- Email: [email protected]
Butt welded joints assessment after fire exposure
Engineering Failure Analysis 106 (2019) 104144
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Butt welded joints assessment after fire exposure ⁎
Dorin Radua, , Teofil Gălățanua, Simon Sedmakb a b
T
Transilvania University of Brașov, Turnului Street No. 5, 500152 Brașov, Romania University of Belgrade, Faculty of Mechanical Engineering, Belgrade, Serbia
A R T IC LE I N F O
ABS TRA CT
Keywords: Structural integrity Fire design Critical crack size Structural steel Failure assessment diagrams
The need of in service life of a steel structure after being exposed to fire raises the problem of strengthening or replacement of some structural elements. In many cases, after structural safety assessment, fire exposed steel structures can be reused. Establishing the post-fire residual capacity of the structure has a major importance in overall assessment process. The paper presents a study for fire behavior of the butt-welded joints subjected to low cycle fatigue loading, taken into account the post fire conditions and the possibility of welding strengthening. There is presented a case study – a building structural element subjected to dynamic loading after fire exposure. For the in case study, post fire investigations revealed several welding flaws including welding surface type flaws. An assessment was needed in order to determine the structural integrity and life assessment of some structural elements. From the fracture mechanics point of view, a Failure Assessment Diagrams FAD - level 2 procedure was applied in order to determine the in service safety of the structure. Several flaws types were assessed and conclusions were taken. The results can be used for damage assessment and strengthening technique of post fire steel structures.
1. Introduction Under fire exposure steel structures are easy to lose strength but post fire assessment reports indicate that in most cases steel structures retain much of their load bearing capacity after cooling [1]. Therefore, in many cases, after structural safety assessment, fire exposed steel structures can be reused. Establishing the post-fire residual capacity of the structure has a major importance in overall assessment process. When having a dynamic low cycling fatigue loaded steel structure, the procedure for assessment of the capacity of the structure, must include also the fracture mechanic approach [2]. Numerous studies on the behavior of fire exposed bolted and welded connections are reported in the literature [3–7]. However, only a few studies on post-fire behavior of dynamically loaded joints are carried out. In the case of welded joints, chemical composition of base metal and weldment are different, and the effects of heating and cooling are similar to tempering and hardening, thus different transformations may occur in the base metal and the weld region. Therefore data of the weld mechanical properties are required in order to assess the whole capacity of fire-exposed steel structure. In many cases the parts affected by the fire are presenting visible deformations thus having an indicator of the tensions' redistribution caused by increased temperature, so much in the assessed element as in the adjacent elements. The evaluation of the general stability of a steel construction affected by the fire is made on the basis of the inspection of building component elements and on the damage degree. According to ASTM E119 [8], carbon steels have to satisfy certain acceptance
⁎
Corresponding author. E-mail address: [email protected] (D. Radu).
https://doi.org/10.1016/j.engfailanal.2019.08.010 Received 25 January 2019; Received in revised form 26 July 2019; Accepted 8 August 2019 Available online 12 August 2019 1350-6307/ © 2019 Elsevier Ltd. All rights reserved.
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
criteria, depending on the structural element's type. The European normative [9] presents recommendations and design rules for connections exposed to different temperatures. The design strength of a full penetration butt weld, for temperatures up to 700 °C, should be taken as equal to the strength of the weaker part of the connection using the appropriate reduction factors for structural steel. According with [9] provisions, the fire resistance of a welded joint is satisfied if the following conditions are met: - the joint load level is less than or equal to the maximum value of the stress level for any of the joined elements; - the resistance of the joint at normal temperature meets the recommendations given in EN 1993-1-8 (the joint is checking from the resistance point of view); - the thermal resistance of the fire protection of the joint is at least equal to the thermal resistance of the minimum protection applied to any of the joined elements. In particular, this means that it is not necessary to check joints in unprotected structures if the other conditions, especially those related to the joint resistance ratio, are met. The grounds commonly provided for this approach is because of the additional material from joints and due to the shadow effect created by the joined elements, joint temperature being inferior to the temperatures in the adjacent elements. Another explanation is related to geometry of the compartment, which leads to lower temperatures in the corners, where the joints are in the usual way. This approach is valid in case of rigid framed structures beam to column joints. In a continuity joint at the bottom of a lattice beam, for example, or a web butt weld of a high height beam, the approach can have different results. Nomenclature a half flaw length for through-thickness flaw, flaw height for surface flaw or half height for embedded flaw B section thickness in plane of flaw Efi,d design force due to the external loads in the with fire scenario kt stress concentration factor km stress magnification factor due to misalignment KI the stress intensity factor (SIF) Kr fracture ratio of applied elastic K value to Kmat Kmat the fracture toughness Lr ratio of stresses Mm stress intensity magnification factor Pb primary bending stress Pm primary membrane stress Rfi,d,0 resistance force of the element in the with fire scenario before the fire start at room temperature Rfi,d,t resistance force in fire situation Q secondary stress Qb residual bending stress Qtb thermal bending stress Qtm thermal membrane stress Qm residual membrane stress μ0 element utilization ratio σY(fy) the yielding resistance of the material σu(fu) the ultimate resistance of the material σmax the maximum tensile stress σx,Rd meridional design buckling stress σθ,Rd circumferential design bucking stress θ temperature of the with fire element (in fire scenario case) θa,cr critical temperature in function of utilization ratio (Y·σ)P contribution of the main stresses (Y·σ)S contribution of the secondary stresses W plate width in plane of flaw Y correction factor ECA Engineering critical assessment FAD Failure assessment diagrams FEM Finite element model NDT non-destructive testing Nevertheless rules or recommendations regarding the post fire behavior of these connections do not exist, and therefore the structures and associated joints are evaluated based on similar experiences and the published literature. The problem arises when trying to choose the solutions of consolidation for some steel structures after fire exposure. The paper is presenting the results from tests on welded joints specimens after exposed to temperatures up to 650 °C high temperatures [1] and furthermore, due to the low cycle fatigue loading, an assessment of the structure from the fracture mechanics 2
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
point of view in order to determine the structural integrity of the post fire structure, considering the discovered flaws. The assessment procedure is an Engineering Critical Assessment (ECA) type which is an analysis based on fracture mechanics principles of whether or not a given flaw is safe from brittle fracture, fatigue, creep or plastic collapse under specified loading conditions. An ECA can be used: during design, to assist in the choice of welding procedure and/or inspection techniques; during fabrication, to assess the significance of known defects which are unacceptable to a given code, or during operation, to assess flaws found in service and to make decisions as to whether they can safely remain, or whether down-rating/repair are necessary [10]. The analysis is carried out in accordance with the British Standard procedure BS 7910 (‘Guide to methods for assessing the acceptability of flaws in metallic structures’) [11]. The ECA concept (also termed ‘fitness-for-purpose analysis’) is widely accepted by a range of engineering industries. For an analysis of a known flaw, the following information is needed:
