International Conference on Applications of Statistics and Probability in Civil Engineering (ICASP) (12th : 2015)

Aspects of code based design of timber structures Kohler, Jochen; Fink, Gerhard Jul 31, 2015

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
53032-Paper_349_Kohler.pdf [ 309.12kB ]
Metadata
JSON: 53032-1.0076305.json
JSON-LD: 53032-1.0076305-ld.json
RDF/XML (Pretty): 53032-1.0076305-rdf.xml
RDF/JSON: 53032-1.0076305-rdf.json
Turtle: 53032-1.0076305-turtle.txt
N-Triples: 53032-1.0076305-rdf-ntriples.txt
Original Record: 53032-1.0076305-source.json
Full Text
53032-1.0076305-fulltext.txt
Citation
53032-1.0076305.ris

Full Text

12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015Aspects of Code Based Design of Timber StructuresJochen KohlerAssociate Professor, Dept. of Structural Engineering, Norwegian University of Scienceand Technology (NTNU), Trondheim, NorwayGerhard FinkResearch Scientist, Empa, Swiss Federal Laboratories for Materail Science andTechnology, Dübendorf, SwitzerlandABSTRACT: The European timber design standard is under development and a new version will beissued at the end of this decade. In this paper the present design standard is critically assessed in regardto its ability to identify design solutions with a consistent level of reliability. The main issues to enhancethe current standards are identified and discussed.1. INTRODUCTIONSustainable development is the important require-ment and goal for modern society and the interna-tional research community is in demand to find so-lutions that provide the foundation for this aim. Therole of structural engineering research is therebyof significant importance. The development ofmethodologies and principles that allows for the op-timal allocation of resources into the structural per-formance and their implementation into the dailyengineering practice constitute the major challengefor ongoing and future research in the field of struc-tural engineering.The broad implementation of newly developedprinciples requires their proper transition into rulesand regulations that constitute the basis for the dailywork of practicing engineers. Thus, rules and reg-ulations as structural design codes constitute themayor interface between structural engineering re-search and practical application and it is of ut-most importance that structural design codes are upto date with the best scientific information avail-able and, at the same time, are simple enough forstraight forward application.This challenge outlined above is general for theentire structural engineering research and profes-sional community. Here, timber and timber basedmaterials might be attributed with a special statussince timber as a natural grown material plays animportant role in the safe, cost efficient and sus-tainable development of our future build environ-ment because of its beneficial properties. Timberis an efficient building material, not least in regardto its mechanical properties but also because it isa highly sustainable material considering all phasesof the life cycle of timber structures: production,use and decommissioning.Timber is a widely available natural resource;e.g. with proper management, there is a potentialfor a continuous and sustainable supply of raw tim-ber material in the future. Because of the low en-ergy use and the low level of pollution associatedwith the manufacturing of timber structures the en-vironmental impact is much smaller than for struc-tures built in other materials.However, besides the beneficial properties oftimber the confident use of timber as a load-bearingmaterial is particularly challenging compared toother common structural materials as steel and con-crete. One of the main reasons for this is that tim-ber is a highly complex material; the proper use instructures actually requires a significant amount ofexpertise in structural detailing.Another main reason is that any prediction of thestructural performance of timber is associated withlarge uncertainties. Timber is by nature a very in-homogeneous material. The material properties de-pend on the specific wood species, the geograph-112th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015ical location and furthermore on the local grow-ing conditions over the entire lifetime of the tree.Timber is an orthotropic material, i.e. it consistsof “high strength” fibres/grains which are predom-inantly orientated along the longitudinal axis of atimber log/ tree and packed together within a “lowstrength” matrix. After a log is sawn into piecesof structural timber, irregularities, such as graindirection or knots, become, in addition to the or-thotropic characteristics mentioned above, highlydecisive for the load-bearing capacity of a timberstructural element. Consequently, the propertiesof solid timber cannot be