Materials & Design

Materials and Design 27 (2006) 11871195 www.elsevier.com/locate/matdes

Short communication

Flash butt welding application on 16MnCr5 chain steel and investigations of mechanical properties
ur Arabaci Cemil C etinkaya *, Ug
Gazi Universty, Techical Education Faculty, Metalurgy Department, 06500 Teknikokullar, Ankara Received 15 October 2004; accepted 1 February 2005 Available online 7 April 2005

Abstract In this study, 16MnCr5 chain steels were welded by ash butt welding and welded samples were annealed. Their microstructure and mechanical properties have been investigated. Upsetting current time was chosen as 1.6 s and build up pressure was chosen as 2, 2.5, and 3 bar. It is pointed out that welding parameters and annealing aected the mechanical properties. For all samples, the highest hardness was measured from weld metal. Non-annealed samples gave the highest tensile strength with the parameter of highest build up pressure. It is also found that annealing reduces the tensile strength of all samples. 2005 Elsevier Ltd. All rights reserved.
Keywords: Flash butt welding; Resistance welding; Mechanical properties; Chain steel

1. Introduction There are some sorts of welding methods according to the materials to weld, processing and aim of the welding. Flash butt welding is a resistance welding method which consists of electrical heating and hydraulic forging [1,2]. Slug residues, porosity and micro strain are disadvantages of the fusion welding. Solid state welding methods have some advantages because no melting occurs and causes little welding defect [3,4]. Surfaces of sample are not roughened. In the touching surface, there are some areas which have high electrical resistance because of not touching areas. This high electrical resistance causes high heat generation [5,6]. Touching points are suddenly melted because of high heat generation. These small melt volumes are thrown to out of welding area by inating. Then, new craters
*

and touching points are formed. This process is repeated until the joining surface is replaced by getting the mobile parts closer. After having the complete surface melted, pressure to the joining surfaces is suddenly increased. At this stage, whole areas in both surfaces touch each other. Friction between surfaces causes sparkling and consequently melting. Sparkling time for a metal is dependent on thermal conductivity, geometry of pats and sparkling cycle [79]. In this study, the ash butt welding of 16MnCr5 steel, which is used mostly in chains, was studied. By holding the build up current time constant, build up pressure is increased. The eect of build up pressure on welding is investigated.

2. Materials and methods 2.1. Materials

Corresponding author. E-mail addresses: [email protected] (C. C etinkaya), uarabaci@ gazi.edu.tr (U. Arabaci). 0261-3069/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2005.02.002

The chemical composition of the 16MnCr5 steel is given in Table 1. The dimensions of the sample are 100 mm

1188 Table 1 Chemical composition of 16MnCr5 Cr (%) 1.04 C (%) 0.143 Mn (%) 1.170

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

Si (%) 0.188

Mo (%) 0.030

Ni (%) 0.061

Co (%) 0.020

P (%) 0.010

Al (%) 0.023

S (%) 0.019

Fe (%) 97.324

of length and 18 mm of diameter. For investigative purposes, ninety welded samples were cut to work on. 2.2. Experimental procedure Cleaned samples were welded by using the SCHALATTER TYPE ash butt welding machine, which is _ FACTORY located in OSowned by YAPAS CHAIN TIM Ankara. Through the welding process, faces are kept clean to prevent the forming of a lm between joining surfaces. The welding parameters mentioned below are kept constant: Voltage Ampermeter Arc heating time Touching time Vice pressure Machine pressure Welding parameters were chosen by taking into account the specications of welding machine. The build up pressures were chosen as 2; 2.5 and 3 bar, while build up current time as 1.6 s. Five samples for microstructural investigation and ve samples for tensile test were prepared by using the parameters given in Table 2.
Table 2 Applied parameters Code Building pressure (bar) 2 2.5 3 2 2.5 3 Building current time (s) 1.6 1.6 1.6 1.6 1.6 1.6 Heat treatment after welding Not applied Not applied Not applied Applied Applied Applied Number of samples 15 15 15 15 15 15

Samples were separated into two groups. One group is heat-treated and the other one is not heat-treated. The heat treatment prole is given in Fig. 1. The heattreated samples are coded as I (Table 2). Samples were heat-treated at 780 C for 4 h, then left cooled in the furnace. Standard metallographic method was used to prepare the samples for microstructural examination. Two percent nital was used as etching media. Prior type optical microscope was used to examine the microstructure. Hardness measurement was carried out by using Instron Wolpert type hardness measurement machine. JSM 500 type JEOL Scanning Electron Microscope was used to photograph the fracture surface. By using the same scanning electron microscope, EDS analysis was also carried out. 3. Results and discussion 3.1. Hardness results In Fig. 3, hardness proles of the welded samples are given. Samples A, B and C were not heat-treated. Welding of these samples was carried by using 1.6 s build up current time. Two bar for A and I-A, 2 bar for B and I-B and 3 bar for C and I-C were used as build up pressures.

