International Journal of Automotive Technology, Vol. 16, No. 4, pp. 619−627 (2015) DOI 10.1007/s12239−015−0063−9
Copyright © 2015 KSAE/ 085−09 pISSN 1229−9138/ eISSN 1976−3832
EFFECT OF HOT-STAMPING PROCESS CONDITIONS ON THE CHANGES IN MATERIAL STRENGTH Y. J. JEON1), M. J. SONG1), H. K. KIM2) and B. S. CHA1)* 1)
Molds and Dies Technology, KITECH, 156 Gaetbeol-ro, Yeonsu-gu, Incheon 406-840, Korea 2) Department of Automotive Engineering, Kookmin University, Seoul 136-702, Korea (Received 4 October 2013; Revised 10 April 2014; Accepted 25 April 2014)
ABSTRACT−According to the recent trend of increasing demand in weight reduction of car bodies for the improvement of vehicle safety and gas mileage, there exists a forming method that makes use of the hot-stamping process as a means that satisfies all of the conditions above. However, there is insufficient systematic technical information on the method of heating material and setting up the forming process in case of the hot-stamping technique. A material heating device was manufactured in this study and the temperature profile for the forming process was created in order to design the material heating technique and the forming process in the hot-stamping method. Furthermore, in order to study the changes in strength according to the hot-stamping forming variables, high temperature forming was conducted according to forming variables such as forming speed, rapid cooling time, and material rolling direction. In order to verify the subsequent results, the springback was measured to assess the material hardness and to check the forming precision. In addition, a stamping analysis and quenching analysis were conducted on the experiment conditions through a commercial finite element analysis program, and the results of comparing the hardness distribution with the test values showed similar trends. Based on the results of the analysis, having the shortest possible material transport time would allow for the production of products with evenness and high strength. KEY WORDS : Hot-stamping, Formability, Quenching, Forming, Press die, Finite element anlysis
1. INTRODUCTION
al., 2011). In particular, it has been reported that the variables are highly sensitive depending on the blank thickness, material composition, forming temperature, and others (Aranda et al., 2003; Mori et al., 2005; Hoffmann et al., 2007). order to apply such hot-stamping process in the actual mass production of cars, studies have been conducted through experiments and simulations on the production of the martensite (Lee et al., 2009), which is the final target structure, according to temperature conditions such as heating, quenching and tempering, studies on the thermal and mechanical material flow characteristics according to the forming speed and the quenching speed (Merklein et al., 2006), studies through experiments and simulations on the flow properties of the material at high temperatures and the heat transfer between the die and the material (Kim et al., 2006), studies on the effect of the forming factor on the mechanical properties during the hot bending process of the boron steel (Kwon et al., 2010), and studies on the springback, residual stress, and others of the product on the final quality (Suh et al., 2010). However, there is still insufficient systematic technical information on the heating method of the material and the set up of the forming process in the hot-stamping method. As such, the aim of the research is to study material strength and material forming precision while varying the forming speed and the quenching rate during the forming
Recently, the demand for the weight reduction and improvement of strength in car bodies for better car safety and gas mileage has increased. Hot-stamping parts with a tensile strength of 1,500 MPa grade are being currently applied in vehicles (Karbasian and Tekkaya, 2010; Kang et al., 2008; Lee et al., 2009). Hot-stamping is a process in which the boron steel that has boron additive is heated to temperatures of at least 900oC after which it is rapidly cooled in the die as it is formed simultaneously. Such hot-stamping products carry the advantages of having excellent formability and satisfactory dimensional precision among others since they are formed at high temperatures. Moreover, there is the characteristic of having improved material strength through the quenching effect of the material as it rapidly cools while being formed simultaneously at high temperature. Hot-stamping process requires expensive equipment and a high level of control technique since the blank is transported at hot conditions and forming is performed fast in the austenite region. Also, there are many variables to consider compared to a regular stamping process (Choi et *Corresponding author. e-mail:
[email protected] 619
620
Y. J. JEON, M. J. SONG, H. K. KIM and B. S. CHA
of the material in order to understand the changes in the material strength as a result of the quenching effect during the hot-stamping forming process and to design the optimal quenching conditions. In addition to the above, the commercial finite element analysis program was used to conduct the stamping analysis and the quenching analysis regarding the test conditions, and to analyze the phase transformation distribution and the strength. Moreover, the material transport time was selected as a process parameter to compare the changes in the strength.