• size, position and orientation of flaw, • stresses acting on the region containing the flaw, • toughness and tensile properties of the region containing the flaw, The fact that knowledge of all these three aspects are necessary, implies a multidisciplinary approach, involving stress analysis, NDT expertise and materials engineering. 2. Case study – low cycle fatigue fire exposed element In order to study the behavior of post fire welded joints elements, there has been suggested an experimental program based on testing butt welded parts specimens extracted from a steel structure subject to the fire. The construction represented a wood industry manufacturing hall with steel columns, truss beams and longitudinal ten tones overhead crane steel beams (Fig. 1). Fire action on the structure has been accidental. The time of exposure to heat was estimated at 60–90 min, with random fire propagation. Usually, the materials inside the on fire building provide accurate information related to temperature regime, by identifying the status of their post-fire melting temperature. Following the fire expertise, after investigating the behavior of surrounding structural elements, it has been estimated an average temperature of 600 °C. It was needed to identify if the main structure (columns, beams and crane beams) was safe and if the structure can be reuse following a consolidation and a reconstruction of the roof. The sampling areas selection it was made taking into consideration all the aspects that could have modified the steel properties, so much as due to high temperatures, as well as to the structure's heavy cooling as a result of water usage in order to extinguish the fire [1]. A number of seven specimens were sampled, with geometry according with [12] (Fig. 2). In order to avoid heat-affected zones or mechanical stresses, all the specimens were cut using water jet cutting. Considering the study of the butt welded joints, there were taken samples from the overhead crane steel beam continuity joint – web area (Fig. 2b – specimens SWF1, SWF2, SWF3), respecting the normative regarding the sampling from an existing structural element [13] - (Annex A). For comparison of the steel behavior, specimens from the web area without butt welded joints were also taken – base material (BM) (Fig. 2a – specimens SPF1, SPF2 and SPF3). Another specimen was taken from a without fire element (an exterior area (canopy) of the same structure element type) – specimen named SPN1. All the specimens have the 16 mm thickness. Following spectral analysis was determined the steel composition (Table 1). Following the chemical composition tests, the steel qualifies for a S355JR type steel [14]. From the point of view of the steel weldability properties, the value was determined based on the published literature. The chemical composition influence on the welding of misaligned and poor aligned steels, it is expressed with the help of the equivalent carbon's concept, determined based on Deardon-O'Niell formula which was modeled after carbon‑manganese steels:
Fig. 1. Building affected by fire - photos taken after fire occurrence. 3
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 2. Specimens for tensile tests (a) base material; (b) butt-welded joint (c) specimen without fire exposure. Table 1 The steel chemical composition. Specimen
C%
Si%
Mn%
P%
S%
Cu%
SPF1 SPF2 SPF3 SPN1
0.159 0.149 0.155 0.172
0.198 0.35 0.188 0.225
0.76 0.85 0.811 0.74
0.02 0.028 0.022 0.02
0.0025 0.0030 0.0027 0.0021
0.171 0.245 0.183 0.163
CE = %C + %Mn/6 + (%Cr + %Mo + %V )/5 + (%Cu + %Ni )/15
(1)
For the specimens material, the value of equivalent carbon is CE = 0.3, which indicates a steel with good properties as it concerns weldability.
2.1. Tensile strength testing results All specimens were tested using hydraulic testing systems for tension applications. There were measured: the minimum yield strength, the ultimate strength, strain and elongation. The test results on the base material (BM) concluded that the steel has a yield strength between 280 and 290 MPa (Table 2), resulting that the used steel does not qualifies for a S355JR type. The base material is the one sampled from the steel structure exposed to flames during the fire and damaged by the rapid changes of temperature. Following the comparison of the tests results, there weren't identified major differences between the base metal (SPF type) specimens and the welded (SWF type) specimens. However, following the force-elongation diagrams (Fig. 3), a substantial differences regarding the ductility can be noticed. 4
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Table 2 Tensile strength results. Specimen
fy (MPa)
fu (MPa)
Elongation (mm)
SPF1 SPF2 SPF3 SWF1 SWF2 SWF3 SPN1
289.44 280.25 281.00 288.23 280.40 280.51 297.09
421.61 393.15 395.12 398.45 410.15 422.42 445.70
49.3 38.5 41.7 27.5 23.1 29.2 33.3
Fig. 3. Diagram force-elongation for base material steel (SPF1 and SPF3 specimens).
The ductility at the material level is expressed through the following requirements [14]: - the ultimate strength fu must be at least 20% higher than the minimum yield strength fy, - the elongation at the failure to be at least equal with 20%. If the ductility of the base steel from the sampled specimens, exceed the limit (Fig. 3), the results obtained for the butt weld joints specimens have shown that the specific elongation between the yielding limit and the ultimate limit decreases [15] (Fig. 4), while the yield limit does not change significantly (Table 2). Following the tensile test results, can be concluded that for the base material, the yield strength of steel does not change considerably as a result of the fire. However, in case of welded joints, the ductility decreases significantly. The reduction it is approximate 28% (Fig. 5), the result being presented in comparison with the post fire base metal specimen (SPF versus SWF). From the comparison the conclusion follows that the decreasing of the ductility is due to the welding process, not due to the fire action. It should be mention that at about 725 °C, the microstructure of metal changes significantly and the lattice structure is rearranged in the process of cooling. Similar results are obtained for butt welds reported in reference [5]. Though the conclusions are the same, there is a loading capacity decreasing request, the maximum decrease of residual bearing capacity reaches 20% when the exposure temperature is around 700 °C [16,17].
Fig. 4. Diagram force-elongation for butt welded joints (SWF1 and SWF2 specimens). 5
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 5. Force-elongation curves for base metal specimen (SPF) and butt welded joint (SWF).
2.2. Charpy test results In order to evaluate the toughness of the used material and to determine the steel type, Charpy test were performed. From additional tensile non-tested specimens, Charpy V notch specimens were manufactured (Fig. 6) and tested at 0 °C. The results revealed high values of toughness (Table 3). Following the Charpy tests, it was concluded that the used steel qualifies for a S355JR steel type, the toughness properties defining an affiliation to the master curve – lower to transition regime. The determination of fracture toughness was done following the Charpy toughness correlation [18], taking into account a transitional behavior – Master curve approach, thus resulting Kmat = 84 MPa·m1/2. 2.3. Determining the stresses in the area of the welded joints According with [9] the weld resistance of a fully penetration butt weld for a temperature up to 700 °C must be taken equal to the strength of the weakest joint component using the corresponding reduction coefficients for the base material (Fig. 7). The standard Eurocode - Design of steel structures - Part 1–2: General rules - Structural fire design [9], indicates also a procedure to determine the critical temperature in function of the utilization ratio of the element/joint. Thus the utilization ratio can be determined as following:
μ 0 = E fi,d /Rfi,d,0
(2)
where Efi,d is the design force due to the external loads in the with fire scenario, and Rfi,d,0 is the resistance force of the element in the with fire scenario before the fire start at room temperature (20 °C). The critical temperature in function of utilization ratio is:
1 ⎤ θa, cr = 482 + 39.19 ln ⎡ ⎢ 0.9674 μ 3.833 − 1⎥ 0 ⎦ ⎣
(3)
In fact, [9] states that there must be a proportionality relation between the resistance effort in fire situation (Rfi,d,t) and the yielding limit of the material, function of temperature (fy(θ)) (4), which is the validity condition for Relation (3). In other words, according with the existing normative, there is a function type interdependence between the temperature of the with fire element (in fire scenario case) and the resistance/design force (4), where m is a proportionality constant.
