Wet-gas flowline: coupon rate versus UT pitting
A 12-inch API 5L X65 flowline carries wet gas from three wells to a slug catcher 3.1 km away. It runs at 62 °C and 78 bar with 6.2 mol% CO2, 40 ppm H2S and 34% water cut; produced water has 18,000 mg/L chloride and the water phase pH is 5.7. Flow in the low sections is stratified at 1.4 m/s, and the route crosses a 40 m rise. A fixed 25 ppm corrosion inhibitor dose is applied, calculated on total liquid. Four weight-loss coupons retrieved at 12 months show 0.03–0.05 mm/y, and this figure has been used to justify continued operation at MAOP. A UT grid survey of the low section then found 4.1 mm of loss in an 8-month window against 9.1 mm nominal wall.
1. Both the coupon data and the UT grid are real measurements. Explain why they differ by two orders of magnitude and name the dominant damage mechanism, including the roles of CO2, H2S and flow.
Read the worked answer
What the numbers say
- pCO2 = 0.062 × 78 bar ≈ 4.8 bar, which is aggressive. At 62 °C and wall pH 5.7 the water is below the FeCO3 saturation needed for a stable film (practically pH 6.0–6.5 at this Fe2+ level), so any film is patchy.
- 40 ppm H2S gives a very low partial pressure (about 3 mbar) and only a thin, poorly protective sulfide film; it does not save the steel.
- The UT loss is 4.1 mm in 8 months, i.e. about 6.2 mm/y, at the 6 o'clock line where water accumulates.
Why the coupon looks benign
Coupons measure mass loss averaged over the whole surface, and they are almost always carried in a vertical run or a top-of-pipe access fitting where the water phase is not continuously wetting the face. 0.03–0.05 mm/y is a credible general rate for that location. It says nothing about penetration rate at a water-wetted bottom under stratified flow at 1.4 m/s - below the critical velocity to sweep the water on a 12-inch line.
Mechanism
CO2-dominated localised film breakdown (mesa/pitting) on the bottom-of-line water layer, with intermittent inhibitor availability at that surface. Averaging destroys the signal.
2. Before any change to operating philosophy is accepted, list four specific data sets or measurements you would require, and state what each one rules in or out.
Read the worked answer
- Water and deposit chemistry from the low point. Sample the free water and the solids at the 6 o'clock section: Fe2+, pH, chloride, sulfate and acid-volatile sulfide. High dissolved Fe2+ with a low pH supports film breakdown; high AVS and sulfide point to a significant H2S/souring contribution. Ruling out or ruling in MIC needs sulfate and sulfide data, not guesswork.
- Bottom-of-line corrosion measurement. A flush-mounted coupon or ER/LPR probe at 6 o'clock in the low section, or a retrieved section of the low point, installed for a defined exposure. This tests whether the 6.2 mm/y is a propagating pit or a single event from a slug.
- Flow-regime and critical-velocity analysis. Multiphase modelling with the actual rates to establish how long the water stays resident between slugs, and whether increased gas rate or pigging removes the holdup. This decides whether chemical treatment alone can ever work.
- High-resolution in-line inspection over the whole 3.1 km, not just the grid. A single grid cannot bound the worst pit or show whether other low points behave the same way.
Also retrieve the coupons' carrier records: position and orientation explain the discrepancy and must go into the report.
3. The operator proposes doubling the inhibitor dose and continuing at MAOP. Give your assessment and the actions you would require before restart at full pressure.
Read the worked answer
Doubling the dose is not an answer to this data. A 25 ppm dose on total liquid is already low, but the failure is a delivery and water-holdup problem: doubling a chemical that is not reaching the bottom of the pipe at the right time buys little. The increase must be justified by a measured response at the 6 o'clock surface, not by the coupon.
Integrity arithmetic. 9.1 - 4.1 = 5.0 mm of wall remains. At 6.2 mm/y, a further 0.8 mm of loss takes barely six weeks. Full MAOP is not defensible on that basis; a fitness-for-service assessment (ASME B31G or API 579) on the measured pit will almost certainly return a curtailed pressure, and should be run now, not after the next survey.
Actions I would require
- Reduce MAOP to the FFS-verified limit, or isolate/replace the affected low section, before returning to full pressure.
- Pig or otherwise remove the water holdup, and set a gas rate or pigging frequency that keeps the low point dry.
- Move inhibitor injection to a point that ensures the low section is treated, and verify with a bottom-mounted monitor.
- Re-inspect within weeks, not the annual cycle, and re-grade the line in the integrity plan.