Metal Fatigue Part 8 pot

194 Chapter IO Table 10.1 Cast numbers and chemical composition Analysis (wt. Oh) Cast No Mean size of particles added (wm) C Si Mn S P Ni Cr Mo Commercial En24 cast JK None 0.40 0.30 0.56 Laboratory En24 cast 1 None 0.42 0.26 0.65 Laboratory En24 casts with angular alumina particles added 55 73 0.45 0.38 0.69 56 65 0.45 0.46 0.68 61 65 0.47 0.23 0.69 63 40 0.47 0.33 0.70 78 30 0.41 0.16 0.54 76 20 0.44 0.23 0.69 77 10 0.46 0.24 0.68 84 73 0.43 0.26 0.66 119 63 0.41 0.47 0.66 118 45 0.43 0.43 0.67 117 28 0.43 0.42 0.67 116 19 0.41 0.37 0.67 115 9 0.39 0.29 0.69 Laboratory En24 casts s11 52 54 s7 S6 S8 with 73 63 45 34 20 10 particles added 0.31 0.57 0.23 0.56 0.25 0.54 0.35 0.61 . . 0.21 0.49 0.26 0.53 0.012 0.022 0.022 0.022 0.021 0.022 0.01 1 0.017 0.01 1 0.014 0.013 0.01 1 0.015 0.017 0.01 1 0.021 0.012 0.012 0.013 0.012 0.017 0.015 0.035 0.024 0.024 0.023 0.023 0.014 0.012 0.017 0.01 1 0.019 0.016 0.017 0.015 0.014 0.033 0.019 0.016 0.01 5 0.016 0.01 5 1.46 1.46 1.55 1.59 1.51 1.54 1.66 1.60 1.56 1.66 1.56 1.56 1.51 1.61 1.56 1.63 1.76 1.74 1.64 1.62 1.53 1.06 1.16 1.28 1.42 1.30 1.11 1.02 1.06 1.02 1.19 1.13 1.16 1.17 1.20 1.23 1.28 1.27 1.31 1.27 1.25 1.25 0.21 0.32 0.34 0.31 0.30 0.36 0.33 0.33 0.33 0.37 0.32 0.34 0.31 0.36 0.35 0.30 0.32 0.30 0.33 0.30 0.33 Table 10.1 shows the cast number, the nominal size of the added alumina particles, and the results of chemical analysis of the ingots. Particles of two shapes were added, angular and spherical. Angular particles were added to casts numbers 55, 56, 61, 63, 76-78, 84 and 115-119, and spherical particles to casts numbers S2, S4, S6-S8 and S 11. All ingots were forged and rolled to a 19.05-mm diameter bar. All test piece blanks were annealed before machining by heating at 650°C for 4 h. They were then rough machined to 0.76 mm oversize in all dimensions. Subsequent heat treatment was heating at 850°C for 1 h, followed by oil quenching and tempering at 200°C for 8 h. The hardness of each specimen was checked on a Vickers hardness testing machine, and specimens then finally ground to the appropriate final dimensions shown in Fig. 10.1. Therefore, the measured value of hardness may be a little higher (%lo%) than that in the final state of the specimens [4]. The purpose of the tests carried out by Duckworth and Ineson was to investigate the effects of the artificially introduced alumina particles on the initiation of fatigue. Therefore, all tests were conducted at a constant stress level, above the fatigue limit, at which the majority of the test pieces would be expected to fail. Thus, a nominal stress of 7 IO MPa was used. Most of the fatigue tests were performed in two-point loading using Wohler-type rotating bending fatigue testing machines. Tension compression fatigue tests were carried out at zero mean stress (R = - 1) in a Losenhausen universal fatigue testing machine, model UHW6, having a maximum capacity of 29.9 kN. After fatigue testing, all the fractured specimens were examined, using an optical Effects of Shape and Size of Artificially Introduced Alumina Particles on En24 Steel 195 c I 196.9 (b) Figure 10.1 Specimen geometry. (a) Rotating bending beam specimen. (b) Tension compression speci- men. Dimensions in millimetres. microscope, at magnifications of x35 and x 100. In every