T-joints of RHS chord and RHS or H-shape branch. CIDECT Program 5AF (March 1981). Korol, R.M., El-Zanaty, M. and Brady, F.J. 1977. Unequal width connections of square hollow sections in Vierendeel trusses. Canadian Journal of Civil Engineering 4(2): 190–201.

Proceedings of the METNET Seminar 2016 in Castellon

Ono, T., Iwata, M. and Ishida, K. 1991. An experimental study on joints of new truss system using rectangular hollow sections. In Tubular Structures. 4 th International Symposium. pp. 344–353. Owen, J.S., Davies, G. and Kelly, R.. 2001. The influence of member orientation on the resistance of cross joints in square RHS construction. Journal of Constructional Steel Research 57(3): 253–278. Simoes, L.M.C. 1996. Optimization of frames with semi-rigid connections. Computers & structures 60(4): 531–539. Snijder, H.H., Boel, H.D., Hoenderkamp, J.C.D. and Spoorenberg, R.C. 2011. Buckling length factors for welded lattice girders with hollow section braces and chords. Proceedings of Eurosteel 2011: 1881–1886. Tong, L., Xu, G., Liu, Y., Yan, D. and Zhao, X.-L. 2015. Finite element analysis and formulae for stress concentration factors of diamond bird-beak SHS Tjoints. Thin-Walled Structures 86(3): 108–120. Wardenier, J. 1982. Hollow Section Joints. Doctoral dissertation, Delft: Delft University of Technology. 544 p. Weynand, K., Jaspart, J.-P. and Steenhuis, M. 1998. Economy studies of steel building frames with semi-rigid joints. Journal of Constructional Steel Research 46(1-3): 85. Yura, J.A., Edwards, I.F. and Zettlemoyer, N. 1981. Ultimate Capacity of Circular Tubular Joints. Journal of the Structural Division 107(10): 1965–1984.

Proceedings of the METNET Seminar 2016 in Castellon

STEEL JOINTS WITH BEAMS OF UNEQUAL DEPTH USING CRUCIFORM ELEMENTS FOR FRAME ANALYSIS Eduardo Bayo University of Navarra The behaviour of joints exerts an important influence on the overall performance of steel and composite structures. Modern codes, including Eurocodes 3 (EC3) and Eurocode 4 (EC4), have included advances in research for modelling of joints to be used in common practice. The method adopted in EC3 and EC4 to characterize the connections is the so-called component method. The development of the component method for the characterization of steel joints (Eurocode 3) has led to the use of mechanical models, composed of springs and rigid bars for frame analyses. However, this adds complexity, numerical overhead â&#x20AC;&#x201D;consequence of the large number of degrees of freedom â&#x20AC;&#x201D; and possible round off errors (due to the rigid bars) in the solution process. One of the most important components is the column web panel in shear, and its behaviour has been investigated thoroughly for the case of single rectangular shear panels arising when the beams have equal depths. However, the case of trapezoidal and double column panels, formed by beams of different depths at each side of the column, has not been researched as much. In order to isolate the shear behaviour from other components, diagonally and horizontally stiffened beam to column connections with commercial sections are experimentally tested. Also, finite element modelling and analyses are carried out to compare results. Mechanical models are proposed and a parametric study is performed by means of calibrated finite elements models to characterize the different components, and define the analytical expressions for the stiffness and resistance of the proposed mechanical model. With the aim of avoiding the problems mentioned above, generalized cruciform finite elements with 4Â nodes and 12 degrees of freedom (for the 2D case) are proposed. These cruciform elements are also suitable for semi-rigid connections and can be used for the global analysis of semi-rigid steel frames. The elements are capable of modelling rectangular, trapezoidal and double rectangular web panels, arising in internal joints with beams of different depth. The characteristics of these elements (stiffness and resistance) are generated from the mechanical models by means of an efficient displacement based modal approach. All the forces and moments coming from the adjacent beams and columns concur at the joint, therefore, the complete force field in the panel

Proceedings of the METNET Seminar 2016 in Castellon

zones is known with no need for a transformation parameter used in EC3-Part 1.8. In addition, since the real dimensions of the joint are being considered, the eccentric moments are automatically taken into account. Also, the interaction between the deep and the shallow sides of column double panels are taken into account. Numerical simulations are carried out that show the validity and efficiency of the proposed approach. The full paper will present the behaviour of the proposed mechanical model. It is shown that this general cruciform element is suitable for simple, rigid and semi-rigid connections and can be used for global analysis steel frames.

