Impeller Wear, Erosion, and Corrosion: Reading the Damage Pattern

Key Formula / Key Facts Box

Overview — What It Is and Why It Matters

Impeller degradation in centrifugal pumps occurs through three inter‑related mechanisms: mechanical wear from solid particles, erosion caused by high‑velocity impacts or cavitation, and electro‑chemical corrosion. Each mechanism reduces hydraulic efficiency, alters the flow‑path geometry, and can precipitate premature failure. Recognising the characteristic damage pattern enables the engineer to select a more suitable alloy, adjust operating conditions, or schedule targeted maintenance before a costly shutdown.

The Method — Derivation and Variants

The Finnie model starts with the kinetic energy of a single particle, E = ½ m V². Assuming a fraction η of that energy is dissipated in plastic deformation, the mass removed per impact is Δm = η E / H. By multiplying Δm with the particle flux Φ (proportional to C_s · ρ_f · V) and integrating over time, the continuous‑time form becomes the empirical expression shown above.

Variants

• Pure mechanical erosion: set α → 0, eliminating the corrosion term, yielding W = K·Vⁿ·C_s.
• Cavitation‑dominated erosion: replace V with the collapse velocity of vapor bubbles (≈300 m/s) and use n ≈ 1.5.
• Corrosion‑adjusted wear: add a separate corrosion term R_c = β·[Cl⁻]·e^{γ(T‑T_ref)} to account for chloride‑induced attack.

All constants (K, α, n, β, γ) are obtained from ASTM‑G76 slurry‑erosion tests or API‑610 qualification data for the specific alloy‑fluid pair.

Worked Example

Example 1 – US Customary Units

A 12‑in stainless‑steel impeller runs at 3,600 rpm in a slurry containing 0.5 lb/1000 lb quartz (hardness ≈ 6 HR). Peripheral velocity:

V = π D N / 60 = π (12 in / 12 ft) × 3,600 rpm / 60 ≈ 15.7 ft/s

Assume K = 0.001 lb/(ft·hr), n = 2, α = 0.05 HR⁻¹.

W = 0.001 × (15.7)² × 0.5 × e^{‑0.05·6}
   = 0.001 × 246 × 0.5 × 0.7408
   ≈ 0.091 lb/hr

The impeller loses roughly 0.09 lb of material each operating hour.

Example 2 – SI Units

A 300‑mm impeller operates at 3,500 rpm in a water‑sand mixture with C_s = 0.0008 kg/kg. Peripheral speed:

V = π D N / 60 = π × 0.300 m × 3,500 / 60 ≈ 55 m/s

Take K = 2 × 10⁻⁶ kg/(m·s), n = 2.5, α = 0.08 HV⁻¹, H = 200 HV.

W = 2e‑6 × 55^{2.5} × 0.0008 × e^{‑0.08·200}
   ≈ 2e‑6 × 1.4e4 × 0.0008 × 1.1e‑7
   ≈ 2.5e‑9 kg/s ≈ 0.009 g/hr

In this low‑abrasive case the wear is negligible, confirming that a material change is unnecessary.

Calculator

For quick estimates, use an online wear‑rate calculator: http://pumpcalcs.com/calculators/wear-rate/

Reference Values & Typical Ranges

• Peripheral velocity for severe slurry service: 10–30 ft/s (3–9 m/s).
• Solids concentration that initiates noticeable erosion: >0.2 lb/1000 lb (0.2 %).
• Typical hardness of common impeller alloys: Cast iron ≈ 150 HRB, 304 SS ≈ 70 HRB, Duplex ≈ 150 HRB.
• Acceptable wear rate for cast‑iron impellers: <0.05 lb/hr (≈0.02 kg/hr).
• Cavitation erosion onset when NPSH_available < 5 ft (1.5 m) for high‑speed pumps.

Application Guidance

During routine disassembly, look for the following visual cues:

• Uniform radial thinning – indicates high‑velocity abrasive wear; consider a larger diameter or a harder alloy.
• Localized pitting on the leading edge – classic of cavitation; verify NPSH margin and redesign inlet geometry if needed.
• Diffuse discoloration or intergranular attack – chemical corrosion; assess fluid pH, dissolved oxygen, and add appropriate inhibitors.

Practical adjustments:

• Increase inspection frequency by 50 % when measured wear exceeds 0.1 lb/hr.
• Apply ceramic or hard‑facing coatings to extend life when hardness‑limited alloys still erode.
• Install suction‑side pressure transducers to detect cavitation‑induced spikes.

Common Mistakes, Limits & Safety Notes

1. Mixing US and SI units in the wear equation – always convert before substitution.
2. Using a single K value for all alloys – K varies widely with material and fluid chemistry.
3. Neglecting temperature effects on corrosion – higher T accelerates chemical attack exponentially.
4. Relying solely on visual inspection; microscopic profilometry is required for early‑stage erosion detection.
5. Ignoring NPSH margin, which can mask true wear rates by inducing cavitation.
6. Failing to lock‑out/tag‑out the pump before impeller removal – rotating parts pose severe entanglement hazards.
7. Applying the empirical model beyond its calibrated range (V < 30 ft/s, C_s < 1 %).