Rfi,d,t = m·f y (θ)
(4)
Fig. 6. (a) Specimens sampling; (b) Charpy specimens; (c) post tests Charpy specimens. 6
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Table 3 Tensile strength results. Specimen
Test results Toughness at 0 °C (J)
P1 P2 P3
55 52 67
Fig. 7. Variation of the reduction coefficients for the high temperature steel stress-elongation curve [9].
For the welded joints, the utilization ratio method and the reduction coefficient method, both indicated by [9], are given the same results. However the results can be used for the in fire elements, the post fire assessed (tested) elements revealing a decreasing of the ductility (ch.2.1), with slight change in the stress area. In the present research was needed to have a more in depth analysis in order to determine the accurate stresses in the specimen, stresses which were needed for further fracture mechanics assessment. Thus was needed for a FEM type analysis. The model was analyzed with Ansys software, and respected tested specimen geometry and material properties, with the dimensions of the butt weld as in the SWF type specimens. The results revealed the almost the same results as the tests, with a yielding stress of 270.55 N/mm2 and a rupture stress of 422.44 N/mm2 (Fig. 8), values obtained in the base material. It worth mentioning that, same as for the specimens tests were the rupture occurred in the base material, also in the FEM analysis results, the rupture values are not in the welding joint. Considering the conservative approach, the FEM stresses results values will be used as an input data in further fracture mechanics assessment. 2.4. Fracture mechanics approach - assessment In case of steel structures, existing of flaws in critical parts of structural elements may lead to failures of the element and in case of lack of redundancy, even to the collapse of the entire structure. For the in case study, there was needed to investigate the flaws in the existing structure and to assess the implication in the overall behavior of the structure. Considering the initial site welding (not a workshop controlled type) of the overhead crane steel beams continuity joints, an in depth NDT testing was performed. Following the visual inspection, there were discovered surface and buried flaws type. An in depth assessment was needed in order to conclude the in service safety of the structure. The used assessment is an ECA type in accordance with the British Standard procedure [11] determining the critical dimensions of the flaw. In order to determine the critical value of the flaw it is used the FAD – level 2 assessment procedure [11]. The method is presenting an assessment line given by an equation of a curve and a cut-off line. If the assessment point is in the interior of the surface limited by the assessment line, the flaw is acceptable and if the assessment point is at the outside area, the flaw is considered unacceptable [11] [19]. The equations which are describing the assessment line are:
δr or Kr = (1 − 0, 14Lr2 ) [0, 30 + 0, 70 exp(−0.65Lr6)] for Lr ≤ Lrmax
(5)
δr or Kr = 0 for Lr > Lrmax
(6) 7
Fig. 8. FEM results – (a) yielding and (b) rupture VonMises stress values.
D. Radu, et al.
Engineering Failure Analysis 106 (2019) 104144
8
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
The cut-of line is fixed in point where Lr = Lrmax where:
Lr max = (σY + σu )/(2σY )
(7)
For the assessment FAD - level 2 is necessary to pass through the following phases (more or less similar with FAD – level 1). As presented at FAD – level 1, the stresses must be known – following a structural analysis, these can be determined. The assessments are considering the real distribution of the stresses in the proximity of the flaws – Pm, Pb, Qm and Qb. The stress intensity factor is determined as FAD – level 1 where for the level 2 the factor:
Y ·σ = (Y ·σ )P + (Y ·σ )S
(8)
in which:
(Y ·σ )P = M·fw ·{ktm·Mkm·Mm·Pm + ktb·Mkb·Mb·[Pb + (km − 1) Pm]}
(9) (10)
(Y ·σ )S = Mm·Qm + Mb·Qb
The correction factor Y is determined according to the level 1 relations function of the defect type [11]. The ratio of stress Lr is determined according with:
Lr = σref / σY
(11)
in which σref is obtain according with a relation specific with the flaw type. The point/points of assessment are represented graphically in (Kr, Lr) coordinates on the FAD level 2. The input data for FAD – level 2 assessment was: yield strength fY = 270 MPa; ultimate strength fu = 420 MPa. The determination of fracture toughness was done following the Charpy toughness correlation [18], taking into account a transitional behavior – Master curve approach, thus resulting Kmat = 84 MPa·m1/2. Considering the assessment reliability, the primary stress was conservatively taken as the yielding value of the stress Pm = 289.44 N/mm2, with value determined following the specimens testing results – SPF1 specimen. Ten types of flaws – surface flaws (FP-SF type) and buried flaws (FP-BF type) are considered with the geometry presented in Table 4. For each type of flaw was determined the limit value (critical value) (Table 4) [20]. Determining the critical value of flaws is important because it serves to a limit value for fatigue further analysis based on fracture mechanics principles, needed for determining the number of cycles for a crack to extend from initial dimension to critical dimension which means the failure of element [21]. The procedure uses FAD – level 2 assessment data, and it gives the critical dimension of the crack. The results are presented in Table 4 and in Fig. 9. Following applying FAD – level 2 procedure, the Lr and Kr values are determined (Table 5). A graphical representation of the assessment results is presented in Figs. 10 and 11, making a comparison in the flaw importance. Thus it was revealed the buried flaw FP-BF-1 type as a critical flaw type. 3. Conclusions The paper presents a study for fire behavior of the butt-welded joints subjected to low cycle fatigue loading, taken into account the post fire behavior. For the in case study, post fire investigations revealed several welding flaws – surface and buried type flaws. It was needed to assess the implications of the fire exposure and the reliability of the structure for further in service life. Following the tensile test results, it can be concluded that for the base material, the yield strength of steel does not change considerably as a result of the fire. However, in case of welded joints, the ductility decreases significantly. The reduction it is approximately 28%, the result being presented in comparison with the post fire base metal specimen (SPF versus SWF). From the comparison can be drawn the conclusion that the decreasing of the ductility is due to the welding process not due to the fire action. A comparison with a FEM type analysis was made in order to validate the results. The yielding and ultimate stresses values were used as Table 4 FAD level 2 – in case - flaws assed: geometry and critical flaw dimensions. Case name
FP-SF-1 FP-SF-2 FP-SF-3 FP-SF-4 FP-SF-5 FP-BF-1 FP-BF-2 FP-BF-3 FP-BF-4 FP-BF-5
B
W
2a0
a0
2c0
p0
r0
h0
tw
Flaw height critic
Flaw length critic
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
16 32.63 200 25 25 16 32.63 200 25 25
200 200 32.63 200 120 200 200 32.63 200 120
5 5 5 5 5
30 30 10 30 30 30 30 10 30 30
5.690 18.261 10.734 14.755 14.505 7.590 10.984 OK 10.190 10.106
34.440 49.395 23.800 42.870 42.420 30.160 30.360 OK 30.365 30.360
5 5 5 5 5
3 3 3 3 3
9
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 9. Critical flaws assessment: (a) FP-SF-1; (b) FP-BF-1; (a) FP-SF-2; (b) FP-BF-2. Table 5 FAD level 2 – in case - flaws assed: geometry and results. Case
FP-SF-1 FP-SF-2 FP-SF-3 FP-SF-4 FP-SF-5 FP-BF-1 FP-BF-2 FP-BF-3 FP-BF-4 FP-BF-5
B
W
2a
a
2c
p
r0
h
tw
mm
mm
mm
mm
mm
mm
mm
mm
mm
16 32.63 200 25 25 16 32.63 200 25 25
200 200 32.63 200 120 200 200 32.63 200 120
5 5 5 5 5
30 30 10 30 30 30 30 10 30 30
5 5 5 5 5
3 3 3 3 3
10
Lr
Kr
0.8330 0.7429 0.7125 0.7644 0.7644 1.1888 0.8504 0.7290 0.9379 0.9379
0.4183 0.3899 0.2808 0.3965 0.3984 0.4075 0.4075 0.2679 0.4075 0.4140
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 10. FP-SF – Group of flaws – assessment.
