FAILURE ANALYSIS OF MINE SHAFT HOIST BRAKE SPRINGS

Transcription

1 Steel Image Inc., Failure Analysis and Metallography 7 Innovation Drive, Suite 155, Flamborough Ontario, Canada, L9H 7H9, (289) FAILURE ANALYSIS OF MINE SHAFT HOIST BRAKE SPRINGS EXAMPLE REPORT OVERVIEW & OUTCOME Two springs apart of a mine shaft hoist brake assembly had failed. The same design has been used without issue at another mine head. Due to the assembly design, these springs are difficult to inspect and therefore, the mine wanted to better understand why these springs had failed in order to prevent re-occurrence. Analysis found that the springs had failed due to a quality issue with the spring steel. The full length of the steel coil used to make these springs would have been at risk of having unacceptable defects. Therefore, in order to guarantee reliability, it was recommended that all springs from this material lot be replaced. - Electronic Copy - Shane Turcott, M.A.Sc. Principal Metallurgist

2 FAILURE ANALYSIS OF MINE SHAFT HOIST BRAKE SPRINGS SUMMARY The two mine shaft brake springs had failed due to material flaws located along the inner diameter of the springs. These flaws acted as stress concentrators resulting in fatigue crack initiation and failure. Compression springs are inherently sensitive to stress concentrators along the springs inner diameter. The springs surfaces consisted of numerous as-manufactured, shallow cracks ranging up to 75µm (0.003 inches). Combined with (1) other springs produced from the same material lot being as risk of having material flaws and (2) the remaining springs having experienced additional loading due to spring failure, it is recommended that the remaining springs be either (a) closely monitored and/or (b) considered for retirement. 1.0 INTRODUCTION Two failed springs from a mine shaft hoist brake were submitted for analysis. As part of a braking assembly comprising of twelve springs per stack, these springs had been in service for over six years before failure of two springs were detected during inspection. Another mine had an almost identical brake system which has been in service for two year longer without incident. Figure 1 displays a photograph of the braking system. No changes in the operation of the mine hoist were reported. Steel Image was requested to determine the nature of failure. 2.0 EXAMINATION 2.1 Visual / Macroscopic Examination Figures 2 and 3 displays the springs submitted for the investigation. For the ease of reference, the springs were allocated as Springs #1 and #2. Spring #1 had failed two and a half turns from one end. Spring #2 had failed one and a half turns from an end. Crack initiation on both springs had occurred at the inner diameter of the springs at material flaws. Figures 4 and 5 display the fracture surface of Spring #1 and material flaw from which crack initiation had occurred. The depth of the material flaw was approximately 0.7mm (0.028 inches). Embedded foreign material was present on both halves of the fracture initiation site within the material flaw cavity. The flaw cavity was painted. Crack initiation at the flaw, and crack growth through approximately a third of the spring s cross-sectional area, had occurred by high cycle fatigue cracking. The crack propagation mode then changed to overload failure for the remaining two-thirds. SEM Failure Analysis of Mine Shaft Hoist Brake Springs Page 1 of 15

3 examination would later confirm this region had failed by brittle fracture. Figure 6 provides a summary of the sequence of failure for Spring #1. Spring #2 exhibited similar fracture features as Spring #1. Crack initiation had occurred at a small material flaw cavity. The depth of the cavity was measured to be approximately 0.3mm (0.012 inches). Crack propagation had initially occurred by high cycle fatigue cracking through approximately a third of the cross-section. The remainder of the fracture comprised of brittle overload fracture features. Figure 7 displays the fracture surface of Spring # SEM Fractography The fracture surface from Spring #1 was cut, cleaned and examined under a scanning electron microscope (SEM). SEM examination found multi-origin fatigue crack initiation to have occurred within the cavity of the material flaw (Figure 8). Fatigue striations were not apparent at the crack initiation site even at 10,000x magnification, indicating very short crack propagation during each loading cycle. As observed macroscopically, embedded material was present within the material flaw cavity. Energy Dispersive Spectroscopy (EDS) analysis of this material detected predominantly aluminium and oxygen (Figure 9). Lesser amounts of nickel, carbon and phosphorous were also present. Based upon its composition and appearance, this embedded material was likely entrained refractory from the steelmaking process. Examination within the center of the fatigue zone found closely spaced fatigue striations as illustrated in Figure 10a,b. These striations were consistent with high cycle fatigue crack propagation. The final failure zone comprised of mixed fracture features, the most predominant feature being intergranular brittle fracture (Figure 10c,d). Spring #2 exhibited nearly identical fracture features. Crack initiation had occurred by multi-origin fatigue cracking at a material flaw cavity. Fatigue striations were present within the first third of the fracture, typical of high cycle fatigue. The remainder of the spring fracture comprised of mixed overload fracture features, the most predominant being brittle, intergranular features. Figure 11 displays the micro-fracture features of Spring #2. No entrained material was observed at the crack initiation site. Failure Analysis of Mine Shaft Hoist Brake Springs Page 2 of 15

