A cast silicon bronze (UNS C86700) impeller that had been severely corroded was submitted for failure analysis. The failed part was used to pump potable water, but service life and chlorine content of the water were unknown. The impeller displayed a Cu-rich red phase on its surfaces and showed a pattern very similar to dezincification. Further investigation to determine the cause of damage using light microscopy and SEM-EDS techniques revealed that the microstructure consisted of multiple phases and that a Si-rich phase was being preferentially attacked, leading to increased porosity. After a thorough examination, it was concluded that the part had failed due to dealloying via desiliconification.

An aluminum bronze bushing that serves as a guide in a crimping machine began to fail after 50,000 cycles or approximately two weeks of operation. Until then, typical run times had been on the order of months. Although the bushings are replaceable and relatively inexpensive, the cost of downtime adds up quickly while operators troubleshoot and swap out worn components. Initially, the quality of the bushings came into question, but after a detailed analysis of the entire crimping mechanism, several other issues emerged that were not previously considered. As a result, the investigation provides information on not only better materials, but also design changes intended to reduce wear and increase service life.

An air system on a marine platform unexpectedly shut down due to the failure of a union nut, which led to an investigation to quantify the material limitations of bronze alloys in corrosive marine environments. The study focused on two alloys: Al-Si bronze, as used in the failed component, and Ni-Al bronze, which has a history of success in naval applications. Material samples were examined using chemical analysis, SEM imaging, and corrosion testing. Investigators also analyzed precracked tension specimens, exposing them to different conditions to quantify stress intensity thresholds for environmentally assisted cracking. Al-Si bronze was found to be susceptible to subcritical intergranular cracking in air and seawater, whereas Ni-Al bronze was unaffected. Both materials, however, are susceptible to cracking in the presence of ammonia, although the subcritical crack growth rate is two to three times higher in Ni-Al bronze. Based on the results of this work, the likelihood of subcritical cracking under various conditions can be reasonably estimated, which, in the case at hand, proved to be quite high.

Coaxial cable connectors made of brass were failing at a high rate after less than one year of service in an outdoor industrial environonment. The observed failures, which consisted of cracks in the body and end cap, were analyzed and found to be brittle fractures due to stress-corrosion cracking. Two common stress-corrosion cracking tests for copper materials were conducted on new connectors from the same manufacturing lot, confirming the initial determination of the fracture mode. Additional testing as was done in the investigation is often helpful when analyzing corrosion failures.

A brackish water pump impeller was replaced after four years of service, while its predecessor lasted over 40 years. The subsequent failure investigation determined that the nickel-aluminum bronze impeller was not properly heat treated, which made the impeller susceptible to aluminum dealloying. The dealloying corrosion was exacerbated by erosion because the pump was slightly oversized. The investigation recommended better heat treating procedures and closer evaluation to ensure that new pumps are properly sized.

Several cases of embrittlement failure are analyzed, including liquid-metal embrittlement (LME) of an aluminum alloy pipe in a natural gas plant, solid metal-induced embrittlement (SMIE) of a brass valve in an aircraft engine oil cooler, LME of a cadmium-plated steel screw from a crashed helicopter, and LME of a steel gear by a copper alloy from an overheated bearing. The case histories illustrate how LME and SMIE failures can be diagnosed and distinguished from other failure modes, and shed light on the underlying causes of failure and how they might be prevented. The application of LME as a failure analysis tool is also discussed.

An air blower in an electric power plant failed unexpectedly when a roller bearing in the drive motor fractured along its outer ring. Both rings, as well as the 18 rolling elements, were made from GCr15 bearing steel. The bearing also included a machined brass (MA/C3) cage and was packed with molybdenum disulfide (MoS 2 ) lithium grease. Metallurgical structures and chemical compositions of the bearing’s matrix materials were inspected using a microscope and photoelectric direct reading spectrometer. SEM/EDS was used to examine the local morphology and composition of fracture and contact surfaces. Chemical and thermal properties of the bearing grease were also examined. The investigation revealed that the failure was caused by wear due to dry friction and impact, both of which worsened as a result of high-temperature degradation of the bearing grease. Fatigue cracks initiated in the corners of the outer ring and grew large enough for a fracture to occur.

Two clamps that support overhead power lines in an electrified rail system fractured within six months of being installed. The clamps are made of CuNiSi alloy, a type of precipitation-strengthening nickel-silicon bronze. To identify the root cause of failure, the rail operator led an investigation that included fractographic and microstructural analysis, hardness testing, inductively coupled plasma spectroscopy, and finite-element analysis. The fracture was shown to be brittle in nature and covered with oxide flakes, but no other flaws relevant to the failure were observed. The investigation results suggest that the root cause of failure was a forging lap that occurred during manufacturing. Precracks induced by the forging defect and the influence of preload stress (due to bolt torque) caused the premature failure.