2. HOT-STAMPING EXPERIMENT 2.1. Experiment Preparation 2.1.1. Construction of the heating furnace of hot-stamping Prior to hot-stamping, the heating furnace for hot-stamping was first constructed. In case of the heating furnace, the experiment was conducted at a constant speed by constructing together a transporting device for the material to be discharged from the heating furnace after it was moved to the press via heating (Figure 1). For the major specification of the heating furnace for the high temperature forming, the temperature elevation was allowed to be heated up to a maximum of 1,000oC for hotstamping by making use of the Kanthal A-1 wire bury heater. Also, for the material inside the furnace, ceramic material was used to prevent the heat from being released to the outside, while K-type thermocouple was used for the sensor measuring the temperature inside the furnace. Moreover, the open/close and the transport devices for the discharge of the heated material in the furnace were pneumatically controlled. 2.1.2. Forming process design for hot-stamping In case of the hot-stamping process, the basic process involves heating → keeping → material discharge →
Figure 2. Temperature-time profile for hot-stamping. specimen transfer → forming and quenching → air cooling. For heating, the material is placed inside the heating furnance and heated to approximately 950oC after which the material is kept at 950oC for a certain period of time. Afterwards, the material heated inside the furnace is extracted and is transported to the die by using the transfer device. The material then undergoes the quenching process and the forming process simultaneously in the die. Based on such forming process, the high-temperature, forming temperature time profile was designed as shown in Figure 2. Here, (I) is the process of heating the material inside the furnace, (II) is the process of keeping the material at a certain temperature inside the furnace, (III) is the process of extracting the material from the furnace to the transfer device, (V) is the process of forming and rapid quenching of the material, and (VI) is the air cooling process. The material was heated to 950oC and then formed after being transferred to the die at an approximate temperature of 650oC. With regard to the heating temperature of the material, the quenching effect cannot be realized if the product is formed after the temperature drops below 600oC, which results in reduced product strength and lower dimensional precision. 2.2. Hot-stamping 2.2.1. Selection of forming targets As shown in Figure 3, hat-type shape was selected for hotstamping. In case of hat type, it is the cross-sectional shape of the side sill that plays the most important role in the side crash protection of cars and, as such, it requires high strength and is one of the car parts where hot-stamping is applied.
Figure 1. Experimental apparatus for the heating furnace.
Figure 3. Shape of the hat-type for hot-stamping.
EFFECT OF HOT-STAMPING PROCESS CONDITIONS ON THE CHANGES IN MATERIAL STRENGTH
2.2.2. Hot-stamping testing method The hot-stamping experiment was conducted based on the forming profile designed above. Table 1 summarizes the test conditions wherein the major conditions are that the material heating temperature is 950oC with three types of forming variables that are 0o, 45o, and 90o according to the material rolling direction. Also, the experiment was conducted by varying each of the two types of forming rates and two types of the material quenching times in order to understand the effects according to the material cooling. Here, the material quenching time refers to the time in which the die presses on the product after product forming.
Table 2. Hardness result for various hot-stamping test conditions (Measurement region: ③ ). Drawing velocity
Quenching time 5s
Rolling direction
Hardness (HRC)
0o
42.5
o
45
41
o
42
90
5.4 mm/s
o
44.7
o
45
44.5
o
90
43
o
45
0 30 s
0 2.3. Analysis of the Hot-stamping Experiment Results 2.3.1. Hardness distribution analysis Based on the test conditions determined above, the hardness was measured to check the material strength of the hot-stamping product. The Rockwell hardness measuring device was used to measure the HRC. As shown in Figure 4, the hardness was measured at five positions (flange areas at both sides, side wall areas, and punch area) to check the hardness distribution in each of the areas. Table 2 summarizes the values of the hardness measurements at position ③ among the measurements. The observation of the results shows that at a forming rate of 51.4 mm/s, hardness values are at least HRC 40 in all quenching times and material rolling directions. The results of observing the changes in the hardness according to each of the forming variables show that at a low forming rate of 5.4 mm/s, the hardness increases as a result of longer quenching time available. At a high forming rate of 51.4 mm/s, there is not much change in hardness according Table 1. Conditions of experimental hot-stamping test. Heating temperature
Rolling direction
Drawing velocity
Quenching time
950oC
0o 45o 90o
5.4 mm/s 51.4 mm/s
5 sec 30 sec
Figure 4. Measurement regions for the hardness of the hattype.