4 2.3 Chemical Analysis Chemical analysis of Spring #1 was performed in accordance with ASTM E1019, E1097(mod) and E1479. Table 1 lists the obtained chemical results The spring material matched AISI 5160 alloy steel. Table 1: Chemical Analysis Results Composition (wt%) C Mn Si S P Cr Ni < B Nb V Cu Sb As <0.001 < <0.01 < Optical Examination Longitudinal and transverse cross-sections were taken near the fracture on Spring #1. Optical examination of the sections revealed the core structure to comprise of tempered martensite. Such a structure is typical of a quenched and tempered heat treated AISI 5160 steel. Figure 12 displays the core structure of Spring #1. The surface of the spring exhibited numerous, shallow cracks beneath the paint (Figure 13). These cracks were filled with a high-temperature oxide. The cracks ranged up to a depth of 75µm (0.003 inches) and followed the prior austenitic grain boundaries. The cracks had been deformed open on the outer diameter surface of the spring while those on the inner diameter were very fine and closed. This indicated that the cracks had preexisted the spring formation operation. Therefore, the cracks had been present on the original bar stock, potentially from the extrusion operation producing the bar. No service-induced cracks were found to have initiated from these pre-existing cracks on any of the metallographic samples examined. A cross-section was taken through the crack initiation site on Spring #1 (location indicated in Figure 14a). Figure 14b,c displays the cavity. Entrained material was present within the cavity. A layer of oxide beneath the paint was present. 2.5 Mechanical Testing Hardness testing was conducted on material taken from both Springs #1 and #2 in accordance with ASTM E18. Table 2 lists the obtained hardness of the springs. The hardness values were typical for spring applications. Table 2: Hardness Results Sample Measurements (HRC) Avg. Hardness (HRC) Spring #1 45.5, 45.5, Spring #2 45.0, 45.0, Failure Analysis of Mine Shaft Hoist Brake Springs Page 3 of 15

5 A sub-standard tensile bar was machined from Spring #1 and tensile tested in accordance with ASTM A370. Table 3 lists the obtained results. Examination of the tensile specimen pieces after testing found the material to exhibit a classic cup-and-cone fracture morphology typical for ductile failure. The spring exhibited high strength expected for such applications. Table 3: Tensile Test Results Yield Strength Sample (ksi) UTS (ksi) Elongation (%) Reduction (%) Spring # DISCUSSION The brake springs had failed by fatigue failure initiating from material flaws. Located at the inner diameter of the springs, these material flaws had acted as stress concentrators, increasing the local stresses experienced during operation and resulting in fatigue crack initiation. Compression springs are inherently sensitive to stress concentrators at the inner diameter. Even mild stress concentrators at this location can result in crack initiation and, ultimately, failure of the spring. The material flaws had existed on the original bar material used to produce the springs. Embedded material was found within the flaw on Spring #1. This material, likely entrained refractory from the steelmaking process, had caused the formation of the cavity which had expanded during further processing of the bar. The result was a geometrical depression that acted as a local stress concentrator. Note that other springs made from the same bar material lot may also have similar material flaws, increasing the risk of additional failures. After the fatigue crack had grown through approximately a third of the springs crosssection, the remainder had failed immediately in a brittle fashion. High strength steels such as AISI 5160 steel are notch sensitive. Therefore, although this material is ductile under pure tensile conditions, the addition of a sharp crack can result in the transition to a brittle fracture. Due to (a) the small size of the fatigue cracks before final failure and (b) lack of deformation/crack opening prior to final failure, these cracks would be difficult to observe during visual inspection. The surface of the springs exhibited numerous, shallow cracks from the production of the spring bar. Although no fatigue cracks were observed to have formed at these features, these cracks would reduce the integrity/reliability of the springs. Also note that the remaining springs would have experienced additional loading after failure of two springs. Combined with the risk that other springs produced from the same bar material lot may Failure Analysis of Mine Shaft Hoist Brake Springs Page 4 of 15

6 also have material flaws, it is recommended that the remaining springs be either (a) closely monitored and/or preferably, (b) considered for retirement. 4.0 CONCLUSIONS The two mine shaft brake springs had failed due to material flaws located along the inner diameter of the springs. These flaws acted as stress concentrators resulting in fatigue crack initiation and, ultimately, causing failure. Compression springs are inherently sensitive to stress concentrators at the inner diameter. The material flaws responsible for failure had been present on the original bar stock used to produce the springs. The flaws were the result of entrained material from the steelmaking process. Other springs produced from the same bar stock may also have similar flaws and, if located along the inner diameter, would increase the risk of failure. In addition to these relatively large material flaws, the entirety of the springs surfaces consisted of numerous, shallow cracks also from the original bar stock material. Although the metallographic audit did not reveal any service fatigue cracks to have initiated from these manufacturing flaws, they would reduce the springs durability/reliability. Combined with (1) other springs produced from the same material lot being at risk of having material flaws and (2) the remaining springs having experienced additional loading after failure of two springs, it is recommended that the remaining springs be either (a) closely monitored and/or (b) considered for retirement. Failure Analysis of Mine Shaft Hoist Brake Springs Page 5 of 15

