International Journal of Engineering and Technology sciences (IJETS) ISSN 2289-4152 Academic Research Online Publisher Research Article Experimental and numerical investigation of tube sinking of rectangular tubes from round section Morteza Hosseinzadeh a, *, Ali Zamani b a Department of Mechanical Engineering, Ayatollah Amoli Branch, Islamic Azad University, Amol, Iran b Department of Mechanical Engineering, Sama Technical and Vocational Training College, Ayatollah Amoli Branch, Islamic Azad University, Amol, Iran * Corresponding author. Tel.: +989125366319; E-mail address: m_hoseinzadeh59@yahoo.com A b s t r a c t Keywords: Tube drawing, Thickness distribution, FEM simulation, Drawing force. Accepted:06 December2014 Metal tubes with various cross-sections are extremely used in several engineering application such as heat exchanger, aerospace and lightweight structure. Nowadays, Tube drawing as desirable approach techniques is used by a lot of industries for producing tubes in various sizes and cross-sections. In this process, tube passes through the conical die to changes cross-section and dimension. Deformation in tube drawing is influenced by a number of parameters; friction coefficient, drawing speed, die semi- angle and reduction in area ratio are the most significant. In this study, single-pass tube sinking is conducted in forming of a square tubular section from round tube of pure copper. In addition, the influence of die semi angle on thickness distribution and drawing force is determined experimentally and numerically by applying finite element simulation. Finally, it was discovered by increasing die semi angle, thickness distribution was increased and drawing force decreased. Academic Research Online Publisher. All rights reserved. 1. Introduction At the present time, tubular materials are widely used in the various fields of industry such as transport vehicles, medical appliances, electric products etc. Cold tube drawing is a technique that can be applied for manufacturing of tubular shafts with several advantages such as high productivity and cost reduction. This process involves reducing the diameter or cross-section by passing through a conical drawing die. Tube drawing process increases mechanical properties by strain hardening and can produce tube with irregular shape [1].
There are three basic types of tube drawing processes consist of tube sinking, plug drawing (fixed plug and floating plug) and moving mandrel drawing which are shown in Figure 1 [2]. Fig. 1: Different methods of tube drawing: A-sinking, B-fixed plug, C- floating plug, D- moving mandrel [2] For each type of tube drawing, there are advantages and disadvantages which are mentioned in the following: Tube sinking (Figure 1-A): In this method there is no internal tooling (internal wall is not supported) and the wall then thicken slightly. Only tube diameter is reduced and tube wall thickness remains almost constant and in some cases may increases a little. The obtained external surface in this method is smooth but the interior surfaces are rough. Sinking drawing is used only for special cases such as the production of columns and thick-wall tubes [1]. Tube drawing with fix plug (Figure 1-B): This method used conical plug to control size and shape of inside section. So it has greater dimensional accuracy than tube sinking. The initial tube after placed on the rod which is fixed at the end of the bar is drawn within a die that its result is reduction of the tube diameter and the wall thickness. This method is used usually for making thin wall and large diameter tubes [1]. Tube drawing with floating plug (Figure1-C): In this method, a taper plug is placed inside the tube in deformation area. As the tube is drawn, the plug and the die act together to reduce both inside and outside dimension and geometry. This method used most for cold drawing of long tube. Using this method, we can produce the tubes of any desired length and reduction of slightly tube wall-thickness [1]. Tube drawing with moving mandrel (Figure 1-D): In this method, draw force is transmitted to the tube by the pull on the exit section and by the friction forces acting along the tube-mandrel interface. The mandrel also imparts a smooth inside finish surface of the tube but mandrel removal disturbs dimensional tolerance [1]. 408 P a g e
Several researcher works on the tube drawing. Bayoumi et al [3] have been used rollers to change circular tube section into the pentagon section. He obtained that by increasing the number of sides, drawing force, the force on the rollers, decreases and by decreasing corners radius, drawing force increases. Furthermore, Karnezis and Farrugia [4] have optimized drawing force and heat distribution by finite element modelling of cold drawing. Also, they found that the residual stress is reduced with increasing distance from the interior surface. Wang et al [5] have evaluated the effect of some parameters such as reduction ratio and drawing speed on the drawing force. Their results indicate that by increasing the reduction ratio, drawing force increased. Optimization of process tool geometry has also been proposed by Bland and colleagues [6] for drawing of aluminium tubes 6063. They developed a new method using a combination of two tube shaping methods with a fixed plug and tube sinking drawing. Their work results in decreasing of the number of shaping steps from two-step to one-step. Reduction ratio and tube initial thickness