FRACTURED BRASS TEE FITTING
Posted on July 25th, 2019 by Met-Tech
ANALYSIS OF A FRACTURED BRASS TEE PLUMBING FITTING
SUMMARY:
A fractured brass tee plumbing fitting was analyzed to investigate the cause of the fracture at one of the brass tee nipples.
Results indicate the brass plumbing fitting fractured due to stress corrosion cracking (SCC). The crack initiated on the ID surface of the brass tee fitting in the area where an outer diameter (OD) compression ring is crimped to secure tubing to the brass fitting. Multiple cracks were observed on the ID surfaces of the two intact brass tee fitting nipples at the same location. Chlorine was detected on the fracture surface and ID surface of a second nipple from the fitting. No significant deformation of the fitting was observed.
A blue oxide heat tint stain was noted on the ID surfaces of the two intact nipples but not on the fractured brass tee nipple. The heat tint appeared to have been machining induced. Residual stress from the machining operation may have also contributed to cracking.
The fractured brass tee fitting composition was similar to Copper Alloy C37700; however, the zinc concentration was elevated. The hardness indicated increased susceptibility to SCC. Brasses with higher zinc concentrations are typically associated with reduced resistance to SCC.
ANALYSIS:
A fractured brass tee fitting was analyzed to investigate the cause of the fracture at one of the tee nipples as shown in Figure 1. This type of fitting is typically used in potable water systems with flexible polyethylene (PEX) tubing. The fitting is slipped into the tubing to the tube stop. A compression ring is crimped approximately 0.125-in. to 0.25-in. from the tube end to secure the tubing. The fracture occurred at the corner of a machined shoulder in the fitting, approximately 0.1-in. from the tube stop, near the crimp area. No deformation of the fitting was observed.
A close-up view of the fitting fracture is presented in Figure 2. No notable yielding was observed at the fracture site. The fracture surface was subjected to low magnification stereomicroscopic examination. The mating half of the fracture (nipple side) is presented in Figure 3. The radial ridges and a dark band around the ID periphery of the fracture surface indicated the crack initiated on the ID of the fitting.
All three nipples from the fitting were subsequently sectioned longitudinally to observe the ID surfaces as shown in Figure 4. A blue tint was observed on the ID surface of the intact nipples. The coloring appeared to be due to heat tinting during the manufacturing process and not the result of corrosion.
The nipple side of the fracture was subjected to high magnification scanning electron microscopic (SEM) examination.
Figure 5 is a low magnification SEM image of the nipple fracture surface prior to cleaning. The higher magnification SEM image in Figure 6 shows a darker area along the ID surface of the nipple fracture. Energy dispersive x-ray spectrographic (EDS) micro-analysis in general accordance to ASTM E1508-98 was performed on the darker area to determine the composition. Mainly copper oxide was indicated in the fracture surface deposit. However, chlorine was detected within the darker area, as shown by the EDS spectrum in Figure 7.
Figure 8 shows a SEM image of the nipple fracture surface after ultrasonic cleaning in a mild detergent solution. Cleavage-like facets were observed over most of the fracture surface. Figure 9 presents a higher magnification image of the fracture surface at the ID edge. The ID edge of the fracture surface was covered with a residual oxide coating. The fracture ridge patterns indicate the fracture initiated at the ID surface. Several similar initiation sites were observed around the ID edge of the nipple fracture. Figure 10 shows a second initiation site.
Figures 11 and 12 show an example of the mid-wall fracture and the fracture surface near the OD, respectively. The surfaces exhibited cleavage, indicative of transgranular stress corrosion cracking. No indications of fatigue cracking or ductile overload were observed.
Figure 13 shows an overview of the ID surface of half of the fractured nipple. The fracture path was irregular with additional cracks branching from the main fracture plane as shown at increased magnification in Figures 14 and 15. The branching is typical for stress corrosion cracking.
A sectioned half of another (intact as-received) nipple from the fitting was examined using the SEM. Figure 16 shows a backscatter (BSE) image of the ID surface of the comparison nipple. The BSE image is more sensitive to compositional and topographical variations. Low atomic number (light) elements are darker and high atomic number (heavy) elements are lighter in the image. Cracks were observed on the oxide coated ID surface at approximately the same location where the broken nipple fractured.
EDS micro-analysis was performed to determine the composition of the oxide on the ID surface of the intact nipple. Figure 17 provides the EDS spectrum of the oxide, which was composed primarily of copper and zinc. No corrosive elements were detected. This indicates the blue tinted deposit on the ID previously observed (Figure 4) is likely heat tint rather than corrosion products.
Near the comparison nipple opening, darker stained areas were observed on the ID surface as shown in Figure 18. EDS micro-analysis was performed on the stains to determine the composition and is presented in Figure 19. Chlorine was detected within the stained region.
The second intact nipple was also sectioned longitudinally, cleaned and examined using the SEM. Figure 20 shows the ID surface of the second comparison nipple. Cracks are observed at the same location where the fractured nipple separated and at the next rib location. The cracks do not follow the machining marks, are somewhat intermittent, and have multiple branches, as shown in Figures 21 through Figures 24.