Fig. 11. FP-BF – Group of flaws – assessment.
a base for the further assessment. From the fracture mechanics point of view, an Engineering Critical Assessment procedure was applied – determining the failure assessment diagrams (FAD - level 2), for each type of flaw. Buried (FP-BF) and surface (FP-SF) type flaws were assessed. Critical dimensions were calculated and the results were compared, thus revealing several problems: - Sensibility of the butt welded joints to the embedded/buried type flaws - The embedded/buried flaw is also a flaw type that is putting under risk the structural element – for the flat plate embedded flaw (FP-BF-1), the fracture ratio Lr is off the safe area and the flaw is considered inadmissible. The assessment results can be used for damage assessment and strengthening technique of post fire steel structures. For the further steps, the next phase of the research is to evaluate the in service fatigue loading and to calculate the structural element remaining service lifetime following an ECA fatigue based type.
11
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Declaration of Competing Interest None. Acknowledgements Part of the present research paper was presented at ECF22 Conference – Belgrade, and published in Procedia Structural Integrity [22]. References [1] T.F. Gălățanu, G. Băetu, C. Cazacu, D. Radu, R. Muntean, F. Tămaș, The study of butt-welded connections after fire exposure, Adv. Eng. Forum 21 (2017) 129–134, https://doi.org/10.4028/www.scientific.net/AEF.21.129. [2] D. Radu, A. Sedmak, R. Băncilă “Determining the crack acceptability in the welded joints of a wind loaded cylindrical steel shell structure”, DOI: https://doi.org/ 10.1016/j.engfailanal.2018.04.032. [3] K.S. Al-Jabri, I.W. Burgess, R.J. Plank, Prediction of the degradation of connection characteristics at elevated temperature, J. Constr. Steel Res. 60 (2004) 771–781. [4] P. Pakala, V. Kodur, System-level approach to predict the transient fire response of double angle connections, J. Constr. SteelRes. 89 (2013) 132–144. [5] L. Chen, Y.C. Wang, Methods of improving survivability of steel beam/column connections in fire, J.Constr.Steel Res. 79 (2012) 127–139. [6] A.S. Daryan, M. Yahyai, Behaviour of welded top-seat angle connections exposed to fire, Fire Saf. J. 44 (2009) 603–611. [7] Y. Mahmood, D.A. Saedi, The study of welded semi-rigid connections in fire, Struct. Des. Tall Spec. 22 (2013) 783–801. [8] ASTM E119, Standard Test Methods for Fire Tests of Building Construction and Materials, ASTM International, West Conshohocken, 2015. [9] Eurocode 3 Design of Steel Structures-Part 1–2: General Rules-Structural Fire Design, European Committee for Standardization, Brussels. [10] Mei-Chun Zhu, Guo-Qiang Li, Behavior of beam-to-column welded connections in steel structures after fire, 6th International Workshop on Performance, Protection & Strengthening of Structures Under Extreme Loading, PROTECT2017, 11-12 December 2017, Guangzhou (Canton), China, Procedia Eng. 210 2017, pp. 551–556. [11] BS 7910/2013, Guide to Methods for Assessing the Acceptability of Flaws in Metallic Structures, BSI British Standards. [12] EN 10002-1:1994, Tensile testing of metallic materials. Method of Test at Ambient Temperature. [13] EN 10025-1:2005, Hot rolled products of structural steels - part 1: general technical delivery conditions. [14] EN 10025-2 Hot rolled products of structural steel – technical delivery conditions for non-alloy structural steels. [15] J. Outinen, P. Makelainen, Mechanical properties of structural steel at elevated temperatures and after cooling down, Second International Workshop “Structures Fire”, Christchurch, 2002, pp. 273–289. [16] G. Zhang, M.C. Zhu, V. Kodur, G.Q. Li, Behavior of welded connections after exposure to elevated temperature, J. Constr. Steel Res. 130 (2017) 88–95. [17] F. Hanus, G. Zilli, J.M. Franssen, Experimental tests and analytical models for welds and grade8.8 bolts under heating and subsequent cooling, J.Struct. Fire Eng. 2 (2011) 181–194. [18] Structural integrity assessment procedure for European industry – SINTAP, Sub-Task 3.3. Report – Final Issue: Determination of Fracture Toughness Following the Charpy Impact Energy: Procedure and Validation. [19] I. Vučetić, S. Kirin, A. Sedmak, T. Golubović, M. Lazić, Risk management of a hydro power plant – fracture mechanics approach, Tech. Gazette 26 (2) (2019) 428–432. [20] TWI CrackWise Software ver.5.0 – Documentation. [21] T. Sedmak, R. Bakić, S. Sedmak, Lj. Milović, S. Kirin, Applicability of risk-based maintenance strategy to a penstock, Struct. Integrity and Life 12 (3) (2012) 191–196. [22] D. Radu, T.F. Gălățanu, S. Sedmak, “Structural Integrity of Butt Welded Connection after Fire Exposure” https://doi.org/10.1016/j.prostr.2018.12.227.