621
5s 54.1 mm/s
45
o
44.5
o
90
45
o
45
o
45
44
o
45
0 30 s
90
to the change in the quenching time. This shows that the effect of the quenching time is insignificant once the forming rate exceeds a certain speed. 2.3.2. Analysis of spring back tendencies The springback was measured to understand the tendencies of the springback reduction in the formed products. For its measurement, the designed shape data was compared with the actual formed product as shown in Figure 5. The punch area springback (pθ) and the flange area springback (fθ) were assessed for the springback measurement. The wall bend (ρ) was not measured since it was present at room temperature forming but not at high temperature forming. The comparison between the designed shape and the amount of change at the positions for springback measurement noted above was conducted. Observation of the springback measurement results shows that the hot-stamping product has decreased springback amount compared to the case of the product formed at room temperature as shown in Figure 6. Furthermore, it
Figure 5. Measurement method of spingback modes in the hat-type.
622
Y. J. JEON, M. J. SONG, H. K. KIM and B. S. CHA
is isothermal. This assumption is called ‘‘isokinetic reaction” by Sheil (1935). The transformed phase fraction at the current time step n+1 is calculated from an equivalent transformation time t′ which produces the phase fraction on the previous curve of time step n (Xi,n) and time increment Δt. ′
ni
X i, n + 1 = Xi, n + 1 ( 1 – exp ( –k t ( t + Δt ) ) ) eq
and
(2)
1---
X i, n⎞ ⎞ nj 1 t = ⎛⎝ – --- ln ⎛⎝ 1 – ------eq ki X i, n⎠ ⎠ ′
(3)
Contrary to the kinetics of ferrite, pearlite, and bainite, the diffusionless transformation of austenite to martensite is modeled based on the equation proposed by (Koistinen and Marburger, 1959). The volume fraction of martensite Xmart at the current temperature T below Ms becomes X mart = X ra ( 1 – exp ( –α ( M s – T ) ) )
Figure 6. Springback result for various drawing velocity and rolling direction. can be observed that the springback amount clearly declines as the forming rate increases. This may be due to the faster material cooling during increasing forming rate.
(4)
where Xra and Xs are the volume fraction of retained austenite and martensite starting temperature, respectively. 3.1.2. Hardness determination The calculation of the Vickers hardness involved the empirical model proposed by Maynier et al. (1978). Martensitic structure:
3. FINITE ELEMENT ANALYSIS 3.1. Theory 3.1.1. Transformation kinetics The transformed fraction of phase i, Xi, decomposed from austenite, is calculated based on the Avrami type equation which considers rate of nucleation and growth of newly formed phases from austenite grain boundaries(Lee et al., 2009). n
Xi = X i ( 1 – exp ( – ki t i ) ) eq
Here, C, Si, Mn, Ni, Cr, Mo, and V are concentrations of a solute element in % and Vr is the cooling rate (oC/hour). Bainitic structure:
(1)
Here, X ieq is thermo-dynamical equilibrium fraction of phase i determined from the equilibrium phase diagram with known temperature and chemical composition, and ki and ni are empirically obtained constants for the phase i. From the previous investigations, the constant ki generally depends on temperature, chemical compositions, and prior austenite grain size. Conversely, the parameter ni is usually known to be constant over the range of temperature and the suggested value is between 1 and 4. The Avrami equation was originally proposed for an isothermal condition, thus the equation cannot be directly utilized for non-isothermal cases such as the transformation on heating or cooling. For non-isothermal transformation kinetics, it is assumed that heating or cooling curves can be divided into small time intervals within which the kinetics
Ferritic structure:
Austenitic structure: Hvaustenite = 100
(5)
3.2. F.E.Modeling and Boundry Condition Based on the test results above, the stamping analysis and the quenching analysis were conducted on the hotstamping process. The finite element analysis can generally be divided into two steps: the stamping analysis and the quenching analysis. As shown in Figure 7, the stamping
EFFECT OF HOT-STAMPING PROCESS CONDITIONS ON THE CHANGES IN MATERIAL STRENGTH
623
Figure 7. F.E. model for stamping analysis.
