-Nnterational Conference on Engineering Technology an Appi Seiners QCETAS) Investigation of the Effects of Re-bending with Varied Radii on Springback in U-Bending Process Hakan Giiriin'"’, Selin Yildirim’, Hakan Dilipak! Abstract: The dimensional accuracy of the products depends on the correct determination of the springback in cold forming process. The most important cause of springhack is the distribution of stresses om the material daring ‘deformation. Springback values vary according to material properties and deformation process applied. In this study, it is aimed to investigate the springback behavior of different materials in U-bending operations. Springback behaviors of the materials were determined with analyzes which were made by using Deform software. DP-600, DP-1000 and JAC270 sheet materials were analyzed using diferent sheet thicknesses and different punch radi. In addition, the U- bending process, which was made atthe first a small radius, was repeated with increasing radi. It has been determined that repeated bending operations using increasing punch radi, reduce the springhack on the sheet material. The effects of bending parameters in the first bending and re-bending process on springhack were evaluated by using variance ‘analysis, Keywords: Bending, Springback, Re-bending, Anova L INTRODUCTION Bending is a forming operation where sheot material is subjected to bending stress, Thus a curve ‘ereates in the sheet material. The bending processes are used in many places in the sheet metal industry. After ‘bending operation, itis important to provide the desired measurement and angle values within the tolerances Springback isan elastic recovery of the sheet metal after the bending operation. To obiain the desired rmieasurements, the values of the springback should also be known, The springback that occurs after bending is shown in Fig. 1. oo Na fe J) NM RY Bending Radi After Spank ss: Paneh Ange Bending “Angle Ate Spangback Fig. 1 Demonstration of springback after bending process ‘There are numerous studies in the literature on the ‘estimation and compensation of the springback which cccur after bending. In some studies, it has been determined that after the bending process completed. to be stand up the punch on the material fora while reduces springback [1-4]. Ozdemir et al. investigated the formability of the wear resistant Hardox 500 sheet material. V-bending was cartied out on materials that had previously been subjected to annealing at diferent temperatures, When examined structurally, it was seen that crack formation was prevented [5-6]. Sharad and Nandedkar, have worked upon prediction of the springback using finite clement analysis. In their study. die radius, sheet thickness and material have been used as variable process parameters. Springback estimations have been developed by using neural network according to the results of the analysis [7]. Angsuseranee et a studied to compare the elficeney of springback and sidewall curl prediction of AHSS grade SPFC9SOY in the U-bending process by the finite element method and artificial neural network approach. Punch radius, die radius, and blank holder force have been used as input data for analysis and neural network. The results have been showed that the finite element method was more efficient than the artificial neural network approach [8] Choi et al. formed a U-shaped channel using double slage drawing process to reduce springback. The results, ofthe double stage forming process have been compared with the conventional U-deaw bending process. Single stage and double stage U-draw bending tests have been conducted for DP980 and TWIP9SO sheets. It has been, found that the springhack was reduced using the double stage Usdraw bending [9]. Yang et al. developed an analytical model to predict the springback in air-bending of AHSS materials. A computer code, which is developed based on classical bending theory, was updated based on the new model [10]. Sumikawa et al measured the noa-linear behaviors of high strength stel sheets under uniaxial and biaxial sess states [11]. How and friends did inplane biaxial tensile tests in the 980, ‘MPa grade mulliphase steel (MP9E0). 