12
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Butt welded joints assessment after fire exposure ⁎
Dorin Radua, , Teofil Gălățanua, Simon Sedmakb a b
T
Transilvania University of Brașov, Turnului Street No. 5, 500152 Brașov, Romania University of Belgrade, Faculty of Mechanical Engineering, Belgrade, Serbia
A R T IC LE I N F O
ABS TRA CT
Keywords: Structural integrity Fire design Critical crack size Structural steel Failure assessment diagrams
The need of in service life of a steel structure after being exposed to fire raises the problem of strengthening or replacement of some structural elements. In many cases, after structural safety assessment, fire exposed steel structures can be reused. Establishing the post-fire residual capacity of the structure has a major importance in overall assessment process. The paper presents a study for fire behavior of the butt-welded joints subjected to low cycle fatigue loading, taken into account the post fire conditions and the possibility of welding strengthening. There is presented a case study – a building structural element subjected to dynamic loading after fire exposure. For the in case study, post fire investigations revealed several welding flaws including welding surface type flaws. An assessment was needed in order to determine the structural integrity and life assessment of some structural elements. From the fracture mechanics point of view, a Failure Assessment Diagrams FAD - level 2 procedure was applied in order to determine the in service safety of the structure. Several flaws types were assessed and conclusions were taken. The results can be used for damage assessment and strengthening technique of post fire steel structures.
1. Introduction Under fire exposure steel structures are easy to lose strength but post fire assessment reports indicate that in most cases steel structures retain much of their load bearing capacity after cooling [1]. Therefore, in many cases, after structural safety assessment, fire exposed steel structures can be reused. Establishing the post-fire residual capacity of the structure has a major importance in overall assessment process. When having a dynamic low cycling fatigue loaded steel structure, the procedure for assessment of the capacity of the structure, must include also the fracture mechanic approach [2]. Numerous studies on the behavior of fire exposed bolted and welded connections are reported in the literature [3–7]. However, only a few studies on post-fire behavior of dynamically loaded joints are carried out. In the case of welded joints, chemical composition of base metal and weldment are different, and the effects of heating and cooling are similar to tempering and hardening, thus different transformations may occur in the base metal and the weld region. Therefore data of the weld mechanical properties are required in order to assess the whole capacity of fire-exposed steel structure. In many cases the parts affected by the fire are presenting visible deformations thus having an indicator of the tensions' redistribution caused by increased temperature, so much in the assessed element as in the adjacent elements. The evaluation of the general stability of a steel construction affected by the fire is made on the basis of the inspection of building component elements and on the damage degree. According to ASTM E119 [8], carbon steels have to satisfy certain acceptance
⁎
Corresponding author. E-mail address: [email protected] (D. Radu).
https://doi.org/10.1016/j.engfailanal.2019.08.010 Received 25 January 2019; Received in revised form 26 July 2019; Accepted 8 August 2019 Available online 12 August 2019 1350-6307/ © 2019 Elsevier Ltd. All rights reserved.
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
criteria, depending on the structural element's type. The European normative [9] presents recommendations and design rules for connections exposed to different temperatures. The design strength of a full penetration butt weld, for temperatures up to 700 °C, should be taken as equal to the strength of the weaker part of the connection using the appropriate reduction factors for structural steel. According with [9] provisions, the fire resistance of a welded joint is satisfied if the following conditions are met: - the joint load level is less than or equal to the maximum value of the stress level for any of the joined elements; - the resistance of the joint at normal temperature meets the recommendations given in EN 1993-1-8 (the joint is checking from the resistance point of view); - the thermal resistance of the fire protection of the joint is at least equal to the thermal resistance of the minimum protection applied to any of the joined elements. In particular, this means that it is not necessary to check joints in unprotected structures if the other conditions, especially those related to the joint resistance ratio, are met. The grounds commonly provided for this approach is because of the additional material from joints and due to the shadow effect created by the joined elements, joint temperature being inferior to the temperatures in the adjacent elements. Another explanation is related to geometry of the compartment, which leads to lower temperatures in the corners, where the joints are in the usual way. This approach is valid in case of rigid framed structures beam to column joints. In a continuity joint at the bottom of a lattice beam, for example, or a web butt weld of a high height beam, the approach can have different results. Nomenclature a half flaw length for through-thickness flaw, flaw height for surface flaw or half height for embedded flaw B section thickness in plane of flaw Efi,d design force due to the external loads in the with fire scenario kt stress concentration factor km stress magnification factor due to misalignment KI the stress intensity factor (SIF) Kr fracture ratio of applied elastic K value to Kmat Kmat the fracture toughness Lr ratio of stresses Mm stress intensity magnification factor Pb primary bending stress Pm primary membrane stress Rfi,d,0 resistance force of the element in the with fire scenario before the fire start at room temperature Rfi,d,t resistance force in fire situation Q secondary stress Qb residual bending stress Qtb thermal bending stress Qtm thermal membrane stress Qm residual membrane stress μ0 element utilization ratio σY(fy) the yielding resistance of the material σu(fu) the ultimate resistance of the material σmax the maximum tensile stress σx,Rd meridional design buckling stress σθ,Rd circumferential design bucking stress θ temperature of the with fire element (in fire scenario case) θa,cr critical temperature in function of utilization ratio (Y·σ)P contribution of the main stresses (Y·σ)S contribution of the secondary stresses W plate width in plane of flaw Y correction factor ECA Engineering critical assessment FAD Failure assessment diagrams FEM Finite element model NDT non-destructive testing Nevertheless rules or recommendations regarding the post fire behavior of these connections do not exist, and therefore the structures and associated joints are evaluated based on similar experiences and the published literature. The problem arises when trying to choose the solutions of consolidation for some steel structures after fire exposure. The paper is presenting the results from tests on welded joints specimens after exposed to temperatures up to 650 °C high temperatures [1] and furthermore, due to the low cycle fatigue loading, an assessment of the structure from the fracture mechanics 2
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
point of view in order to determine the structural integrity of the post fire structure, considering the discovered flaws. The assessment procedure is an Engineering Critical Assessment (ECA) type which is an analysis based on fracture mechanics principles of whether or not a given flaw is safe from brittle fracture, fatigue, creep or plastic collapse under specified loading conditions. An ECA can be used: during design, to assist in the choice of welding procedure and/or inspection techniques; during fabrication, to assess the significance of known defects which are unacceptable to a given code, or during operation, to assess flaws found in service and to make decisions as to whether they can safely remain, or whether down-rating/repair are necessary [10]. The analysis is carried out in accordance with the British Standard procedure BS 7910 (‘Guide to methods for assessing the acceptability of flaws in metallic structures’) [11]. The ECA concept (also termed ‘fitness-for-purpose analysis’) is widely accepted by a range of engineering industries. For an analysis of a known flaw, the following information is needed:
• size, position and orientation of flaw, • stresses acting on the region containing the flaw, • toughness and tensile properties of the region containing the flaw, The fact that knowledge of all these three aspects are necessary, implies a multidisciplinary approach, involving stress analysis, NDT expertise and materials engineering. 2. Case study – low cycle fatigue fire exposed element In order to study the behavior of post fire welded joints elements, there has been suggested an experimental program based on testing butt welded parts specimens extracted from a steel structure subject to the fire. The construction represented a wood industry manufacturing hall with steel columns, truss beams and longitudinal ten tones overhead crane steel beams (Fig. 1). Fire action on the structure has been accidental. The time of exposure to heat was estimated at 60–90 min, with random fire propagation. Usually, the materials inside the on fire building provide accurate information related to temperature regime, by identifying the status of their post-fire melting temperature. Following the fire expertise, after investigating the behavior of surrounding structural elements, it has been estimated an average temperature of 600 °C. It was needed to identify if the main structure (columns, beams and crane beams) was safe and if the structure can be reuse following a consolidation and a reconstruction of the roof. The sampling areas selection it was made taking into consideration all the aspects that could have modified the steel properties, so much as due to high temperatures, as well as to the structure's heavy cooling as a result of water usage in order to extinguish the fire [1]. A number of seven specimens were sampled, with geometry according with [12] (Fig. 2). In order to avoid heat-affected zones or mechanical stresses, all the specimens were cut using water jet cutting. Considering the study of the butt welded joints, there were taken samples from the overhead crane steel beam continuity joint – web area (Fig. 2b – specimens SWF1, SWF2, SWF3), respecting the normative regarding the sampling from an existing structural element [13] - (Annex A). For comparison of the steel behavior, specimens from the web area without butt welded joints were also taken – base material (BM) (Fig. 2a – specimens SPF1, SPF2 and SPF3). Another specimen was taken from a without fire element (an exterior area (canopy) of the same structure element type) – specimen named SPN1. All the specimens have the 16 mm thickness. Following spectral analysis was determined the steel composition (Table 1). Following the chemical composition tests, the steel qualifies for a S355JR type steel [14]. From the point of view of the steel weldability properties, the value was determined based on the published literature. The chemical composition influence on the welding of misaligned and poor aligned steels, it is expressed with the help of the equivalent carbon's concept, determined based on Deardon-O'Niell formula which was modeled after carbon‑manganese steels:
Fig. 1. Building affected by fire - photos taken after fire occurrence. 3
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 2. Specimens for tensile tests (a) base material; (b) butt-welded joint (c) specimen without fire exposure. Table 1 The steel chemical composition. Specimen
C%
Si%
Mn%
P%
S%
Cu%
SPF1 SPF2 SPF3 SPN1
0.159 0.149 0.155 0.172
0.198 0.35 0.188 0.225
0.76 0.85 0.811 0.74
0.02 0.028 0.022 0.02
0.0025 0.0030 0.0027 0.0021
0.171 0.245 0.183 0.163
CE = %C + %Mn/6 + (%Cr + %Mo + %V )/5 + (%Cu + %Ni )/15
(1)
For the specimens material, the value of equivalent carbon is CE = 0.3, which indicates a steel with good properties as it concerns weldability.
2.1. Tensile strength testing results All specimens were tested using hydraulic testing systems for tension applications. There were measured: the minimum yield strength, the ultimate strength, strain and elongation. The test results on the base material (BM) concluded that the steel has a yield strength between 280 and 290 MPa (Table 2), resulting that the used steel does not qualifies for a S355JR type. The base material is the one sampled from the steel structure exposed to flames during the fire and damaged by the rapid changes of temperature. Following the comparison of the tests results, there weren't identified major differences between the base metal (SPF type) specimens and the welded (SWF type) specimens. However, following the force-elongation diagrams (Fig. 3), a substantial differences regarding the ductility can be noticed. 4
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Table 2 Tensile strength results. Specimen
fy (MPa)
fu (MPa)
Elongation (mm)
SPF1 SPF2 SPF3 SWF1 SWF2 SWF3 SPN1
289.44 280.25 281.00 288.23 280.40 280.51 297.09
421.61 393.15 395.12 398.45 410.15 422.42 445.70
49.3 38.5 41.7 27.5 23.1 29.2 33.3
Fig. 3. Diagram force-elongation for base material steel (SPF1 and SPF3 specimens).
The ductility at the material level is expressed through the following requirements [14]: - the ultimate strength fu must be at least 20% higher than the minimum yield strength fy, - the elongation at the failure to be at least equal with 20%. If the ductility of the base steel from the sampled specimens, exceed the limit (Fig. 3), the results obtained for the butt weld joints specimens have shown that the specific elongation between the yielding limit and the ultimate limit decreases [15] (Fig. 4), while the yield limit does not change significantly (Table 2). Following the tensile test results, can be concluded that for the base material, the yield strength of steel does not change considerably as a result of the fire. However, in case of welded joints, the ductility decreases significantly. The reduction it is approximate 28% (Fig. 5), the result being presented in comparison with the post fire base metal specimen (SPF versus SWF). From the comparison the conclusion follows that the decreasing of the ductility is due to the welding process, not due to the fire action. It should be mention that at about 725 °C, the microstructure of metal changes significantly and the lattice structure is rearranged in the process of cooling. Similar results are obtained for butt welds reported in reference [5]. Though the conclusions are the same, there is a loading capacity decreasing request, the maximum decrease of residual bearing capacity reaches 20% when the exposure temperature is around 700 °C [16,17].
Fig. 4. Diagram force-elongation for butt welded joints (SWF1 and SWF2 specimens). 5
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 5. Force-elongation curves for base metal specimen (SPF) and butt welded joint (SWF).
2.2. Charpy test results In order to evaluate the toughness of the used material and to determine the steel type, Charpy test were performed. From additional tensile non-tested specimens, Charpy V notch specimens were manufactured (Fig. 6) and tested at 0 °C. The results revealed high values of toughness (Table 3). Following the Charpy tests, it was concluded that the used steel qualifies for a S355JR steel type, the toughness properties defining an affiliation to the master curve – lower to transition regime. The determination of fracture toughness was done following the Charpy toughness correlation [18], taking into account a transitional behavior – Master curve approach, thus resulting Kmat = 84 MPa·m1/2. 2.3. Determining the stresses in the area of the welded joints According with [9] the weld resistance of a fully penetration butt weld for a temperature up to 700 °C must be taken equal to the strength of the weakest joint component using the corresponding reduction coefficients for the base material (Fig. 7). The standard Eurocode - Design of steel structures - Part 1–2: General rules - Structural fire design [9], indicates also a procedure to determine the critical temperature in function of the utilization ratio of the element/joint. Thus the utilization ratio can be determined as following:
μ 0 = E fi,d /Rfi,d,0
(2)
where Efi,d is the design force due to the external loads in the with fire scenario, and Rfi,d,0 is the resistance force of the element in the with fire scenario before the fire start at room temperature (20 °C). The critical temperature in function of utilization ratio is:
1 ⎤ θa, cr = 482 + 39.19 ln ⎡ ⎢ 0.9674 μ 3.833 − 1⎥ 0 ⎦ ⎣
(3)
In fact, [9] states that there must be a proportionality relation between the resistance effort in fire situation (Rfi,d,t) and the yielding limit of the material, function of temperature (fy(θ)) (4), which is the validity condition for Relation (3). In other words, according with the existing normative, there is a function type interdependence between the temperature of the with fire element (in fire scenario case) and the resistance/design force (4), where m is a proportionality constant.