Figure 8. F.E. model for quenching analysis. analysis was comprised of three elements, which were the punch, blank, and die, wherein the punch and the die were set as the shell elements. Also, the quenching analysis was modeled by making the punch and the die solid elements as illustrated in Figure 8. This is to observe the effects of the cooling channel since the quenching effect varies partly according to the position of the cooling channels. As such, the cooling channel of the die that is identical to the one used in the experiment was used for modeling. For the conditions of the analysis, the quenching time after forming was set at 30 seconds with consideration of 15 seconds for the transfer time. The initial temperature of the blank was set at 950oC with forming speed of 51.4 mm/s. Based on such conditions, Pam-stamp, which is a commercial code, was utilized to analyze the hot-stamping process (Yoon and Kim, 2013; Bok et al., 2010; Bok et al., 2011; Kim et al., 2012; Kim and Kang, 2010). 22MnB5, which is included in the material library of the Pam-stamp, was applied for the related thermal and mechanical properties (ESI Group, 2012). 3.3. Hot-stamping Analysis Figure 9 shows the result after the completion of the press process before beginning the quenching process in the hotstamping process. Figure 9 (a) shows the general thinning rate while Figure 9 (b) illustrates the plastic strain. Maximum thinning of 2.3% and maximum plastic strain of 6.4% were observed through which it can be understood
Figure 9. Results of hot stamping analysis.
that such results are highly satisfactory against fracture similar to the test results in case of simple hat-type shapes as used in this study. As shown in Figure 10 (a) for the temperature
Figure 10. Thermal & metallurgy distribution of hot stamping analysis.
624
Y. J. JEON, M. J. SONG, H. K. KIM and B. S. CHA
Figure 11. Temperature distribution after quenching process. distribution, a distribution of 483oC was observed as the minimum temperature at a maximum of 654oC in the punch flange area. Also, it can be understood as shown in Figure 15 (b) that austenite is generally distributed at above 93% since the general temperature of the sheet before the quenching process is distributed at temperatures that exceed the temperature of martensite phase transformation.
Figure 13. Phase fraction at measuring point during quenching process.
3.4. Quenching Analysis For the quenching analysis, the analysis was conducted by importing the blank that had completed forming from the quenching analysis process. Here, the blank was a data that includes the thickness, temperature, and phase transformation. As shown in Figure 8, the quenching die is a 3D solid die that includes the cooling channels wherein the initial temperature of the die was set at 20oC and the temperature of the cooling water in each channel was all set at 15oC. Figure 11 illustrates the temperature distribution after the completion up to the quenching process in which it can be observed that the general temperature distribution is in the range of 65oC ~ 77oC. In order to investigate the temperature distribution from a wider perspective of the general quenching process, the temperature distribution according to position is illustrated in Figure 12 by selecting three points of hardness
measurement from Figure 4. As shown in Figure 12, the temperature rapidly decreased to a martensite phase transformation temperature of 410oC after an elapse of approximately 5 seconds. After an elapse of 7 seconds, the temperature dropping rate relatively decreased compared to before. Also, the temperature difference according to position was small since the shape was relatively small. Figure 13 is a graph that illustrates the phase transformation distribution according to the quenching times at points ①, ②, and ③ as shown in Figure 4. Although there are some differences according to the positions, it can be observed that the martensite phase transformation begins after an elapse of 5 seconds, which is completed at 15 seconds. It can also be known that as the martensite is completed, the austenite is barely distributed. Figure 14 shows the phase fraction after the completion of the quenching process wherein (a) is the residual austenite fraction, (b) is the ferrite fraction, (c) is the martensite fraction, and (d) is the bainite fraction. It can be observed from the figure that the martensite structure is generally distributed with isolated occurrences of bainite (6.2%) and ferrite (3.8%). However, because such
Figure 12. Temperature distribution after quenching process.
Figure 14. Phase fraction at measuring point during quenching process.
EFFECT OF HOT-STAMPING PROCESS CONDITIONS ON THE CHANGES IN MATERIAL STRENGTH
625
Figure 16. Temperature profile at point ① after quenching process.