case, the fracture appearance of the test specimens containing added alumina particles was different from that observed on the specimens without artificial inclusions. In the former, a circular area of lighter colour than the remainder of the fracture surface, the so-called fish eye, was usually observed, and in the centre of this fish eye, an inclusion very often remained in one half of a specimen. In some cases, the inclusion had fractured, leaving part in each half, and in the few remaining situations, the inclusion had completely shattered, leaving a hole in both fracture surfaces. 10.2 Rotating Bending Fatigue Tests without Shot Peening Table 10.2 shows the fatigue test results. 196 Chapter 10 Table 10.2 Rotating bending fatigue test results, angular particles in specimens not shot peened Distance Nominal Inclusion from stress at Fatigue limit predicted Specimen Cycles to size, darea surface, inclusion, by equations (6.1), (6.2), No failure, Nt (pm) h (pm) u' (MPa) (6.3), GlMPa) o'l~. Cast No 55, Hv = 606 A1 1.05 X 106 A5 4.80 X 109 A9 1.01 x 106 A3 1.57 x 107 A6 6.56 x 104 AI 2 2.83 x 105 Cast No 56, Hv = 614 A1 4.33 x 106 A4 1.41 x 105 A6 7.11 x 104 A7 1.96 x 105 A9 7.02 x 104 A10 8.16 x 104 A1 2 3.58 x io4 A13 1.95 X lo* A14 9.76 x 104 A1 5 6.47 X lo6 Cast No 61, Hv = 610 A9 5.97 x 106 A10 1.56 X los A12 7.84 x 105 AI 3 8.50 x lo6 A14 3.31 x 105 Cast No 63, Hv = 610 A4 1.29 x 105 A5 2.21 x 105 A6 4.57 x 107 A7 8.96 x 10' A8 4.52 x 107 A12 2.81 x 105 A1 5 6.20 x loa AI 7 5.57 x 10' Cast No 78, Hv = 602 A3 1.64 x lo7 A5 4.68 X 10" A3 3.11 X 10" A6 4.02 x 105 A7 7.21 x 104 A9 3.43 x 10" A12 4.07 x lod A14 4.39 x 107 A1 5 5.34 x 106 A17 2.13 x 107 Cast No 76, Hv = 606 A3 8.34 x 105 A7 6.71 x lo6 A8 1.07 x 104 A10 1.42 X lo6 Cast No 77, Hv = 610 A3 7.86 X IO6 A7 1.57 x 105 A8 8.91 x 107 A9 7.38 X 106 77.9 88.8 31.4 30.1 55.5 65.2 47.6 35.9 62.0 51.0 51.7 66.8 53.2 77.7 56.2 51.4 20.4 38.9 38.8 39.6 64.2 15.7 14.7 29.5 30.2 24.1 26.6 28.0 14.6 19.7 17.2 19.7 33.2 19.8 15.6 15.3 29.1 35.8 25.1 11.5 21.7 15.3 8.86 14.3 12.5 11.5 11.5 290 327 0 15 1 03 122 118 40 0 30 41 71 32 45 50 219 just breaks free surface just breaks free surface 134 48 59 0 0 20 58 0 58 iust breaks free surface 0 0 0 13 58 0 0 0 151 0 115 22 10 0 breaks surface break free surface break free surface 20 15 656 649 710 707 691 687 688 703 710 704 702 697 704 702 701 669 710 710 685 701 699 710 710 706 699 710 699 710 710 710 710 708 699 710 710 710 682 710 689 706 708 710 710 710 710 706 707 548 (6.3) 536 (6.3) 584 (6.1) 580 (6.2) 580 (6.3) 565 (6.3) 601 (6.3) 630 (6.3) 528 (6.1) 537 (6.2) 593 (6.3) 568 (6.3) 534 (6.2) 501 (6.21 585 (6.3) 594 (6.3) 623 (6.2) 559 (6.2) 619 (6.3) 617 16.3) 569 (6.3) 660 (6.1) 667 (6.1) 586 (6.2) 645 (6.3) 614 (6.1) 659 (6.3) 591 (6.2) 668 (6.1) 635 (6.11 643 (6.1) 619 16.2) 628 (6.3) 628 (6.1) 653 (6.1) 655 (6.1) 642 (6.3) 718 (6.1) 620 (6.3) 662 (6.3) 754 (6.3) 622 (6.1) 650 (6.2) 661 (6.2) 676 (6.2) 758 (6.3) 758 16.71 1.20 1.21 1.22 1.22 1.19 1.22 1.14 1.11 1.35 1.31 1.18 1.23 1.32 1.40 1.20 1.13 1.14 1.27 1.11 1.14 1.23 1.08 1.06 1.21 1.08 1.16 1.06 1.20 1.06 1.12 1.10 1.14 1.11 1.73 1.09 1.08 1.06 0.99 1.1 1 1.07 0.94 1.14 1.09 1.07 1.05 0.93 0.93 Effects of Shape and Size of Artificially Introduced Alumina Particles on En24 Steel 197 1.0 -(050.9- 2 0.8- 0.7- 0.6- 