Fatigue Crack Growth Rate Behaviour of HSLA Steel at Varying Load Amplitudes

ScienceDirect

Available online at www.sciencedirect.com

ScienceDirect

Structural Integrity Procedia 00 (2018) 000–000

Available online www.sciencedirect.com Available online at at www.sciencedirect.com Structural Integrity Procedia 00 (2018) 000–000

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

ScienceDirect ScienceDirect

Procedia Structural Structural IntegrityIntegrity Procedia1400(2019) (2016)330–336 000–000

www.elsevier.com/locate/procedia 2nd International Conference on Structural Integrity and Exhibition 2018

2nd International ConferenceRate on Structural Integrityof andHSLA Exhibition 2018 at Fatigue Crack Growth Behaviour Steel VaryingRate LoadBehaviour Amplitudes Fatigue Crack Growth of HSLA Steel at

XV Portuguese Conference on Fracture, PCF 2016, 10-12 February 2016, Paço de Arcos, Portugal

Loadb, Amplitudes Sachin Bandgara, Varying Chiradeep Gupta Gaurav Raoa, Pranshu Malik c,

bof a a Thermo-mechanical modeling high pressure turbine blade of an andb,K. Sridhar R.N.Singh Sachin Bandgara, Chiradeep Gupta Gaurav Raoa, Pranshu Malik c, airplaneBARC, gas turbine engine NMRL,DRDO,Ambernath-421506,India; b Trombay-400085; Western a Naval Command, Indian Navy a

b

c

R.N.Singh and K. Sridhar a c a c NMRL,DRDO,Ambernath-421506,India; Western Naval*Command, Indian Navy P. Brandãob BARC, , V. Trombay-400085; Infanteb, A.M. Deus

a

Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal IDMEC, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugalto propagation of fatigue cracks, as majority of the failures Steels for ship building applications has to possess adequate resistance Abstract c CeFEMA, of Mechanical Instituto Superior Lisboa, Av.within Rovisco 1, 1049-001 in service are Department due to metal fatigue. In Engineering, this work, the fatigue crack Técnico, growth Universidade rate (FCGR)debehaviour thePais, Paris regime ofLisboa, two Portugal high strength low alloy (HSLA) steels were studied. In particular the experiments were directed to reveal the effect of load