designed or produced bymeans of some recipe but may be ensured to fulfilgiven requirements only by quality control proce-dures implemented during the production processfor sawn timber. Timber material for structural pur-pose is generally associated to a certain grade orstrength class. However, there are various differ-ent ways how quality control is implemented in theproduction process and the properties of timber ofa certain strength class are highly sensitive to thequality control scheme applied to the timber.Timber is a viscoelastic and hygroscopic mate-rial. When using timber as a load-bearing ele-ment in a structure it is of high interest how theload-bearing performance is developing over time,i.e. how the building environment with its variableloads, temperature and moisture is influencing thetimber structural element.The high importance of structural timber andtimber products for the sustainable development ofour build infrastructure together with the fact thatmany features of the structural behaviour of timberare not known with accurate precision underlinesthe urgent need for extensive and coordinated re-search in this field. Furthermore it is necessary thatcurrent and future knowledge about timber and tim-ber based materials load- bearing behaviour is rep-resented in the current design standards in a sensi-ble way.In Europe the design of structures is regulated bythe Eurocodes, a suite of consistent standards forstructural design covering all relevant load scenar-ios and building materials. They were developedunder the supervision of the European Committeeof Standardization (CEN) and regulate to a largeextent the performance criteria of the build envi-ronment being reliability, serviceability and safetyof structures. The Eurocodes had been introducedin the 1980s and are by now compulsory for struc-tural engineering design in most European coun-tries. Until 2020 a revision and update of the Eu-rocodes is planned. Thus, this constitutes an excel-lent opportunity to critically reflect the design pro-cedures prescribed in the Eurocode 5 – “Design ofTimber Structures” in the light of recent scientificdevelopments.2. BASIC PRINCIPLES OF RELIABILITYBASED CODE CALIBRATIONModern design codes, such as the Eurocodes(2002), are based on the so-called load and resis-tance factor design (LRFD) format. Next, the prin-ciple of LRFD is explained for the case of twoloads; one that is constant and one that is variableover time. The LRFD equation is given in Eq. (1).Here Rk, Gk and Qk are the characteristic values ofthe resistance R, the permanent load G, and the timevariable load Q. γm,γG and γQ are the correspond-ing partial safety factors. z is the so-called designvariable, which is defined by the chosen dimensionsof the structural component.zRkγm− γGGk− γQQk = 0 (1)The characteristic values for both load and resis-tance are in general defined as fractile values ofthe corresponding probability distributions. In Eu-rocode 5 (2004) the following characteristic valuesare defined: Rk is the 5% fractile value of a Log-normal distributed resistance, Gk is the 50% frac-tile value (mean value) of the Normal distributedload (constant in time), and Qk is the 98% fractilevalue of the Gumbel distributed yearly maxima ofthe load (variable in time).The corresponding partial safety factors can becalibrated to provide design solutions (z) with anacceptable failure probability Pf (Eq. 2). Here R,G, and Q are resistance and loads represented asrandom variables, z∗ = z(γm,γG,γQ) is the designsolution identified with Eq. (1) as a function of theselected partial safety factors, and X is the model212th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015uncertainty.Pf = P{g(X ,R,G,Q) < 0}with g(X ,R,G,Q) = z∗XR−G−Q = 0(2)Often the structural reliability is expressed withthe so-called reliability index β (Eq. 3). A com-mon value for the target reliability index is β ≈ 4.2which corresponds to a probability of failure Pf ≈10−5 (JCSS, 2001).β =−Φ−1(Pf ) (3)In general, different design situations are rele-vant; i.e. different ratios between G and Q. Thiscan be considered using a modification of Eq. (1)–(2) into Eq. (4)–(5). αi might take values between0 and 1, representing different ratios of G and Q. Rˆ,Gˆ, and Qˆ are normalized to a mean value of 1. Foreach αi one design equations exists, thus altogethern different design equations have to be considered.ziRˆkγm− γGαiGˆk− γQ(1−αi)Qˆk = 0 (4)gi(X , Rˆ, Gˆ, Qˆ) = z∗i XRˆ−αiGˆ− (1−αi)Qˆ = 0 (5)Afterwards, the partial safety factors (γm,γG, andγQ) can be calibrated by solving the optimisationproblem give in Eq. (6).minγ[n∑j=1(βtarget−β j)2](6)The reliability based code calibration is brieflyintroduced to illustrate the influence of uncertain-ties (load and resistance), in respect to codes.Please find more information in (e.g. JCSS, 2001;Faber and Sørensen, 2003).The application of the above sketched frameworkconstitutes the basis for reliability based calibrationof the partial safety factors of a load and resistancefactor design format. And it entirely depends ona realistic representation of loads, resistances andmodel accuracy by the random variables R, Q, G,and X .12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12-15, 2015  4line with squares. It can be observed that the reliability indices of the design solutions according to the Eurocode tend to be too low compared to the target reliability index, especially for small values of α. The line with the diamonds is obtained when all partial safety factors are subject to optimization, the resulting set of partial safety factors is 1.29,m  1.3,G   1.57Q  . However, it is the philosophy of the Eurocodes that the partial safety factors for the loads are material independent; therefore it is reasonable to fix G  and Q  and perform the optimization only subject to m . The line with the circles in Figure 1 is representing the corresponding result ( 1.33m  ). An enhancement in fit to the target reliability can be observed for both calibrated solutions.    Figure 1: Reliability Index over different design situations alpha for solid timber in bending. The different lines represent different sets of partial safety factors.  The above example demonstrates the validity of the Eurocode 5 design safety concept for timber load -bearing elements under the assumption that the parameters given in Figure 1 represent the real situation with sufficient accuracy. In the following it will be discussed in which ways the actual load load-bearing behavior derivate from the assumptions in Table 1. It is demonstrated and quantified how the corresponding deviation affects the reliability of design situations and it is discussed how recent research results might integrated in the further developed issue of Eurocode 5.  3. PARTICULARITIES IN TIMBER MATERIAL MODELLING  In this Section specific characteristics of timber and engineered wood products in respect to code based design are discussed.  3.1. Different “material properties” Timber is a rather complex building material. Its properties are highly variable, spatially and in time. In structural engineering, material properties of timber are in general understood as the stress and stiffness related properties of standard test specimen under given (standard) loading and climate conditions and the timber density. Test configurations are prescribed in e.g. ISO 8375 and any statement about stress and stiffness related properties of structural timber is conditional to the corresponding test configuration. In general it is distinguished between the different loading modes and “material properties” are given corresponding to the loading direction relative to the main fiber direction of a beam shaped element (Figure 2).    Figure 2: Different “material properties” dependent on the loading mode.  The “material properties” have different statistical properties and when using the design criterion in Eq. 1 and applying the same partial safety factor m , as it is practiced in the 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.844.14.24.34.4TargetEurocodeOptimized, ,m G Q  mOptimizedFormatted: FirstparagraphFigure 1: Reliability Index over different design situ-ations alpha for solid timber in bending. The differentlines r prese t diffe nt sets of partial safety factors(Kohler and Fink, 2012).2.1. xampleThe design equation for a structural component canbe calibrated according to the procedure describedin above. The chosen variables of Eq. (4) and Eq.(5) are summarized in Table 1. Using this valuesthe situation could represent a solid timber bendingbeam loaded by constant (e.g. self-weight of beamand installations) and variable (e.g. live load).In the presented example, which is explainedin more detail in Kohler and Fink (2012), therange α = [0.1,0.2, ...,0.8] is chosen, to excluderather unrealistic design situations. The calcula-tions was performed with the software CodeCal(JCSS, 2001). In Figure 1 the chosen target reli-ability index of β = 4.2 (red line) is compared withthe design solutions for the structural componentobtained according to the current version of the Eu-rocode (γm = 1.3,γG = 1.5,γQ = 1.5); representedby the line with squares. The reliability indicesof the design solutions according to the Eurocodetend to be too low compared to the target reliabil-ity index, especially for small α . The line with thediamonds is obtained when all partial safety fac-tors are subject to optimization: γm = 1.29,γG =1.30,γQ = 1.57. However, it is the philosophy ofthe Eurocodes that the partial safety factors for the312th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015Table 1: Chosen