IFG
Integrity Field Guide Damage mechanisms, fitness-for-service, and the inspection methods that find them
All articles Open the guide
GuideArticles › API 579

After a fire: deciding what can return to service

Fire damage assessment starts as detective work rather than inspection. Before any method is chosen, somebody has to establish how hot each component actually got — and the answer is written in what melted, what did not, and what colour the scale went.

API 5798 min read

The instinct after a fire is to inspect for damage: look for cracks, take thickness readings, check for distortion. All of that happens, but not first, because it answers the wrong question. A component can come out of a fire dimensionally perfect, free of cracks, at full wall thickness, and be metallurgically ruined.

Part 11 of API 579 therefore starts somewhere else. Before anything is measured, every component is placed in a heat exposure zone according to the maximum metal temperature it reached. Everything downstream — which components need evaluating, which methods, whether replacement is even arguable — follows from that classification.

Reading the temperature off the wreckage

Nobody had a thermocouple on the shell. The temperature is inferred, and the plant is full of instruments that were not intended as instruments. Each material has a melting or softening point, and finding out what melted and what merely charred brackets the temperature at that spot.

EvidenceTells you the metal reached
Paint blistered but not burned offA few hundred degrees at most
Common thermoplastics slumped or goneAbove roughly 100–150 °C
Zinc coating or galvanising meltedAbove about 420 °C
Aluminium fittings, ladders or cladding meltedAbove about 660 °C
Glass sight glasses softened and deformedAround 700 °C and above
Brass or bronze valve components meltedAbove roughly 900 °C
Heavy scale, distortion, sagging under own weightWell into the red-heat range

These are physical constants, not estimates, and used together they bracket a temperature quite tightly. An aluminium junction box that ran and a glass gauge that did not soften puts the metal between roughly 660 and 700 °C. That is a more defensible number than any eyewitness account of how big the flames were.

What temperature does to carbon steel

The threshold that matters for carbon steel sits around 730 °C, where the steel begins to transform to austenite. Below it, and away from the very long exposures that cause creep, the steel's microstructure is broadly unchanged and the assessment is mostly about distortion and any thinning. Above it, the material you have is no longer the material that was designed.

What it becomes depends entirely on how it cooled. Cooling slowly in still air after the fire burned out is one outcome. Being hit with a fire monitor while red hot is another, and it is a quench — the same operation a heat treater would use deliberately to make steel hard. Hard, in this context, means brittle: reduced toughness, sometimes dramatically, in a component that looks entirely normal.

The trap

A quenched component has no crack to find, no wall loss to measure and no distortion to see. Every conventional inspection passes it. The only thing that reveals it is a measurement of the material itself — and if it is put back into service and later sees a cold start-up, you have assembled the conditions for brittle fracture out of a fire that appeared to do no damage.

The order the work goes in

  1. Zone the whole affected area

    Walk it with the melting-point evidence and mark up a drawing. This is the deliverable that everything else refers back to, and it degrades every day as the site is cleaned up. Do it before anything is moved.
  2. Visual and dimensional survey

    Bulging, sagging, out-of-roundness, distorted supports, pulled anchor bolts, cracked welds, damaged insulation and its condition. Distortion is assessed against the tolerances in the relevant part of API 579 for shell distortion, and it is also evidence in its own right that the metal was hot and loaded at the same time.
  3. Field hardness testing

    The workhorse of fire damage assessment. It is quick, portable and non-destructive, and it detects both the hard quenched condition and the soft over-tempered one. Anomalous hardness against a baseline from the same material outside the fire zone is the signal.
  4. Replication where hardness is anomalous

    Field metallography puts a name to what hardness only flagged: a transformed structure, spheroidisation, decarburisation, incipient melting at grain boundaries.
  5. Thickness and cracking, now

    Only after the metallurgy is understood does conventional inspection add value: scale loss reduces wall, and quench cracking and thermal fatigue cracking need surface methods to find.
  6. Destructive testing, if the stakes justify it

    A removed coupon tested for toughness and strength is the only way to be certain about a critical component in a high zone.

Judgement, honestly stated

Two things are worth saying plainly at the end of a fire assessment.

The first is that replacement is often the right answer and the cheap one. For piping, small vessels and standard fittings that sat in a high zone, the engineering effort required to defend reuse costs more than new material and carries residual risk that new material does not.

The second is that insulation and fireproofing hide both good news and bad. Removing it is expensive and it is where the assessment budget goes. But insulated components in a fire zone are the ones you know least about, and a zone classification that assumes the insulation protected them is an assumption, not a finding.

Try it

Hardness, replication and the other metallurgy methods

Fire damage is assessed with techniques that look at the microstructure rather than the wall. The methods list explains what each of them can and cannot resolve.

Open the detection methods
Read next

Who writes this. A mechanical engineer with twelve years in oil and gas — in-line inspection, fired heater and furnace inspection, and pipeline integrity. What is here comes from the published standards and from what those years in the field actually looked like. It is not written by an API-certified inspector.

This is not an assessment. Nothing on this site may be used to justify a decision about real equipment. Assessing plant requires the current editions of the applicable codes, data from a licensed source, and a competent engineer who signs for the answer. · Integrity Field Guide