Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 144 (2016) 1260 – 1269
12th International Conference on Vibration Problems, ICOVP 2015
Fluid-Structure Interaction based Adhesion Failure analysis of Bonded Tubular Socket joints R.R.Dasa*, N. Baishyab, K. Vermab, A. Choudharyb a
Department of Mining Machinery Engineering, Indian School of Mines, Dhanbad- 826004 School of Mechanical Engineering, KIIT-University, Bhubaneswar, Odisha, India, 751024
b
Abstract Computational Fluid Dynamics (CFD) and Fluid-Structure Interaction (FSI) based modelling and simulation techniques have been developed in the present research to investigate adhesion failure analysis in bonded tubular socket joints. Joint parameters like gap between the adherends, adhesive thickness, and overlap length have been optimized so as to minimize stress concentration effects localized to the joint edges under turbulent flow modelled through a realizable k-ε model. Adherendadhesive and socket-adhesive interfaces in the joint region have been identified to be the critical bond line interfaces prone to failure under turbulence. Three-dimensional out-of-plane stresses (σrr, τθr, τzr) critically responsible for adhesion failure have been studied leading to evaluation of Parabolic yield criterion used to identify the failure prone zones. © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license © 2016 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-reviewunder underresponsibility responsibility of the organizing committee of ICOVP Peer-review of the organizing committee of ICOVP 2015 2015. Keywords: Adhesion failure; CFD; Tubular Socket Joint
1. Introduction Piping systems form an integral part of construction and energy industries. Due to manufacturing constraints joints are inevitable in piping systems. Adhesive bonding is one of the most attractive methods for assembling pipes as it enhances a smooth load transfer with minimum stress concentration effects. Adhesive failure in bonded pipe joints, arising from fluid stress, direction of flow and other chemical agents present in the fluid, may manifest itself
* Corresponding author. Tel.: +91-8895556016; fax: +91-326-2296563. E-mail address:
[email protected]
1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICOVP 2015
doi:10.1016/j.proeng.2016.05.113
1261
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
in various forms such as adhesion failure, cohesion failure etc. Fracture study of these pipe joints has been a concern for many industrial practices, since this may shorten the service life of the pipes. Previous researches in this direction have majorly focused on flow induced vibration deformation [1], flow erosion [2] kind of problems. Zou and Taheri [5] found the shear stress distribution in the adhesive layer of bonded sand-wich pipe joints subjected to torsional moments effective strategy for analyzing the adhesive failure pattern and mechanism as a function of fluid pressure and flow direction within bonded pipe joints. The effectiveness of a computational model in predicting the aforementioned factors and hence the service life of pipes has also been analyzed in details. 2. Methodology They performed their analysis both analytically and computationally using the Finite Element method (FEM). Esmaeel and Taheri [6] found the effect of the pre-existing delamination on the state of stresses developed within the adhesive layer, focusing mainly on the adhesive layer shear and peel stresses. Adhesion failure analysis of a composite pipe joint under structural loading conditions has been studied by Das and Pradhan [3]. However, literature investigating the effect of pressure developed during fluid flow and direction of flow field on bonded joint failure is limited. In the present study, the effectiveness of epoxy resin as an adhesive in the steel pipes has been investigated. The properties of the adhesive as well as pipes and socket are given below, Table 1. Material properties of the epoxy resin used as adhesive S. No.
Property
Value
Units
1.
Density
1186
kg/m3
2.
Young’s Modulus
2800
MPa
3.
Poisson’s Ratio
0.4
4.
Bulk Modulus
4.7
GPa
5.
Shear Modulus
1
GPa
6.
Tensile Yield Strength
65
MPa
7.
Compressive Yield Strength
84.5
MPa
Table 2. Material properties of the structural steel S. No.
Properties
Values
Units
1.
Density
7850
kg /m-3
2.
Coefficient Of Thermal Expansion
1.2E-05
C-1
3.
Reference Temperature
22
°C
4.
Young's modulus
200
GPa
5.
Poisson's ratio
0.3
6.
Bulk modulus
166.67
GPa
7.
Shear modulus
76.9
GPa
8.
Tensile yield strength
250
MPa
9.
Compressive yield strength
250
MPa
10.
Tensile ultimate strength
460
MPa
In the present analyses, two steel pipes bonded to a steel socket through a thin layer of epoxy has been considered to be subjected to a fluid flow under turbulent boundary conditions as considered in the work of Hongjun et al. [4]. Geometry and boundary conditions of the bonded joint specimen along with finite element mesh has been shown in Fig. 1. Multizone hexagonal elements have been implemented for the FE simulation. The two pipes are similar with respect to geometry (length of each pipe (l) = 700mm, outer (do) and inner diameters (di) of each tube is 92 mm and
1262
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
87.62 mm, respectively). The steel socket has outer radius (R o) of 53.7 mm and a thickness (h2) of 4.38 mm. Gap between the adherends (g), coupling length (2c), and adhesive thickness (h 0) for the bonded joint has been optimized in the present analyses for improving the performance of the joint under turbulent flow conditions.