representation of the model uncertainty X, the bending strength R, the permanent load G and thevariable load Q.X R G QMean value 1 1 1 1Standard deviation 0.1 0.25 0.1 0.4Distribution type Lognormal Lognormal Normal GumbelFractile - 0.05 0.5 0.98Characteristic value - 0.647 1 2.037loads are material independent. Thus, γG and γQ arefixed and the optimization is performed only sub-ject to γm. The line with the circles in Figure 1 isrepresenting the corresponding result (γm = 1.33).The above example demonstrates the validity ofthe Eurocode 5 (2004) design safety concept fortimber load-bearing elements under the assumptionthat the parameters given in Figure 1 represent thereal situation with sufficient accuracy. In the fol-lowing it will be discussed in which ways the ac-tual load load-bearing behavior derivate from theassumptions in Table 1. It is demonstrated andquantified how the corresponding deviation affectsthe reliability of design situations and it is discussedhow recent research results might integrated in thefurther developed issue of Eurocode 5 (2004).3. PARTICULARITIES IN TIMBER MATE-RIAL MODELLING3.1. Different “material properties”Timber is a rather complex building material. Itsproperties are highly variable, spatially and in time.In structural engineering, material properties oftimber are in general understood as the stress andstiffness related properties of standard test spec-imen under given (standard) loading and climateconditions and the timber density. Test configura-tions are prescribed in e.g. ISO 8375 (1985) and anystatement about stress and stiffness related proper-ties of structural timber is conditional to the corre-sponding test configuration. In general it is distin-guished between the different loading modes and“material properties” are given corresponding to theloading direction relative to the main fiber directionof a beam shaped element (Figure 2).The “material properties” have different statisti-cal properties and when using the design criterionintroduced before and applying the same partialsafety factor γm, as it is practiced in the Eurocode,the reliability of the corresponding design solutionsdiffer.12th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12-15, 2015  4line with squares. It can be observed that the reliability indices of the design solutions according to the Eurocode tend to be too low compared to the target reliability index, especially for small values of α. The line with the diamonds is obtained when all partial safety factors are subject to optimization, the resulting set of partial safety factors is 1.29,m  1.3,G   1.57Q  . However, it is the philosophy of the Eurocodes that the partial safety factors for the loads are material independent; therefore it is reasonable to fix G  and Q  and perform the optimization only subject to  . The line with the circles in Figure 1 is representing the corresponding result ( 1.33m  ). An enhancement in fit to the target reliability can be observed for both calibrated solutions.    Figure 1: Reliability Index over different design situations alpha for oli  timber in bending. The different lines represent different sets of partial safety factors.  The above example demonstrates the validity of the Eur cod  5 de gn safety concept for timb r load -bearing elements under the assumption that the parameters given in Figure 1 represent the real situation with sufficient accuracy. In the following it will be discussed in which ways the actual load load-bearing behavior derivate from the assumptions in Table 1. It is demonstrated and quantified how the corresponding deviation affects the reliability of design situations and it is discussed how recent research results might integrated in the further developed issue of Eurocode 5.  3. PARTICULARITIES IN TIMBER MATERIAL MODELLING  In this Section specific characteristics of timber and engineered wood products in respect to code based design are discussed.  3.1. Different “material properties” Timber is a rather complex building material. Its properties are highly variable, spatially and in time. In structural engineering, material properties of timber are in general understood as the stress and stiffness related properties of standard test specimen under given (standard) loading and climate conditions and the timber density. Test configurations are prescribed in e.g. ISO 8375 and any statement about stress and stiffness related properties of structural timber is conditional to the corresponding test configuration. In general it is distinguished between the different loading modes and “material properties” are given corresponding to the loading direction relative to the main fiber direction of a beam shaped element (Figure 2).    Figure 2: Different “material properties” dependent on the loading mode.  