# A Crack Width Survey Can Pass Every Reading and Still Miss the Crack That's Letting Water Through
Author: Nitin Verma
Author URL: https://insights.spans.co.in/author/nitin-verma
Published: 2026-11-04
Meta Title: Crack Width Survey vs Active & Through-Cracks | Spans
Meta Description: A crack width survey on an RCC tank confirms each crack's width falls under the code limit at the moment it was measured. It does not confirm whether that crack is still moving, or whether it runs through the full wall section, and both are what actually decide the risk.
URL: https://insights.spans.co.in/crack-width-survey-active-through-crack-wastewater

Every RCC tank at an [ETP](https://spans.co.in/effluent-treatment-plant-etp/) or [STP](https://spans.co.in/sewage-treatment-plant-stp/) eventually gets a crack width survey, a tape or gauge reading against each visible crack, checked against the code-permissible limit for a water-retaining structure, typically around 0.2 mm for a crack that passes fully through the section. A survey that comes back with every crack under that limit is a real result and a necessary one. It does not answer the two separate questions that actually decide whether any of those cracks are a problem, because a single width reading taken on one day tells you nothing about whether the crack is still moving, and nothing about whether it runs all the way through the wall.

![A crack width survey report shows three cracks all passing under the 0.2mm limit, next to two unanswered questions: is the crack still opening, and does it go all the way through to the wetted face](https://spans.co.in/images/blog/crack-width-survey-active-through-crack-inline.png)

## A width reading and a risk assessment are different claims

A crack width measurement answers: how wide is this crack, on the face where it was measured, right now. It does not answer whether the crack is active, still opening under ongoing movement such as differential settlement or shrinkage that has not finished, or dormant, having opened once during early curing and stabilized. An active crack measured today at a comfortably passing 0.1 mm can be wider than the code limit within months if nothing is tracking its trend, while a dormant crack sitting at the exact same 0.1 mm poses no ongoing risk at all, and a single reading cannot tell the two apart. The measurement also says nothing about depth: whether the crack penetrates the full wall thickness, giving water and dissolved chlorides a direct path to the reinforcement, or stops short of the liquid face as a surface-only crack. Two cracks of identical measured width can carry completely different risk depending on which of these conditions holds, and a crack width survey, performed correctly against its own narrow brief, was never built to distinguish either one.

## Why a passing survey and real risk can point in opposite directions

A crack width survey is usually a point-in-time exercise: a surveyor walks the tank, measures each visible crack once, and records a pass or fail against the code limit. That snapshot is genuinely useful for catching a crack that is already too wide. It cannot, by its own design, catch a crack that passes today but is actively widening, because detecting activity requires a second reading taken deliberately later and compared against the first, something a one-time survey never does. It also cannot distinguish a through-crack from a surface crack from the outside face alone, because that distinction depends on whether the crack extends through the full section to the wetted face, a condition a width gauge pressed against the dry side cannot see.

This matters because the two failure modes the code limit actually exists to prevent, ongoing leakage and reinforcement corrosion, are driven specifically by through-cracks and active cracks, not by crack width in isolation. [Durability of reinforced concrete](https://en.wikipedia.org/wiki/Durability_of_reinforced_concrete_structures) depends on keeping water and chloride ions away from the steel, and the [concrete cover](https://en.wikipedia.org/wiki/Concrete_cover) that is supposed to provide that protection is exactly what a through-crack bypasses; a crack that penetrates to the reinforcement does far more damage at a given width than a surface crack that never reaches it, which is exactly the distinction a single-face width reading cannot make. A tank that passes every width reading on a survey report can still be carrying an active, through-thickness crack that the survey format was never designed to flag, because nothing about recording "0.15 mm, pass" captures either its trend or its depth.

This is the same shape of mistake we described in [the piece on concrete corrosion in the H2S headspace](https://insights.spans.co.in/concrete-corrosion-h2s-headspace-wastewater): a measurement that is correct and genuinely useful for its narrow purpose gets mistaken for a complete answer to a broader durability question it was never built to cover. There, a visual concrete condition check missed an active acid-attack mechanism happening above the waterline. Here, a crack width reading that is accurate for the instant it was taken says nothing about whether that crack is moving toward the limit or already past it on the inside face, because width alone was never the full risk picture to begin with.

Picture a tank wall that develops a hairline crack during early curing, surveyed at 0.12 mm and logged as a clear pass. Unmonitored, the crack is in fact still active, driven by ongoing differential settlement under one corner of the foundation, and widens gradually over the following two years, crossing the 0.2 mm through-crack limit well before the tank's next scheduled survey. By the time a visible damp patch on the outside face prompts a re-inspection, the crack has been an active leakage and corrosion path for months, while the only record on file says the tank passed its crack width check.

The same gap shows up for a different reason in [the piece on anchor bolt vibration loosening](https://insights.spans.co.in/anchor-bolt-vibration-loosening-pump-foundation-wastewater): a one-time torque check and a one-time crack width reading are both snapshots of a condition that is defined by how it changes over time, not by where it sits on the day someone happened to measure it. A bolt torqued correctly at installation can still loosen under months of vibration nobody re-checked for. A crack measured within limit at survey time can still be actively opening under movement nobody re-measured against. Both checks are correct and necessary at the moment they are performed, and both become misleading the instant they are treated as a final answer rather than a first reading in a series.

## What actually assessing crack risk requires

The fix is not abandoning the crack width survey, which remains the correct first screen for catching an already-too-wide crack. It is adding the two checks a single reading cannot provide: a tell-tale or crack gauge across any crack near the limit, re-read on a defined schedule, so an active crack reveals itself as a trend rather than a single number, and a wetted-face inspection or dye test where feasible to confirm whether a crack of concern actually penetrates the full section rather than assuming width alone settles the question.

Treating a crack survey as complete only once both checks have been applied to any crack close to its limit turns a snapshot into an actual risk assessment, rather than a reading that happens to look fine on the day someone measured it. Against the [repair and remediation cost](https://spans.co.in/wastewater-treatment-cost/) of discovering an active through-crack only after it has been leaking for months, a tell-tale gauge and an occasional wetted-face check are a small, recurring expense rather than a significant one, and the gap between the two costs widens every month an active crack goes untracked.

If your plant's crack inspection regime stops at a one-time width reading, with no tell-tale tracking and no check on whether a crack actually goes through the wall, that gap is worth closing with [Spans' structural engineering team](https://spans.co.in/contact/) before a crack that passed its survey becomes the leak nobody saw developing.


---
This blog is powered by Superblog. Visit https://superblog.ai to know more.
---