The effect of dynamic bending moments on the ratchetting behavior of stainless steel pressurized piping elbows

Transcription

1 International Journal of echanical Engineering and Alications 2014; 2(2): Published online ay 30, 2014 (htt:// doi: /j.ijmea The effect of dynamic bending moments on the ratchetting behavior of stainless steel ressurized iing elbows S. J. Zakavi, V. Golshan * Faculty of ech.eng, Uni.of ohaghegh Ardabili, Ardabil, Iran address: zakavi@uma.ac.ir (S. J. Zakavi), Vahid.golshan@yahoo.com (V. Golshan) To cite this article: S. J. Zakavi, V. Golshan. The Effect of Dynamic Bending oments on the Ratchetting Behavior of Stainless Steel Pressurized Piing Elbows. International Journal of echanical Engineering and Alications. Vol. 2, No. 2, 2014, doi: /j.ijmea Abstract: In this aer the ratchetting behavior of four airs of stainless steel, long and short radius welding elbows is studied under conditions of steady internal ressure and in-lane, resonant dynamic moments that simulated seismic excitations. The finite element analysis with the nonlinear kinematic hardening model has been used to evaluate ratchetting behavior of the elbow under mentioned loading condition. Stress strain data and material arameters have been obtained from several stabilized cycles of secimens that are subjected to symmetric strain cycles. The results show that the maximum ratcheting strain occurred mainly in the hoo direction at flanks. Ratcheting strain rate increases with increase of the bending loading level at the constant internal ressure. The results show that the FE method gives over estimated values comaring with the exerimental data. Keywords: Ratchetting, Pressurized Elbows, Cyclic Bending oment, Strain Hardening odel, Stainless Steel 1. Introduction Ratchetting, namely the accumulation of lastic deformation, occurs when the structures are subjected to a rimary load with a secondary cyclic load if the alied loads are high enough to make the structures yield. In the materials or structures subjected to a cyclic stressing with non-zero mean stress, a cyclic accumulation of inelastic deformation will occur if the alied stress is high enough which is called ratcheting. Plastic analysis of structures is usually studied using the isotroic and kinematic hardening theories. Ratcheting descrition in terms of the conventional equations is mainly related to kinematic hardening. Pressurized iing as the most basic structures in chemical industries and ower lants are subjected to variable mechanical and thermal loads which often have a cyclic nature. Accurate determination of the lastic strain rate in each cycle is imortant subject in ressurized iing of ower lant and chemical industries. Collase, ratcheting, and fatigue interaction may occur and lead to failures. Thus, during the design of ressurized iing today, esecially which used in chemical industries and ower lant comonents, ratcheting and ratcheting fatigue must be under consideration. The literature review shows that accurate closed form solutions may not be found to analyse the ratchetting behavior of the ressurized ies under cyclic bending loading which can be caused by seismic loads. However, aroximate solutions have been develoed by Tasnim et al. (1992, 2008), Rahman et al. (2008), Zakavi et al. (2010), Chaboche (1989,1991,2008), Chen et al. (2005,2006, 2013), Abdel-Karim (2005), Ohno et al. (1993), Bari and Hassan (2000,2001,2002) which can be used to calculate the induced incremental lastic strains caused by ratchetting. Exerimental works to study the ratchetting of the elbow ies have also been carried out by Chen et al. (2013) and Yahiaoui et al., (1996). 2. aterials and ethods In this aer, a finite element code, ABAQUS, is used to study the ratchetting of stainless steel ressurized elbow subjected to cyclic bending loading. In the exerimental tests (Yahiaoui et al., 1996), series of tests have been undertaken subjecting ressurized elbow secimens to rising amlitude dynamic (5 Hz, the resonant frequency) bending moments. Then, by conducting a series of finite element runs based on the nonlinear kinematic hardening model using the ABAQUS, the exerimental tests are modeled and ratchetting data obtained. Then the two sets of results are