7 Figure 1: Provided photographs displaying the spring brake assembly at the head of a mine shaft. Failure Analysis of Mine Shaft Hoist Brake Springs Page 6 of 15

8 Spring #1 Spring #2 Figure 2: Photograph displaying the two springs submitted for the investigation. a) Spring #1 Figure 3: b) Spring #2 Photographs displaying the failed springs. Failure Analysis of Mine Shaft Hoist Brake Springs Page 7 of 15

9 a) Spring #1, Fracture Crack Initiation b) Spring #1, Fracture Figure 4: Crack Initiation Photographs displaying the fracture surface of Spring #1. Fatigue crack initiation had occurred at a material flaw. Failure Analysis of Mine Shaft Hoist Brake Springs Page 8 of 15

10 a) Spring #1, 10x Crack Initiation Entrained Material Entrained Material Figure 5: b) Spring #1, 10x Macrographs displaying the material flaw from which crack initiation had occurred. The ~0.7mm depression was painted. Within the depression was foreign material embedded into the steel. Spring #1 Brittle Overload Failure High Cycle Fatigue Fatigue Crack Initiation at Material Flaw Figure 6: Photograph marked with the details of failure. Fatigue crack initiation had occurred at a material flaw at the ID of the spring. The crack propagated by high cycle fatigue through a third of the spring and then the remainder failed by brittle overload failure. Failure Analysis of Mine Shaft Hoist Brake Springs Page 9 of 15

11 a) Spring #2, Fracture b) Spring #2, Fracture d) Spring #2, Fracture Fatigue Crack Initiation at this Material Flaw Figure 7: Photographs displaying the fracture on Spring #2. Similar to Spring #1, fatigue crack initiation had occurred at a material flaw located at the inner diameter of the spring. Failure Analysis of Mine Shaft Hoist Brake Springs Page 10 of 15

12 Figure 10c,d Figure 10a,b Figures 8 & 9 a) Spring #1, Location of Images b) Initiation, 50x Entrained Material c) Initiation, 200x d) Initiation, 1000x Figure 8: SEM images displaying crack initiation to have occurred by multi-origin, high cycle fatigue at the embedded material/depression. SE1, 20kV. Entrained Refractory Figure 9: EDS analysis of the entrained material at the crack initiation site. The embedded material at the initiation site was consistent with entrained refractory from the steelmaking process. SE1, 15kV. Failure Analysis of Mine Shaft Hoist Brake Springs Page 11 of 15

13 Fatigue Striations a) Fatigue Zone, 1000x b) Fatigue Zone, 10,000x Intergranular Fracture (Brittle Fracture) c) Final Failure, 500x d) Final Failure, 1000x Figure 10: SEM images displaying the fracture surface of Spring #1 within the (a,b) high cycle fatigue region and (c,d) the final, brittle overload region. Figure 8a indicates the locations from where the images were taken. Failure Analysis of Mine Shaft Hoist Brake Springs Page 12 of 15

14 Figure 11e,f Figure 11c,d Fatigue Crack Initiation at Material Flaw Figure a) Spring #2, Location of Images 11b b) Initiation, 50x Fatigue Striations c) Fatigue Zone, 1000x d) Fatigue Zone, 10,000x Intergranular Fracture (Brittle Fracture) e) Final Failure, 500x f) Final Failure, 1000x Figure 11: SEM images of Spring #2 displaying features similar to Spring #1. Crack initiation had occurred by mutli-origin high cycle fatigue at a material flaw cavity. Fatigue striations were present within the fatigue zone which changed to mixed, intergranular brittle fracture features. SE1, 20kV. Failure Analysis of Mine Shaft Hoist Brake Springs Page 13 of 15

15 a) Core, 400x b) Core, 1000x Figure 12: Micrographs displaying the core structure of Spring #1 to comprise of tempered martensite typical for a quenched and tempered 5160 alloy steel. Etched using 3% nital. Paint a) OD Surface, 100x b) OD Surface, 400x c) ID Surface, 100x d) ID Surface, 400x Figure 13: Micrographs displaying cracks along the length of the spring material (taken approximately 1 inch from the fracture). These cracks had been in the original bar material. During forming of the bar, the cracks on the outer diameter had been opened and exposed to the atmosphere during heat treatment. As-polished condition. Failure Analysis of Mine Shaft Hoist Brake Springs Page 14 of 15

16 Fracture Surface a) Spring #1, Location of Cross-Section Fracture Surface Entrained Material b) Crack Initiation, 100x c) Crack Initiation, 400x Figure 14: Micrographs displaying the cross-section taken through the material flaw on Spring #1 from which fatigue crack initiation had occurred from. Entrained material was present, likely being the cause of the flaw. Failure Analysis of Mine Shaft Hoist Brake Springs Page 15 of 15