The “material properties” have different statistical properties and when using the design criterion in Eq. 1 and applying the same partial safety factor m , as it is practiced in the 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.844.14.24.34.4TargetEurocodeOptimized, ,m G Q  mOptimizedFormatted: FirstparagraphFigure 2: Different “mater al properties” dep dent onthe loading mode.The influence of different “material properties”was investigated in Kohler and Fink (2012). There,the distribution functions and the associated vari-ability for different types of “material properties”were chosen, as recommended in the ProbabilisticModel Code JCSS (2006), see also Köhler (2006)for background information. The results are sum-marized in Table 2. The obtained scatter in partialsafety factors suggests a rather differentiated treat-ment of the different design situations in future de-velopments of design codes.The most extreme deviation from the values pro-posed in the Eurocode γm = 1.30 is obtained forthe load case tension perpendicular to the grainγm = 3.05. This also indicated that if a structuralelement for this load case is designed with the cur-rent safety factor of γm = 1.30, very low reliabilityindices, in the order of magnitude of 3.1 are ob-tained. However, the results concerning this par-ticular load case have to be considered with spe-412th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015Table 2: Calibrated partial safety factors for the re-sistance, for constant γG = 1.35 and γQ = 1.50 (fromKohler and Fink, 2012).Ultimate limit state γmBending strength 1.33Tension strength parallel to the grain 1.40Tension strength perp. to the grain 3.05Compression strength parallel to the grain 1.24Compression strength perp. to the grain 1.20Shear strength 1.33cial care. In fact the material capacity under thisloading mode is specified by EN 338 (2010) with anominal value that does not correspond to the 5%-fractile value taken from the statistical distributionthat is derived from test data for the same load-ing mode. It is rather a value well below the 5%-fractile value. Furthermore, in best practice timberengineering design this loading mode at its limit isavoided due to the high sensibility to aspects thatare not directly controlled in design, as e.g. mois-ture induced stresses and macro and micro cracksin the timber.3.2. Timber as a graded materialDue to the special way timber material propertiesare ensured by means of grading in the produc-tion line, special considerations must be made whenmodeling their probabilistic characteristics. Previ-ous work on this subject is reported in (e.g. Rouger,1996; Pöhlmann and Rackwitz, 1981). Further as-sessment of the probabilistic modelling on the prop-erties of graded timber material was presented inFaber et al. (2004); Sandomeer et al. (2008). Inthe latter references it is reported that the scatter ofstrength related material properties is highly sen-sitive to the grading procedure applied and to theproperties of the original ungraded material. Thisobservation is confirmed by a large experimentalcampaign that took place recently in Europe in con-nection to the Gradewood project. Here a largenumber of graded samples have been tested anda large between sample variability has been ob-served. Furthermore it has been shown that it ishighly uncertain whether a sample that is graded toa specific grade actually meets the correspondingrequirements in regard to minimum 5% fractile val-ues of strength properties.It is continued along the example introducedabove, assuming that the grading accuracy directlyaffects the coefficient of variation of the timberbending capacity. The material partial safety fac-tor are calibrated for different grading schemes thatcorrespond to different coefficients of variation inthe range from 0.2 – 0.4. The corresponding par-tial material safety factors rage between γm = 1.2−1.65 depending on the applied grading procedure.These results suggest a better differentiation of thegrading procedure in future design codes. Alterna-tively, if no information about the accuracy of tim-ber grading is utilized a larger coefficient of vari-ation for representing the bending capacity shouldbe used.3.3. Non linear design equationsFor common design equations a linear compari-son of load effects and component resistance as inEquation (1) is not sufficient. One example is thedesign of slender columns where strength and stiff-ness properties and creep effects play an importantrole for assessing the stability. For the analysis ofsingle members, standards generally give simpli-fied calculation models that do not require a 2ndorder ultimate limit state analysis. However, forthe analysis of more complex systems like unbracedframe structures, a 2nd order structural analysis ismore appropriate and accurate and an