-7 -5

Hardness Value (Vickers)

A B C I-A I-B I-C

Fig. 2. Hardness measurement points from welded samples (0 is weld center).

A B

C I-A

I-B I-C

800 640

Heat

480 320 160 0 0 4 8 12 16 20 24 28 32 36 40 44

360 340 320 300 280 260 240 220 200 180 160 140 120 100 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8

Distance to welding center (mm)

Time (Hour)

Fig. 1. Heat treatment prole applied after welding.

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

1189

It is seen from Fig. 3 that the highest hardness value is measured from weld metal of non-heat-treated samples. The highest hardness values of weld metal are 298, 330 and 344 Hv for the samples A, B, and C, respectively. The increase of hardness can be contributed to the deformation that was caused by increased pressure. It is reported by Ylmaz [10] that maximum hardness was measured from the HSS side of joining line and hardness of joining line and adjacent area was increased because of deformation. It is also reported by Cive lu [11] that the highest hardness value was mealekog sured from the welding zone of ash butt welded dierent couples. Hardness was reduced while measuring point changing from welding center to base metal. As it is shown in Fig. 2, on the measuring point of #5, 250, 248 and 251 Hv hardness values were measured

for the samples A, B and C, respectively. As schematically shown in Fig. 2, hardness was reduced from measuring point of #5 to base metal, however, this reducing is not as high as it is in the welding zone. When the hardness graphs of heat-treated samples I-A, I-B and I-C were examined, it is seen that hardness of all areas was reduced because of heat treatment, however, hardness of weld zone was higher than that of other zones. Lamellar cementite in the pearlite was transformed into granular cementite by heat treatment of welded sample. This transformation of cementite reduced the hardness of welded sample. In the welded zone, hardness values were reduced from 298 to 175 Hv, from 330 to 180 Hv and from 340 to 188 Hv for the samples A, B and C, respectively. The lowest hardness value was measured from measuring point 5

Fig. 4. Microstructural pictures of sample A (280): (a) Base metal, (b) HAZ and (c) Weld metal.

Fig. 5. Microstructural pictures of sample I-A (280): (a) Base metal, (b) HAZ and (c) Weld metal.

1190

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

(Fig. 2). The lowest values were 153, 152 and 151 Hv for the samples A, B and C, respectively. Hardness increases 23 Hv while going to base metal from weld metal same as it isfor the non-heat-treated samples. 3.2. Microstructural results In this section, microstructures of base metal, HAZ and weld metal will be examined. Eect of build up pressure and heat treatment on the microstructure of weld metal and HAZ will be discussed. In the pictures of A and I-A dark, areas represent the pearlite phase and grey areas represent the ferrite phase (Figs. 4 and 5). As it is seen from Fig. 4(a), the micro-

structure of base metal of the sample A, there are pearlite colonies among the large ferrite grains. These pearlite colonies are not distributed homogeneously but gathered on some areas. The microstructure of HAZ is shown in Fig. 4(b). In the HAZ zone, pearlite colonies were smaller and distributed homogenously because of heat and pressure. Weld metal microstructure in Fig. 4(c) presents ne and long grains. These ne and long grains are asicular ferrite [12]. Heat treatment eect on welded samples is shown in Fig. 5. Lamellar pearlite transformed to spherical cementite after heat treatment. When both Figs. 4(a) and 5(a) are examined, it is seen that heat treatment caused small grains and pearlite colonies are scattered

Fig. 6. SEM pictures of fracture surfaces of: (a) Sample A, (b) sample I-A1.

Fig. 7. EDS results of sample A: (a) From point 1, (b) from point 2 in Fig. 6(a).