effect on thickness distribution and inside/outside surface quality have investigated by Yoshida et al [7] in the three types of drawing process (fixed plug, floating plug and tube sinking). Bui, Q.H., et al [8] have investigated the formability limit of aluminium tubes drawn with variable wall thickness experimentally and numerically. They also found that in this process, the tube diameter has more effect compared with the tube thickness on drawing force. Producing tube with several cross sections by using of tube drawing process has some advantages in compare with other process such as extrusion. It consist of good quality of internal and external surface, producing tubes with long length, cutting down of the consumption of energy, the reduction in production cost and etc. Also, It can be used for small amount of production, tubes with large diameters and this process will done at cold situation therefore temperature controlling is more simpler than the hot working process. In the present paper, the effect of die semi-angle on the thickness distribution and drawing force is conducted for the problem of cold tube sinking of rectangular tubular section from round tube numerically and experimentally. 2. Experimental Procedures and Tooling Pure copper tubes with 12.7 x 0.7 mm (mean outer diameter x mean wall thickness) were reduced to square cross section with side length 7 x 7 mm, which represents a drawing pass with 28.2% of area reduction. To determine mechanical properties, uni-axial tensile test was derived at room temperature according to ASTM-A370 standard. The samples were drawn up to fracture point at constant peed. Based on load-displacement curve obtained from tensile test, true stress-strain curve for this material 409 P a g e
were obtained and presented in Figure 2. The chemical composition in wt.% of this tube was obtained by quantum test and is given in Table 1. Table 1: chemical composition of copper tube Sn=0.002 Fe=0.0 P=0.0 Zn=0.011 Ni=0.030 Si=0.0 Pb=0.001 Al=0.0 Mn=0.009 Cr=0.00 Mg=0.003 As=0.00 Cu= 99.9 Fig. 2: true stress - strain curve of pure copper tube Tube drawing was realized using a small drawing bench up to maximum stroke of 2 m illustrated in Figure 3. It equipped by chain puller driver, an electrical motor with 3 KW power, a gear box joint to electro motor and an inverter to control drawing speed. The swaged end of tube passes through the dies and is clamped to the jaws of the draw head is moved by a chain drive and so, the drawing bench can draw it in several speeds. Fig. 3: Tube drawing bench and its components. 410 P a g e
Schematic view of used die geometry is shown in Figure 4. The region A is die entrance with a conical transition from circular section to square section and with die semi-angle (α). The region B is bearing length with square cross section. In this area final geometry and dimension of tube cross section is formed. The region C is die exit with a relief angle. This angle is higher than die semi-angle to reduce surface contact and friction between tube and die wall. Fig. 4: Schematic view of the used die As shown in Figure 5, for performing experimental test four dies were used, all of tungsten carbide with different die semi-angle consists of 2.5, 5, 7.5 and 10 degree and bearing length of 4 mm. All of these dies have circular input section and square output section. Tubes were drawn at constant speed of 7.5 m/min. Fig. 5: A Used dies with different half-angles B The input and output section of used dies The specifications of used die are given in Table 2. 411 P a g e
Table 2: The specifications of used dies Characteristic Value bearing length Die semi-angle Relief semi-angle 4 mm 2.5-5-7.5-10 degree 30 degree Diameter of input circle 14 mm 3. Finite Element Simulation 3D finite element model based on ABAQUS EXPLICIT 6/11 code has been carried out to simulate the cold drawing process. To obtain steady-state conditions, away from the tube ends, a piece of tube 200 mm long was modeled. In the simulation, an isotropic elastic-plastic deformable objective was chosen as a material model for the tube and the die was assumed to be as discrete rigid. To introduce tube properties to the software, true stress-strain data obtained from uniaxial tensile test were used. The tube is modelled using four-node tetrahedral shell element (S4R4). The Coulomb friction model with a surface to surface contact is used to analysis of contact. The friction coefficient between tube and die is considered 0.2. The FE example of tube sinking are shown in Figure 6. (a) Fig. 6: a) Created finite element model in software b) tube formation steps (b) 412 P a g e
4. Results and Discussions The above experimental tooling and FEM analysis is applied to investigate different parameters effect in tube sinking. Figure 7 presents the initial and the drown tube. Fig. 7: Initial and drawn tube Fig. 8: Measurement route of thickness distribution In order to measure the thickness distribution, all drawn tubes were cut into two parts by using the wire cut machine from one of their symmetry plane and the thickness was measured along axial and circumferential axis. This is illustrated in Figure 8. Firstly, in order to validate the simulation, obtained experimental thickness distribution along axial direction is compared with simulation results in die semi-angle 2.5 degree. Figure 9 shows this comparison. Fig. 9: Comparison of experimental and simulation results along axial direction As it can be found from Figure 9, the trend of curves is similar and it indicates the experimental and numerical results are in fair agreement. Also it is understood that tube thickness increased due to area reduction and lack of plug. 413 P a g e
