Pagsusuri ng Pagkabigo

Hydrogen embrittlement failure is characterized by a sudden, brittle fracture that occurs under static load, often hours or days after installation. The fracture surface, when viewed under a Scanning Electron Microscope (SEM), shows 'intergranular' cracking, where the grain boundaries have separated. Unlike fatigue failure, there are no 'beach marks' or striations. If a batch of carbon steel springs fails prematurely in a static application, the plating process should be the first area of investigation, specifically checking the time-to-bake and bake temperature parameters.

Relaxation is the loss of load $P$ over time while the spring is held at a constant work height $H$. At elevated temperatures, the thermal energy allows dislocations in the crystal lattice to move more easily, resulting in plastic deformation even if the nominal stress is below the yield point. The rate of relaxation follows the Arrhenius equation. In failure analysis, this is observed as a 'loss of free height' $H_f$. If a spring designed for $400^{\circ}F$ is accidentally used at $600^{\circ}F$, it may lose $20-30\%$ of its load within hours, leading to assembly looseness and failure of the preloaded bearing.

Fretting fatigue occurs when there is minute oscillatory motion between the wave spring and its mating surfaces (the 'points of contact'). This motion, often caused by vibration, breaks down the protective oxide layer of the metal, leading to micro-pits. These pits act as stress risers ($K_t$), where fatigue cracks initiate. During failure analysis, fretting is identified by the presence of fine reddish or black debris (depending on the material) and a 'pitted' appearance at the wave crests. Increasing the preload or applying a solid-film lubricant can mitigate this by preventing the relative motion.

If the frequency of system vibration matches the natural frequency of the retaining ring, the ring can undergo high-amplitude oscillations, leading to a loss of cling force. The fundamental natural frequency $f_n$ for a circular ring is $f_n = _x000c_rac{k}{2 ext{π}} ext{sqrt} _x000c_rac{E I}{m R^3}$. In a resonant state, the ring 'dances' in the groove, causing rapid fretting wear or even 'jumping' out of the groove entirely. This is common in reciprocating compressors. To solve this, engineers 'tune' the ring by changing its mass (thickness) or its stiffness (width), or by using a multi-turn ring which has higher internal damping due to inter-turn friction.

Shear failure of the ring itself is rare and characterized by a clean, $45°$ or vertical 'shear lip' across the entire radial wall of the ring, with the ring remaining in the groove but 'sliced' into two or more layers. 'Groove Yield' is more common and is characterized by the ring being ejected from the groove, while the groove itself appears 'mushroomed' or flared out. Shear failure only happens when the groove and the retained part are both extremely hard (e.g., HRC $50+$), allowing the full axial force to act as a 'guillotine' on the ring. This is solved by increasing the ring's thickness $T$ or the number of turns.

The failure signature of an excessively dished ring includes: 1) A permanent 'conical' shape to the ring when removed, 2) Wear marks only on the inner edge of the ring's face and the outer edge of the groove's back-wall, and 3) A 'rolled' or rounded edge on the housing groove. This indicates that the axial load exceeded the groove's material yield strength or the ring's bending stiffness. Forensic analysis involves measuring the 'dish angle.' If the angle is $>5°$, the assembly was significantly overloaded. The fix involves increasing the groove depth $d$ or using a higher-strength housing material.

In applications with reversing or oscillating axial loads, the spiral ring will micro-move within the groove. If the ring's material hardness is too close to the housing hardness, 'Galling' (cold welding) occurs. This is seen as torn metal on the groove face. For an aluminum housing, a steel ring will quickly chew into the groove. This failure is analyzed by looking at the contact surfaces under $20 imes$ magnification. To mitigate this, engineers should either: 1) Increase the preload to eliminate movement, 2) Use a harder groove material (e.g., a steel bushing), or 3) Apply a hard-anodize or plating to the housing.

'Ring walk' or 'Spiral-out' occurs when the ring rotates relative to the groove and the end of the ring catches on a groove imperfection or is 'pumped' out by fluid flow/vibration. In centrifugal pumps, this is often caused by the hydrodynamic forces of the fluid acting on the 'gap' of the ring. If the ring's cling force is insufficient to overcome these forces, it begins to expand. Diagnosis involves looking for wear marks on the OD of the ring. The solution is to increase the ring's 'cling' by reducing its free diameter or using a self-locking design that prevents radial expansion altogether.

'Beach marks' (also known as clamshell marks) are macroscopic evidence of progressive fatigue crack growth. Each mark represents a period of crack extension followed by a period of rest or change in load. In a wave spring, these marks typically originate at the surface of the ID or OD at the peak of a wave, where the tensile bending stress is maximum. The presence of these marks, followed by a rough 'final fracture' zone, confirms that the spring failed due to cyclic loading rather than a single over-stress event. Analysis of the spacing between marks can sometimes help estimate the number of cycles to failure and the magnitude of the stress cycles.

While 316 stainless steel is 'marine grade,' it is still susceptible to pitting in stagnant seawater environments where chloride ions can breach the passive chromium-oxide layer. Wave springs are particularly vulnerable because the contact points between the waves and the housing create 'crevices.' These crevices lead to localized oxygen depletion, causing a breakdown of the passive film and forming an anodic cell. The failure is characterized by small, deep holes. For subsea applications, materials with a higher Pitting Resistance Equivalent Number (PREN) such as Inconel 625 or Super Duplex stainless steel are recommended.

Hysteresis is the difference between the loading and unloading curves of a spring. In wave springs, this is primarily caused by friction between the spring and the bore (or rod) and internal friction in multi-turn/nested designs. The area inside the hysteresis loop represents energy dissipated as heat. While some damping is often desirable in mechanical systems to suppress vibrations, excessive hysteresis indicates binding or bore interference. If the hysteresis is non-linear or 'jerky,' it suggests stick-slip behavior, which can lead to fatigue failure. For precision applications, reducing the radial wall $b$ or improving the surface finish can minimize these effects.