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

1191

homogenously. The same thing can be said for the HAZ zone too. In the weld metal zone, widmanstaten ferrite phase was formed and the amount of ferrite was increased after heat treatment. It is seen from Fig. 6(a) that the fracture surface of sample A is not smooth. There are several craters on the fracture surfaces. The fracture surface of annealed sample is rather smooth (Fig. 6(b)). EDS analysis of point 1 in Fig. 6(a) is shown in Fig. 7(a). Elemental analysis of point 1 in Fig. 6(a) shows 57.096% Fe and 42.456% Mn. The other EDS analysis of point 2 in Fig. 6(a) is shown in Fig. 7(b). That point contains 96% Fe, 2.2667% Cr and 1.203% Mn. Such a dierence can be contributed to the dierent distance to the welding point.

Applied force was increased with increasing build up pressure. The microstructure of base metal, shown in Fig. 8(a), is similar to the microstructure shown in Fig. 4(a). Ferrite grains are gathered while pearlite colonies scattered randomly among ferrite grains of sample A. After heat treatment, ferrite grains become smaller and are distributed regularly in the HAZ of sample A (Fig. 8(b)). Weld metal of sample A exhibits a mixture of pin-like and dendiritic grains (Fig. 8(c)). After heat treatment, base metal of sample A (Fig. 9(a)) has smaller ferrite grains than that of sample B (Fig. 8(a)). Base metal of sample A exhibits a homogenous distribution of ferrite grains. Pearlite ratio reduced in the HAZ zone of heat-treated sample I-B. In the zone of weld metal of heat-treated sample, widmanstaten ferrite ratio increased more than does in the zone

Fig. 8. Microstructural pictures of sample B (280): (a) Base metal, (b) HAZ and (c) Weld metal.

Fig. 9. Microstructural pictures of sample I-B (280): (a) Base metal, (b) HAZ and (c) Weld metal.

1192

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

of not heat-treated weld metal (Fig. 8(c)). Laminar widmanstaten grain structure is also seen. SEM micrograph of fracture surface of HAZ of samples is shown in Fig. 10(a). EDS analysis of point 1 from micrograph in Fig. 10(a) is shown in Fig. 11(a). The composition of this point is 77.393 wt% Al, 11.545 wt% Fe and 10.585 wt% Mn. In point 2 of micrograph in Fig. 10(a), Fe amount is increased considerably (98.153 wt%). Cr, Mn and Si ratios are 0.873, 0.621 and 0.353 wt%, respectively. It can be seen from SEM picture of fracture surface of this sample that heat treatment reduced the mean grain size of HAZ

(Fig. 10(b)). Black points are cavities on the fracture surface. Samples were also welded by increasing the build up pressure to 3 bar from 2.5 bar. Microstructure of this sample, shown in Fig. 12(a), resembles the microstructure shown in Figs. 4(a) and 8(a). The dominant phase here is the ferrite. Little amount of pearlite phase also exists. In the HAZ shown in Fig. 12(b), ferrite grains are smaller than those in the base metal. Pearlite colonies are uniformly located in the grain boundaries of ferrite phase. Weld metal area (Fig. 12(c)) has needleshaped grains. There exist same big ferrite grains locally.

Fig. 10. SEM pictures of fracture surfaces of: (a) Sample B, (b) sample I-B1.

Fig. 11. EDS results of sample B: (a) From point 1 and (b) from point 2 in Figure.

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

1193

Fig. 12. Microstructural pictures from sample C (280): (a) Base metal, (b) HAZ and (c) Weld metal.

Fig. 13. Microstructural pictures from sample I-C (280): (a) Base metal, (b) HAZ and (c) Weld metal.

Fig. 14. SEM pictures of fracture surface of HAZ: (a) Sample C, (b) Sample I-C.

1194

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

Fig. 15. EDS results of sample C: (a) From point 1 and (b) from point 2 in Fig. 14.

It is seen from the picture in Fig. 12(c) that weld metal area has small amount of dendritic grains. The annealing aected the shape and size of all the three zones of welded samples. When the annealed and unannealed welded samples were compared, it is seen that ferrite grains are distributed homogenously in the annealed one (Fig. 13(a)). Microstructure of HAZ of annealed samples showed that annealing increased the pearlite amount (Fig. 13(b)). Grain size of HAZ of annealed sample is smaller than that of unannealed sample. Weld metal of the annealed sample has mostly nodular ferrite grains and little amount of laminar grains (Fig. 13(c)). The widmanstaten ferrite phase ratio is gradually increased as seen in Figs. 5(c), 9(c) and 13(c) and consequently hardness is increased. There are several dark points on the fracture surface of welded sample C (Fig. 14(a)). EDAX analysis showed that chemical composition of point 1 is 84.83% Fe, 9.583% Mo, 3.651% Mn and 1.003% Cr. SEM micrograph of the same sample (Fig. 14(b)) showed that the grain size of this sample got ner after annealing. EDAX analysis of this point in Fig. 14(b) showed that samples compositions is 97.591% Fe and small amount of other components. Numerical amounts of elements are given in Fig. 15.