Rfi,d,t = m·f y (θ)
(4)
Fig. 6. (a) Specimens sampling; (b) Charpy specimens; (c) post tests Charpy specimens. 6
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Table 3 Tensile strength results. Specimen
Test results Toughness at 0 °C (J)
P1 P2 P3
55 52 67
Fig. 7. Variation of the reduction coefficients for the high temperature steel stress-elongation curve [9].
For the welded joints, the utilization ratio method and the reduction coefficient method, both indicated by [9], are given the same results. However the results can be used for the in fire elements, the post fire assessed (tested) elements revealing a decreasing of the ductility (ch.2.1), with slight change in the stress area. In the present research was needed to have a more in depth analysis in order to determine the accurate stresses in the specimen, stresses which were needed for further fracture mechanics assessment. Thus was needed for a FEM type analysis. The model was analyzed with Ansys software, and respected tested specimen geometry and material properties, with the dimensions of the butt weld as in the SWF type specimens. The results revealed the almost the same results as the tests, with a yielding stress of 270.55 N/mm2 and a rupture stress of 422.44 N/mm2 (Fig. 8), values obtained in the base material. It worth mentioning that, same as for the specimens tests were the rupture occurred in the base material, also in the FEM analysis results, the rupture values are not in the welding joint. Considering the conservative approach, the FEM stresses results values will be used as an input data in further fracture mechanics assessment. 2.4. Fracture mechanics approach - assessment In case of steel structures, existing of flaws in critical parts of structural elements may lead to failures of the element and in case of lack of redundancy, even to the collapse of the entire structure. For the in case study, there was needed to investigate the flaws in the existing structure and to assess the implication in the overall behavior of the structure. Considering the initial site welding (not a workshop controlled type) of the overhead crane steel beams continuity joints, an in depth NDT testing was performed. Following the visual inspection, there were discovered surface and buried flaws type. An in depth assessment was needed in order to conclude the in service safety of the structure. The used assessment is an ECA type in accordance with the British Standard procedure [11] determining the critical dimensions of the flaw. In order to determine the critical value of the flaw it is used the FAD – level 2 assessment procedure [11]. The method is presenting an assessment line given by an equation of a curve and a cut-off line. If the assessment point is in the interior of the surface limited by the assessment line, the flaw is acceptable and if the assessment point is at the outside area, the flaw is considered unacceptable [11] [19]. The equations which are describing the assessment line are:
δr or Kr = (1 − 0, 14Lr2 ) [0, 30 + 0, 70 exp(−0.65Lr6)] for Lr ≤ Lrmax
(5)
δr or Kr = 0 for Lr > Lrmax
(6) 7
Fig. 8. FEM results – (a) yielding and (b) rupture VonMises stress values.
D. Radu, et al.
Engineering Failure Analysis 106 (2019) 104144
8
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
The cut-of line is fixed in point where Lr = Lrmax where:
Lr max = (σY + σu )/(2σY )
(7)
For the assessment FAD - level 2 is necessary to pass through the following phases (more or less similar with FAD – level 1). As presented at FAD – level 1, the stresses must be known – following a structural analysis, these can be determined. The assessments are considering the real distribution of the stresses in the proximity of the flaws – Pm, Pb, Qm and Qb. The stress intensity factor is determined as FAD – level 1 where for the level 2 the factor:
Y ·σ = (Y ·σ )P + (Y ·σ )S
(8)
in which:
(Y ·σ )P = M·fw ·{ktm·Mkm·Mm·Pm + ktb·Mkb·Mb·[Pb + (km − 1) Pm]}
(9) (10)
(Y ·σ )S = Mm·Qm + Mb·Qb
The correction factor Y is determined according to the level 1 relations function of the defect type [11]. The ratio of stress Lr is determined according with:
Lr = σref / σY
(11)
in which σref is obtain according with a relation specific with the flaw type. The point/points of assessment are represented graphically in (Kr, Lr) coordinates on the FAD level 2. The input data for FAD – level 2 assessment was: yield strength fY = 270 MPa; ultimate strength fu = 420 MPa. The determination of fracture toughness was done following the Charpy toughness correlation [18], taking into account a transitional behavior – Master curve approach, thus resulting Kmat = 84 MPa·m1/2. Considering the assessment reliability, the primary stress was conservatively taken as the yielding value of the stress Pm = 289.44 N/mm2, with value determined following the specimens testing results – SPF1 specimen. Ten types of flaws – surface flaws (FP-SF type) and buried flaws (FP-BF type) are considered with the geometry presented in Table 4. For each type of flaw was determined the limit value (critical value) (Table 4) [20]. Determining the critical value of flaws is important because it serves to a limit value for fatigue further analysis based on fracture mechanics principles, needed for determining the number of cycles for a crack to extend from initial dimension to critical dimension which means the failure of element [21]. The procedure uses FAD – level 2 assessment data, and it gives the critical dimension of the crack. The results are presented in Table 4 and in Fig. 9. Following applying FAD – level 2 procedure, the Lr and Kr values are determined (Table 5). A graphical representation of the assessment results is presented in Figs. 10 and 11, making a comparison in the flaw importance. Thus it was revealed the buried flaw FP-BF-1 type as a critical flaw type. 3. Conclusions The paper presents a study for fire behavior of the butt-welded joints subjected to low cycle fatigue loading, taken into account the post fire behavior. For the in case study, post fire investigations revealed several welding flaws – surface and buried type flaws. It was needed to assess the implications of the fire exposure and the reliability of the structure for further in service life. Following the tensile test results, it can be concluded that for the base material, the yield strength of steel does not change considerably as a result of the fire. However, in case of welded joints, the ductility decreases significantly. The reduction it is approximately 28%, the result being presented in comparison with the post fire base metal specimen (SPF versus SWF). From the comparison can be drawn the conclusion that the decreasing of the ductility is due to the welding process not due to the fire action. A comparison with a FEM type analysis was made in order to validate the results. The yielding and ultimate stresses values were used as Table 4 FAD level 2 – in case - flaws assed: geometry and critical flaw dimensions. Case name
FP-SF-1 FP-SF-2 FP-SF-3 FP-SF-4 FP-SF-5 FP-BF-1 FP-BF-2 FP-BF-3 FP-BF-4 FP-BF-5
B
W
2a0
a0
2c0
p0
r0
h0
tw
Flaw height critic
Flaw length critic
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
mm
16 32.63 200 25 25 16 32.63 200 25 25
200 200 32.63 200 120 200 200 32.63 200 120
5 5 5 5 5
30 30 10 30 30 30 30 10 30 30
5.690 18.261 10.734 14.755 14.505 7.590 10.984 OK 10.190 10.106
34.440 49.395 23.800 42.870 42.420 30.160 30.360 OK 30.365 30.360
5 5 5 5 5
3 3 3 3 3
9
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 9. Critical flaws assessment: (a) FP-SF-1; (b) FP-BF-1; (a) FP-SF-2; (b) FP-BF-2. Table 5 FAD level 2 – in case - flaws assed: geometry and results. Case
FP-SF-1 FP-SF-2 FP-SF-3 FP-SF-4 FP-SF-5 FP-BF-1 FP-BF-2 FP-BF-3 FP-BF-4 FP-BF-5
B
W
2a
a
2c
p
r0
h
tw
mm
mm
mm
mm
mm
mm
mm
mm
mm
16 32.63 200 25 25 16 32.63 200 25 25
200 200 32.63 200 120 200 200 32.63 200 120
5 5 5 5 5
30 30 10 30 30 30 30 10 30 30
5 5 5 5 5
3 3 3 3 3
10
Lr
Kr
0.8330 0.7429 0.7125 0.7644 0.7644 1.1888 0.8504 0.7290 0.9379 0.9379
0.4183 0.3899 0.2808 0.3965 0.3984 0.4075 0.4075 0.2679 0.4075 0.4140
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Fig. 10. FP-SF – Group of flaws – assessment.
