Wall Thinning Rate and Remaining Life from Metal Loss

This page is bookkeeping, and it is the bookkeeping that turns a laboratory result or an inspection finding into a date on a maintenance plan. A mass of metal has gone. Divide by the metal's density and it is a volume; divide by the area it came off and it is a thickness; divide by the time it took and it is a rate. Divide the wall you can still afford to lose by that rate and you have the remaining life.

To read the thinning rate rather than the life, set the life to exactly one year and solve for the allowance: the answer is the metal lost per year, in millimetres or mils, which is the number an inspection report actually wants. An allowance and a rate are the same measurement wearing different hats, which is why one relation serves both.

The assumption of uniform loss is the one that breaks first, and it breaks hard. Erosion is almost never uniform. It concentrates on the outer radius of a bend, the shoe of a tee, the first few diameters downstream of a choke or an orifice, the root of a weld where the bore steps. A component perforates at its thinnest point, not at its average, and spreading a localised groove's weight loss over a whole spool can overstate the life by an order of magnitude. If the wall shows a groove, a wear scar or a polished patch, the life is set by the deepest point, and finding that point is an inspection problem rather than a calculation. The same warning applies to corrosion coupons, where pitting rates run five to fifty times the general rate measured on the same coupon.

The assumption of a constant rate is the second. Erosion tracks the solids production and the velocity, and both change over a well's life — sand production is often negligible for years and then rises sharply as a completion degrades, and velocity changes as pressures fall and the mixture density falls with them. A rate measured in year two is not a rate for year ten.

And the third is the one people miss entirely: erosion and corrosion are not additive. Flow and impacting particles strip whatever protective film the metal had; the exposed metal then corrodes at its bare rate rather than its protected one; the film starts to re-form and is stripped again. The synergy term can be larger than either of the individual contributions, which is why the phenomenon has its own name — erosion-corrosion — and its own characteristic morphology of directional, undercut, horseshoe-shaped pits pointing downstream. If you have a mechanical erosion rate from a wear model and an electrochemical rate from a coupon, adding them gives a number that is comfortably too small.

One practical note on measurement. The uncertainty in a rate is the weighing or thickness-measurement uncertainty divided by the metal actually lost, so a short exposure gives a rate that means nothing at all. Leave the coupon in long enough to lose metal well above the balance's noise and above whatever the cleaning procedure removes, and remember that three readings over a long baseline are worth far more than two close together.