10.2 PRD Causes of Failure, Shop Testing & Overhaul
Key Takeaways
- The primary root causes of PRD failure include corrosion of internal trim/springs, deposit fouling/polymerization, galling of guiding surfaces, spring relaxation/cracking, seat leakage from particulate entrapment, and destructive chattering.
- As-received pop testing must be performed on the test stand prior to any cleaning or disassembly to verify in-service pop pressure; results must be compared against the Cold Differential Test Pressure (CDTP), accounting for temperature and backpressure corrections.
- For conventional PRVs with constant superimposed backpressure, the CDTP equals the stamped set pressure minus superimposed backpressure, multiplied by the manufacturer temperature correction factor: CDTP = (Set Pressure - P_back) * K_temp.
- Refurbishing PRV seating surfaces requires precision lapping using optical flats and monochromatic light; acceptable seat flatness is typically within 1 to 2 light bands (11.6 to 23.2 microinches) with zero radial scratches.
- Seat tightness testing per API Standard 527 measures leakage at 90% of CDTP using a 5/16 in. OD x 0.049 in. wall tube submerged 1/2 in. in water, counting bubbles per minute against code maximums.
PRD Causes of Failure, Shop Testing & Overhaul (API RP 576 & API 527)
Pressure-relieving devices operate in harsh chemical, thermal, and mechanical environments. Unlike active process control valves that stroke continuously throughout normal operations, a pressure relief valve remains seated in a static closed state for months or years, yet must operate flawlessly within milliseconds of an overpressure event. When a PRV fails to open at its design set point, the pressure vessel faces catastrophic rupture; conversely, if a PRV leaks or opens prematurely, it causes substantial product loss, environmental emissions, and plant downtime.
API Recommended Practice 576 (Inspection of Pressure-Relieving Devices) establishes rigorous guidelines for diagnosing failure mechanisms, executing as-received pop testing, conducting full valve teardown and overhaul, inspecting critical internal components, and performing standardized seat tightness testing in strict accordance with API Standard 527 (Seat Tightness of Pressure Relief Valves).
1. Primary Degradation Mechanisms & PRD Failure Modes
API RP 576 Section 5 classifies PRD failures into distinct operational and mechanical categories:
+-----------------------------------------------------------------------------+
| PRD DEGRADATION & FAILURE TAXONOMY |
| |
| [1. CORROSION & CHEMICAL ATTACK] |
| - Pitting & crevice corrosion on nozzle/disc seating interfaces |
| - Sulfide stress cracking (SSC) and chloride SCC on helical springs |
| - General thinning of guiding surfaces and body wall |
| |
| [2. FOULING, COKING & POLYMERIZATION] |
| - Viscous hydrocarbon tars and coke deposition in nozzle throat |
| - Monomer polymerization (styrene, butadiene) bonding disc to guide |
| - Particulate bridging and corrosion scale buildup |
| |
| [3. MECHANICAL GALLING & SEIZURE] |
| - Metal-to-metal pickup between disc holder and guide sleeve |
| - Caused by excessive piping strain, thermal binding, or tight clearances |
| |
| [4. SPRING DEGRADATION & THERMAL RELAXATION] |
| - Loss of spring temper / load relaxation from sustained heat |
| - Mechanical fatigue cracking from hydraulic pulsation / vibration |
| |
| [5. CHATTER & FLUTTER DAMAGE] |
| - Rapid cycling and violent disc-to-nozzle seat pounding |
| - Destroys lapped seating surfaces, shears spindles, ruptures bellows |
+-----------------------------------------------------------------------------+
Corrosion & Metallurgical Attack
Corrosion is the single most prevalent cause of PRV malfunction:
- Seat Pitting: Minute corrosion pits across the narrow seating contact boundary destroy the optical micro-seal, causing continuous simmer and leakage.
- Spring Failure: Helical springs are subjected to high static compressive stress. In sour ($H_2S$) service, atomic hydrogen charging induces Sulfide Stress Cracking (SSC); in chloride environments, austenitic stainless steel springs suffer rapid Chloride Stress Corrosion Cracking (Cl-SCC). A fractured spring causes the valve to blow wide open immediately.
Galling of Guiding Surfaces
Guiding surfaces maintain precise axial alignment between the disc holder and the body nozzle. If guiding clearances are compromised:
- Galling: High sliding contact pressure between similar stainless steel alloys causes microscopic adhesive welding (pickup) and surface tearing.