Figure 15. Hardness distribution after quenching process. occurrences are small, it is considered in case of this material product that martensite structure is formed overall. Figure 15 (a) shows the results of the hardness analysis. Uneven distribution with irregular hardness differences was observed in some areas. Such uneven distribution was considered to be due to the insufficient quenching effect at high temperatures. Figure 15 (b) shows the comparison between the Rockwell hardness, measured at a rolling direction of 0 degrees, and the Vickers hardness based on the analysis results. Although there are differences in measuring methods, both cases similarly show the lowest hardness values at the flange area on both sides and the highest hardness values at the side wall areas of the product. 3.5. Differences in Phase Transformation according to the Material Transfer Time When discharging the material and transferring it to the press after heating the boron steel to temperatures that exceed 900oC, the material is exposed to the room temperature and is cooled rapidly. Even in this experiment, the material cooled at room temperature by approximately 300oC from 950oC to 650oC in a span of about 15 seconds, which is the time frame for extracting and transferring. After such temperature drop at room temperature, the quenching effect declines, which prevents obtaining products of good quality. Accordingly, the time for discharge and transport was set at 5 seconds for carrying out the analysis after which the measuring point ① (flange
Figure 17. Comparison of martensite & hardness distribution after quenching process. area) from Figure 4 was compared. The results show that at 15-second discharge and transfer, a pre-forming temperature of 650oC was observed whereas in the case of the 5-second discharge and transfer, a temperature of 820oC was observed. After conducting the stamping analysis afterwards, the results of the quenching analysis were compared as shown in Figure 16. As shown in the figure, the temperature distributions follow a relatively similar pattern in being quenched from the base point of approximately 400oC, which is the temperature of martensite phase transformation. Figure 17 shows the comparison results of analyzing the martensite and hardness. When quenching from 820oC, martensite phase transformation of above 99% was observed with relatively higher hardness of 474 with even distribution. Studying the results of analysis shows that suppressing the heat transfer during the discharge and transfer of the material as much as possible is important during the hot-stamping process.
626
Y. J. JEON, M. J. SONG, H. K. KIM and B. S. CHA
4. CONCLUSION In this research, hot-stamping experiment was conducted according to the forming variables of hot-stamping process design, forming speed, quenching time, and material rolling direction on the hat-type shape, which was the sectional shape of side sills. The effects of each forming variable on the material strength and product shape were also studied. Such study will be able to provide data which forming researchers can utilize in the future through conducting further experiments with more varied test conditions. (1) The observation of the changes in the material strength according to the forming variables showed that the strength increases at a low forming speed condition of 5.4 mm/s since the material quenching time increases. However, the effect of the quenching time at a high forming speed condition of 54.1 mm/s is insignificant. (2) The results of springback measurement according to the forming variables show that the amount of springback decreases in hot-stamping products compared to the case of products formed in room temperature. It can also be observed that the amount of springback clearly reduces as the forming speed increases. (3) The stamping analysis and quenching analysis were conducted under the same conditions as in the experiment and they were done based on the results that the martensite structure can generally be obtained. (4) Similar tendencies of hardness after hot-stamping were observed in both the experiment and the analysis. The results of conducting the analysis showed that even martensite phase fraction and even hardness distribution could be obtained if the discharge and transfer time were shortened since the quenching effect would improve. ACKNOWLEDGEMENT−This research was conducted as part of a research program for the Industrial Source Technology Development Program (project number: Km-12-0062).