2 32 S. J. Zakavi and V. Golshan: The Effect of Dynamic Bending oments on the Ratchetting Behavior of Stainless Steel Pressurized Piing Elbows comared with each other. 3. Nonlinear Kinematic Hardening odels The kinematic hardening models are used to simulate the inelastic behavior of materials that are subjected to cyclic loading. The use of lasticity material models with isotroic tye hardening is generally not recommended since they continue to harden during cyclic loading. The isotroic hardening model always redicts shakedown behavior, if cree is not considered (ahbadi et al., 2006). The kinematic hardening lasticity models are roosed to model the inelastic behavior of materials that are subjected to reeated loading. For examle, the Armstrong- Frederick (1996) kinematic hardening model is suggested for the nonlinear strain hardening materials. Based on the Armstrong-Frederick nonlinear kinematic hardening rule, many constitutive models have been constructed to simulate the unaxial and multiaxial ratcheting of materials characterized by cyclic hardening or cyclic stable behaviors. The results of these models are discussed for structures under various tyes of cyclic loads in references. The classical linear kinematic hardening rule and different nonlinear kinematic hardening models are available for the lastic analysis of structures. The nonlinear kinematic hardening model was first roosed by Armstrong and Frederick (1996). Nonlinearities are given as a recall term in the Prager rule. So that the transformation of yield surface in the stress sace is different during loading and unloading. This is done by assuming different hardening modulus in loading and unloading conditions. The yield function for time indeendent lasticity, using the von-ises yield criterion, is exressed as (Lemaitre and Chaboche, 1994): f = J ( X ) k 2 σ (1) where X is the back stress tensor, k the initial size of the yield surface, and denotes the von-ises distance in the deviatoric stress sace: 1 3 J 2 ( σ X ) = σ X : 2 ( σ X ) 2 where σ and X are the stress and back stress tensors, and σ and X are the stress and back stress deviatoric tensors in the stress sace, resectively. The nonlinearities are given as a recall term in the Prager rule: (2) 2 (3) dx = Cdε γxdε 3 where dε is the equivalent lastic strain rate, C and γ are two material deendent coefficients in the Armstrong Frederick kinematic hardening model, and γ = 0 stands for the linear kinematic rule. The normality hyothesis and the consistency condition df = 0 lead to the exression for the lastic strain rate (Lemaitre and Chaboche, 1994): ( f ) f H f f d ε = d = : dσ σ h σ σ where H denotes the Heaviside ste function: H ( f ) = 0 if f 0, H ( f ) = 1 if f 0 and the symbol denotes the accauley bracket, i.e., = hardening modulus h becomes: σ X h = C γ X : 2 k u ( u + u ) 2 (4). The 3 (5) In the case of tension comression, the criterion and the equations of flow and hardening can be exressed in the form (Lemaitre and Chaboche, 1994): f = σ X k = 0 (6) σ X σ X dσ ε = 1 dσ = (7) h k k h d P dx = Cdε γx dε h = C γ XSgn σ ( X ) The evolution equation of hardening can be integrated analytically to give: c X = ν + X γ c ν ex γ [ νγ ( ε ε ] 0 0 (8) (9) (10) where ν = ± 1 according to the direction of flow, and ε 0 and X 0 are the initial values. For examle at the beginning of each lastic flow. 4. Review of Exerimental Set-U (Yahiaoui et al., 1996) The exerimental set-u for testing iing elbows under in-lane bending has been reorted in reference (Yahiaoui et al., 1996). The nominal ie size was 2 inch NPS corresonding to an outside diameter of 60.3 mm. Elbow geometry and the arameters and section of the results is shown in Fig. 1. The comonent identification and relevant dimensions of the elbows are given in Table 1. Tyical stress-strain curves for the stainless steel materials (304L) are shown in Fig. 2. For stainless steel (SS304) material, secification and roerties obtained by tensile tests data is