4.1. Effect of Die Semi-Angle on Thickness Distribution Figure 10 A and B shows the effect of die semi-angle on the axial thickness distribution obtained from experimental and simulation test. a b Fig. 10: Comparison of thickness distribution results along axial axis A experimental and B simulation As the die semi angle increases to 10 degree, as shown in Figure 10, it was understood that the tube thickness increases. It can be explained by the fact that by reducing die semi-angle, the contact length between the die and the tube increases. This causes higher friction and then further thickness reduction. To better explanation, the strain distribution was obtained along axial axis numerically. The curves in the Figure11 show the influence of the die semi-angle on the strain distribution. 414 P a g e
Fig. 11: The strain distribution along axial axis After studying the effect of die semi-angle on the axial thickness distribution, the thickness distribution along the tube circumferential has been investigated. The sample curve of circumferential thickness distribution has been illustrated in the Figure 12. Fig. 12: Comparison of experimental and simulation results of thickness distribution along tube circumferential To better analysis, this curve is divided into several areas. Figure 13 shows these areas related to divisions on the tube cross section profile. 415 P a g e
Fig. 13: divided areas corresponding to the Figure 12 As it is observed in Figure12, the thickness distribution curve has several peaks which are related to the regions A, C, E and G in the Figure 13. With changing from circle to square cross section, folding occurs and material exposes in excessive compression force and thickness increases. In the other side, the minimum thickness distribution is in the regions H, B, D and F. In these areas because of contact between tube and die and lack of folding, the tube thickness doesn't increase considerably. Figure14 A and B shows the influence of die semi-angle on the thickness distribution along tube circumferential experimentally and numerically respectively. As the curves indicate, by increasing the die semi-angle, a significant change does not occur in the thickness distribution and it increases slightly. 416 P a g e
(a) (b) Fig. 14: The comparison of thickness distribution along circumferential, A experimental and B simulation 4.2. Effect of Die Semi-Angle on the Drawing Force Figure 15 shows the effect of die semi-angle on the drawing force. As it can be seen, reducing of die semi-angle, increases drawing force. The die with semi-angle 2.5 degree has dedicated the maximum force to itself. 417 P a g e
Fig. 15: Comparison of drawing force in different die semi angles Furthermore, in Figure 16 the maximum value of drawing force by considering various die semiangles is shown. As it is shown, the increase of die semi-angle causes the reducing of drawing force. This is due to the shorter contact length between die and tube and so lower friction force. Also it is observed that with increasing from 2.5 to 5 degree, the curve's slope is very high whereas in die semiangle more than 5 degree it is low. It is because of, in die semi angle more than 5 degree, friction decreases but tube forming in the die becomes harder. 5. Conclusions Fig. 16: Comparison of maximum drawing force in different die semi angles In this paper, the effect of die semi-angle on some parameters such as thickness distribution, and drawing force has been investigated. The following conclusions can be drawn from this research: i. With increasing the die semi-angle, the amount of thickness has significant increase along axial axis but it is slightly increase along circumferential of tube. ii. The increase of die semi-angel causes increasing of drawing force. iii. Tubes drawn with die semi angle of 7.5 degree indicate better thickness distribution and lower drawing force. 418 P a g e
6. References [1]-Toiserkani H. Principle of Metal Forming book, Third Edition, 2009, Isfahan Industrial University Press. [2]-Bihamta R et al. A new method for production of variable thickness aluminum tubes: Numerical and experimental studies. Journal of Materials Processing Technology 2011; 211(4): 578-589. [3]- Bayoumi LS. Cold drawing of regular polygonal tubular sections from round tubes. International Journal of Mechanical Sciences 2001; 43(11):2541-2553. [4]- Karnezis P, Farrugia DCJ. Study of cold tube drawing by finite-element modelling. Journal of Materials Processing Technology 1998; 80 81(0):690-694. [5]- Wang ZT, Luan GF, Bai GR. Study of the deformation velocity field and drawing force during the dieless drawing of tube. Journal of Materials Processing Technology 1999; 94(2 3):73-77. [6]- Béland J F et al. Optimization on the cold drawing process of 6063 aluminum tubes. Applied Mathematical Modelling 2011; 35(11):5302-5313. [7]-Yoshida K, Furuya H. Mandrel drawing and plug drawing of shape-memory-alloy fine tubes used in catheters and stents. Journal of Materials Processing Technology 2004;153 154(0):145-150. [8]-Bui Q H et al. Investigation of the formability limit of aluminium tubes drawn with variable wall thickness. Journal of Materials Processing Technology 2011; 211(3):402-414. 419 P a g e