An axial cross-section through the fractured nipple and the first comparison cracked nipple was cut, then prepared for metallographic analysis, in accordance with ASTM E3-01. Etching, in accordance with ASTM E407-99, revealed the microstructure that was examined using an optical microscope, in accordance with ASTM E883-02.
Figures 25 and 26 present optical photomicrographs of the etched longitudinal cross-sections through one side of the nipple fracture. The optical photomicrographs in Figures 27 and 28 show the etched cross-section of the opposite side of the nipple fracture. Branching of the cracks, indicating stress corrosion cracking, was observed. The cracking was transgranular. Secondary cracks were observed adjacent to the main fracture. A cast brass microstructure of alpha plus beta phase was observed.
Figures 29 through 31 present optical photomicrographs of the etched longitudinal cross-sections through one side of the comparison cracked nipple. The cracking initiated from the ID surface. The crack morphology, branching, and multiple fingers, indicate stress corrosion cracking. A large, elongated alpha plus beta grain structure was noted.
The fractured brass tee fitting was subjected to chemical analysis to verify the chemical composition. Results indicate the fitting is made from a leaded brass alloy similar to Copper Alloy C37700, however the copper and lead content were low, and the zinc and iron contents were elevated. Chemical analysis was performed using an optical emission spectrograph in accordance with MTI Procedure No. MTI-0241, Rev. 0. The results of the analysis are compared to ASTM specifications and are summarized below:
|
Element |
Tee Fitting
(wgt %) |
ASTM B-124
Alloy C37700 (wgt %) |
| Copper | 54.8 | 58.0-61.0 |
| Tin | 0.29 | – |
| Lead | 1.16 | 1.5-2.5 |
| Zinc | 43.3 | Remainder(1) |
| Iron | 0.39 | 0.30 max |
| Nickel | 0.08 | – |
| Aluminum | 0.012 | – |
| Manganese | 0.001 | – |
| Sulfur | 0.002 | All others <0.50 |
| Phosphorus | 0.001 | – |
| Silicon | 0.002 | – |
| Antimony | 0.007 | – |
| Arsenic | 0.016 | – |
Note 1: Typical zinc content is 36.0 to 40.0%.
Microhardness testing of the fractured brass tee nipple was performed in accordance with ASTM E384-99e1. Four separate readings were taken at random, widely-spaced locations on the nipple cross-section for an average hardness value. Results were converted from Knoop 500-gram load to Rockwell B (HRBW) using the ASTM E140-05 conversion table. The fractured nipple exhibited a hardness of 80 HRB. The hardness level indicated increased susceptibility to SCC.
CONCLUSIONS:
The brass tee fitting fractured due to stress corrosion cracking. The fracture initiated on the ID surface of the fitting in the area where the compression ring is crimped to secure tubing to the fitting. Chlorine was detected on the fracture surface and ID surface of an adjacent nipple. Similar partial SCC cracks were found in both intact nipples of the fitting. The fitting was manufactured from a leaded brass alloy with a non-standard composition (similar to C37700 with elevated zinc and iron and decreased copper and lead). The hardness indicated increased susceptibility to SCC.
The elevated zinc and high hardness typically contribute to increased SCC of copper alloys. Residual stresses from machining may have also contributed to cracking.
IMAGES:
Figure 1:View of the fractured brass tee fitting received for analysis. Red arrows point to the fracture site. The fracture occurred at a machined shoulder. (Fractures will be identified as nipple side or fitting side, as identified in the photo).
Figure 2:A close-up view of the fracture on the brass fitting. Stereomicroscope examination of the fracture indicated the fracture initiated at the ID surface of the fitting.
Figure 3: A close-up view of the mating nipple half of the fitting fracture. Darker areas (at arrows) around the ID circumference of the fracture and fracture ridges indicate the crack initiated from the ID surface.
Figure 4: A close-up view of the ID surfaces of the fitting nipples after sectioning longitudinally. The yellow arrows identify the fracture surface. The ID surface of the intact nipples exhibited a bluish color. Subsequent analysis of the surfaces determined the blue coloring was the result of heat tinting from manufacturing.