Proceedings of the METNET Seminar 2016 in Castellon

CRUCIFORM FINITE ELEMENTS FOR THE MODELLING OF STEEL JOINTS IN FRAME ANALYSIS Eduardo Bayo University of Navarra The development of the component method for the characterization of steel joints (Eurocode 3) has led to the use of mechanical models, composed of springs and rigid bars for frame analyses. However, this adds complexity, numerical overhead — consequence of the large number of degrees of freedom — and possible round off errors (due to the rigid bars) in the solution process. With the aim of avoiding these problems generalized cruciform finite elements with 4 nodes and 12 degrees of freedom (for the 2D case) are proposed. These cruciform elements are also suitable for semi-rigid connections and can be used for the global analysis of semi-rigid steel frames. The elements are capable of modelling rectangular, trapezoidal and double rectangular web panels, arising in internal joints with beams of different depth. The characteristics of these elements (stiffness and resistance) are generated from the mechanical models by means of an efficient displacement based modal approach. All the forces and moments coming from the adjacent beams and columns concur at the joint, therefore, the complete force field in the panel zones is known with no need for a transformation parameter used in EC3-Part 1.8. In addition, since the real dimensions of the joint are being considered, the eccentric moments are automatically taken into account. Also, the interaction between the deep and the shallow sides of column double panels are taken into account. Numerical simulations are carried out that show the validity and efficiency of the proposed approach.

Proceedings of the METNET Seminar 2016 in Castellon

RECENT RESEARCH AND DESIGN DEVELOPMENTS IN STEEL AND COMPOSITE STRUCTURES AT UPC Mirambell E., Real E. and ChacĂłn R. Department of Civil and Environmental Engineering Universitat PolitĂ¨cnica de Catalunya, BarcelonaTech Recent research and design developments of the Steel and Composite Division group of the Department of Civil and Environmental Engineering (UPC, Barcelona Tech) encompasses several areas of expertise. Particular research focus includes i) Structural behavior of stainless steel, ii) Stability and design developments of plated structures and iii) Steel-concrete composite members. The research in steel and composite structures covers experimental, numerical and theoretical studies on the behavior and analysis for safe and sustainable design. Stainless steel design has been studied for two decades by the group within the frame of several European and National research projects. The projects include the Theoretical and experimental study on instability of stainless steel structures (DGES PB95-0772), Structural applications of plated stainless steel girders (MAT 2000-1000) and Structural analysis of ferritic stainless steel members such as tubes, slabs and hat-sections (BIA 2012-036373 and SAFSS 2010 RFS-PR-09032). On the other hand, the Steel and Composite Division group has been involved in several valorization projects funded by RFCS-European Commission: Development of the use of stainless steel in construction (ECSC Contract 1999-7215-PP/056), Structural design of coldworked austenitic stainless steel (RFS2-CT-2005-00036), and Promotion of new Eurocode rules for structural stainless steels (RFCS-2015, RFCS RFSAM-709600). Research has resulted in a considerable amount of scientific production (journal papers, conference proceedings, design manuals and participation in European committees for standardization and design such as EN1993-1-4 Working Group). Plated structures have been actively studied for more than two decades as part of separate research activities. National projects related to instability problems in plates (DGICYT PB90-0604), to hybrid steel plate girders (BIA 2004-04673) and to steel tapered plate girders (BIA 2008-01897), among others, have provided a robust platform to the group for the development of theoretical, experimental and numerical expertise in the field. Research has also resulted in a considerable amount of scientific output (journal papers, conference proceedings, design manuals and participation in European committees for standardization and design such as TC8 and TWG 8.3 of ECCS, and EN 1993-1-1 Working Group WG 1 and EN 1993-1-5 Working Group WG 1-5).

Proceedings of the METNET Seminar 2016 in Castellon

Composite structures have also been studied as part of research projects related to concrete-filled steel tubes (Safety and serviceability of integral railway bridges in front of accidental actions: Research for design criteria and construction, 2007-2011, Project 51/07), to stainless steel-concrete composite slabs (SAFSS project) and to a vast array of experimental applications including push-out and pull-out tests, composite action with shear connectors and thermally-induced deformations within composite elements. Research has resulted in journal papers, conference proceedings and experimental expertise. In summary, the research expertise includes: instability-related problems (buckling), new metallic materials (stainless steel), numerical simulations of construction processes, time-dependent effects in composite structures, plate girders and the systematic usage of FEM in steel structures design. Researchers work in close collaboration with international institutions, universities and research centers and they participate actively in the development of the Spanish Steel Code (EAE-12) as in the ongoing development of European Standards.