Abstract b

Steels for ship building applications has to possess adequate resistance to propagation of fatigue cracks, as majority of the failures ratio(R) on the Paris constants for the two grades of (FCGR) HSLA steels. Results indicated that thereofistwo an in service(Tension-Tension) are due to metal fatigue. In thislaw work, the fatigue crack growth rate behaviour within the Paris regime increase in Paris slope 'm' and decrease in Y intercept 'C' with increase in load ratio for both the steels. Fractography study was high strength low alloy (HSLA) steels were studied. In particular the experiments were directed to reveal the effect of load Abstract carried using SEM at locations to various of grades stress intensity factor in order to reveal possible ratio(R)out (Tension-Tension) on thecorresponding Paris law constants for values the two of HSLA steels. Results indicated that reasons there is for an acceleration of crack growth with change in R ratio from 0.1 to 0.5. It was found that secondary cracks are predominant in Steel Duringintheir modern aircraft components are subjected to increasingly operating study conditions, increase Parisoperation, slope 'm' and decrease in Yengine intercept 'C' with increase in load ratio for both thedemanding steels. Fractography was A at a load R=0.1 as compared to R=0.5. However for Steelof B,cause secondary cracks at reveal bothtypes the load ratios. From especially the of high pressure turbine (HPT) blades. Such conditions these parts towere undergo different of time-dependent carried out ratio using SEM at locations corresponding to various values stress intensity factor in found order to possible reasons for the plot of crack growth 'a' vs no. of cycles 'N', it became evident, the number of cycles for the same range of crack length is degradation, one of which is creep. A model using the finite element method (FEM) was developed, in order to be able to predict acceleration of crack growth with change in R ratio from 0.1 to 0.5. It was found that secondary cracks are predominant in Steel lesser for R=0.1 than that for R=0.5 for both Steel A and Steel B. Limiting values of ‘∆K’ and ‘K’ max has been obtained for the creep behaviour of HPT blades. Flight data records (FDR) for a specific aircraft, provided by a commercial aviation A at a load ratio of R=0.1 as compared to R=0.5. However for Steel B, secondary cracks were found at both the load ratios. From various growth rates atofdifferent load for data steel A and B. of flight were used to'a'‘da/dN’ obtain and'N', mechanical for three different order to create the length 3D model thecompany, plotcrack of crack growth vs no.thermal cycles it ratio became evident, thesteel number cyclescycles. for theInsame range of crack is needed for the FEM analysis, a HPT blade scrap was scanned, and its chemical composition were lesser for R=0.1 than that for R=0.5 for both Steel A and Steel B. Limiting values of ‘∆K’ and ‘K’and maxmaterial has beenproperties obtained for obtained. The data that gathered was fedload intoratio the FEM model andsteel different various crack growth rateswas ‘da/dN’ at different for steel A and B. simulations were run, first with a simplified 3D rectangular block shape, in order to better establish the model, and then with the real 3D mesh obtained from the blade scrap. The © 2018 The Authors. Published by Elsevier B.V. in Elsevier terms ofB.V. displacement was observed, in particular at the trailing edge of the blade. Therefore such a © overall 2019 Theexpected Authors.behaviour Published by This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) model useful in the goalthe ofCC predicting turbine blade life, given a set of FDR data. This is an can openbeaccess article under BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection andAuthors. peer-review under responsibility of Peer-review under responsibility of the SICE 2018 organizers. © 2018 The Published by ElsevierofB.V. Selection and peer-review under responsibility Peer-review under responsibility of the SICE 2018 organizers. © 2016 Published Elsevier B.V. This is an The openAuthors. access article underbythe CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the Scientific Committeeunder of PCF 2016. Selection and peer-review under responsibility of Peer-review responsibility of the SICE 2018 organizers.

Keywords: FCGR; Paris region; HSLA steel; load ratio; Stress intensity factor.

Keywords: High Paris Pressure Turbine Blade; FiniteStress Element Method; 3D Model; Simulation. Keywords: FCGR; region; HSLA steel;Creep; load ratio; intensity factor. * Corresponding author. Tel.: 7387116029 Email address: [email protected]

* Corresponding author. Tel.: 7387116029

Email address: [email protected] 2452-3216 © 2018 The Authors. Published by Elsevier B.V.

This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of Peer-review under responsibility of the SICE 2018 organizers. 2452-3216 © 2018 The Authors. Published by Elsevier B.V.

This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under * Corresponding author. Tel.: +351responsibility 218419991. of Peer-review under responsibility of the SICE 2018 organizers. E-mail address: [email protected]

2452-3216 © 2016 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the Scientific Committee of PCF 2016.

2452-3216  2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of Peer-review under responsibility of the SICE 2018 organizers. 10.1016/j.prostr.2019.05.041

2

Sachin Bandgar et al. / Procedia Structural Integrity 14 (2019) 330–336 Sachin V Bandgar/ Structural Integrity Procedia 00 (2018) 000–000

