To decide how long a pipe or vessel will last, an engineer needs corrosion expressed not as a total amount but as a speed, and the most common way to state that speed is mils per year. A mil is a thousandth of an inch, so a corrosion rate in mils per year, or MPY, says how many thousandths of an inch of wall are being lost every year. This guide defines the metric and its metric-unit twin, shows the mass-loss and Faraday formulas behind it, explains how coupon, electrical-resistance, and LPR data become a rate, and covers the severity bands and how a trended rate projects remaining wall life.
Corrosion Rate (MPY) in one line: Corrosion rate in mils per year, abbreviated MPY, is the average penetration or wall-thickness loss of a metal expressed in thousandths of an inch per year, with the metric equivalent stated in millimetres per year. It is calculated from mass loss over time for a coupon, from resistance change for an electrical-resistance probe, or electrochemically from current for an LPR probe. The rate is grouped into severity bands from low to severe, and a reliably trended rate is used to project the remaining life of the wall and to set inspection and monitoring intervals.
A corrosion rate turns metal loss into a speed by dividing the amount lost by the time it took. In United States practice that speed is stated in mils per year, where a mil is one thousandth of an inch, so a value of 5 MPY means the wall is thinning by five thousandths of an inch each year. The rest of the world usually states the same idea in millimetres per year, and the two convert directly, so the choice is a matter of units rather than meaning. Both describe an average penetration rate over the period measured.
For a weight-loss coupon the underlying formula is a mass-loss relationship. The coupon is weighed before exposure, cleaned and reweighed after a known time, and the mass it lost is combined with its exposed surface area, the density of the metal, and the exposure time to yield a uniform penetration rate. Intuitively, more mass lost, less area, a lighter metal, or a shorter exposure all translate to a faster rate. The result is an average over the whole exposure, which is why coupon rates smooth out short-term swings.
For an electrochemical measurement the link runs through Faraday's law, which ties the electric current from corrosion to the quantity of metal dissolved. An LPR probe measures a corrosion current, and Faraday's law, together with the metal's equivalent weight and density, converts that current into a penetration rate. This is what lets an LPR probe report a rate almost instantly, whereas a coupon has to wait for measurable mass loss. An electrical-resistance probe sits between the two, deriving a rate from how fast the resistance of a thinning element climbs over time.
Because a bare number means little without context, corrosion rates are commonly sorted into qualitative severity bands, typically labelled low, moderate, high, and severe, with thresholds that rise from a fraction of a mil per year up to many mils per year. These bands give operators a quick read on whether a system is behaving benignly or eating through its wall, and they help translate a monitoring value into an action, since a rate creeping from moderate into high is a signal to intervene even before a hard remaining-life limit is reached. The exact thresholds depend on the material and service, so bands are used as a guide rather than a universal rule.
A subtlety that trips up new users is that short-term and long-term corrosion rates can diverge sharply, and both are meaningful. A long-term rate, averaged over months or a full coupon exposure, captures the steady background behaviour of the system. A short-term rate, taken over days or from a fast-reacting probe, captures what is happening right now and can spike far above the long-term average during an upset, a chemistry change, or a lapse in inhibitor injection, then fall back. Neither is more correct; they answer different questions, one about the settled trend and one about the current condition.
This is why a good program watches both. The long-term rate anchors the remaining-life estimate and the inspection plan, while the short-term rate is the early-warning signal that something has changed. A short-term spike that fast probes catch may never move the long-term average much if it is brief, but it flags an event worth investigating, and a short-term rate that stays elevated is what eventually drags the long-term rate up. Reading the two together tells an operator both how the system normally behaves and whether today is normal.
The practical purpose of a corrosion rate is to answer how much service life is left, and that comes from a simple projection: the remaining allowable wall, meaning the metal above the minimum required thickness, divided by the corrosion rate gives the years remaining before that limit is reached. The quality of that estimate depends entirely on how representative the rate is, which is why a rate trended over time is far more trustworthy than a single reading. A stable, well-established rate gives a confident projection, while a noisy or rising rate signals that the life estimate is moving and needs closer attention.
Bringing the rate into a monitoring system is what makes trending practical. On a cloud SCADA platform such as Merobix a corrosion transmitter's output can be logged continuously, so LPR and electrical-resistance rates are recorded alongside the process variables that drive them, such as temperature, water cut, and inhibitor injection. Seeing the corrosion rate and its causes on the same trend lets an operator connect a rising rate to a specific change and respond, rather than discovering the drift only at the next manual coupon pull. Alarms on a high or rising rate turn the metric into an active safeguard instead of a number reviewed after the fact.
The trend also drives the inspection schedule. A system holding a low, steady rate can justify a longer interval between detailed inspections, while one whose rate has climbed into a higher band warrants shortening the interval and confirming the readings with a coupon or direct measurement. For remote and unmanned sites, surfacing the trended MPY through cloud SCADA means this judgement no longer waits on a site visit, since the rate history is available centrally and an adverse trend raises a notification. Used this way, a trended corrosion rate becomes the link between what a probe measures today and how the asset is managed over its life.
A mil is one thousandth of an inch, so a corrosion rate in mils per year, or MPY, states how many thousandths of an inch of wall are lost each year. A rate of 5 MPY means the wall is thinning by five thousandths of an inch annually. The metric equivalent expresses the same idea in millimetres per year, and the two units convert directly, so they describe the same penetration speed in different measurement systems.
A long-term rate is averaged over months or a full coupon exposure, so it captures the steady background behaviour of the system, while a short-term rate is taken over days or from a fast-reacting probe and captures what is happening right now. During an upset or a lapse in inhibitor injection the short-term rate can spike well above the long-term average and then fall back. Both are meaningful because they answer different questions, one about the settled trend and one about the current condition.
Remaining life is projected by dividing the remaining allowable wall, meaning the metal above the minimum required thickness, by the corrosion rate to get the years left before the wall reaches its limit. The estimate is only as good as how representative the rate is, so a rate trended over time gives a far more reliable projection than a single reading. A rising rate shortens the estimated life and is a signal to shorten inspection intervals and investigate the cause.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.