0.5- 0.4- 0.3- 0.2- 0.1- 0 W ca 0 0 Surface inclusion A Subsurface inclusion CI Interior inclusion I I I I 90 80 70 5 ,","I 40- k 30- 20': 10 0- Fig. 10.2 shows a comparison between the ratio of the applied stress, o', at an inclusion to the calculated fatigue limit, uk, with the number of cycles to failure, Nf. In this figure, the symbols 0, A and indicate the location of inclusions. There are no data points located below a'/uk = 0.9, showing the high accuracy of the evaluation method. and the number of cycles to failure for specimens fractured from a surface or a subsurface inclusion. The data trend suggests that the larger the value of l/area, the shorter the fatigue life. This figure indirectly verifies the utility of the geometrical parameter, In this figure, the fracture data from internal inclusions were not plotted because the fatigue crack propagation from Fig. 10.3 shows the relationship between - - A 0 Surfaceinclusion - A Subsurface inclusion 0 A A A OA 43 A A 0 Ao &a 0 I I I 1 A. B. I /\"I - Figure 10.4 Qpical alumina particles, as added to ingots. (A) Typical spherical alumina particles. (a) 73 pm nominal size. (b) 40 ym nominal size. (c) 10 pm nominal size. (B) Typical angular alumina particles. (a) 73 ym nominal size. (b) 40 pm nominal size. (c) 10 pm nominal size (Duckworth and Ineson 9 6 ill). % 5( L. 0 Effects of Shape and Size of Artificially Inrroduced Alumina Particles on En24 Steel 199 40 30 20 an internal inclusion would cause a variation in the ,/ZEZ-Nf relationship due to the difference in the location of the inclusions. - - - 10.3 Rotating Bending Fatigue Tests on Shot-Peened Specimens 10 The purpose of these experiments was to examine the effects of the shape of inclusions on fatigue strength. Fig. 10.4 shows typical spherical and angular alumina particles, as added to the ingots. The nominal size does not necessarily indicate the true size of each inclusion, because of variation in the sizes of inclusions of the same nominal size. After examining Fig. 10.4, a typical common answer to the question “Which is more detrimental, a spherical inclusion or an angular inclusion?” would be “An angular inclusion.” However, reality does not correspond to this answer. Since the sizes of the inclusions found on the fracture surface have large scatter, all the results are classified separately in terms of for spherical and angular inclusions at fracture origins. Fig. 10.5 shows histograms of ,hZZ at the fracture surface for (a) spherical alumina and (b) angular alumina. Although the values of .Jarea do show a larger scatter, there is no significant difference in the distribution of 1/.ye. for spherical and angular inclusions. - - - - % Oo in 20 30 40 ! Ja (a) Number 24 Mean value 57.504 Standard deviation 25.510 L70 80 90 100 110 120 130 ?a, pm I I I1 Number 36 Mean value 66.294 Standard deviation 22.954 -l Jarea, Prn Figure 10.5 Histograms of e inclusions at fracture origins. (a) Spherical alumina particles. (b) An- gular alumina particles. 