Proceedings of the METNET Seminar 2016 in Castellon

CHARACTERIZATION OF RHS-IPE BEAM-COLUMN JOINTS M. Lozano, M.A. Serrano, C. Lรณpez-Colina, F. Lรณpez-Gayarre Department of Construction and Manufacturing Engineering. University of Oviedo One of the main priorities of research when referring to steel structures in the recent years has been the characterization of structural joints. This is due to the great importance of a good knowledge of their resistance and the fact that they behave differently depending on the stiffness of the joint. It is very important to implement the correct stiffness of the joint to carry out a reliable structural analysis. Nevertheless, sometimes, it is not so easy to know the actual stiffness of a joint. The component method is the most important and widely recognized theoretical approach for the evaluation of the stiffness and resistance properties of a wide range of joint configurations. It has been adopted for steel structures by Eurocode-3 as its main tool for the description of the behaviour of joints between H or I sections. The component method is a powerful tool to determine the performance of the connections in terms of strength and stiffness. It is the framework for many researches based on connections. The difficulty in studying altogether the elements and parameters that condition the connections is the reason for focusing on a part or component. The structural hollow sections are many times a very good proposal for columns in building structures. However, despite the clear advantages of tubular profiles, joints that involve structural hollow sections (RHS or CHS) have been traditionally excluded from the application of the component method. It is obvious that the possibility of applying the component method to this kind of joints would be a significant advance allowing as well the consideration of semi rigid joints. A global saving would be reached adding to the possibility of using well characterized partially rigid joints, thus saving material that may involve the use of tubular columns. The present project we are working on, intends to apply the component method to joints involving square (SHS) or rectangular hollow sections (RHS) as tubular columns, due to their multiple economic and material saving advantages. For the beam members a pair of IPE sections will be used due to their excellent behaviour when bending is just in the vertical plane which is a common situation in buildings. The research is presented in a comprehensive way, initiating the process with the characterization of the components that are identified in a RHS column as a part of a symmetrically loaded doublesided beam-to-column joint.

Proceedings of the METNET Seminar 2016 in Castellon

DIAPHRAGM STABILIZATION OF STEEL BUILDINGS IN FIRE Olli Ilveskoski HAMK University of Applied Sciences The objective is to study the literature available on the diaphragm stabilization of steel buildings at elevated temperatures. Bracing used in structural systems generally serve two primary functions. They resist secondary loads on structures (e.g. wind bracing) and increase the strength of individual members by resisting deformation in the weakest direction. The use of self-supporting sandwich panels as stabilizing elements in normal temperature for single steel members such as beams or columns has been recently researched and Recommendations on the Stabilization of Steel Structures by Sandwich Panels published. The European Recommendations give instructions about diaphragm stabilization and connection forces in normal temperature. This short survey is made to evaluate the present references dealing the diaphragm stabilization of steel buildings at elevated temperature.

Proceedings of the METNET Seminar 2016 in Castellon

Virdi & Tenhunen (Editors)

METNET seminars and workshops deal with technical aspects of metal construction as well as issues of concern to industry on management, planning and sustainability of projects. This book covers papers presented and submitted for the tenth annual METNET seminar in October 2016 held at Universitat Jaume I (UJI), Castellon, Spain. The seminar continued the METNET tradition of presenting scientific and development papers of high calibre.

Proceedings of the METNET Seminar 2016 in Castellon

The METNET network aims to enhance the development potential and the research and innovation capacities of regional innovation systems and wider communities of Europe. METNET provides an international innovation environment for its members and the possibility to expand their regional innovation networks internationally.

Proceedings of the METNET Seminar 2016 in Castellon

METNET Annual Seminar in Castellon, Spain, on 11 – 12 October 2016

ELECTRONIC ISBN 978-951-784-785-8 (PDF) ISSN 1795-424X HAMKin e-julkaisuja 6/2016

Kuldeep Virdi and Lauri Tenhunen (Editors) HAMK

PRINTED ISBN 978-951-784-784-1 ISSN 1795-4231 HAMKin julkaisuja 4/2016