- Misalignment / Piping Strain: Excessive external mechanical loads transferred from un-supported discharge piping bend the valve body, pinching the guide sleeve and causing the valve to hang up (fail to open) or fail to reclose.
Chatter vs. Flutter: Physics, Causes & Remediation
+-----------------------------------------------------------------------------+
| VALVE DYNAMICS: CHATTER VS. FLUTTER |
| |
| [FLUTTER (Abnormal High-Frequency Oscillation)] |
| - Rapid aerodynamic oscillation of disc without contacting nozzle seat. |
| - Causes: Flow rate far below valve rated capacity; turbulence. |
| - Consequence: Severe bellows fatigue, guide wear, capacity loss. |
| |
| [CHATTER (Violent High-Frequency Impact Cycling)] |
| - Rapid, full-stroke opening and closing where disc violently hammers seat|
| - Frequencies typically range from 10 to 40 Hz (cycles per second). |
| - Primary Cause 1: Excessive INLET PRESSURE DROP (> 3% of Set Pressure). |
| * Valve opens -> inlet friction drops pressure below reseat -> closes; |
| * Flow stops -> inlet pressure recovers to set point -> opens again. |
| - Primary Cause 2: Excessive BUILT-UP BACKPRESSURE in discharge piping. |
| - Primary Cause 3: Valve SEVERELY OVERSIZED for actual process relief load.|
| - Consequence: Destroys seating surfaces, shears spindle, bellows failure.|
+-----------------------------------------------------------------------------+
[!IMPORTANT] The 3% Inlet Piping Pressure Drop Rule (API 520 Part II & API 576): Total non-recoverable pressure drop in the piping between the pressure vessel nozzle and the PRV inlet flange must not exceed $3%$ of the stamped set pressure at rated capacity. Excessive inlet piping pressure drop is the leading operational cause of catastrophic PRV chattering.
2. As-Received Pop Testing & CDTP Adjustment Protocols
When a PRV is removed from a process unit for shop servicing, it must undergo an As-Received Pop Test before any cleaning, solvent flushing, bolt loosening, or mechanical adjustments are performed.
+-----------------------------------------------------------------------------+
| AS-RECEIVED POP TESTING PROTOCOL |
| |
| [STEP 1: INITIAL VISUAL & INLET INSPECTION] |
| - Inspect inlet nozzle and outlet bore for deposits, coking, fouling |
| - Record as-found condition (dry, wet, gummy, scale buildup) |
| - Verify test medium: DRY AIR / NITROGEN for vapor; WATER for liquid |
| |
| [STEP 2: TEST STAND CLAMPING & SAFETY VERIFICATION] |
| - Clamp valve securely to test bench with rated test flange |
| - Install safety shield; ensure all personnel wear ear/eye protection |
| - Verify spring compression screw seal wire is intact |
| |
| [STEP 3: SLOW PRESSURE APPLICATION & POP VERIFICATION] |
| - Increase inlet pressure at controlled rate (<= 2 psi/sec near set point)|
| - Note SIMMER pressure (first audible/visible leakage) |
| - Record POP PRESSURE (sharp audible pop or continuous discharge) |
| |
| [STEP 4: MULTI-POP VALIDATION] |
| - Pop valve 2 to 3 times to verify repeatability |
| - Compare against Cold Differential Test Pressure (CDTP) specification |
+-----------------------------------------------------------------------------+
Cold Differential Test Pressure (CDTP) Engineering Mathematics
A PRV operating in a 600°F distillation tower with constant 20 psig superimposed backpressure behaves differently on an ambient 70°F atmospheric test stand. The Cold Differential Test Pressure (CDTP) is the exact pressure to set on the shop test stand.
Two engineering corrections must be applied:
- Temperature Correction Factor ($K_{\text{temp}}$ or $C_t$): At elevated service temperatures, the valve body and spring expand, and the spring material modulus of elasticity decreases ($E$ decreases), weakening the spring force. To ensure the valve opens at stamped set pressure at operating temperature, the spring must be compressed more tightly on the cold test bench ($K_{\text{temp}} > 1.00$).
- Superimposed Backpressure Correction ($P_{\text{back}}$): For conventional PRVs with bonnets vented to the discharge side, constant superimposed backpressure assists the spring. On an atmospheric test bench, this backpressure is absent, so the backpressure must be subtracted from the set pressure.