200 Chapter 10 U A 0 Rotating bending A Angular particles 0 Spherical particles Tension-compression 0 Spherical particles 104 105 106 107 108 Number of cycles N, Figure 10.6 Comparison between failure stress and predicted fatigue strength for shot-peened rotating bending specimens and for shot-peened tension compression specimens. Fig. 10.6 shows the relationship between d/a; and Nf in rotating bending fatigue for angular and spherical particles. All data points have values of o'/D; L 0.89. The maximum evaluation error is approximately lo%, which may be caused by a higher estimate for HV than the actual value, as described above. Thus, the evaluation accuracy is sufficient for practical purposes, regardless of inclusion shape. If we look at the histograms of Fig. 10.5a and b from the viewpoint of, there is very little difference. This is the reason why the shape of inclusion is not significant at lower stress levels in Fig. 10.6, rather it is that is the crucial geometrical factor. Many researchers may find this conclusion difficult to accept if we concentrate our attention on 'stress concentration factors' of small defects and inclusions. If we try to solve a problem of this kind by stress concentration factors, we shall not be able to reach a complete solution. Although the angular shape of TiN inclusions has been thought to be the cause of their detrimental effect, we can understand from the experiments by Duckworth and Ineson that this widely accepted viewpoint is not correct with respect to fatigue limits. However, at higher stress levels the specimens fractured from angular alumina particles do tend to show slightly shorter fatigue lives compared with those fractured from spherical alumina particles, as shown in Fig. 10.6. The most likely reason is that cracks nucleate earlier from angular inclusions than from spherical Efects of Shape and Size of Art$cially Introduced Alumina Particles on En24 Steel 20 1 Table 10.3 Rotating bending fatigue test results, angular particles in shot-peened specimens Nominal stress at Fatigue limit predicted by Inclusion Cycles to size, qarea Distance from inclusion, equations (6.1), (6.2)and (6.3). Cast No 84. Hv = 581 failure, N, (pm) surface, h (pm) U' (MPa) d(MPa) U'/C& 1.16 X 10' 52.0 374 641 566 (6.3) 1.13 1.57 X lo6 68.0 686 583 541 (6.3) 1.08 6.02 x lo5 93.0 327 649 514 (6.3) 1.26 7.67 X lo5 74.2 257 662 533 (6.3) 1.24 8.96 X lo5 90.8 318 651 516 (6.3) 1.26 2.55 x 105 93.4 375 640 513 (6.3) 1.25 4.12 X lo5 69.1 41 8 632 540 (6.3) 1.17 6.97 X lo6 56.0 453 626 559 (6.3) 1.20 1.45 x lo6 79.4 449 627 527 (6.3) 1.19 4.01 X los 97.0 52 1 613 510 (6.3) 1.20 3.04 X lo5 87.9 445 627 519 (6.3) 1.21 Cast No 119, Hv = 579 4.92 X lo6 52.7 325 650 563 (6.3) 1.15 1.63 X lo6 63.8 437 629 546 (6.3) 1.15 1.66 x lo6 81.3 36 1 643 524 (6.3) 1.23 8.31 x los 83.8 335 648 521 (6.3) 1.24 2.13 x lo6 55.4 364 642 558 (6.3) 1.15 3.66 X lo5 85.6 427 631 519 (6.3) 1.21 2.98 x lo6 63.3 276 659 546 (6.3) 1.21 Cast No 118, HV = 581 2.80 X 10' 56.7 582 602 558 (6.3) 1.08 6.84 x lo6 53.9 57 5 603 563 (6.3) 1.07 2.43 X lo6 63.9 473 622 547 (6.3) 1.14 1.05 x 107 44.2 522 613 582 (6.3) 1.05 3.68 X lo6 73.1 766 568 535 (6.3) 1.06 1.56 X lo7 57.1 557 607 557 (6.3) 1.09 2.62 X lo6 56.7 464 624 558 (6.3) 1.12 8.28 X lo7 33.2 402 635 610 (6.3) 1.04 4.68 X lo7 51.7 424 631 567 (6.3) 1.11 1.05 x lo8 26.9 367 642 632 (6.3) 1.02 Cast No 117, Hv = 581 2.57 X lo6 54.5 382 639 562 (6.3) 1.14 4.43 X 10' 72.1 803 561 536 (6.3) 1.05 1.75 X lo7 46.6 521 613 576 (6.3) 1.06 Cast No 116, Hv = 574 9.08 x lo7 36.2 277 659 595 (6.3) 1.11 4.78 X lo5 125.9 510 61 5 484 (6.3) 1.27 1.61 x lo7 37.6 