Fig. 11. FP-BF – Group of flaws – assessment.
a base for the further assessment. From the fracture mechanics point of view, an Engineering Critical Assessment procedure was applied – determining the failure assessment diagrams (FAD - level 2), for each type of flaw. Buried (FP-BF) and surface (FP-SF) type flaws were assessed. Critical dimensions were calculated and the results were compared, thus revealing several problems: - Sensibility of the butt welded joints to the embedded/buried type flaws - The embedded/buried flaw is also a flaw type that is putting under risk the structural element – for the flat plate embedded flaw (FP-BF-1), the fracture ratio Lr is off the safe area and the flaw is considered inadmissible. The assessment results can be used for damage assessment and strengthening technique of post fire steel structures. For the further steps, the next phase of the research is to evaluate the in service fatigue loading and to calculate the structural element remaining service lifetime following an ECA fatigue based type.
11
Engineering Failure Analysis 106 (2019) 104144
D. Radu, et al.
Declaration of Competing Interest None. Acknowledgements Part of the present research paper was presented at ECF22 Conference – Belgrade, and published in Procedia Structural Integrity [22]. References [1] T.F. Gălățanu, G. Băetu, C. Cazacu, D. Radu, R. Muntean, F. Tămaș, The study of butt-welded connections after fire exposure, Adv. Eng. Forum 21 (2017) 129–134, https://doi.org/10.4028/www.scientific.net/AEF.21.129. [2] D. Radu, A. Sedmak, R. Băncilă “Determining the crack acceptability in the welded joints of a wind loaded cylindrical steel shell structure”, DOI: https://doi.org/ 10.1016/j.engfailanal.2018.04.032. [3] K.S. Al-Jabri, I.W. Burgess, R.J. Plank, Prediction of the degradation of connection characteristics at elevated temperature, J. Constr. Steel Res. 60 (2004) 771–781. [4] P. Pakala, V. Kodur, System-level approach to predict the transient fire response of double angle connections, J. Constr. SteelRes. 89 (2013) 132–144. [5] L. Chen, Y.C. Wang, Methods of improving survivability of steel beam/column connections in fire, J.Constr.Steel Res. 79 (2012) 127–139. [6] A.S. Daryan, M. Yahyai, Behaviour of welded top-seat angle connections exposed to fire, Fire Saf. J. 44 (2009) 603–611. [7] Y. Mahmood, D.A. Saedi, The study of welded semi-rigid connections in fire, Struct. Des. Tall Spec. 22 (2013) 783–801. [8] ASTM E119, Standard Test Methods for Fire Tests of Building Construction and Materials, ASTM International, West Conshohocken, 2015. [9] Eurocode 3 Design of Steel Structures-Part 1–2: General Rules-Structural Fire Design, European Committee for Standardization, Brussels. [10] Mei-Chun Zhu, Guo-Qiang Li, Behavior of beam-to-column welded connections in steel structures after fire, 6th International Workshop on Performance, Protection & Strengthening of Structures Under Extreme Loading, PROTECT2017, 11-12 December 2017, Guangzhou (Canton), China, Procedia Eng. 210 2017, pp. 551–556. [11] BS 7910/2013, Guide to Methods for Assessing the Acceptability of Flaws in Metallic Structures, BSI British Standards. [12] EN 10002-1:1994, Tensile testing of metallic materials. Method of Test at Ambient Temperature. [13] EN 10025-1:2005, Hot rolled products of structural steels - part 1: general technical delivery conditions. [14] EN 10025-2 Hot rolled products of structural steel – technical delivery conditions for non-alloy structural steels. [15] J. Outinen, P. Makelainen, Mechanical properties of structural steel at elevated temperatures and after cooling down, Second International Workshop “Structures Fire”, Christchurch, 2002, pp. 273–289. [16] G. Zhang, M.C. Zhu, V. Kodur, G.Q. Li, Behavior of welded connections after exposure to elevated temperature, J. Constr. Steel Res. 130 (2017) 88–95. [17] F. Hanus, G. Zilli, J.M. Franssen, Experimental tests and analytical models for welds and grade8.8 bolts under heating and subsequent cooling, J.Struct. Fire Eng. 2 (2011) 181–194. [18] Structural integrity assessment procedure for European industry – SINTAP, Sub-Task 3.3. Report – Final Issue: Determination of Fracture Toughness Following the Charpy Impact Energy: Procedure and Validation. [19] I. Vučetić, S. Kirin, A. Sedmak, T. Golubović, M. Lazić, Risk management of a hydro power plant – fracture mechanics approach, Tech. Gazette 26 (2) (2019) 428–432. [20] TWI CrackWise Software ver.5.0 – Documentation. [21] T. Sedmak, R. Bakić, S. Sedmak, Lj. Milović, S. Kirin, Applicability of risk-based maintenance strategy to a penstock, Struct. Integrity and Life 12 (3) (2012) 191–196. [22] D. Radu, T.F. Gălățanu, S. Sedmak, “Structural Integrity of Butt Welded Connection after Fire Exposure” https://doi.org/10.1016/j.prostr.2018.12.227.
12