Axisymmetric Stress Analysis and Strength Evaluation of Bonded Shrink Fitted Joints of Circular Pipes Subjected to Internal Pressure and Tensile Loads

2000 ◽  
Author(s):  
Masahide Katsuo ◽  
Toshiyuki Sawa ◽  
Masahiro Yoneno

Abstract This study deals with the stress analysis and the strength evaluation of a bonded shrink fitted joint of circular pipes subjected to an internal pressure and a tensile load. In the analysis, two pipes and the adhesive are replaced with finite hollow cylinders, and the stress distributions in the joint are analyzed by using the axisymmetric theory of elasticity. From the numerical calculations, the following results are obtained: (1) Both the compressive and shear stresses at the interface between the adherend and the adhesive increase as Young’s modulus of the adherend increases. (2) The stress becomes singular at the edges of the interfaces. (3) The joint strength can be evaluated using the compressive and shear stresses near the edge of the interface. In the experiments, bonded shrink fitted joints consisting of dissimilar circular pipes were manufactured, and rupture tests of the joints were carried out by applying an internal pressure, and a tensile load to the joints. From the results, the joint strength of the bonded shrink fitted joint was found to be greater than that of the shrink fitted joint. Furthermore, the numerical results are in fairly good agreement with the experimental ones.

2004 ◽  
Author(s):  
Masahide Katsuo ◽  
Toshiyuki Sawa

The interface stress distributions between the coupling collar, the adhesive and the pipes of the joint subjected to an internal pressure, a temperature change and bending moments are analyzed by using the elastic finite element method (FEM). The experiment of the rupture test of the joints manufactured by pipes made of structural steel (S45C, JIS) and epoxide adhesive was carried out by applying the above loads to the joints. From the numerical calculations, the following results were obtained: (1) the stress distributes uniformly at the interface except near the edges, (2) the stress becomes singular at the edges of the interfaces and (3) the stress distribution at a half part of the interface increase as the external bending moments increase and also Young’s modulus of the adhesive increases. From the experiments, the following results were obtained: (1) the joint strength (evaluated as a 95% non-rupture probability) under both the internal pressure and the temperature change increases as the coupling length increases and (2) the joint strength under both the internal pressure and the temperature change decreases when the external bending moment is applied to the joint. Furthermore, the numerical results are in fairly good agreement with the experimental results.


Author(s):  
Toshiyuki Sawa ◽  
Yosuke Akita

This paper deals with a two-dimensional stress analysis of adhesive butt joints filled with elastic circular fillers in an adhesive subjected to an external tensile load. Similar adherends and an adhesive bond are replaced with finite strips in the analyses. Stress distributions in adhesive joints are analyzed exactly using the two-dimensional theory of elasticity. The effects of stiffness and number of fillers on the interfaces stress distributions and around the fillers with higher Young's modulus are shown in the numerical computations. It was seen that as an amount of number of fillers increased, the strength of joints was able to be more improved than that of the joints without a filler. For verification, experiments were carried out to measure the strains. The analytical results are in fairly good agreements with the experimental ones. In addition, for verification of the interface stress distribution, the analytical results are also compared with those obtained from the two-dimensional and three-dimensional FEM calculations. Fairly good agreements are seen between the analytical and the two-dimensional FEM results. However, the result from the present analysis is different from that of the three-dimensional FEM.


