Available online at www.sciencedirect.com
Procedia Engineering
Procedia Engineering 00 (2009) 000–000 Procedia Engineering 2 (2010) 1823–1828 www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Fatigue 2010
Evaluation of dynamic defect detection in FSSW welded joints under fatigue tests Malafaia, A. M. Sa *, Milan, M.T.b, Oliveira, M. F.a, Spinelli, D.a b
a Universidade de São Paulo, Av. Trabalhador São-carlense, 400, São Carlos CEP 13566-590, Brazil Materials Institute of Brazil, R. Paulino Botelho de Abreu Sampaio, 881, São Carlos CEP 13561-060, Brazil
Received 8 March 2010; revised 10 March 2010; accepted 15 March 2010
Abstract The present study aims to evaluate the dynamic defect monitoring technique (CVM – comparative monitoring vacuum) during fatigue tests in FSSW welded joints. This method is recent and has been applied to aeronautic industry. It works basically by pressure difference measurements between two positions, one firstly evacuated and another at ambient pressure. When a crack or defect links different pressure regions, a pressure sensor allows the defect detection. In this study, the specimens were instrumented with polymer sensors allowing for the detection of passing cracks in the joint region. In the present configuration the system can detect a passing crack, but not the crack nucleation stage. A computer program saves the sensor signals with time allowing the detection of the exact moment at which a crack crosses the joint. During the high-load tests the crack detection was observed almost at the final failure. However, for medium and low loads, the method was quite efficient, detecting defects in good advance before the final failure. Thus, assuming project conditions in which loads are always low, regarding fatigue limit, the method can be considered a good monitoring tool for crack detection in FSSW welded joints. c 2010 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Keywords: FSSW; fatigue; riveted joints; failure modes.
1. Introduction
1.1. Friction Stir Spot Welding (FSSW) Friction Stir Spot Welding process (FSSW) was developed in 1991 and has been mainly studied for applications in automotive, aeronautic and other industries [1]. This welding technology derives from Friction Stir Welding process (FSW), and the main difference is the type of joint. In FSSW, the plates form a lap-joint and the tool penetrates the plates only in a point, making a punctual bond. In the FSW process the plates are arranged in a buttjoint configuration and the tool moves towards the joint direction. Therefore, the FSSW process can be explained in three distinct steps: plunging, stirring and retracting [2]. The parameters mainly studied are plunge rate, rotational
* Corresponding author. Tel.: +55-16-8812-6534; fax: +55-16-3373-9590. E-mail address:
[email protected].
c 2010 Published by Elsevier Ltd. Open access under CC BY-NC-ND license. 1877-7058 doi:10.1016/j.proeng.2010.03.196
1824 2
A.M.S. Malafaia et al. / Procedia Engineering 2 (2010) 1823–1828 A. M. S Malafaia et al. / Procedia Engineering 00 (2010) 000–000
speed, plunge depth and dwell time. Some authors have also measured the axial force generated during the FSSW process [3-5]. The FSSW studies correlate the mechanical properties and the parameters used to make the joint, aiming to find better parameter sets. The mechanical properties measured in most works are restricted to tensile shear tests [3, 6, 7], however cross-tension tests have also been cited [8]. Regarding fatigue tests, some authors have performed studies in AA 5000 and AA 6000 aluminum alloys series, determining fatigue lives, failure modes, microstructural analysis and lives prediction model [9-11]. The use of two different tool geometries, flat and concave, has also been evaluated and the concave tool has generated the best results [9, 10]. 1.2. Comparative Vacuum Monitoring (CVM) Comparative vacuum monitoring (CVM) technique was developed in 1995 by SMS (Structural Monitoring Systems) company and has been largely used as structural health monitoring (SHM), mainly in the aeronautic industry [12]. Other possible applications to CVM is monitoring buildings, bridges, trains and subway vehicles, mining structures, railroad cars, trucks and other heavy machinery, pressure vessels, oil recovery equipment, pipelines, steel transmission towers, ships, tanks and a wide array of military structures [13]. CVM consists in the measurement of the difference of pressure between narrow channels evacuated and channels maintained at ambient pressure. When a failure appears in the structure and connects two channels, an air flow due to the atmosphere pressure starts to enter an evacuated channel [12]. These channels are formed in the contact region of the sensor (made of self-adhesive polymer) and the monitored structure, i.e. the component, or structure makes part of the sensor. Another possibility is to construct the sensor inside the component. Kousourakis et al [14, 15] introduced long and narrow interlaminar galleries in aerospace carbon/epoxy laminates to avoid sensor application. They perceived the existence of a critical size that allowed the same compressive and tensile resistance in the laminates. Thus, galleries with diameter larger than the critical size affected the mechanical properties. Figure 1 shows a scheme of the top view of a surface monitored by a CVM sensor. As the evacuated channels and the channels at ambient pressure are disposed in an alternated way, the sensor’s sensitivity is inversely proportional to the channels distance and thickness. The same figure illustrates a crack joining an evacuated channel (white) and a channel at ambient pressure (black), allowing the air to enter the evacuated channels [16].