Fig. 1. Sectional view of the bonded tubular socket joint subjected to turbulent flow.
Fluent and ANSYS mesh-generators were employed to perform the fluid and solid geometry generation and meshing respectively. As shown in Fig. 2 (a) and 2 (b) are fluid mesh whereas (c) and (d) are for solid structure mesh. Suitable grid density is reached by repeating computations until a satisfactory independent grid is found.
a
b
c
d
Fig. 2.(a)-(b): Fluent FEM mesh, (c)-(d): Ansys FEM mesh
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
1263
3. Results and Discussion 3.1. Validation In order to verify the modeling procedure described in the present study, the results were compared with the solution developed by Hongjun Zhu et al. [4]. The solution developed based on the fluid flow in a T joint pipe has been validated. Hence, similar procedure was followed for the analysis of fluid flow in bonded tubular socket joint pipes as described in the previous section. The validated results are presented in Fig. 3.
a
b
c
d
Fig. 3. (a)-(b) pressure obtain by Hongjun Zhu et al. [4], (c)-(d) present results
3.2. Optimization of Adhesive layer Figure 4 depicts plots comparing failure index at different adhesive-adherend interfaces, viz., inner adhesiveadherend layer, mid adhesive layer and outer adhesive-adherend layers of the composite pipe. It was observed that middle and outer layers exhibited negligible failure index, while inner layer showed considerably higher values for the same. This may be interpreted as increased chance of failure in the inner layer. Failure index of composite pipes with different adhesive thicknesses have been calculated and are represented in Fig. 5. Failure index have been plotted against overlap length and the intensity of failure has been determined. Here the adhesive thickness is taken as from Fig. (a) to (e) as 0.10, 0.15, 0.20, 0.25, and 0.30 mm respectively. Failure index of various layers have been plotted in Fig. 6. Adhesive thicknesses varying from 0.1 to 0.3 mm were studied for all the layers – inner, middle and outer. When adhesive thickness was 0.15 mm, the failure index was found to be minimum in all the cases. Hence, adhesive thickness of 0.15 mm was considered for further studies.
1264
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
a
b
c
d
Fig. 4. Two dimensional plot for failure index vs overlap length for (a)inner layer, (b) middle layer, (c) outer layer and (d) comparision of all
a
b
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
c
d
e
Fig. 5. Two dimensional plot of failure index for different adhesive thickness(a) 0.1mm, (b) 0.2mm, (c) 0.3mm,( d) 0.15mm and (e) 0.25mm
a
b
1265
1266
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
c
Fig. 6. Two dimensional optimized plot for different adhesive thickness at different layer.
3.3. Optimization of overlap length a
b
c
d
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
e
f
g
h
i
j
1267
1268
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
k
Fig. 7. Two dimensional plot of failure index for different overlap lengths (a) 20cm, (b) 22cm (c) 24cm, (d) 26cm, (e) 28cm, (f) 30cm, (g) 32cm, (h) 34cm,( i) 36cm, (j) 38cm, and (k) 40cm
The failure index is compare with a changing overlap length varying from 20cm to 40cm with a increment of 2cm each time. From the Fig 8. it was seen that the failure index was minimum when the overlap was 32 cm. So it was considered that the optimized overlap length is 32cm for the present study.
Fig. 8. Two dimensional optimized plot for different overlap length.
4. Conclusion The thickness of the adhesive layer and overlap length of the socket have been optimized using ANSYS 14. The computational model used in the study has been validated. Failure index of the considered steel pipes using epoxy as adhesive was found to be minimum in case of adhesive thickness of 0.15 mm and socket overlap length of 32 cm. References [1] S.W. Myoung, I.L. Hoe, S.K.You, A study of the deformation of flexible pipes buried under model reinforced sand, KSCE J. Civ. Eng. 8 (2004) 377–385. [2] Y.M. Ferng, L. Binhong, Predicting the wall thinning engendered by erosion–corrosion using CFD methodology, Nucl. Eng. Des. 240 (2010) 2836–2841.
R.R. Das et al. / Procedia Engineering 144 (2016) 1260 – 1269
[3] R.R. Das, B. Pradhan, Finite Element Based Design and Adhesion Failure Analysis of Bonded Tubular Socket Joints Made with Laminated FRP Composites, J. Adhes. Sci. Technol. 25 (2011) 41–67 [4] H. Zhu, W. Zhang, G. Feng, X. Qi, Fluid–structure interaction computational analysis of flow field, shear stress distribution and deformation of three-limb pipe, Engg. Fail. Ana. 42 (2014) 252–262. [5] G.P. Zou, F. Taheri, Stress analysis of adhesively bonded sandwich pipe joints subjected to torsional loading, Int. J. Sol. Struc. 43 (2006) 5953–5968. [6] R.A. Esmaeel, F. Taheri, Influence of adherend’s delamination on the response of single lap and socket tubular adhesively bonded joints subjected to torsion, Compos. Struc. 93 (2011) 1765–1774.
1269