Author(s):  
Yuya Omiya ◽  
Toshiyuki Sawa

The pipe flange connections have been widely used in mechanical structures and these are mainly used under internal pressure and elevated temperature. Some investigations have been performed on mechanical behaviors such as the contact gasket stress distribution of pipe flange connection under internal pressure. However, the effects of elevated temperature on the sealing performance and mechanical behavior in the pipe flange connection have not been analyzed exactly. To design a pipe flange connection, it is necessary to know the effect of elevated temperature and the detailed flange strength for the hub and the contact gasket stress distributions in the connection. In this paper, the stress analysis of a pipe flange connection under elevated temperature and internal pressure is carried out taking into account a non-linearity and hysteresis of the gasket using the axi-symmetric theory of elasticity. The contact gasket stress distributions at the interfaces, the hub stress, in the pipe flange connection under elevated temperature and internal pressure are analyzed. Using the obtained contact gasket stress distributions, the amount of gasket leakage (He gas) is estimated. Leakage tests of the pipe flange connections were also conducted and the temperature of pipe flange connection was measured using thermoelectric couples. As a result, the difference of temperature between inner and outer is about 10%. The contact gasket stress was increased as the heated temperature was increased.


Author(s):  
Toshiyuki Sawa ◽  
Satoshi Nagata ◽  
Naofumi Ogata

This paper deals with the stress analysis of a pipe flange connection with a spiral wound gasket using the elasto-plastic finite element method taking account the hysteresis and the non-linearity in the stress-strain curve of the spiral wound gasket, when an internal pressure is applied to the pipe flange connections with the different nominal diameters from 2 to 20. The effects of the nominal diameter of the pipe flange on the contact stress distributions at the interfaces are examined. Leakage tests of the pipe flange connections with 3 and 20 nominal diameters were conducted and measurement of the axial bolt force was also performed. The results by the finite element analysis are fairly consistent with the experimental results concerning the variation in the axial bolt force. By using the contact stress distributions and the results of the leakage test, the new gasket constants are evaluated. As a result, it is found that the variations in the contact stress distributions are substantial due to the flange rotation in the pipe flange connections with the larger nominal diameter. In addition, a method to determine the bolt preload for a given tightness parameter is demonstrated.


Author(s):  
Toshiyuki Sawa ◽  
Kengo Kuwaki ◽  
Yukio Morozumi ◽  
Masahiko Okumura

In designing bolted joints, it is necessary to know the stress distributions in bolted joints. Recently, high strength bolts have been used with a higher bolt preload. As the results, the permanent set occurs sometimes at the bearing surfaces of clamped parts in a bolted joint. In addition, when an external load is applied to the bolted joint, the permanent set can be extended at the bearing surfaces. As the permanent set increases, the reduction in the bolt preload increases. Thus, it is important to estimate the reduction in the bolt preload from the reliability stand point. However, no study on the permanent set at the bearing surface under the external loading has been carried out. In this study, the stress distribution and the extension of the permanent set at the bearing surface of the bolted joint under the external tensile loading are examined using finite element Method (FEM), where two hollow cylinders are clamped with a hexagon bolt and a nut. The spring constants for the hexagon bolt and the clamped parts are analyzed using an axi-symmetrical theory of elasticity. Using the obtained results, an increment in the axial bolt force and the reduction in the bolt preload are estimated. For verification of the FEM stress analysis, the load factor of hexagon bolt was measured. The FEM results of the load factor (the increment in the axial bolt force) and the axial bolt force are in a fairly good agreement with the experimental results and the reduction of the axial bolt force. Finally, discussion is made on the appreciate bolt preload.


Author(s):  
Toshiyuki Sawa ◽  
Yuya Omiya

In designing a bolted joint, it is important to estimate an increment in axial bolt force when an external tensile load is applied to an assembly. The ratio of the increment Ft in the axial bolt force to the external tensile load W is called the load factor φ(= Ft/w). The formula φ = Kt/(Kt+Kc) proposed by Thum has been applied for estimating the value of the load factor φ, where Kt is the spring constant of bolt-nut system and Kc is the compressive spring constant of clamped parts. It has been found that the value of the load factor varies with the position of load application to the assembly. Then, a method to compensate Thum’s formula was proposed. However, this compensation is made empirically and the theoretical background is not made clear. In this paper, the concept of the tensile spring constant Kpt for clamped parts is introduced newly when an external load is applied to the outer circumference of clamped parts (hollow cylinders) and a method for estimating the value of the load factor exactly is proposed by using Kpt. The value of Kpt is analyzed using an axisymmetric theory of elasticity. For verification of the proposed method, experiments were carried out to measure the load factor. A fairly good agreement is seen between the analytical and the experimental results of the values of the load factor while the values of the load factor obtained from Thum’s formula were so different with the experimental results. The reason why the difference in the values of the load factor is substantial between values and the values obtained from Thum’s formula is elucidated. It is found that the value of the load factor decreases as the outer diameter of the hollow cylinder increases and the as thickness of the clamped parts decreases. In addition, FEM calculations for the load factor are carried out. The FEM results are in a fairly good agreement with the theoretical results.