alternativedesign procedure is given e.g. in the Eurocode(2004). Compared to the simple design format aspresented in Equation (1), the design equations forslender columns are more complex containing un-certain properties as strength, stiffness and load ec-centricity in non-linear combination. The problemwas addressed in Köhler et al. (2008) and quite un-even reliabilities for different column slendernesshave been reported. Figure 3 the reliability index ofdesign solutions with different slenderness-ratio arepresented. The different colors correspond to dif-ferent design frameworks; I. EN 1995-1-1 (2004),2nd order method with the stiffness considered asthe mean modulus of elasticity;. II. DIN 1052(2004), 2nd order method with the stiffness consid-512th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 201512th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12 Vancouver, Canada, July 12-15, 2015  63.2.1. Example cont.  It is continued along the example introduced in chapter 2.1.1. It is assumed that the grading accuracy directly affects the coefficient of variation of the timber bending capacity. The material partial safety factor will be calibrated for different grading schemes that correspond to different coefficients of variation in the range from 0.2 – 0.4. The results are presented in Figure 3 and it can be seen that the corresponding partial material safety factors rage between 1.2 and 1.65 depending on the applied grading procedure. These results suggest a better differentiation of the grading procedure in future design codes. Alternatively, if no information about the accuracy of timber grading is utilized a larger coefficient of variation for representing the bending capacity should be used.    Figure 3: Range of calibrated partial material factors for different grading qualities.   3.3. Non linear design equations For common design equations a linear comparison of load effects and component resistance as in Equation 1 is not sufficient. One example is the design of slender columns where strength and stiffness properties and creep effects play an important role for assessing the stability. For the analysis of single members, standards generally give simplified calculation models that do not require a 2nd order ultimate limit state analysis. However, for the analysis of more complex systems like unbraced frame structures, a 2nd order structural analysis is more appropriate and accurate and an alternative design procedure is given e.g. in the Eurocode (EN 1995). Compared to the simple design format as presented in Equation (1), the design equations for slender columns are more complex containing uncertain properties as strength, stiffness and load eccentricity in non-linear combination. The problem was addressed in Kohler et al. (2008) and quite uneven reliabilities for different column sle der ss have been reported (figure). ‘  Figure (Kohler et al. 2008)   CIB paper Köhler/Steiger  3.4.  Duration of load effect and moisture induced stress The capacity of a timber structural element is highly dependent on the time duration of the load effect to which it is exposed to. E.g. the capacity of a bending beam continuously loaded is only 60% of that of a similar beam exposed to an instant load, as it was already observed by (?). Timber is a hygroscopic material, i.e. it adsorbs and desorbs moisture from the surrounding air. Variations in moisture content in Reliability Index effiλ = 0 50 100 150 200 2503.844.24.44.64.855.2I.II.III.Figure 3: Reliability Index over slenderness for designsolutions according to different design formats (Köhleret al., 2008).ered as the 5% fractile of the modulus of elasticity,and; III. EN 1995-1-1 (2004) / DIN 1052 (2004) ac-cording to the so called simplified equivalent lengthapproach. For more details compare Köhler et al.(2008).In future code safety formats design strength andstiffness should be clibrated in order to obtain con-sistent reliability levels for different design situa-tions.3.4. Duration of load effect and moisture inducedstressThe capacity of a timber structural element ishighly dependent on the time duration of the loadeffect to wh ch it is expo ed to. E.g. th capacity ofa bending beam continuously loaded is only 60%of that of a similar beam exposed to an instant load(Wood, 1947).Timber is a hygroscopic material, i.e. it ad-sorbs and desorbs moisture from the surroundingair. Variations in moisture content in the surround-ing air will, with a corresponding time lack, leadto variations in moisture content in the timber,this affects the mechanical properties of the tim-ber but more importantly it will induce stresses dueto shrinkage and swelling alongside the moisturegradients in the timber. These moisture inducedstresses have been a matter of intensive discussionin the timber engineering