Figure 5: Low magnification SEM image of the nipple side fracture surface prior to cleaning. The area at the arrow is shown at higher magnification in Figure 6. (SEM Image, Mag: 6X)
Figure 6: Higher magnification SEM image of the as-received nipple side fracture surface. EDS analysis was performed on the deposit covered darker area (circled) along the ID edge of the fracture. The resulting spectrum is presented in Figure 7. (SEM Image Mag: 85X)
Figure 7: EDS analysis of the darker area along the ID edge of the nipple fracture. The deposit is mainly copper oxide. Chlorine (Cl) was also detected. (Spectrum)
Figure 8: SEM image of the typical fracture surface observed on the nipple side fracture after cleaning. The fracture initiated at the ID surface. The arrow points to the ID surface, which is shown at higher magnification in Figure 9. (SEM Image Mag: 50X)
Figure 9: Higher magnification SEM image of the ID edge (at red arrow) of the nipple fracture shows the surface is oxidized (circled area) along the ID. The fracture ridge patterns indicate the fracture initiated at the ID surface and progressed in the direction of the yellow arrows. (SEM Image, Mag: 250X)
Figure 10:A second initiation area on the nipple side fracture shows the fracture initiated at the ID edge (red arrow) and progressed in the direction of the yellow arrows toward the OD surface. The surface along the ID edge is oxidized. (SEM Image, Mag: 250X)
Figure 11: SEM image of the fracture surface at approximately mid-wall shows a cleavage-type fracture surface, indicative of transgranular stress corrosion cracking. (SEM Image, Mag: 250X)
Figure 12: SEM image of the OD edge of the fitting fracture surface shows a transgranular cleavage-type fracture surface. (SEM Image Mag: 200X)
Figure 13: Low magnification SEM image of the ID surface of half of the nipple side fracture. The boxed area is shown at higher magnification in Figure 14. (SEM Image Mag: 10X)
Figure 14: Higher magnification SEM image of the ID surface of the nipple side fracture. Note the crack branching off the main fracture (at arrows). The boxed area is shown at higher magnification in Figure 15. (SEM Image Mag: 100X)
Figure 15: Higher magnification SEM image of the branching cracks (arrows) off the main fracture surface. (SEM Image Mag: 400X)
Figure 16: Low magnification SEM backscatter image of the ID surface of a comparison nipple from the fractured fitting. Cracks are observed in the blue tinted surface (observed in visual examination) at each of the arrow locations. EDS analysis of the blue oxide on the surface is presented in Figure 17. (SEM Image Mag: 35X BSE)
Figure 17: EDS analysis of the oxide at the crack locations on the comparison nipple ID surface. The blue oxide is composed primarily of copper (Cu) and Zinc (Zn). No contaminants or corrosive elements were detected. (Spectrum)
Figure 18: SEM backscatter image of dark stained areas on the comparison nipple ID surface near the nipple opening. EDS analysis of the stain in the circled area is presented in Figure 19. (SEM Image Mag: 13X)
Figure 19: EDS analysis of the stained area on the comparison nipple ID surface. The stain exhibits mainly copper and zinc oxide; however, chlorine was also detected. (Spectrum)
Figure 20: Low magnification SEM image of the second comparison nipple ID surface. Cracking is observed just beyond the tube stop (red arrows) and at the rib location (yellow arrows). The cracks exhibit multiple branching and do not follow the machining marks on the ID surface. Higher magnification images of cracks are shown in Figures 21 through 24. (SEM Image Mag: 10X)
Figure 21:Higher magnification SEM image of the second comparison nipple crack (arrows) just beyond the tube stop location. Machining smears (vertical ripples) are noted in the ID surface. (SEM Image Mag: 100X)
Figure 22:Higher magnification SEM image of the second comparison nipple crack (arrows) just beyond the tube stop location. The crack crosses the circumferential ID machining marks. (SEM Image Mag: 100X)
Figure 23: Higher magnification SEM image of the second comparison nipple crack (arrows) at the rib location. The crack has multiple branches. (SEM Image Mag: 100X)
Figure 24: Higher magnification SEM image of the second comparison nipple crack (arrows) at the rib location. (SEM Image Mag: 50X)
Figure 25: Optical photomicrograph of the etched longitudinal cross-section through one side of the nipple fracture. The crack initiated on the ID of the fitting at the yellow arrow. A secondary crack is observed at the white arrow. The main fracture exhibited branching as indicated at the red arrows. (Mag: 100X, Swiss etch)
Figure 26: Higher magnification optical photomicrograph of the etched longitudinal cross-section through the nipple side fracture initiation. No unusual conditions in the microstructure are observed in the alpha plus beta microstructure. (Mag: 500X, Swiss etch)
Figure 27: Optical photomicrograph of the etched longitudinal cross-section through the opposite side of the nipple fracture. The crack initiated on the ID of the fitting at the yellow arrow. (Mag: 100X, Swiss etch)
Figure 28: Higher magnification optical photomicrograph of secondary cracks above the fracture location at the yellow arrows. (Mag: 500X, Swiss etch)
Figure 29: Optical photomicrograph of the etched longitudinal cross-section through a comparison nipple from the fractured fitting. A crack initiating on the ID of the fitting (yellow arrow) extends across approximately three-quarters of the wall thickness ending at the red arrow. A large, elongated grain structure is noted. (Mag: 50X, Swiss etch)
Figure 30: Higher magnification optical photomicrograph of the transgranular branching of the crack through the comparison nipple. (Mag: 200X, Swiss etch)
Figure 31:Higher magnification optical photomicrograph of multi-branched transgranular crack tips (arrows) in the partially cracked nipple cross-section. (Mag: 500X, Swiss etch)