Author(s):  
Toshiyuki Sawa ◽  
Takashi Kobayashi ◽  
Hirokazu Tsuji ◽  
Satoshi Nagata

This paper deals with the stress analysis of a pipe flange connection with a spiral wound gasket using the elasto-plastic finite element method when an internal pressure is applied to the pipe flange connections with the different nominal diameters from 2″ to 20″. The effects of the nominal diameter of the pipe flange on the contact stress distributions at the interfaces and the hub stress are examined. Leakage tests of the pipe flange connections with 3″ and 20″ nominal diameters were conducted and measurement of the axial bolt force was also performed. The results by the finite element analysis are fairly consistent with the experimental results concerning the variation in the axial bolt force (Load factor). By using the contact stress distributions and the results of the leakage test, the modified gasket constants are proposed and compared with PVRC values. As a result, it is found that the variations in the contact stress distributions are substantial due to the flange rotation in the pipe flange connections with the larger nominal diameter. The hub stress has been overestimated by ASME method. In addition, a method to determine the bolt preload for a given tightness parameter and a rational design method for pipe flange connections are demonstrated.


Author(s):  
Kazunari Kotajima ◽  
Yasuhisa Sekiguchi ◽  
Takashi Kobayashi ◽  
Yuta Ueda ◽  
Toshiyuki Sawa

This paper deals with an FEM stress analysis of stepped-lap adhesive joints of similar hollow cylinders under static tensile loadings. The effects of Young’s modulus ratio between the adherends and adhesive, the thickness of the adhesive, scarf angle, the number of steps, and singular stress on the interface stress distributions are calculated using FEM. The code of FEM employed is ANSYS. The singular stress is found to occur at the edge of the interfaces. The singular stress at the inside edge is larger than that at the outside edge. It is shown that the maximum principal stress at the edge of the interface decreases as Young’s modulus ratio between the adherend and the adhesive and the adhesive thickness decreases while it decreases as the number of steps increases. Using the obtained interface stress distribution, we can predict the joint strength. For verification of the strength prediction, experiments to measure the joint strength were carried out. The numerical results of the joint strength are in a fairly good agreement with the experimental results.


1998 ◽  
Vol 120 (5) ◽  
pp. 608-613 ◽  
Author(s):  
J. J. Garcia ◽  
N. J. Altiero ◽  
R. C. Haut

Stress analysis of contact models for isotropic articular cartilage under impacting loads shows high shear stresses at the interface with the subchondral bone and normal compressive stresses near the surface of the cartilage. These stress distributions are not consistent, with lesions observed on the cartilage surface of rabbit patellae from blunt impact, for example, to the patello-femoral joint. The purpose of the present study was to analyze, using the elastic capabilities of a finite element code, the stress distribution in more morphologically realistic transversely isotropic biphasic contact models of cartilage. The elastic properties of an incompressible material, equivalent to those of the transversely isotropic biphasic material at time zero, were derived algebraically using stress-strain relations. Results of the stress analysis showed the highest shear stresses on the surface of the solid skeleton of the cartilage and tensile stresses in the zone of contact. These results can help explain the mechanisms responsible for surface injuries observed during blunt insult experiments.


Sign in / Sign up

Export Citation Format

Share Document