Fig. 1. CVM sensor application in a schematic view.
1825
A.M.S. Malafaia et al. / Procedia Engineering 2 (2010) 1823–1828 A. M. S Malafaia et al. / Procedia Engineering 00 (2010) 000–000
3
Comparing with other non destructive techniques (NDT) as radiography, ultrasonic, liquid penetrant, magnetic particles inspection and eddy current testing, the CVM presents some facilities. CVM can operate in real-time, is not an invasive technique, does not require disassembly and does note require operator skill and/or high level of results interpretation [17]. On the other hand, the main disadvantages are the limitation to detect the crack position and the limitation to detect only surface cracking [18]. In addition to defect detection, CVM was used monitoring crack growth in welded T-butt joints tested under fatigue and allowed control the crack growth in a rate of 50,000 cycles/mm [18]. However, as this technique has not been tested in the monitoring of welded joints by the FSSW process, the present study aims to evaluate the use of these two techniques together. Thus, FSSW welded joints were evaluated in fatigue tests [19] using CVM sensors to detect cracks and evaluate the prediction efficiency of the technique. 2. Materials and Methods Two sets of lap shear specimens welded by FSSW, named WA and WB were fabricated and evaluated in fatigue tests in a previous study [19]. During fatigue tests, CVM sensors were placed on the characteristic holes produced by the FSSW process, as showed in Figures 2a and b. Figure 2a presents a specimen in the MTS fatigue machine mounted for the test with its two spot welds. The wires connected the sensors to the transducer and allowed the pressure measurement. Figure 2b shows a schematic view of the sensor applied on the welded joint surface. In this configuration there are no channels at ambient pressure, and only one evacuated region. Thus, the pressure variation is detected when a crack crosses the welded region between the plates interface and the evacuated hole.
(a)
(b)
Fig. 2. Details from CVM sensor installation. (a) specimen with sensors mounted in the MTS test machine; and (b) schematic view representing the sensor operating.
The crack monitoring system worked with a vacuum regulator operating at a –20 kPa pressure, and a pressure monitor connected to the evacuated hole. When an increase in the flow rate occurred, the pressure monitor detected the pressure variation, and consequently the crack presence. In all tests two sensors were placed in the specimens (one in each hole), and they were in parallel connected to the pressure monitor. Thus, the detection occurred always in the first passing crack, however it did not identify which of the welded point failed. It is important to note that in this configuration the system does not detect the crack initiation, but only the first crack crossing the welded region. 3. Results and Discussion Figure 3 shows the fatigue results for the WA specimens in a load x life curves. The curves present the fatigue lives (cycles to failure) and the detection lives (cycles until crack detection). The same results are presented for the
1826
A.M.S. Malafaia et al. / Procedia Engineering 2 (2010) 1823–1828 A. M. S Malafaia et al. / Procedia Engineering 00 (2010) 000–000
4
WB specimens in Figure 4. As shown in the figure the detection curves followed the life curves trends. However, in the WB specimens the detection curve was close to the total joint life, demonstrating shorter lives remaining after the defects detection. As the sensors only detect when a crack crosses the total welded joint length, the difference in the prediction efficiency between specimens WA and WB was probably caused due to the WB larger plunge depth in the FSSW process. This larger plunge depth probably promotes a weld extension higher, justifying the longer time to detect the first passing crack in the WB specimens. Another important observation is the WB specimen that achieved the run-out fatigue criterion few cycles after the detection time (in the lowest load used). If the test had been continued, the curves would have been more separate at this point.