Stator vanes (cast from a Cu-Mn-Al alloy) in a hydraulic dynamometer used in a steam-turbine test facility were severely eroded. The dynamometer was designed to absorb up to 51 MW (69,000 hp) at 3670 rpm, and constituted an extrapolation of previous design practices and experience. Its stator was subject to severe erosion after relatively short operating times and initially required replacement after each test program. Although up to 60 cu cm (3.7 cu in.) of material was being lost from each vane, it only reduced the power-absorption capacity by a small amount. Analysis supported the conclusion that the damage was due to liquid erosion, but it could not be firmly established whether it was caused by cavitation or by liquid impact. Recommendations included making a material substitution (to Mo-13Cr-4Ni stainless steel) and doing a redesign to reduce susceptibility to erosion as well as erosion-producing conditions.

Several fuses made of nickel silver (57 to 61% Cu, 11 to 13% Ni, bal Zn) exposed to air containing ammonium and nitrate ions failed by SCC. Test solutions of 1 N ammonium nitrate (NH4NO3) and a 1:1 mixture of 1 N sodium nitrate (NaNO3) and 1 N calcium nitrate (Ca(NO3) 2) were prepared. In addition, stressed fuses made of nickel silver and of cupro-nickel (80Cu-20Ni) were exposed to a drop of corrosive solution in the stressed area. All nickel silver specimens failed after two days of exposure to NH4NO3 solution. However, 17% of them failed and 67% showed crack initiation but no failure after 42 days of exposure to NaNO3 + Ca(NO3)2 solution. None of the cupro-nickel specimens failed, but among those exposed to NH4NO3, 17% displayed crack initiation and 83% showed partial dealloying after 42 days. Based on the test results, the fuse material was changed from nickel silver to cupro-nickel, solving the SCC problem.

Electrical contact-finger retainers blanked and formed from annealed copper alloy C65500 (high-silicon bronze A) failed prematurely by cracking while in service in switchgear aboard seagoing vessels. In this service they were sheltered from the weather but subject to indirect exposure to the sea air. About 50% of the contact-finger retainers failed after five to eight months of service aboard ship. Investigation (visual inspection, 250x images etched with equal parts NH4OH and H2O2, emission spectrographic analysis, and stereoscopic views) supported the conclusion that the cracking was produced by stress corrosion as the combined result of: residual forming and service stresses; the concentration of tensile stress at outer square corners of the pierced slots; and preferential corrosive attack along the grain boundaries as a result of high humidity and occasional condensation of moisture containing a fairly high concentration of chlorides (seawater typically contains about 19,000 ppm of dissolved chlorides) and traces of ammonia. Recommendations included redesign of the slots, shot-blasting the formed retainers, and changing the material to a different type of silicon bronze-copper alloy C64700.

Failures of various types of hydraulic couplings used to connect pipes in a naval vessel are described and used to illustrate some of the general procedures for failure analysis. Cracking of couplings, which were manufactured from nickel-aluminum- bronze extruded bar, occurred in both seawater and air environments. Cracks initiated at an unusually wide variety of sites and propagated in either longitudinal or circumferential directions with respect to the axis of the couplings. Fracture surfaces were intergranular and exhibited little or no sign of corrosion (for couplings cracked in air), and there was very limited plasticity. Macroscopic progression markings were observed on fracture surfaces of several couplings but were not generally evident. At very high magnifications, numerous slip lines, progression markings, and striations were observed. In a few cases, where complete separation had occurred in service, small areas of dimpled overload fracture were observed. It was concluded from these observations, and from comparisons of cracks produced in service with cracks produced by laboratory testing under various conditions, that cracking had occurred by fatigue. The primary cause of failure was probably the unanticipated presence of high-frequency stress cycles with very low amplitudes, possibly due to vibration, resonance, or acoustic waves transmitted through the hydraulic fluid. Secondary causes of failure included the presence of high tensile residual stresses in one type of coupling, undue stress concentrations at some of the crack-initiation sites, and overtorquing of some couplings during installation. Recommendations on ways to prevent further failures based on these causes are discussed.

The failure of a 90-10 cupronickel heat exchanger tube resulted in flooding of the vessel and subsequently sinking it. The corrosion of the cupronickel alloy was facilitated by the high sulfur content of the seawater in which it operated. The failure modes were anodic dissolution and copper reprecipitation.

A very large diameter worm gear that had been in service in a dam for more than 60 years exhibited cracks and was removed. It was reported that the high-strength, low-ductility cast bronze gear was only rarely stressed during service, associated with infrequent opening and closing of gates. Due to the age of the gear and the time frame of its manufacture, no original material specifications or strength requirements could be located. Likewise, no maintenance records of possible repairs to the gear were available. Investigation (visual inspection, chemical analysis, tension and hardness testing, 119x SEM images, and potassium dichromate etched 297x metallographic images) supported the conclusion that the bronze gear cracked via mixed-mode overload, rather than by a progressive mechanism such as fatigue or stress-corrosion cracking. The cracking was not associated with regions that would be highly stressed and did not appear to be consistently correlated to casting imperfections, repair welds, or associated heat-affected zones. Cracking across the gear face suggested that bending forces from misalignment were likely responsible for the cracking. Recommendations included further review of the potential root cause.