REFERENCES Altan, T. (2006). Hot-stamping boron-alloyed steels for automotive parts, part I: Process methods and uses. Stamping J., 4041. Aranda, L. G., Ravier, P. and Chastel, Y. (2003). Hot stamping of quenchable steels: Material data and process simulations. IDDRG 2003 Conf. Proc., 64166. Bok, H. H., Lee, M. G., Kim, H. D. and Moon, M. B. (2010). Thermo-mechanical finite element analysis incorporating the temperature dependent stress-strain response of low alloy steel for practical application to the hot stamped part. Met. Mater. Int. 16, 2, 185−195. Bok, H. H., Lee, M. G., Pavlin, E., Barlat, F. and Kim, H. D. (2011). Comparative study of the prediction of microstructure and mechanical properties for a hot-
stamped B-pillar reinforcing part. Int. J. Mechanical Sciences, 53, 744−752. Choi, H. S., Lim, W. S., Kang, C. G. and Kim, B. M., (2011). A local softening sethod for reducing die load and increasing service life in trimming of hot stamped part. Trans. Mater. Process 20, 6, 427−431. Choi, Y. S., Jung, W. S., Park, I. M., Ko, K. W., Jee, Y. K. and Yoo, I. S. (1996). Boron distribution behavior and property of boron treated steel. The Korean Society of Mechanical Engineers Annuals Spring Conf., 740−744. Engels, H., Schalmin, O. and Muller-Bollenhagen, C. (2006). Controlling and monitoring of the hot-stamping process of boron-alloyed heat-treated steels. Proc. Int. Conf. "New Development in Sheet Metal Forming Technology", Stuttgart, Germany, 135−150. ESI Group (2012). Pam-Stamp 2G 2012 User’s Guide. ESI Group. Hoffmann, H., So, H. and Steinbeiss, H. (2007). Design of hot stamping tools with cooling system. Ann. Girp. 56, 1, 262−272. Kang, K. P., Lee, K. H. and Kim, Y. S. (2008). Prediction of phase transformation of boron steel during hot press forming using material properties modeler and DEFORM-HT. Trans. Mater. Process 17, 4, 249−256. Karbasian, H. and Tekkaya, A. E. (2010). A review on hot stamping. J. Materials Processing Technology, 210, 2103−2118. Kim, H. G., Son, H. S. and Park, S. H. (2006). Development of thermal-mechanical coupled simulation skills for hot press forming tool design. Posco Tech. Rep. 9, 1, 117−125. Kim, J. T., Kim, B. M. and Kang, C. G. (2012). Blank shape design process for a hot stamped front pillar and its experimental verification. Trans. Materials Processing 21, 3, 186−194. Kim, N. H. and Kang, C. G. (2010). The prediction of interfacial heat transfer coefficient according to contact time and pressure in forging and casting die materials for the hot press forming. Trans. Mater. Process 19, 6, 378− 386. Koistinen, D. P. and Marburger, R. E. (1959). A general equation prescribing the extent of the austenite– martensite transformation in pure iron–carbon alloys and plain carbon steels. Acta Metall 7, 1, 59−60. Kwon, K. Y., Sin, B. S. and Kang, C. G. (2010). The effect of forming parameter on mechanical properties in hot bending process of boron steel sheet. Trans. Mater. Process 19, 4, 203−209. Lee, D. H., Kim, T. J., Lim, J. D. and Lim, H. J. (2009). Development of high strength steel body by hot stamping. Trans. Mater. Process 18, 4, 304−309. Lee, J. S., Chae, M. S., Park, C. D. and Kim, Y. S. (2009). Mechanical and forming characteristics of high-strength boron-alloyed steel with hot forming. Trans. Mater. Process 18, 3, 236−244. Lee, M. G., Kim, S. J., Han, H. N. and Jeong, W. C. (2009).
EFFECT OF HOT-STAMPING PROCESS CONDITIONS ON THE CHANGES IN MATERIAL STRENGTH
Implicit nite element formulations for multi-phase transformation in high carbon steel. Int. J. Plasticity 25, 9, 1726−1758. Maynier, B. J. and Dollet, J. (1978). Creusot-loire system for the prediction of the mechanical properties of low alloy steel products. Doane, D. V. and Kirkaldy, J. S. (edn), Hardenability Concepts with Applications to Steel, AIME, Warrendale, PA, 518. Merklein, M., Lechler, L. and Geiger, M. (2006). Characterization of the flow properties of the quenchenable ultra high strength steel 22MnB5. CIRP Ann 55, 1, 229−233.
627
Mori, K., Maki, S. and Tanaka, Y. (2005). Warm and hot stamping of ultra high strength steel sheets using resistnace heating. Ann. Girp 54, 1, 209−212. Sheil, E. (1935). Arch. Eisenhuttenwes, 12, 565−567. Suh, Y. S., Ji, M. W., Lee, K. H. and Kim, Y. S. (2010). Application and verification of virtual manufacturing to hot press forming process with boron steel. Trans. KSAE. 18, 2, 61−66. Yoon, S. C. and Kim, D. H. (2013). Analysis of phase transformation and temperature history during hot stamping using the finite element method. Trans. Materials Processing 22, 3, 123−132.