3.3. Tensile test results Tensile test parameters are tabulated in Table 3. Pulling speed was kept at 2 mm/s. The results are shown in Fig. 16. All fractures were taken place in the HAZ. That shows that weld metal has higher strength than HAZ does. It is seen from the graph in Fig. 16 that, increasing build up pressure increases the tensile strength because of deformation. Samples A, B and C have the tensile strengths of 670, 684 and 698 N/mm2, respectively. Annealing of the welded samples reduced the tensile strength but the relationship between tensile strength and build up pressure appears to be like unannealed
Table 3 Test parameters and tensile strength of samples Sample A B C I-A I-B I-C Pressure (bar) 2 2.5 3 2 2.5 3 Build up time (s) 1.6 1.6 1.6 1.6 1.6 1.6 Maximum tensile strength (N/mm2) 670 684 698 479 485 490

C. C etinkaya, U. Arabaci / Materials and Design 27 (2006) 11871195

1195

750

annealed

Max. Tensile strenght (N/mm2)

700 650 600 550 500 450 400 1.5 2 2.5 3

3. Full annealing of welded samples reduced the hardness of base metal and weld metal reduced the hardness dierence between base metal and weld metal. 4. Annealing reduced the tensile strength of welded samples. References
[1] Tu i K, Ylmaz M. Flash-butt welding, Welding lbentc _ world. Istanbul: Birsen Co.; 1990. p. 1425. . Weldability of welding techniques of aluminum and [2] C alis kan O aluminum alloys. M.Sc. Thesis, Marmara University, Institute of _ Science and Technology, Istanbul, Turkey; 1994, p. 565. [3] Suzuki M, Endo Shigero, Yosetso GR. Eect of PWHT on TMCP steel ash butt welded joint. Q J Jap Weld Soc 1990;8(2):5864. _ [4] Ank S. Handbook for welding: methods and elements, 1. Istanbul, Turkey: Birsen Co.; 1983. p. 4050. [5] Sprow DL. Flash-butt welding new and future pipeline application. In: Proceedings of the pipeline technology confer stend, Belgium. p. 1128. ence, O [6] Ylmaz M, Karago z S. Micro analysis application in the fussion welding. In: XI national electron microscopy symposium, Edrine, Turkey. p. 1635. [7] Kuchuk S. Experince of ash-butt welding application in pipeline construction in the USSR. In: Proceedings of the Pipeline stend, Belgium. p. 379. Technology Conference, O [8] Trompler Jr Ruben. Flash butt welding of crane rail. Iron Steel Eng 1996;73(10):424. [9] Thomas K, Michailov V, Wohlfahrt H. Toughness investigations on narrow process- aected zones of resistance-butt. Weld Cut 2000;52:25462. [10] Ylmaz M, Kaluc E, Tu i K. Investigation of weld zone of lbentc ash-butt welded C45 plain carbon steel and (HSS 6-5-2) high speed steel. Sci Technol Weld Join 1993:2868. lu F. Examination of welding parameters and investi[11] Civelekog gation of weld zone of ash-butt welded Ck60 plain carbon steel and (S-6-5-2) high speed steel. A thesis for the degree of Assoc. Prof., Istanbul Technical University, Institute of Science and _ Technology, Istanbul. Turkey; 1971, p. 759. _ _ [12] Yu kler A Irfan. Weld metal. Istanbul, Turkey: Erdini Co.; 1979. p. 1760.

Buildup pressure (Bar)

one. Micropictures of the annealed samples suggest that transforming of pearlite to spherical cementite because of annealing reduces the hardness consequently reducing the tensile strength.

4. Results The results of ash butt welding of 16MnCr5 steel are given as follows: 1. Increasing build up pressure increases the hardness of weld zone. 2. The highest hardness was measured from the weld metal for all samples.