defect detection and total lives to specimens WA maximum load (kN)
3.50 total life defect detection
3.00 2.50 2.00 1.50 1.00 0.50 0.00 1.0E+03
1.0E+04
1.0E+05
1.0E+06
1.0E+07
N (cycles) Fig. 3. Total lives and defect detection curves under fatigue test to the WA specimens.
defect detection and total lives to specimens WB 3.50
total life defect detection
maximum load (kN
3.00 2.50 2.00 1.50 1.00 0.50 0.00 1.0E+03
1.0E+04
1.0E+05
N (cycles) Fig. 4. Total lives and defect detection curves under fatigue test to the WB specimens.
1.0E+06
1.0E+07
1827
A.M.S. Malafaia et al. / Procedia Engineering 2 (2010) 1823–1828 A. M. S Malafaia et al. / Procedia Engineering 00 (2010) 000–000
5
The defect detection for WB specimens presented considerable scattering in the results. Due to the imperfection sensor application to some tests, and some run-out tests, there is no enough data recorded for a complete prediction analysis. Thus, only for the WA specimens an evaluation of the percentage lives before and after the defect detection was performed for each load level (Figure 5). For the highest load level, as suspected, the life after the defect detection is shorter (in percentage) than those of lower load levels, due to the difficulty for supporting high loads with the presence of defects. On the other hand, for the lowest load levels the prediction was efficient, detecting defects with good antecedence. In such cases, the average life after detection was always higher than 80% of the total life. Since, in real applications, repairs in spot welded joints are not usual, the life prediction system here evaluate could indicates the component replacement before its catastrophic failure. 100% 90% 80%
average life after defect detection average life before defect detection
70% 60% 50% 40% 64.8
30% 20% 10%
12.3
0% 1.09
18.9 6.8 1.64
load (kN)
2.18
2.73
Fig. 5. Average percentage lives, after and before defect detection by the CVM technique during fatigue tests.
The present results demonstrated the possibility of CVM technique be used in joints welded by FSSW process and other applications in which a hole or another site receive a passing crack. The configuration is simple and the sensor only needs to close the site allowing the evacuation. Thus, when a passing crack connects the ambient pressure with the evacuated place, the CVM technique allows the defect detection. Furthermore, the defects were detected in advance, mainly to lower load levels (more often used in projects). So, the CVM technique is also able to real-time monitoring in this type of defects, besides the well-known surface cracks case. 4. Conclusions The CVM method was used to detect defects in FSSW welded joints and presented good results concerning defect detection and prediction of detection. Two specimen sets (WA and WB) were tested and detection defect and fatigue lives curves were obtained. The detection defects curves followed the fatigue live trends and presented better results to the WA specimens. The life percentage, before and after the defect detection, was evaluated in the WA specimens showing good anticipation to the final failure. To the three lower load levels the detection occurred before 20% of the total life, and to the highest level the defects were detected after near around 65% of the total life. As in normal applications the load levels are low, these results showed the CVM potential to real-time monitoring in this joint type. The CVM sensor configuration, although innovative, allowed detect passing cracks joining and evacuated site to the ambient pressure on the other side of the structure observed.
1828
A.M.S. Malafaia et al. / Procedia Engineering 2 (2010) 1823–1828 A. M. S Malafaia et al. / Procedia Engineering 00 (2010) 000–000
6
Acknowledgements The authors would like acknowledge FAPESP (07/02093-5) and CNPq for the financial support, Prof. Dr. R. T. Coelho (OPF-EESC-USP) for allow the use of the milling machine used to produce the FSSW joints, and the Materials Institute of Brazil (Brazilian company representative of SMSystems) for support in the CVM application.