331

1. Introduction Ship hull structures and their integrity are of paramount importance for ship building industries as these materials are subjected to high fatigue loading during operational conditions. This is particularly true for high strength low alloy steels which are being used nowadays for hulls and superstructures due to their enhanced properties and lowercost fabrication practices [Czyryca EJ. (1990)]. High Strength Low Alloy steel (HSLA) or mircoalloyed steel have properties of plain carbon steel as well as high tensile steels with better mechanical properties, higher load carrying capacity, lighter weight(High strength to load ratio) and good resistance to corrosion than plain carbon steels. Due to these factors, HSLA steels are used in heavy constructions like ship building, oil and gas transmission lines, and offshore drilling platforms. But, majority of HSLA steels are prone to fatigue failure in service [Htayaung (2007)]. Also, it is reported that 90% of mechanical failure is due to metal fatigue. The fatigue properties of a high strength low alloy steel are governed by its chemical composition, processing history, microstructural features, nature of loading and the test environment to which they are exposed to in service. Hence the fatigue crack growth rate (FCGR) behaviour in air of two different HSLA steels (with varying microstructures) at different load ratios are studied in this paper. Of the two HSLA steels, steel A has Ferritepearlite microstructure and steel B has Tempered martensite. The chemical composition of steel A and steel B is mentioned in table1 and 2.Mechanical properties are mentioned in table 3. Table 1. Composition of Steel A Composition

C

S

P

Mn

Si

Cr

Ni

Steel A

0.09

0.004

0.004

1.45

0.38

0.07

0.072

Table 2. Composition of Steel B Composition Steel B

C 0.073

S

P

0.01

0.02

Mo 0.28

Si

Cr

0.28

0.45

.

Fig 1. (a) Steel A-Ferrite+ pearlite ; (b) Steel B-Tempered martensite Table 3. Mechanical properties of Steel A and Steel B Material

YS (MPa)

UTS (MPa)

%EI

Steel A

450 + 10

520 + 20

25 + 10

Steel B

655 + 10

720 + 20

15 + 10

Ni 1.88

Sachin Bandgar et al. / Procedia Structural Integrity 14 (2019) 330–336 Sachin V Bandgar/ Structural Integrity Procedia 00 (2018) 000–000

332

3

There are three basic factors necessary to cause fatigue: a maximum tensile stress of sufficiently high value, a large enough variation or fluctuation in the applied stress and a sufficiently large number of cycles of the applied stress. The process of fatigue consists of three stages:  Initial fatigue damage (stress concentrations) leading to crack nucleation and crack initiation.  Progressive cyclic growth of a crack (Crack propagation) until the uncracked cross section of a part becomes too weak to sustain the loads imposed.  Final, sudden fracture of remaining cross section.

2. Experimental Procedure ASTM E647 standard is followed to perform FCGR test. The testing was done on make: MTS 100KN machine. The Compact Tension (CT) specimen is used for FCGR tests as shown in Fig 2. Precracking and actual FCGR test of the CT sample was done at 10 Hz. The maximum load for all specimens kept constant i.e. 12 KN with different load ratio 'R'.

Fig 2. CT sample drawing

3. Results and discussion 3.1. FCGR at load ratio of 0.1 FCGR test of steel A was conducted at constant load corresponding to ∆K value of 27 MPa*m0.5with a load ratio of 0.1. The data obtained from COD gauge and load cell were processed with available post processing software to generate da/dN vs ∆K. The plotted data of da/dN vs ∆K is shown in Fig 3. The data obtained was smoothened by sixth order polynomial and then fitted by power law to obtain 'm' and 'C' values. The 'm' and 'C' values are calculated from the graph is mentioned in the table 6. Similar procedure was followed for obtaining FCGR of steel B with a load ratio of 0.1. From the table it is evident that there is marginal change in the value of 'm' for steel A and steel B at load ratio of 0.1, whereas the value of intercept 'C' increased considerably for steel B. The FCGR at R=0.1 is higher in steel A than steel B for almost same value of crack length at different ∆K, which is shown in table 4 and 5. Similarly for R=0.5, FCGR for steel A is higher than steel B at different ∆K. Table 4. Steel A at R=0.1 ∆K

30

35

40

45

50

da/dN(mm/cycle) a(mm)

2.04 x 10-4 16.99

3.06x 10-4 19.978

4.35 x 10-4 22.43

5.96 x 10-4 24.47

7.38 x 10-4 26.12

4

Sachin Bandgar et al. / Procedia Structural Integrity 14 (2019) 330–336 Sachin V Bandgar/ Structural Integrity Procedia 00 (2018) 000–000

