CP Shielding: When the Coating Protects the Corrosion

A foreign line is going in, and it crosses yours. Not your project. Somebody else's pipe, somebody else's crew, somebody else's excavator. They run their one-call locates, dig the crossing, and expose your line where the two cross. You're called out for what comes next: climb down into the open ditch, connect test leads to both lines so the crossing can be monitored later, and set a couple of magnesium anodes on your line in case you ever need to drain interference down the road. Nobody called you because something's wrong. The pipe isn't backfilled and the connections are brand new, so there's nothing to read yet. This part of the visit is visual: the condition assessment of the exposed pipe that a crossing like this usually requires.

Nothing about the segment says there's a problem, either. A nearby test station read fine at the last annual, and the last close interval survey over this spot was clean. Going in, you'd have called the coating here good.

The coating is polyethylene tape. At a glance it still looks intact. Then you notice a wrinkle, a spot where the tape has lifted off the steel. You press on it and water weeps out from underneath. You peel back a corner, and the steel under it isn't bright and it isn't just discolored. It's pitted and actively corroding, under a coating that never showed a holiday to any survey that passed over it.

The meter said protected. The steel says otherwise. Both are telling the truth. This article is about how that happens, and what it costs you to misread it.

Pull the readings and lay them out honestly.

Start with the on-potential at that station: −1.82 V CSE. Nobody reads protection off an on-potential, and you don't either. It's measured with current flowing, so it carries IR drop, the voltage your meter picks up across the soil between the reference cell and the pipe. How much it carries depends on the current and the soil, not on whether a rectifier happens to be nearby. A −1.82 V on-reading is common enough. It isn't excess protection and it isn't a problem. It's just an on-reading, which is why you interrupt and take the off.

The off-potential is the number that means something: the instant-off, read the moment current stops and the IR drop collapses to zero. Here it was −0.950 V CSE. More negative than −0.850, so it passes. The station was half a mile back, so you pull the close interval survey for this exact spot, and its off-trace runs right around −0.95 V through here too. Two independent reads, same answer: protected.

Neither off-potential was inflated, and neither was a measurement error. Both were clean, defensible, passing readings, and the steel a few inches away was losing metal.

Here's why that can be true. Every potential you read, on or off, describes the steel that's electrically connected to the soil and trading current with it: the bare metal at holidays, the surface your CP actually reaches. Your reference cell reads the potential field that reachable steel produces. The steel sealed under a sheet of intact, disbonded tape is screened out of that circuit. It adds almost nothing to the field, so the cell reports almost nothing about it. A passing off-potential describes the steel cathodic protection can reach. It says nothing about steel hiding behind the shield.

Start with what a good pipeline coating is, electrically. It's a dielectric, an insulator. That's the entire point. A sound coating blocks current from passing through it, so cathodic protection ignores the coated steel and concentrates on the few bare spots, the holidays, where it's actually needed. The better the coating insulates, the less CP current the line draws and the longer the system lasts. Dielectric strength is a virtue. The coatings literature is blunt about it: to be effective, a coating must be dielectric, so it keeps cathodic current from passing through to the steel.

Now disbond that coating without breaking it.

Polyethylene tape is the textbook case, and it's no accident this happened on tape. Tape systems have wrestled with the same failure since they went in the ground: water migrating under the tape during service, soil stress dragging at the wrap as the ground swells and shifts, marginal adhesion from the day it was applied. The coatings industry spent decades engineering around it, with higher-adhesion overlaps and double-faced designs, because tape's known weakness is losing its grip on the steel while staying otherwise intact. When that happens, the tape lifts into a tent or a wrinkle, and a thin gap opens between coating and pipe. Groundwater finds the gap. Now you have an electrolyte-filled crevice with a roof over it.

And that roof is still a dielectric. The same insulating quality that made the coating good is now working against you. It blocks CP current from reaching the steel underneath just as well as it once kept current off the steel it was bonded to. Except now there's bare steel under there, wet, with nothing protecting it. The coating has stopped being a barrier that helps and become a shield that hurts.

This failure mode has a name. SP0169, the same standard behind the −0.850 V criterion, calls it cathodic shielding, and defines it plainly: disbonded coatings can prevent CP current from reaching the pipe, leading to undetected corrosion. Notice where the standard puts the fix. Not on the rectifier, but on the coating: select coatings that resist disbondment, and hold the line on application and repair quality. We'll come back to why the rectifier can't bail you out.

Inside that crevice, the chemistry goes where we covered it in "Oxidation Does Not Mean Oxygen." The trapped water gives up its oxygen and nothing replaces it, but the absence of oxygen doesn't stop the corrosion cell. Iron still gives up electrons:

Fe → Fe²⁺ + 2e⁻.

The environment stagnates, the chemistry and pH drift, and oxygen-starved ground is exactly where sulfate-reducing bacteria set up shop and accelerate the attack. The crevice does everything an aggressive corrosion site does, and the one tool built to stop it is locked outside the door.

A measured word on cathodic disbondment, since it belongs on the list of causes but is easy to over-weight. At a true holiday, the cathodic reaction creates a strongly alkaline, hydrogen-rich environment that can creep under the coating edge and peel it back. That's cathodic disbondment, and resistance to it is something good coatings are tested for. Excessive CP can make it worse, which is one more reason not to chase a high on-potential with more current. But on tape, the dominant driver of disbondment in the ground is mechanical, water and soil stress, not the current. Don't pin this failure on that −1.82 V on-reading. The number is loud, but it isn't the culprit here.

Here's the catch: almost every routine tool you'd reach for to find this is blind to it.