community in the lastyears.Both, duration of load effect and moisture in-duced stresses are highly relevant phenomena totake into account in structural design. They arealso challenging phenomena since the underlyingphysical mechanisms are not fully understood andempirical evidence is scarce. However, in practicaldesign, as in the Eurocode 5 (2004), the effect ofmoisture on the duration of load effect is consid-ered with the joint modification factor kmod whichis given for different climate exposures in designcodes. Values for this factor are prescribed in a ma-trix for three different so-called service classes, i.e.different climate scenarios, and five different loadclasses, i.e. load scenarios.This format appears to be oversimplified and fur-ther research and enhancement of the level of detailin structural design should be developed.3.5. Volume and length effectsOne major topic that is continuously discussedwithin the research community is the appropriaterepresentation of size effects on strength in solidtimber. For most loading modes as tension paral-lel or perpendicular to grain, shear or bending, tim-ber predominately presents brittle failure behaviour.A (perfect) brittle material is defined as a materialthat fails if a single particle fails (see e.g. Bolotin,1969). The strength of the material is thus governedby the strength of the ‘weakest’ particle; there-fore the model for ideal brittle materials is alsocalled the weakest link model (Weibull, 1939). Thismodel was applied to the different failure modes intimber, the model parameters have been calibratedbased on experimental evidence on the differentfailure modes. A literature review can be founde.g. in Kohler et al. (2013). There it is concludedthat the size effects in timber are better representedwith a model that takes into account the multi scalevariability of structural timber and a correspondingmodel framework is suggested.In present code formats size effects are often notcompletely taken into account or neglected. This612th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015is particularly critical when large scale engineeredtimber sections are used in modern timber construc-tion. In a revision of the codes this aspect shouldearn appropriate attention and current research re-sults should be implemented.3.6. JointsFor timber structures, the structural performancedepends to a considerable part on the connectionsor joints between different timber structural mem-bers; joints can govern the overall strength, service-ability and fire resistance. Despite their importancetimber joint design frameworks are not based on aconsistent basis compared to the design regulationsof timber structural components.Explanations for this difference in progress ofdesign provisions for members and joints can befound in the relative simplicity of characterizingmechanical behaviour of members, as compared toconnections. A diversity of joints types is used inpractice and these types have infinite variety in ar-rangement. This usually precludes the option oftesting large numbers of replicas for a reliable quan-tification and verification of statistical and mechan-ical models. The main and most important groupof joints corresponds to the joints with dowel typefasteners, i.e. joints with dowels, nails, screws andstaples belong to this group.Different failure modes can be observed fordowel type fastener joints and the modes are partlycaptured by a simple mechanical model based onthe works of Johansen (1949); Meyer (1957). Thesemodels build the basis for the current European de-sign framework for dowel type connectors in theEurocode. However, different failure modes cor-respond to different failure behavior and conse-quences (brittle or ductile). In Köhler (2006) ishas also been observed that model uncertainty andmodel bias for the different failure modes is signif-icantly different. This is not considered in the cur-rent version of the European design standard andshould be subject for further investigation.3.7. Consequence classesIn the previous chapter it was mentioned that differ-ent failure modes in dowel type fastener joint leadto different magnitudes of consequences. This is inprinciple true for all failure modes in timber struc-ture. In Chapter 3.1 different failure modes of tim-ber components have been compared to the sametarget reliability, implying that the consequencesfor all failure modes are classified uniformly. How-ever, if a failure scenario for tension or bending fail-ure is visualized and compared with a typical fail-ure scenario for compression perpendicular to thegrain, it might be agreed that the consequences arequite different and correspondingly the target relia-bility should be defined separately for the differentcases.4. CONCLUSIONSTimber will play an important role in the