3 International Journal of echanical Engineering and Alications 2014; 2(2): given in Table 2. Pairs of 90 welding elbows were, for symmetry, tested simultaneously. The test comonents were ressurized indeendently to their design ressure, calculated using the ASE Code formula. The internal ressure was closely monitored and ket constant during testing. The frequency and design ressure of elbows is given in Table 3. Hoo strains at the crown reading from the gauges attached to the elbows at the crown ositions(fig. 1). a = axial direction h = hoo direction ϕ = angular osition around mid-circumference section containing E, C and I = 0 at C and ositive towards E C = crown ositions(ϕ = 0, 180 ) E = extrados(ϕ = +90 ) F = flank regions (defined by ϕ = ±45 about the crowns) I = intrados (ϕ = -90 ) Fig 1. Elbow geometry, stress directions, angular coordinate and definition of imortant locations around the bend(yahiaoui et al., 1996). Fig 2. Stainless steel used to manufacture the tubular secimens. 5. Finite Element Arrangement For all secimens the finite element code, ABAQUS, was used to study ratcheting behavior of ressurized elbows under simulated seismic bending moments. The elbows have a 1.50 m long iework modeled by 28 elements. The most accurate element in the ABAQUS code for this tye of structural system considering beam elements, ie elements and elbow elements is the elbow element. Four tyes of elbow elements are available in the ABAQUS library, of which the two-noded element ELBOW31 was found to give the best results. ELBOW31 rovides accurate redictions of the behavior of elbows under monotonic loading. In this aer, in order to redict the results, the tensile secimens were roduced accordance with the materials roerties that is given in Table 2. The cyclic nonlinear constitutive model used in the FE analyses in this study is derived from multiaxial formulations by Armstrong and Frederick. It is recommended that the model be calibrated with exerimental data that is close to the exected strain range and loading history of the alication. Stress-strain data is obtained from several stabilized cycles of secimens that are subjected to symmetric strain cycles. The material arameters C and γ determine the kinematic hardening comonent of the model. Three different ways of roviding data for the kinematic hardening comonent of the mode can be given: using half-cycle test data, using single stabilized cycle, or test data obtained from several stabilized cycles. Stress-strain data can be obtained from several stabilized cycles of secimens that are subjected to symmetric strain cycles. The calibration rocedure consists of several cylindrical bar tests, one of which subjected to monotonic tension until necking and others were under symmetric strain-controlled exeriments with different strains. During these calibration tests, the stress state must remain uniaxial. From symmetric strain-controlled exeriments, the equivalent lastic strain equals the summation of the absolute value of the change in longitudinal lastic strains: where ε = εi total strain, i ε( i) = εi σ i ex. E (11) σ ex is the measured stress and E is the elastic modulus. The equivalent back stress, X, equals one-half of the difference in yield stress between the end of the tensile loading and first yield of the subsequent comressive loading. These results, corresonding ( X, ε ) data airs may be lotted, and the kinematic hardening arameters, C and γ, may be calculated by fitting Equation (10) to the data and selecting arameters minimize the sum of the square of the error between Equation (10) and the data. For symmetric strain-controlled exeriments, a tyical curve with strain amlitude ± 0.75 is shown in Fig. 3. The results gained exerimentally and from FE using kinematic hardening model with below. C = Pa, γ = are detailed

4 34 S. J. Zakavi and V. Golshan: The Effect of Dynamic Bending oments on the Ratchetting Behavior of Stainless Steel Pressurized Piing Elbows Fig 3. Tyical cyclic loading curve with strain amlitude ± 0.75 for SS304 L. Table 1. Comonent identification and geometry (Stainless steel) (Yahiaoui et al., 1996) Fig 4. FE analysis dynamic bending moment resonses For SSSI at a dynamic bending moment of Comonent identification* Thickness (mm), (schedule) Bend Radius (mm) Bend characteristic H = t R/r2 Radius ratio b = R/r SLSI 3.91, (40) SLXI 5.54, (80) SSSI 3.91, (40) SSXI 5.54, (80) *Comonents are labelled by a four-character coding: First character: C for carbon or S for stainless steel; Second character: L for long or S for short radius bends; Third character: S for standard weight or X for extra strong; Fourth character: I is used here to denote in-lane bending to differentiate