333

Table 5. Steel B at R=0.1 ∆K

30

da/dN(mm/cycle) a(mm)

1.25 x 10 17.01

35 -4

3.04x 10 19.97

40 -4

2.79 x 10 22.41

45 -4

4.67 x 10 24.43

50 -4

4.38 x 10-4 26.16

3.2. FCGR at load ratio of 0.5 FCGR test of steel A was conducted at constant load corresponding to ∆K value of 15 MPa*m0.5 with a load ratio of 0.5. The plotted data of da/dN vs ∆K is shown in Fig 4. The scatter data obtained smoothened by sixth order polynomial and then fitted by power law to obtain 'm' and 'C' values. Similar procedure was followed for obtaining FCGR of steel B with a load ratio of 0.5. From the table it is evident that there is decrease in the value of 'm' for steel B than steel A at load ratio of 0.5.Whereas the value of intercept 'C' increased by one order of magnitude considerably for steel B.

Fig 3.da/dN vs ∆K (a) SteelA R=0.1; (b) Steel A R=0.5

Fig 4.da/dN vs ∆K (a) Steel B R=0.1; (b) Steel B R=0.5

3.3. Effect of load ratio In case of Steel A, FCGR test at R= 0.1 and 0.5 reveals that the value of Paris slope 'm' increases with increase in 'R' ratio. Similar behaviour is found in the case of steel B. However, the increase in the absolute value of 'm' was found to be higher (around 70% more than that of R=0.1) for steel A which has Ferrite-pearlite microstructure. Whereas in the case of steel B, the increase in 'm' value was around 50% more than that of R=0.1 which has

334

Sachin Bandgar et al. / Procedia Structural Integrity 14 (2019) 330–336 Sachin V Bandgar/ Structural Integrity Procedia 00 (2018) 000–000

5

Tempered martensite microstructure. As regards it was found that the decrease in intercept 'C' when load ratio was increase to 0.5 is significant in case of steel A than steel B. This indicates that effect of 'R' ratio is significant for steel A than steel B. Table 6. 'm' and 'C' values Material  Steel A  Steel A  Steel B    Steel B 

Load  ratio R=0.1 R=0.5 R=0.1 R=0.5

Paris  slope(m) 2.353 4.008 2.223 3.363

Y intercept(C) 8.59E‐8 5.39E‐10 1.03E‐7 2.28E‐9

Fig 5.a vs N (a) Steel A at R=0.1 &R=0.5; (b) Steel B at R=0.1 &R=0.5

Fig 6. ΔK vs Kmax (a) Steel A at R=0.1 & R=0.5; (b) Steel B at R=0.1& R=0.5

3.4. Effect of Kmax In the conventional approach, for a fatigue crack growth to propagate, ΔK is often identified as effective driving force representing intrinsic material behaviour. However in an Unified Approach, proposed by Sadananda and Vasudevan [K. Sadananda (2004)], both parameters Kmax and ΔK are considered and they contribute to two crack tip driving forces . It must be noted that both Kmax and ΔK has a threshold value which must be met for the crack to grow. From this it can be further deduced that in the case of Paris region, for crack to grow at a given growth rate, a

6

Sachin Bandgar et al. / Procedia Structural Integrity 14 (2019) 330–336 Sachin V Bandgar/ Structural Integrity Procedia 00 (2018) 000–000