futuredevelopments towards a more sustainable buildingsector. However, many stakeholders are still skep-tical when it comes to the technological maturity ofthe material, especially compared to concrete andsteel. The structural design regulations in generalcan be seen not only as the main interface connect-ing the state of knowledge in the engineering re-search community with the implementation of thereal build environment; design standards are alsothe precondition for the implementation of buildingmaterial on a high technological level.In the present paper the major challenges for thefuture development of timber design standards havebeen highlighted from a European perspective; i.e.taking the Eurocodes as references. The challengesare hereby related to both, the further developmentof the knowledge basis for the behavior of timberin structures and the implementation of this knowl-edge into practicable rules in the future standards.5. REFERENCESBolotin, V. V. (1969). Statistical Methods in Struc-tural Mechanics. Holden-Day Series in MathematicalPhysics.DIN 1052 (2004). “Bemessungsregeln für Holzkon-struktionen. Normenausschuss Bauwesen (NABau)im DIN (Deutsches Institut für Normung e.V.), Berlin,Germany.EN 1990 (2002). “Eurocode 0: Basis of structural de-sign. European Committee for Standardization, Brus-sels, Belgium.EN 1995-1-1 (2004). “Eurocode 5: Design of timberstructures – Part 1-1: General – Common rules and712th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP12Vancouver, Canada, July 12-15, 2015rules for buildings; German version. European Com-mittee for Standardization, Brussels, Belgium.EN 338 (2010). “Structural timber – Strength classes;German version. European Committee for Standard-ization, Brussels, Belgium.Faber, M. H., Köhler, J., and Sørensen, J. D. (2004).“Probabilistic modelling of graded timber materialproperties.” Structural safety, 26(3), 295–309.Faber, M. H. and Sørensen, J. D. (2003). “Reliabilitybased code calibration – The JCSS approach.” Appli-cations of statistics and probability in civil engineer-ing : proceedings of the 9th International Conferenceon Applications of Statistics and Probability in CivilEngineering, San Francisco, USA, July 6–9, 927–935.ISO 8375 (1985). “Solid timber in structural sizes: De-termination of some physical and mechanical proper-ties. International Standards Organization, Geneva,Switzerland.JCSS (2001). “Probabilistic Model Code Part I - Basisof Design.JCSS (2006). “Probabilistic Model Code Part III - Re-sistance Models (3.05 Timber).Johansen, K. (1949). “Theory of timber connections.”International Association of Bridge and StructuralEngineering, Publication No. 9, pp. 249-262, Bern,Switzerland.Köhler, J. (2006). “Reliability of timber structures.”Ph.D. thesis, ETH Zurich, Zurich, Switzerland, ETHZurich, Zurich, Switzerland.Köhler, J., Andrea, F., and René, S. (2008). “On the roleof stiffness properties for ultimate limit state designof slender columns.” Proceedings of the 41th Meet-ing, International Council for Research and Innova-tion in Building and Construction, Working Commis-sion W18 - Timber Structures, St. Andrews, Canada,CIB-W18, Paper No. 41-1-1.Kohler, J., Brandner, R., Thiel, A., and Schickhofer, G.(2013). “Probabilistic characterisation of the lengtheffect for parallel to the grain tensile strength of cen-tral european spruce.” Engineering Structures, 56,691–697.Kohler, J. and Fink, G. (2012). “Reliability BasedCode Calibration of a Typical Eurocode 5 DesignEquations.” World Conference on Timber Engineer-ing 2012 (WCTE 2012) : The Future of Timber Engi-neering : Auckland, New Zealand : 15-19 July 2012.Volume 4, P. Quenneville, ed., Red Hook, NY, CurranAssociates, Inc., 99–103.Meyer, A. (1957). “Die Tragfähigkeit vonNagelverbindungen bei Statischer Belastung.”Holz als Roh- und Werkstoff, 15(2), 96–109.Pöhlmann, S. and Rackwitz, R. (1981). Zur Verteilungs-funktion der Festigkeitseigenschaften bei kontinuier-lich durchgeführter Sortierung. Materialprüfung 23,Hanser, Munich, Germany.Rouger, F. (1996). “Application of a modified statisti-cal segmentation method to timber machine strengthgrading.” Wood and Fibre Science, 28(4).Sandomeer, M. K., Köhler, J., and Faber, M. H. (2008).“Probabilistic output control for structural timber -modelling approach.” Proceedings of the 41th Meet-ing, International Council for Research and Innova-tion in Building and Construction, Working Commis-sion W18 - Timber Structures, St. Andrews, Canada,CIB-W18, Paper No. 41-5-1.Weibull, W. (1939). A statistical theory of strength ofmaterials. Nr. 151, Royal Swedish Institute for Engi-neering Research.Wood, L. (1947). “Behaviour of wood under continuedloading.” Eng. News-Record, 139(24), 108–111.8

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.53032.1-0076305/manifest

Comment

Related Items