After six years of service, three water shut-off valves on a copper water line in a residential building were found to be inoperative. Macroscopic examination of the valves after disassembly revealed that all three failed at the key that holds the spindle in the gate. In addition, the color near the key changed from yellow to red-brown. The gate was made from leaded red brass (85-5-5-5) while the spindle was made from silicon brass. It was concluded that the valves failed by dezincification resulting from bimetallic galvanic corrosion. It is common in the valve industry to use components made of different alloys in the same valve, but this is not the best approach for all applications.

After about 17 years in service, copper alloy C27000 (yellow brass, 65% Cu) innercooler tubes in an air compressor began leaking cooling water, causing failure and requiring replacement. The tubes were 19 mm in diam and had a wall thickness of 1.3 mm (0.050 in.). The cooling water that flowed through the tubes was generally sanitary (chlorinated) well water; however, treated recirculating water was sometimes used. Analysis (visual inspection, 9x and 75x unetched micrographs, and spectrochemical analysis) showed a thick uniform layer of porous, brittle copper on the inner surface of the tube, extending to a depth of about 0.25 mm (0.010 in.) into the metal, plug-type dezincification extending somewhat deeper into the metal. This supported the conclusion that failure of the tubes was the result of the use of an uninhibited brass that has a high zinc content and therefore is readily susceptible to dezincification. Recommendations included replacing the material with copper alloy C68700 (arsenical aluminum brass), which contains 0.02 to 0.06% As and is highly resistant to dezincification. Copper alloy C44300 (inhibited admiralty metal) could be an alternative selection for this application; however, this alloy is not as resistant to impingement attack as copper alloy C68700.

After 14 months of service, cracks were discovered in castings and bolts used to fasten together braces, posts, and other structural members of a cooling tower, where they were subjected to externally applied stresses. The castings were made of copper alloys C86200 and C86300 (manganese bronze). The bolts and nuts were made of copper alloy C46400 (naval brass, uninhibited). The water that was circulated through the tower had high concentrations of oxygen, carbon dioxide, and chloramines. Analysis (visual inspection, bend tests, fractographs, 50x unetched micrographs, 100x micrographs etched with H4OH, and 500x micrographs) supported the conclusions that the castings and bolts failed by SCC caused by the combined effects of dezincification damage and applied stresses. Recommendations included replacing the castings with copper alloy C87200 (cast silicon bronze) castings. Replacement bolts and nuts should be made from copper alloy C65100 or C65500 (wrought silicon bronze).

The horizontal heat-exchanger tubes made of copper alloy C70600, in one of two hydraulic-oil coolers in an electric power plant, leaked after 18 months of service. River water was used as the coolant in the heat-exchanger tubes. Several nodules on the inner surface and holes through the tube wall, which appeared to have formed by pitting under the nodules, were revealed by visual examination. Steep sidewalls, which indicated a high rate of attack, were revealed by microscopic examination of a section through the pit which had penetrated the tube wall. The major constituent of reddish deposit on the inner surfaces of the tubes was revealed to be iron oxide and slight manganese dioxide. Effluent from steel mills upstream was indicated by the presence of these and other constituents to be the source of most of the solids found in the tubes. It was concluded that the tubing failed by crevice corrosion. The tubing in the cooler was replaced, and cooling-water supply was changed from river to city water, which contained no dirt to deposit on the tube surfaces. An alternate solution of installing replacement tubes in the vertical position to make deposition of solids from river water less likely was suggested.

Persistent leakage was experienced from copper tube heaters which formed part of dairy equipment. Metallurgical examination of the brazed joints showed them to have suffered a preferential corrosion attack. This resulted in the phosphide phase of the brazing alloy being corroded away, leaving a weak, porous residual structure. The brazing alloy was of type CP 1 as covered by BS 1845. Header and tube materials were basically copper-nickel alloys for which the use of a phosphorus bearing brazing alloy is not recommended owing to the possibility of forming the brittle intermetallic compound, nickel phosphide. The use of a brazing alloy containing phosphorus was unsuitable on two counts and a quaternary alloy containing silver, copper, cadmium and zinc, such as those in group AG1 or AG2 of BS 1845 would be more suitable. However, because corrosive problems experienced in these units indicated severe service conditions, a proprietary alloy similar to AG1, but containing 3% nickel, was recommended.

A copper alloy C51000 (phosphor bronze, 5%A) failed prematurely during life testing of several such springs. The wire used for the springs was 0.46 mm (0.018 in.) in diam and was in the spring-temper condition. The springs were revealed to be subjected to cyclic loading, in the horizontal and vertical planes during the testing. The fracture was revealed to have occurred in bend 2. An indentation, presumably caused by the bending tool during forming, at the inner surface of the bend where fracture occurred was revealed by microscopic examination. Spiral marks produced on springs during rotary straightening were observed. A crack that had originated at the surface at the inside bend and had propagated toward the outside of the bend was revealed by microscopy of a longitudinal section taken through bend 2. The small bend radius was interpreted to contribute to spring fatigue as a result of result in straining at the bend zone. The spring was concluded to have failed in fatigue. It was recommended that the springs should be made of wire free from straightener marks and the bending tool should be redesigned so as not to indent the wire.