Typical Temperature Correction Factors ($K_{\text{temp}}$ for Carbon Steel Springs)
| Operating Temperature Range | Temperature Multiplier ($K_{\text{temp}}$) | Adjustment Rationale |
|---|---|---|
| Below $150^\circ\text{F} (65^\circ\text{C})$ | $1.000$ ($0.0%$ increase) | Negligible spring modulus reduction |
| $150^\circ\text{F} \text{ to } 450^\circ\text{F} (65 - 232^\circ\text{C})$ | $1.015 \text{ to } 1.025$ ($+1.5% \text{ to } +2.5%$) | Modest thermal expansion and spring softening |
| $451^\circ\text{F} \text{ to } 750^\circ\text{F} (233 - 399^\circ\text{C})$ | $1.030 \text{ to } 1.050$ ($+3.0% \text{ to } +5.0%$) | Significant spring relaxation; high-temp alloy recommended |
| $751^\circ\text{F} \text{ to } 1000^\circ\text{F} (400 - 538^\circ\text{C})$ | $1.060 \text{ to } 1.080$ ($+6.0% \text{ to } +8.0%$) | Tungsten alloy spring / open bonnet mandatory |
Worked Example: CDTP Calculation
A conventional spring-loaded PRV has a nameplate set pressure of $200\text{ psig}$, operates at $500^\circ\text{F}$ (temperature correction factor $K_{\text{temp}} = 1.030$), and discharges into a header with a constant superimposed backpressure of $25\text{ psig}$. What is the required shop CDTP?
3. Valve Disassembly, Component Inspection & Overhaul
Following the as-received pop test, the PRV is completely disassembled for cleaning, non-destructive examination, and dimensional verification.
+-----------------------------------------------------------------------------+
| PRV OVERHAUL & REFURBISHMENT FLOW |
| |
| [1. TEARDOWN & CLEANING] |
| - Measure spring free length and scribe spring adjustment screw height |
| - Degrease all components; bead blast cast iron/steel parts (NOT seats!) |
| |
| [2. SPRING INSPECTION] |
| - Check spring squareness on precision surface plate with square (< 0.5°) |
| - Measure free length vs. manufacturer specifications (replace if sagged) |
| - Perform Magnetic Particle (MT) or Liquid Penetrant (PT) for cracks |
| |
| [3. SPINDLE & GUIDE DIMENSIONAL CHECK] |
| - Check spindle (stem) straightness on V-blocks with dial indicator (TIR) |
| - Measure guide-to-disc-holder diametrical clearance (API 576 tables) |
| |
| [4. SEAT LAPPING & OPTICAL FLAT FLATNESS CHECK] |
| - Machine damaged seats; hand-lap with cast iron plates & diamond compound|
| - Inspect seat flatness using MONOCHROMATIC LIGHT & OPTICAL FLAT |
| - Acceptance: 1 to 2 light bands (11.6 to 23.2 microinches); zero radial |
+-----------------------------------------------------------------------------+
Critical Component Inspection Checklist
- Spring Inspection:
- Free Length: Measure uncompressed overall length. A permanent set (reduction in free length exceeding manufacturer tolerance, typically $> 1.5%$) mandates spring replacement.
- Squareness / Perpendicularity: Place the spring vertically on a precision granite surface plate against a machinist's square. The ends must be square within $0.5^\circ$ (or $1/64\text{ in. per foot}$ of diameter).
- Crack Detection: Fluorescent Wet Magnetic Particle Testing (WFMT) or Liquid Penetrant Testing (PT) to detect surface cracks.
- Spindle (Stem) Runout: Mount the spindle between centers or on precision V-blocks. Measure Total Indicated Runout (TIR) using a dial test indicator. Any spindle exceeding $0.003\text{ in.} - 0.005\text{ in.}$ bend must be replaced (straightening is prohibited).
- Guiding Surfaces: Inspect disc holder guide surface and body guide sleeve for galling, scoring, or corrosion. Clean with fine emery cloth. Measure clearances against manufacturer tolerances.
Precision Seat Lapping & Monochromatic Light Interference Bands
PRV seating surfaces are subjected to extreme unit loading. Achieving a bubble-tight seal requires optical-grade surface finish:
- Lapping Technique: Ring laps (cast iron lapping blocks) charged with progressively finer aluminum oxide or diamond lapping paste (from 14 micron down to 1 micron) are oscillated in a smooth figure-eight motion with light downward pressure. Lapping blocks must be constantly conditioned on a flat lapping plate.
- Optical Flat & Monochromatic Light Testing:
- An optical flat (a precision-ground quartz or Pyrex glass disc) is placed directly on the cleaned, dry, lapped seat.