Start with the coating survey. In "ACVG vs DCVG" we laid out both methods honestly, including the same blind spot in each. ACVG can't read a disbonded-but-unbroken coating, because no current is leaving the pipe there and there's no gradient to detect. DCVG has the same hole from the other side: with no breach, there's nothing for the survey to see until a holiday opens. Both surveys hunt for current crossing the coating. A shield's whole nature is that current isn't crossing it. The survey walks right over active corrosion and logs nothing.

The coupon won't save you either. A coupon is a coating holiday you build on purpose, a known piece of bare steel taking on CP current the way a real holiday would. That's exactly why it's blind here. The coupon is bare, CP reaches it, and it reads protected. It models the holidays, which are fine. It can't model the shielded steel, which isn't. Your coupon hands you a clean instant-off while the corrosion runs untouched a few feet away.

Even your close interval survey, the over-line workhorse, reads the polarized potential of the steel CP can reach. The shielded steel doesn't broadcast its distress to the surface, any more than it did to the test station. And here's the worst tell of all: a disbonded-but-intact tape coating keeps the line's current demand deceptively low. The pipe still looks well-coated to your rectifier, because electrically it still is. The current just can't get to the steel that needs it. Low, stable current demand on an old tape-coated line is not proof of a healthy line. It can be the sign of a line quietly corroding under its own shield.

So what works? Three things, in rough order of certainty.

Direct examination is the ground truth. Every bell hole is a free coating inspection: integrity digs, tie-ins, foreign-line crossings like this one. When the pipe is exposed, get your hands on the coating. Don't trust tape that looks intact. Push on it, feel for tenting and wrinkles, look for water tracking and soft spots. The disbondment you can find with a gloved hand is one your survey already missed.

In-line inspection doesn't care why the steel is thinning. Magnetic-flux-leakage and ultrasonic tools measure metal loss directly, shield or no shield. On a tape-coated line with a disbondment history, ILI is often the only practical way to find shielded corrosion across miles of pipe you'll never dig. If the line is piggable, the integrity program and the corrosion program need to be reading each other's data.

Write down more than the number. Shielding corrosion isn't random. It tracks coating type, age, and ground conditions, so what turns up in one bell hole tells you something about every similar foot of line. That's what makes the hole worth more than the potential reading you came for. Log the location, the coating and its condition, the disbondment, the soil, and what the steel looked like, not just the pipe-to-soil number. Whether this tape was a surprise or something you already knew was out there, it's a data point either way, and it only pays off if it's written down. Do that across enough digs and the picture fills in: one accidental find at a crossing becomes a map of where to look on purpose.

The reflex, when a reading looks short, is to turn the rectifier up. Understand why that fails here, because it's the whole point of the title.

Cathodic protection suppresses corrosion by supplying electrons from outside, so the steel never has to give up its own. That only works where the current arrives. Behind a disbonded dielectric coating, there's no low-resistance path for current to arrive. That's what a shield is. Turn the output up and you raise the potential on the steel CP can already reach, pushing the accessible surface toward genuine over-protection, while the shielded steel sits right where it was, at its own freely corroding potential, losing metal at its own pace. You won't deliver a single electron to the face that's actually corroding.

That's the trap in one line: the missing oxygen didn't stop the corrosion, and the coating stops the cure. You can't energize your way past a shield. The fix is mechanical, and it's going to take elbow grease.

When you find shielded corrosion, the response is hands-on and unglamorous.

Take the coating back to sound steel. Don't repair only the spot you can see. Disbondment runs under coating that still looks attached, so probe outward and remove every bit that isn't tightly bonded, back to where adhesion is genuinely good. The coating-repair literature is direct about this: overcoating poorly adhered coating just propagates the failure under your new material. If the bond beyond the defect is marginal, it all comes off.

Assess the steel before you re-cover it. Clean the corroded area, measure pit depth and remaining wall thickness, and document it. If the metal loss is significant, that's an integrity question, not a coating one, and how to remediate it belongs to the integrity engineer and the operator's procedures, up to and including a cut-out. Either way it outranks the coating repair. Don't bury a problem you haven't measured.

Prepare the surface and recoat with a system that won't repeat the failure. Surface preparation to the proper standard, a near-white blast and the right anchor profile, is what makes a repair last, and it's almost always the step that gets shortchanged. Choose a recoat material suited to the conditions in that ground, and on a cathodically protected line, favor a well-bonding, non-shielding system over more of what just failed. Just make sure whatever you pick is compatible with the parent coating it ties into. That comes up constantly on tape: many operators won't spec it anymore, so you end up patching a mostly-tape line with something else, and the two have to bond cleanly where they meet. The goal is a repair that, if it ever disbonds again, won't wall the steel off from its CP.

Verify, then widen the search. Holiday-detect the repair before backfill. Then step back. A shielded find on a tape-coated segment is a screening flag for the whole vintage, not a closed ticket. Feed it to the integrity program, schedule ILI or targeted digs on similar pipe, and treat the accidental discovery as the start of a deliberate look.

The pipe at the crossing got lucky. Somebody else's project put eyes on it before it leaked. The rest of that line doesn't have a crossing scheduled. That's the part that matters, and the part worth acting on.

Established in 2011, Roberts Corrosion Services, LLC delivers comprehensive, turn-key cathodic protection and corrosion control solutions nationwide. Our end-to-end expertise encompasses design and inspection, installation and repair, surveys and remedial work. We provide drilling services for deep anode installations and a full laboratory for analysis of samples and corrosion coupons, as well as custom CP Rectifier manufacturing.

While our initial focus was on the Appalachian Basin area, we complete field work all over the US. We are a licensed contractor in many states and can complete a wide range of services.

Our biggest strength is in our flexibility for our clients. Solutions and Results.

Let us know how we can help.

Our Training Site

Newsletter Site

Website

LinkedIn

Location: 39.251882, -81.047440

(304) 869-4007