from another rogramme concerned with out-of-lane 0 Loading. Table 2. aterial (SS304) roerties obtained by tensile tests(yahiaoui et al., 1996) Young s modulus Ultimate stress aterial roerties 2% Proof stress Elongation at failure (%) 1 2 S m = in σult, σ y GPa 597Pa 292Pa 81% 161Pa Table 3. The frequency and Design ressure(yahiaoui et al., 1996) Comonent identification* Frequency (Hz) Design ressure (Pa) SLSI SLXI SSSI SSXI Exerimental and FE results Detailed results will be resented for four of the secimens tested (SSLS, SLXI, SSSI and SSXI) and summary results will be given for all tests conducted. It is erhas useful to resent, bending moment resonse obtained by the FE analysis is shown in Figure. 4. Tyical strain resonse in resence of ratcheting in FE results is shown in figure. 5. Fig 5. Tyical strain resonse in resence of ratcheting in FE results for secimen SSSI at a dynamic bending moment of 4090 N.m. All of the exerimental ratchetting results and FE analysis on secimens SSSI, SLSI, SLXI and SSXI are lotted in Figs. 6 a, b and 7 a, b, resectively. Here, the strain for each cycle has been calculated as the average over the eriod of the test and lotted against /0.2. Figs. 6 and 7 show the data recorded for the crown surface. A tyical set of results for secimen SLSI is shown in Fig. 8, which includes results from the FE analysis using with the nonlinear kinematic hardening model. Here, the ratchet strain er cycle averaged over the first 20 s of excitation has been lotted against increasing /0.2 ratios. The same information obtained for secimens SSSI, SLXI and SSXI are illustrated in Figs In Table 4, the ratchet strains found exerimentally over a 20 s test eriod and by FE analysis, for the same eriod, for secimen SSSI are summarized. 7. Results, Discussion and Conclusions Results from tests on four airs stainless steel some long and short radius iing elbows with two different

5 International Journal of echanical Engineering and Alications 2014; 2(2): thicknesses have been resented. The comonents were subjected to steady internal ressure and resonant, dynamic in-lane bending moments. The exerimental work reorted here rovides reliable data which can be used to judge the value of the FE analysis using the ABAQUS ackage. However, it should be noted that the exerimental work used a rising amlitude technique which may effectively reduce the ratchet strain at any articular dynamic bending moment. It is ossible that those tests conducted at low amlitude will harden the material sufficiently to reduce the ratchet strains observed at higher amlitudes. It is not ossible to quantify the ossible magnitude of this effect. This ossible effect would not have influenced the dynamic bending moment at which ratchetting was first observed. Tyical data obtained exerimentally and from FE model for secimens SLSI to SSXI on the crown surface are shown in Fig. 6 and 7. The exerimental and FE results illustrated by Figs 8-11 show the onset of ratchetting for stainless steel elbow secimens occur at 0.6 / l 0.7 and 0.5 / l.75, resectively. Comlete set of data for secimens SSSI is (a) (b) resented in Table 4. In this study, Stress-strain data and material arameters have been obtained from several stabilized cycles of secimens that are subjected to symmetric strain cycles. The rate of ratchetting deends significantly on the magnitude of the internal ressure, dynamic bending moment and material constants for nonlinear kinematic hardening model. The results show that initial the rate of ratchetting is large and then it decreases with the increasing of cycles. The FE model redicts the hoo strain ratchetting rate to be greater than that found exerimentally. Therefore, The results obtained from the FE method gives over estimated values comaring with the exerimental data. Acknowledgments Areciation is exressed to the technical staff of the Alied echanics Division of the Deartment of echanical Engineering at the University of ohaghegh Ardabili (Iran) for their assistance with the work. Fig 6. (a) Exerimental ratchetting data (Yahiaoui et al., 1996) and (b) FE analysis against moment levels. (a) (b) Fig 7. (a) Exerimental ratchetting data (Yahiaoui et al., 1996) and (b) FE analysis against moment levels