980 MPa was compared with stetching behavior since implemented HilL48, Barlat89, Barla2000 function including flexing criterion. Against actual stress curves were done stess- ‘compression tests. For springback behavior guess, were done finite element simulations with LS-DYNA [12], In this study, springhack values were determined as a result of finite element (FE) analyzes using diferent materials (DP1000, DP600, JAC270), sheet metal thicknesses (Imm, 1.Smm, 2mm), and punch radii (Smm, 10mm, 15mm) in U-bending process. After the bending with RS, R10, R1S mm punch redi, FE analyzes were repeated using increasing punch radii. The parts bended KSmm punch radius were re-bended using R10 mm and RIS mm punch radii, Same process was used fer RLO mim punch radius using only RIS mm punch 273radius, The re-bending process, which is applied by increasing the bending radius, has reduced the springback that occurs inthe bended workpiece 2 ANALYSIS OF U-BENDING AND SPRINGBACK Ja this study, U-bending analyzes were performed on sheet materials by using punches with diferent beading radi. A knockout was used inorder to keep the upper part ofthe sheet constant during bending process. Sam, 10mm, and 1Smm values were used as Punch radi (R,). The fist bending analyzes were done by using RS, R10, RIS mm punch radi (Ry). Then, analyzes where the punch radive was RSmm, were repeated with different punch radii (RlOmm and RIS mam). Likewise, analyzes where the pusch radius was 10mm, were repeated with R15 mam punch radius (R) ‘The punch dimensions were changed depend on th ‘het material thickness used, The die dimensions were kept constant. Analyzes were performed by DEFORM. 2D sofeware. The springback values forthe first and re- ‘ending. process were meaustred ftom the analyzes resulis. The die model generated for conducting of the analyzes in the Deform software and the die parameters used are given in Fig. 2 and Table Fig. 2. Process sequence for the generated die model Table J. Die parameters used in the analyzes Springbacks were determined a a result of the analysis by using Deform software. After the first bending operation, the material was allowed to springback and the re-bending process was applied after the springback using a bigger punch radius. The Schematic. presentation of the part shapes after first bending and re-bending are shown in Fig. 3 and Fig. 4 respectively. Senate ig. 3. The schematic of first bending Tig 4. The schematic ofre-bending ‘3 RESULTS AND DISCUSSION Springback values have been reduced using increasing punch radii forthe re-bening operation aller gst bending. As a result of the study, springback values decreased in all three materials used. The springback values depending on the sheet thickness and punch radius in the re-bending process for DP1000, DP600, and JAC270 materials are given in the Fig. 5, Fig. 6, and Fig T respectively ‘The maximum change in the springback values was seen for the DPLOOO material with the Imm Pavameters Sybal Value thiskness The Teast change’ was seen in JACZ70 Dimension of ie suse Die Cay Fah Dimension 7 Fomming Lengih Punch-Die Cleszance z Die Comer 3 Radius Re 16 7 Punch Comer] R,=Smm, TOmm, Y il Radius 15mm Varied Bik Helder P 5000 8 ‘ ae 3 i Die Velosin Vv Somme Steet hans DP 6b, DP 1000, Vs Fi oFingback results for sheet materi Sheet Material | PP 00. DFI Varied ig 5 Sprghack veal Tox DPVOOD ae atonal sheet Thickawss | T= TSMR TVs When the graphics examined, i wat ten that springback values were reduced ‘sgpificantly byapplying the re-bending process, It was determined that the DP1000 material is the most affected material from the re-bending process. The DP 600 material was affected less than DPIOO0. The JAC270 is the least material affected from re-bending process hetween these thee materials, ‘The bending parameters determined as a result of the first bending explain the springback at the raio of RSq (adj) = 85.67%, Itis seen that material and bending radius are effective on springback The effect of the material was determined as 61.13% and the effect of the punch radius as 18.64%. ‘The variance analysis table obtained with the 98% reliability as a result of the re-bending analyzes is aiven below (Table 3) Table 3. The Anova table obtained asa result of re-bending analyzes — Fig: 7 Springback results for SAC2TO sheet material ‘The springback values increased with the increasing punch radii in the Grslly bending process which were used RS, RIO and R1Smm bending radi. It was observed that springback values were decreased hile the sheet thickness was inereasing. It was seen that, effects of the punch radius and sheet thickness in the re- bending are very low compared to the first bending. 4SANOVA The effects of bending parameters on springback were evalusted by using variance analysis (ANOVA) with the 95% reliability. The Anova table obtained as a result of first bending is given Table 2. The "ay" leer is referred to the material, "U” sheet thickness, + punch radius, and "x" springhack in the Table 2 Table 2. The Anova table obtained as a result of ist ending analyzes Ali of Vance ist benciagy ‘Soar | [ a4) 88 | Adi Ms [w-Vate | poValoe | Perent [2 pasar [uso] ssas | 0] otis 7 [2 [ar [sir | am | oom | 920 Ero [20] 33 | 26s 1198 Teal [36 | 07 70.0 GE RSGIWNS — RSghd) SATE wee rte [aa of Varian bending) Source] DE] Aays5 [Adi is | Value [pValue | Percent m [2 fawmi[2ess| aso | sia0 [2 [asses [aease | rar | 000 | 9s iar t [2 [iasase | iasae] 2200 | 0 | 200 Fig, 6 Springback resulls for DPA0O sheet material Beer [20] sraot | sta0e