Fretting corrosion appears as reddish-brown (for steel) or black (for stainless) debris at the contact points between the wave peaks and the mating surfaces. It is caused by microscopic oscillatory movement (slip) under load, often due to high-frequency vibration. Under SEM (Scanning Electron Microscopy), the surface will show small pits and oxide build-up. Mitigation strategies include: 1) Increasing the axial preload to prevent 'lift-off' and slip, 2) Applying a dry-film lubricant (e.g., $MoS_2$) to reduce friction, or 3) Increasing the number of waves $n$ to distribute the load across more points, thereby reducing the local contact pressure.

'Settling' or permanent set occurs when the spring is compressed to a stress level exceeding the material's elastic limit. In failure analysis, this is diagnosed by measuring the free height $H_f$ after use and comparing it to the original specification. If $H_f$ has decreased, the spring has plastically deformed. Likely causes include: 1) Operating temperatures exceeding the material's limit, 2) Design stress $\sigma$ exceeding the yield strength $S_y$ at solid height, or 3) Improper heat treatment (e.g., incomplete transformation of 17-7PH). To fix this, one should consider 'presetting' the spring during manufacture or switching to a higher-strength material like Inconel X-750.

Fracture during installation is almost always due to 'Over-Straining' or 'Material Brittleness.' If the material is high-carbon steel, check for Hydrogen Embrittlement if it was recently plated. If it is 17-7PH, check the heat treatment records; 'Over-aging' or 'Under-aging' can result in improper ductility. Geometrically, if the installation mandrel was too large, the fiber stress at the ring's outer diameter would exceed the Ultimate Tensile Strength (UTS). The calculation $\sigma = _x000c_rac{E imes t imes (D_{exp} - D_i)}{(D_i imes D_{exp})}$ should be used to verify that the strain $\epsilon$ during expansion does not exceed the material's 'Elongation at Break' value (typically 5-10% for these alloys).

Stress Corrosion Cracking (SCC) manifests as sudden, brittle fracture without significant deformation, even in ductile materials like 17-7PH. It occurs due to the simultaneous presence of tensile stress (the installation 'cling' stress) and a corrosive medium (chlorides). Microscopic inspection of the fracture surface will show 'branched' transgranular or intergranular cracks. In retaining rings, the 'ends' of the wire and the 'removal notch' are common initiation points due to higher localized stress. To prevent SCC, engineers should ensure proper passivation, consider switching to a more resistant alloy like Inconel 718, or reduce the installation stress by optimizing the groove diameter.

In this failure mode, the shaft's rotational speed exceeded the 'Clinging Speed' of the ring. As the speed increased, the centrifugal force $F_c = m r imes ext{omega}^2$ overcame the elastic 'cling' force of the ring. Once the ring expanded and lost contact with the groove bottom, it began to spin independently of the shaft. This friction caused localized heating, reducing the material's yield strength and causing further expansion. Eventually, the ring 'popped' out of the groove due to the lack of radial constraint. The fix involves using a 'Self-Locking' ring design or increasing the ring's cross-section to increase its mass-to-stiffness ratio favorably.

Groove wall deformation in soft materials like Aluminum (e.g., 6061-T6) is caused by the bearing stress exceeding the compressive yield strength of the material. The thrust load $P$ is concentrated on a small area $\pi D d$. Because Aluminum has a much lower yield strength than the steel ring, the 'corner' of the groove will round off, allowing the ring to tilt (dish) and eventually fail. The solution is to: 1) Increase the groove depth $d$; 2) Increase the shaft/bore diameter $D$; 3) Use a 'Hard Anodize' coating to increase surface hardness; or 4) Use a steel 'Thrust Washer' between the retained part and the ring to distribute the load more evenly.

'Walking' occurs when cyclic axial impact loads cause the ring to vibrate and momentarily lose contact with the groove wall. If the ring has a slight 'dish' or if the groove has a radius, the axial force creates a radial component that 'pushes' the ring out of the groove. This is compounded if the ring has low 'cling' force. Analysis usually involves checking the 'Groove Squareness' and 'Ring Flatness.' Mitigation includes: 1) Increasing the initial interference fit (cling); 2) Using a deeper groove to increase the 'Shoulder' height; 3) Using a 'Self-Locking' spiral ring where the turns are mechanically linked to prevent radial expansion.

Galling on the housing bore is a sign that the wave spring's radial expansion is causing excessive force against the housing wall during compression. This is common when the spring material is similar in hardness to the housing material (e.g., a stainless steel spring in a stainless steel bore). Diagnosis involves inspecting the bore for material transfer and scoring marks aligned with the wave peaks. The solution is to: 1) Increase the clearance between the spring $D_{out}$ and the bore; 2) Harden the housing surface (e.g., nitriding or anodizing for aluminum); 3) Use a spring with 'Turn-down' ends to ensure the sharp edges of the wire do not dig into the bore.

Fretting corrosion occurs at the contact points (peaks) between adjacent turns of a multi-turn spring or between the spring and the mating plates. It is caused by micro-oscillations (vibration) that break down the protective oxide layer of the metal. In stainless steels, this exposes fresh metal which then oxidizes, creating a cycle of wear and oxidation that results in 'pitting.' This can lead to stress concentration and eventual fatigue failure. To prevent this, we apply anti-fretting coatings like Silver plating or PTFE. Additionally, increasing the initial preload can 'lock' the contact points in place, preventing the relative micro-movement that drives fretting.