6 36 S. J. Zakavi and V. Golshan: The Effect of Dynamic Bending oments on the Ratchetting Behavior of Stainless Steel Pressurized Piing Elbows Fig 8. Exerimental and FE ratchet strains(comonent SLSI) Fig 9. Exerimental and FE ratchet strains(comonent SSSI) Fig 10. Exerimental and FE ratchet strains(comonent SLXI) Fig 11. Exerimental and FE ratchet strains(comonent SSXI) Table 4. Exerimental and FE ratchetting data for secimen SSSI. Dynamic bending moment (N.m) ye l Exerimental ratchetting data ( µε/cycle) FE analysis against moment levels ( µε/cycle) Notation t r E y Cylinder Thickness Pie mean radius Young's modulus Dynamic bending moment Yield moment l S Limit moment of elbow m Allowable design stress intensity y Thickness correction factor = 0.4 σult Tensile stress σ y Yield stress f h J2 σ σ X X k C,γ ε P Yield surface Bend characteristic = tr/r2 Von-isses yield function Stress tensor Stress deviatoric tensor Back stress tensor Back stress deviatoric tensor Initial size of the yield surface aterials constants for kinematic hardening Plastic strain tensor ε P Equivalent lastic strain ϕ Angular osition around the circumference of the bend ( = 0 at the crown)

7 International Journal of echanical Engineering and Alications 2014; 2(2): References [1] Abdel-Karim,., Numerical integration method for kinematic hardening rules with artial activation of dynamic recovery term. Int. J. of Plasticity, 21: [2] Armstrong, P.J., Frederick, C.O., A mathematical reresentation of the multi axial Bauschinger effect. CEGB Reort RD/B/N 731, Central Electricity Generating Board. The reort is reroduced as a aer: aterials at High Temeratures, 24(1):1-26. [3] Bari, S., Hassan, T., Anatomy of couled constitutive models for ratcheting simulation. Int. J. of Plasticity, 16: [4] Bari, S., Hassan, T., Kinematic hardening rules in uncouled modeling for multiaxial ratcheting simulation. Int. J. of Plasticity, 17: [5] Bari, S., Hassan, T., An advancement in cyclic lasticity modeling for multiaxial ratcheting simulation. Int. J. of Plasticity, 18: [6] Chaboche, J.L., Time-indeendent constitutive theories for cyclic lasticity. Int. J. of Plasticity, 2: [7] Chaboche, J.L., On some modifications of kinematic hardening to imrove the descrition of ratcheting effects. Int. J. of Plasticity, 7: [8] Chaboche, J.L, A review of some lasticity and viscolasticity constitutive theories, Int. J. of Plasticity, 24: [9] Chen X, Gao B, Chen G.,2005. ultiaxial ratcheting of ressurized elbows subjected to reversed in-lane bending. J Pres Eq Syst;3: [10] Chen X, Gao B, Chen G.,2006. Ratcheting study of ressurized elbows subjected to reversed in-lane bending. J Pres Ves-Trans ASE;128: [11] Chen, Xiaohui., Chen, Xu., Yu, Dunji., Gao, Bingjun., Recent rogresses in exerimental investigation and finite element analysis of ratcheting in ressurized iing, Int. J. of Pressure Vessels and Piing, 101: [12] Lemaitre, J and Chaboche, J.L, echanics of Solid aterials, Published by Cambridge University Press, ISBN , , 584 ages. [13] ahbadi, H. and Eslami,.R., Cyclic loading of thick vessels based on the Prager and Armstrong Frederick kinematic hardening models. Int. J. of Pressure Vessels and Piing, 83: [14] Ohno, N., Wang, J.D., Kinematic hardening rules with critical state of dynamic recovery, art I: formulations and basic features for racheting behavior. J. of Plasticity, 9: [15] Rahman, S.., Hassan, T., Corona, E., 2008, Evaluation of cyclic lasticity models in ratcheting simulation of straight ies under cyclic bending and steady internal ressure. Int. J. of Plasticity, 24: [16] Tasnim, H., Kyriakides, S., Ratcheting in cyclic Plasticity, art I: uniaxial behavior. Int. J. of Plasticity, 8: [17] Tasnim, H., Corona, E., Kyriakides, S., Ratcheting in cyclic lasticity, art II: multiaxial behavior. Int. J. of Plasticity, 8: [18] Tasnim, H., Lakhdar, T., Shree, K., Influence of non-roortional loading on ratcheting resonses and simulations by two recent cyclic lasticity models, Int. J. of Plasticity, (24) [19] Yahiaoui, K., offat, D.G., oreton, D.N., Resonse and cyclic strain accumulation of ressurized iing elbows under dynamic in lane bending, J. of strain analysis vol 31 No 2. [20] Zakavi, S.J., Zehsaz,., Eslami,.R. (2010) The ratchetting behavior of ressurized lain iework subjected to cyclic bending moment with the combined hardening model. Nuclear Engineering and Design, 240(4),