References [1] Hancock, R. Friction welding of Aluminum Cuts Energy Cost by 99%. Welding Journal 2004;83:40. [2] Awang, M.; Mucino, V. H.; Feng, Z.; David S. A. Thermo-Mechanical Modeling of Friction Stir Spot Welding (FSSW) Process: Use of an Explicit Adaptive Meshing Scheme, 2005 SAE International, 2005-01-1251 [3] Lathabai, S.; Painter, M.J.; Cantin, G.M.D.; Tyagi, V.K. Friction spot joining of an extruded Al–Mg–Si alloy. Scripta Materialia 2006;55(10):899-902. [4] Gerlich, A.; Su, P.; North, T. H. Tool penetration during friction stir spot welding of Al and Mg alloys. Journal Of Materials Science 2005;40:6473–81. [5] Gerlich, A.; Su, P.; North, T.H.; Bendzsak, G.J. Friction Stir Spot Welding Of Aluminum and Magnesium Alloys. Materials Forum 2005;29:290-4. [6] Wang, D.-A.; Lee, S.-C. Microstructures and Failure Mechanisms of Friction Stir Spot Welds of Aluminum 6061-T6 Sheets. Journal of Materials Processing Technology 2007;186(1-3): 291-7. [7] Mitlin, D.; Radmilovic, V.; Pan, T.; Chen, J.; Feng, Z.; Santella, M.L. Structure–properties relations in spot friction welded (also known as friction stir spot welded) 6111 aluminum. Materials Science and Engineering 2006;A441:79–96. [8] Feng, Z.M.; Santella, L.; David, S.A.; Steel, R.J.; Packer, S.M.; Pan, T. el al. Friction Stir Spot Welding of Advanced High-Strength Steels – A Feasibility Study. 2005 SAE International, 2005-01-1248. [9] Lin, P.-C.; Pan, J.; Pan, T. Failure modes and fatigue life estimations of spot friction welds in lapshear specimens of aluminum 6111-T4 sheets. Part 1: Welds made by a concave tool. International Journal of Fatigue 2008;30(1):74-89. [10] Lin, P.-C.; Pan, J.; Pan, T. Failure modes and fatigue life estimations of spot friction welds in lapshear specimens of aluminum 6111-T4 sheets. Part 2: Welds made by a flat tool. International Journal of Fatigue 2008;30(1):90-105. [11] Tran, V.-X.; Pan, J.; Pan, T. Fatigue behavior of aluminum 5754-O and 6111-T4 spot friction welds in lap-shear specimens. International Journal of Fatigue 2008;30(12):2175-90. [12] WISHAW, M.; BARTON, D. P. Comparative vacuum monitoring: a new method of in-situ real time crack detection and monitoring. In: Proceding of 10th Asia-Pacific Conference On Nondestructive Testing, 2001. Brisbane. [13] Roach, D. Real time crack detection using mountable comparative vacuum monitoring sensors. Smart Structures and Systems 2009; 5(4):317-28. [14] Kousorakis, A.; Mouritz, A.P.; Bannister, M.K. Interlaminar properties of polymer laminates containing internal sensor cavities. Composite Structures 2006;75: 610–18. [15] Kousorakis, A.; Bannister, M.K.; Mouritz, A.P. Tensile and compressive properties of polymer laminates containing internal sensor cavities. Composites: Part A 2008;39:1394–403. [16] Malafaia, A.M.S. Avaliação das propriedades mecânicas de tração e fadiga, com monitoramento de trincas, de juntas de Al AA2024-T3 soldadas a ponto por fricção-mistura. Master’s thesis, University of São Paulo, 2009. Available at: http://www.teses.usp.br/teses/disponiveis/88/88131/tde-19032009-155633/publico/Disser_Artur_Malafaia.pdf [17] Wishaw, M. Notes on the Industrial Technologies for Crack Detection and Condition Monitoring. Perth, Structural Monitoring Systems Limited 2001;1-9. [18] Choi, C. H. New Vacuum Sensor for Detecting Surface Cracks on Welds. Maritime Platform Division, DSTO, 2002. [19] Malafaia, A.M.S.; Milan, M.T.; Oliveira, M. F.; Spinelli, D. Fatigue Behavior of Friction Stir Spot Welding and riveted joints in an Al alloy. Procedia Engineer 2010;XX:xx-xx