287 657 591 (6.3) 1.11 9.04 X lo5 116.0 860 550 490 (6.3) 1.12 5.24 x lo7 33.6 668 586 603 (6.3) 0.97 inclusions, resulting in shorter fatigue lives at higher stress levels, although the fatigue limit is determined by the condition for non-propagation of a crack emanating from an inclusion. Generally speaking, the compressive stress on the specimen surface produced by shot peening makes the effective distance, h, of the fatal inclusion from the surface deeper, as seen by comparing Tables 10.3 and 10.4 with Table 10.2. In shot-peened specimens, high compressive residual stresses exist on the surface, and tensile residual stresses exist in the interior. Therefore fractures, on the whole, initiate from internal inclusions or defects. The reason why some values of o'/m;, in Fig. 10.6 are a little lower than 1.0 202 Chapter 10 Table 10.4 Rotating bending fatigue test results, spherical particles in shot-peened speamens Inclusion Cycles to size, .\/area Distance from failure, N, (pm) surface, h (pm) Cast No S11, Hv = 556 1.27 X lo6 59.5 341 2.64 x los 137.4 1100 7.88 x io6 76.7 51 5 1.60 x lo6 93.1 830 Cast No S2, Hv = 560 6.77 x lo7 47.8 420 6.21 X lo7 57.6 460 2.85 x lo7 49.2 470 Cast No S4. Hv = 554 3.07 x lo7 40.8 390 2.69 X 10’ 112.6 1300 2.34 X lo7 46.1 375 7.11 x lo6 53.2 470 3.34 x 107 46.1 450 6.39 x lo6 51.4 357 1-54 x 107 44.3 675 Cast No 57, Hv = 566 3.06 x lo7 55.8 1200 3.07 x lo7 34.1 500 6.58 x lo7 33.8 310 2.34 x 107 41.7 320 Cast No S8, Hv = 550 5.55 x 107 54.9 680 2.37 x 107 58.5 655 2.53 X lo6 72.5 440 8.61 X lo6 40.3 56 2.73 x lo7 46.1 390 4.87 X lo7 26.6 41 5 Nominal stress at inclusion, equations (6.1),(6.2)and (6.3), u‘ (MPa) (MPa) atla&. Fatigue limit predicted by 647 506 614 556 632 625 623 638 469 640 623 626 644 585 487 617 652 651 584 588 628 700 638 633 534 (6.3) 1.21 464 (6.3) 1.09 512 (6.3) 1.20 495 (6.3) 1.12 557 (6.3) 540 (6.3) 554 (6.3) 1.13 1.16 1.12 567 (6.3) 1.13 478 (6.3) 0.98 555 (6.3) 1.15 542 (6.3) 1.15 555 (6.3) 1.13 545 (6.3) 1.18 559 (6.3) 1.05 547 (6.3) 594 (6.3) 595 (6.3) 575 (6.3) 536 (6.3) 630 (6.3) 512 (6.3) 564 (6.3) 552 (6.3) 605 (6.3) 0.89 1.04 1.10 1.13 1.09 1.1 1 1.23 1.24 1.16 1.05 may be the result of tensile residual stress in the interior, in addition to an overestimate of the hardness, Hv, at a fracture origin. The method of evaluating the effects of residual stress on the fatigue strength was explained in Chapter 8. The method is not used in the present chapter because residual stress values are not included in the data of Duckworth and Ineson. If we assume the residual stress at the fracture origin to be a, = +200 MPa, then the fatigue strength, a;, predicted using Eqs. 6.3 and 6.4 is a 7% overestimate, and the ratio of d/aL is underestimated by 7%. Thus, if we did consider the effect of residual stresses, then the evaluation error of u‘/aL would be expected to decrease. 