Ta "Tout | 26 [aon 07 100.00 . SIGNET RSGENTM RSqGa)= oO When these two variance analysis tables were examined, it was seen thatthe values ate very close to each other. However, two parameters have changed. As a result of the rechending process, the effect of the material decreased to 51.20% and the effect of the punch radius increased to 27% on the springback. The effect of the punch radius in the first bending operation is 18.64%, while at the end of re-bending 27%. Thus, the effect of punch radius on springback increased the ratio. of 145%, 5 CONCLUSIONS Springback values have decreased for all material used when sheet thicknesses are increased Punch radius has a significant effect on springback behavior. The maximum springback value was seen in the DP1000 material and 1 mm sheet thickness. The te bending process was found as an effective method to reduce springback values forall materials used. Inthe r= bending process, the effect of sheet thickness is insignificant compared tothe fist bending. In order to able to better explain the effect of the re-bending process, Anova was applied 10 the analysis sesulls. The effect of the material on t springback decreased about 16% by applying the re- bending process. The effect of punch radius on springback increased approximately 145% with the use of the re-bending process. As a result, it can be said that a re-bending operation, which is caried out by increasing the punch radius, is highly effective on preventing springback.REFERENCES a a ol (a (1 te a (8) Tekaslan, 0, Soker, U, Gndemir, A. (2006) Determining ‘springback amount of steel sheet metal has 0) mm thickness in bending. dis Materials & Design, 27. 251-258, Tekaslan, O., Seker, U. Gexger, N., (2008). Determination of Copper sheet metal in “V" bending dies. Journal of The Faculy of Engineering and Architecture of Gai University, 25, 201-238 Tekaslan, ., Geeger, N.. Seker, U., (2008). Determination of spring-back of slaniest steel shect metal in “V" bending dies, Materials & Design, 29, 1043-1050. Tekiner,Z, (2004). An experimental study of the examination of spring-back of sheet metals with several thicknesses and properties in bending dics Journal of Materials Processing Technology, WS, 109-117 Gxdemir, M, Gokmeye, H., Dilipak, H., Yilmaz, V., Girt, H., (2015). Effect On The Micro- Structural And The Behavior Against Formabshty ‘Of Wear Resistant Sheet Material, 118815, 1-3 April 2015, 740-744, Karbuk, Turkey Grdemic, M., Gokmese, H., Dilipak. H., Yilmaz, V. Q015). The examination of microstucural characterization and. formability of the wear resisted sheet material, GU J SeiPartsC, 30), 385- 393 Sharad, G., Nandedker. V. M., (2014). Springback in Sheet Metal U Bending Fea and Neural Network Approach, 3rd Imemarinal Conference fon Materials Processing and Characterization (ICMPC 2014) Procedia Materials Science 6 (2014) 835-839, doi 10.1016, mspro 2014.07.10 Angsuseranee, —-N,—Pluphrach,—G., ‘Watcharasresomroeng, B., Songkrob, A., (2018) Springback and sidewall curl prediction in U- bending process of AHSS through finite element method and artificial neural network approach, Songklanakarin J Sc. Technol., 40 (3), 534-539 [91 Choi, 1, Lee, J, Bong, JH, Lee, G. M., Barlat F, Q018). Advanced constitutive modeling of advanced high strength steel sheets for springback prediction after double stage U-draw bending International Journal of Solids and Structures, 151, 152-164, DOK: 10.1016) .ijsolst 2017.09.030 [10] Yang, X., Choi, C., Sever, K.N., Altan, T., (2016). Prediction of springhack in ait-bending of ‘Advanced High Strength steel (DP780) ‘considering Young's modulus variation and with piecewise hardening function, International Journal of Mechanical Sciences, 108, 266-272, DOK: 10.1016/j.imecsei.2015.11.028 [11] Sumikawa, S., Ishiwatari, A, Hiramoto, J. (2017) Improvement of springhack prediction accuracy by cconsideringnonlinear elastoplastic behavior after stress reversal, Journal of Materials Processing Technology, 241 46-53, DOL 10,1016), jmasprotec.2016.11.005 [12] How, ¥., Min, J, Lin, J, Liv, Z, Carsley, BJ, Stoughton, H. T., (2017). Springhack prediction of sheet metals using improved material models, Procedia Engineering, 207, 173-178, DOL 10.1016) proeng.2017.10.757 ‘Authors addresses "Hakan Gurun, 'Selin Yildirim, ‘Hakan Dilipak Investigation of the Effects of Re-bending with Varied Radi on Springback in U-Bending Process, Gazi University, Technology Faculty, Department _ of Manufacturing Engineering, —Telikokallar, 06500, Ankara, Turkey, selinnyildirim@ gmail.com, Iigurun@ gaci-educsr, hdilipak@ gaci.eduuir Contact person “Hakan Girin, Investigation of the Effects of Re- bending with Varied Radii on Springhack in U-Bending Process, Gazi University. Technology Faculty, Department of Manufacturing Engineering. Teknikokullar, 06500, Ankara, Turkey, +903122028615, hagurunt@ gaci edu 276