- The assembly is illuminated under a monochromatic helium light source (wavelength $\lambda = 23.2\text{ microinches} [0.589,\mu\text{m}]$).
- Air wedge interference produces visible alternating dark and light interference bands (fringes). Each band represents a surface height deviation of half a wavelength ($11.6\text{ microinches}$ or $0.295,\mu\text{m}$).
- Acceptance Criteria: Seating surfaces must exhibit 1 to 2 straight, parallel light bands across the entire sealing width. Any curved, irregular, or radial cross-scratch patterns indicate non-flatness and will cause seat leakage.
4. Seat Tightness Testing per API Standard 527
After reassembly, set pressure adjustment, and ring positioning, every PRV must undergo a rigorous Seat Tightness Test in accordance with API Standard 527.
+-----------------------------------------------------------------------------+
| API 527 STANDARD BUBBLE TEST APPARATUS |
| |
| [Safety Relief Valve Under Test] |
| | |
| v |
| +------------------------------------------+ |
| | Valve Discharge Flange Sealed with Cover | |
| +------------------------------------------+ |
| | |
| | Flexible Tubing |
| v (5/16" OD x 0.049" Wall) |
| +------------------------------------------+ |
| | Clear Beaker filled with Clean Water | |
| | | |
| | Tube Submerged EXACTLY 1/2" (12.7 mm) | |
| | ===~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~= | |
| | ( o o o Bubbles ) | |
| +------------------------------------------+ |
+-----------------------------------------------------------------------------+
API 527 Testing Specifications
- Test Medium: Clean dry air or industrial nitrogen.
- Test Pressure: Maintained steadily at $90%$ of the Cold Differential Test Pressure (CDTP). For valves with CDTP $\le 50\text{ psig}$, test pressure is held at $5\text{ psi}$ below CDTP.
- Apparatus Dimensions: A test tubing assembly having an outside diameter (OD) of $5/16\text{ in.} (7.9\text{ mm})$ and a wall thickness of $0.049\text{ in.} (1.2\text{ mm})$ is connected to the sealed discharge flange. The discharge end of the tube is cut square and submerged $1/2\text{ in.} (12.7\text{ mm})$ below the surface of a water container.
- Observation Time: Held at test pressure for a minimum observation period before counting:
- Orifice sizes $D, E, F$: 1 minute minimum.
- Orifice sizes $G, H, J$: 2 minutes minimum.
- Orifice sizes $K, L, M, N, P, Q, R, T$: 5 minutes minimum.
Maximum Allowable Leakage Rates (API 527)
| Stamped Set Pressure Range | Orifice Sizes D, E, F (Bubbles / Min) | Orifice Sizes G, H, J (Bubbles / Min) | Orifice Sizes K through T (Bubbles / Min) |
|---|---|---|---|
| $15 \text{ to } 1000\text{ psig}$ | 40 bubbles/min | 20 bubbles/min | 0 bubbles/min (approx. negligible) |
| $1500\text{ psig}$ | 60 bubbles/min | 30 bubbles/min | Special criteria |
| $2000\text{ psig}$ | 80 bubbles/min | 40 bubbles/min | Special criteria |
| Soft-Seated (O-Ring) PRVs | 0 bubbles / min (Zero leakage for all sizes and pressures) | 0 bubbles / min | 0 bubbles / min |
A conventional spring-loaded PRV in hot gas oil service has a stamped nameplate set pressure of 300 psig and an operating temperature of 600°F (manufacturer temperature correction factor K_temp = 1.040). The valve exhausts into a closed flare system maintaining a constant superimposed backpressure of 50 psig. What is the CORRECT Cold Differential Test Pressure (CDTP) to adjust on the ambient shop test bench?
During in-service operation, a pressure relief valve begins violent, rapid cyclic opening and closing, pounding the disc against the nozzle seat at high frequency (chatter). According to API RP 576 and API 520, which of the following is the MOST PROBABLE piping design deficiency causing this phenomenon?
An instrument technician is lapping the disc and nozzle seating surfaces of a carbon steel safety relief valve. When checking the lapped surface under a monochromatic helium light source using an optical flat, which observation indicates an ACCEPTABLE surface flatness per API RP 576?
When performing a standard seat tightness test on a newly reassembled metal-to-metal seated PRV with a CDTP of 100 psig per API Standard 527, what test pressure and apparatus arrangement must be utilized?