10.4 Tension Compression Fatigue Tests In Fig. 10.6, the values a’/a; for tension compression fatigue (symbol 0) are evidently larger than those for rotating bending fatigue (symbols A and 0). We can easily understand the reason for this if we take into consideration the fact that all the fatigue tests were conducted at the same nominal stress. In tension compression, Efects of Shape and Size of Artificially Introduced Alumina Particles on En24 Steel 203 a greater volume of a specimen is subjected to high stress than in rotating bending fatigue, with its concomitant stress gradient. Accordingly, the value of l/are.,,, of the maximum inclusion in tension compression is larger than that in rotating bending, reducing the fatigue limit, a;, for tension compression, or increasing the ratio a’/ak,. Prediction of .Jnl;ea,,, for inclusions contained in a particular number of specimens can be made by the method based on extreme value statistics, as explained in previous chapters. According to extreme value statistics, the expected value of 1/.r ,,, increases with increasing test volume, or number of specimens. For example, the mean values of of inclusions at the fracture origin are 34.6 1l.m for non-shot-peened specimens in rotating bending (Fig. 10.2), 62.6 Km for shot-peened specimens in rotating bending (Fig. 10.6), and 76.8 vm for tension compression (Fig. 10.6). 10.5 References 1. W.E. Duckworth and E. Ineson: The effects of externally introduced alumina particles on the fatigue life of En24 steel, Clean Steel, Iron Steel Inst., Sp. Rep., 77 (1963) 87-103. 2. Y. Murakami, K. Kawakami and W.E. Duckworth: Quantitative evaluation of effects of size and shape of artificially introduced alumina inclusions on the fatigue strength of Ni-Cr-Mo steel, Tetsu to Hagane. 77(1) (1991) 163-170. Y. Murakami, K. Kawakami and W.E. Duckworth: Quantitative Evaluation of Effects of Shape and Size of Artificially Introduced Alumina Particles on the Fatigue Strength of 1.SNi-Cr-Mo (En24) Steel, Int. J. Fatigue, 13(6) (1991) 489-499. 3. M. Sumita. I. Uchiyama and T. Araki: A model experiment on relationship between fatigue properties of steel and size, shape, and distribution of inclusions, Tetsu to Hagane, 57(2) (1971). 335-354. 4. Y. Murakami and H. Usuki: Rediction of fatigue strength of high-strength steels based on statistical evaluation of inclusion size, Trans. Jpn. SOC. Mech. Eng. A, 55(510) (1989). 213-221. [...]... Alloys f 402 I 4 I 479 I 522 3B17 407 476 519 L (2OOg.ave.) (3g.ave.) (3g.min.) 176 8. 26 1 28 87 6B17 409 441 453 2.53 175 126 100 20A11 223 332 380 18. 1 112 84 68 20A17 396 456 5 08 15.1 161 1 I9 100 16.9 115 85 64 13 .8 167 123 90 7 15A11 15C17 3 78 3.57 I 164 I 1 18 I 89 a,,,, : 0.2%offset yield strength, MPa % or : Ultimate tensile strength, MPa Q : True fracture stress, MPa : Reduction of area, YO HV : Vickers... Chapter 11 Table 11.2 Fatigue test results for peeimens containing an array of three notches (Endo [U]) X Failure 60 76 66 FCD60 (265MPa) FCD70 ( 280 MPa) Number of 0Ran out 0 0 0 Materials (Test stress) cvcles 0 95 85 95 120 145 120 I 04 104 101 133 133 126 30 38 46 38 33 33 32 42 42 40 1 07 X 9 .83 X lo5 0 0 0 7. 084 X lo5 X 0 0 0 0 X 107 9.6X105 0 54 171 0 50 54 I 58 171 0 X - 2. 685 X IO5 Jarea;,, containing... 