335

limiting value of both Kmax and ΔK must be met as per the stated unified approach. Hence based on the results and the range of data obtained in our work, plots of two significant parameters ΔK vs Kmax was plotted for various crack growth da/dN for Steel A and Steel B Fig. 6. From the plot in Fig.6 (a), in the case of steel A, it is evident that limiting Kmax value for crack to grow at da/dN = 2x 10-4 mm/c with a load ratio of R=0.1 is 33.06 MPam1/2 and R=0.5 is 50.834 MPam1/2. This indicates the limiting value of Kmax significantly increases with increasing R ratio. From the experimental results, limiting Kmax values for other crack growth rates viz. 3x 10-4 mm/c and 4x 10-4 mm/c are also shown in the fig.6. Similar behaviour of increased Kmax with increasing R-ratio has been observed for Steel B as well. 3.5 Fractographic analysis Scanning Electron Microscopy (SEM) of fractured surface is done at different ∆K to study the effect of stress intensity factor on crack morphology. It was found that secondary cracks are predominant in steel A at a load ratio of R=0.1 as compared to R=0.5. However for steel B, secondary cracks were found at both the load ratios. A representative fractograph of steel A and steel B at ∆K of 35 MPa*m0.5 is shown in Fig 7 and 8.

Fig 7. (a) Steel A at R=0.1at ∆K=35 MPa*m0.5; (b) R=0.5 at ∆K=35 MPa*m0.5

Fig 8. (a) Steel B at R=0.1 for ∆K=35 MPa*m0.5; (b) at R=0.5 for ∆K=35 MPa*m0.5

4. Conclusion  Results indicated that there was increase in Paris slope 'm' and decrease in Y intercept 'C' with increase in load ratio for both steel A and B.  In respect of fracture mechanics parameters, the value of 'm' was 70% more and around 50% more than that of

336

Sachin Bandgar et al. / Procedia Structural Integrity 14 (2019) 330–336 Sachin V Bandgar/ Structural Integrity Procedia 00 (2018) 000–000

7

R=0.1 in case of steel A and steel B respectively.  In the case of intercept 'C', it was found that the decrease in intercept 'C' when load ratio was increase to 0.5 is significant in case of steel A than steel B.  Limiting values of Kmax and ∆K for various crack growth rates has been found for steel A and steel B.  From fractography study, it was found that secondary cracks are predominant in steel A at a load ratio of R=0.1 as compared to R=0.5. However for steel B, secondary cracks were found at both the load ratios.  Number of cycles for a given crack length was found to be lesser for R=0.1 than that of R=0.5 for both steel A and steel B.  Fatigue crack growth rate was found to be higher for steel A than steel B at both load ratio R=0.1 and 0.5. References Aung, Htay, 2007, An analysis of the study of mechanical properties and microstructural relationship of HSLA steels used in ship hulls,World Maritime University Dissertations190 Campbell F.C, 2012, Fatigue and fracture- Understanding the basics, Published by ASM International, Ohio, USA Czyryca Ernest J, 1990, Development of low-carbon, copper-strengthened HSLA steel plate for naval ship construction. David Taylor Research Center, Report DTRC-SME-90/21. Czyryca EJ, Vassilaors MMG,1993, Advances in low carbon, high strength ferrous alloys, Naval SurfaceWarfare Center, Report – CARDEROCKDIV-SME-92/64. Kim B.C, Lee S, Lee D.Y,Kim N.J, 1991,In situ fracture observations on tempered martensite embrittlement in an AlSl 4340 steel, Metallurgical Transactions A, Volume 22, Issue 8,1889–1892 Kwai S.Chan, Yi-MingPan, David Davidson and R. CraigMcClung, 1997,Fatigue crack growth mechanisms in HSLA-80 steels, Materials Science and Engineering, Volume 22, Issue 1, 1-8 Sadananda K, Vasudevan, A.K, 2003, Fatigue Crack growth mechanisms in Steels, International Journal of Fatigue Vol.25 Iss.9-11, 899-914. Sadananda K, Vasudevan, A.K, 2004, Crack tip driving forces and crack growth representation under fatigue, International Journal of Fatigue, Vol.26, Issue 1,39-47 ASTM E647-15e1, 2015, Standard Test Method for Measurement of Fatigue Crack Growth Rates, ASTM International, PA, USA.