12.2 Fatigue Mechanism Fig 12.3 shows S-N curves obtained by rotating bending and tension compression fatigue tests The S-N curves show fracture at >lo8 cycles, and no definite fatigue limit Although there are some differences in fatigue life between rotating bending and tension compression, no substantial difference in fatigue mechanism was found in observations of fatigue crack growth processes 2 18. .. torsion), Trans Jpn Soc Mech Eng A, 51(464) (1 985 ) 12 08- 1214 23 H Nisitani and S Tanaka: Initiation and propagation of fatigue crack in cast irons (rotating bending fatigue tests of FC25 and FCD45), Trans Jpn SOC Mech Eng A, 51(465) (1 985 ), 1442-1447 24 P Clement, J.P Angeli and A F’ineau: Shot crack behaviour in nodular cast iron fatigue, Eng Mater Shuct., 7(4) (1 984 ), 251-265 25 Y Hirose, T Kurobe, M Tsuda,... shape on fatigue strength of spheroidal graphite cast iron, IMONO (J Cast Inst Jpn.), 51(5) (1979), 281 - 286 8 M Sofue: Evaluation of role of graphite nodules in fatigue strength spheroidal graphite cast iron, IMONO (J Cast Inst Jpn.), 51(3) (1979) 159-163 9 M Sofue: Influence of microstructural constituents on the fatigue strength of spheroidal graphite cast iron, IMONO (J Cast Inst Jpn.), 53(ll) (1 981 ),... nodular graphite cast iron, Trans Jpn Soc Mech Eng A, 51(467) (1 985 ) 1660-1667 18 M Endo: Effects of graphite shape, size and distribution on the fatigue strength of spheroidal graphite Mater Sci Jpn., 38( 433) (1 989 ), 1 139-1 144 cast irons, J SOC 19 M Endo: Fatigue Strength Prediction of Nodular Cast lrons containing Small Defects, MD-Vol 28, Impact of Improved Material Quality on Properties, Product... Trans Jpn SOC Mech Eng A, 53(490) (1 987 ), 1000-1006 13 M Buchanan, J.W Provan and J.E Gruzleski: An apparatus for four-point bending fatigue testing of materials in a scanning electron microscope and its application to nodular cast iron, Metallography, 20 (1 987 ), 125-143 14 K Igawa and Y Tanaka: Microstructure and fatigue strength of cast irons, Bull Jpn Inst Metals, 13(9) (1974), 665-673 15 M Yano:... of graphites on fatigue strength of malleable iron, Trans Jpn SOC Mech Eng A, 51(461) (1 985 ) 132-135 16 M Yano: Effects of elasticity and compactness of graphites on the fatigue strength of spheroidal graphite cast iron, Trans Jpn SOC.Mech Eng A, 52( 481 ) (1 986 ), 2150-2154 17 T Ogawa, H Kobayashi, T Koide, H Nakazawa and E Kometani: Near-threshold characteristics and crack closure in fatigue crack growth... on fatigue strength of S.G cast iron, IMONO (J Cast Inst Jpn.), 42 (8) (1970), 634-6 38 2 I Niimi, M Ohashi, Y Komatsu and Y Hibino: Influence of graphite nodules on the fatigue strength of S.G cast iron, IMONO (J Cast Inst Jpn.), 4x2) (1971), 101-107 3 H Nisitani and Y Murakami: Role of nodules on bending and torsional fatigue of nodular cast iron, Sci Mach., 25(4) (1973), 543-546 4 M Sofue: On the fatigue. .. IMONO (J Cast Inst Jpn.), 48( 7) (1976) 441-447 5 A.G Fuller: Effect of graphite form on fatigue properties of pearlitic ductile irons, AFS Trans., 85 (19 78) , 527-536 6 T Shiota and S Komatsu: Influence of graphite nodule diameter on fatigue strength and crack propagation behavior of femtic spheroidal graphite cast irons under rotational bending, IMONO (J Cast Inst Jpn.), 54(7) (1 982 ),434-439 7 M Sofue: . 606 A3 8. 34 x 105 A7 6.71 x lo6 A8 1.07 x 104 A10 1.42 X lo6 Cast No 77, Hv = 610 A3 7 .86 X IO6 A7 1.57 x 105 A8 8. 91 x 107 A9 7. 38 X 106 77.9 88 .8 31.4 30.1. 66 .8 53.2 77.7 56.2 51.4 20.4 38. 9 38. 8 39.6 64.2 15.7 14.7 29.5 30.2 24.1 26.6 28. 0 14.6 19.7 17.2 19.7 33.2 19 .8 15.6 15.3 29.1 35 .8 25.1 11.5 21.7 15.3 8. 86. 83 .8 335 6 48 521 (6.3) 1.24 2.13 x lo6 55.4 364 642 5 58 (6.3) 1.15 3.66 X lo5 85 .6 427 631 519 (6.3) 1.21 2. 98 x lo6 63.3 276 659 546 (6.3) 1.21 Cast No 1 18, HV = 581 2 .80