Eddy Current Testing of Heat Exchanger Tubes: Field Lessons from 316L and Brass Condensers

Eddy current testing of heat exchanger tubes is one of those methods that looks deceptively simple in a brochure and turns into a genuine engineering problem in the field. This article shares practical observations from inspection campaigns on 316L acid heat exchangers and brass condensers — the kind of work that shaped my approach and eventually became a conference paper presented at the 2019 South Australia NDT Conference.

The Setup

Heat exchangers fail in predictable ways — pitting from chloride attack, erosion at tube inlets, fretting at baffle supports, stress corrosion cracking near U-bends. Eddy current testing is the industry standard for detecting and sizing these defects in non-ferromagnetic tubes because it’s fast, quantitative, and gives you a full length record of every tube in a single pass.

For 316L tubes handling acid service, the primary concern is usually pitting. For brass condensers, the story is different: impingement attack at the inlet, dezincification in certain waters, and mechanical damage from debris.

Same method, two very different inspection strategies.

Probe Selection Matters More Than People Admit

For differential bobbin probe inspection of 316L tubes, I typically use frequencies in the 200–400 kHz range for primary analysis, with a lower frequency around 50 kHz for dimensional verification and tube support location. The bobbin probe is efficient — a single pull gives you 100% volumetric coverage — but it has a fundamental limitation: it’s insensitive to circumferential defects and can miss small pitting on the OD if the defect is not near the probe’s sensitivity peak.

For brass tubes, the conductivity is higher, so the frequency selection shifts. And for any tube where pitting is the concern, I’ve found that supplementing bobbin with a rotating probe (RPC or motorised rotating pancake coil) on any suspect tubes is worth the extra time. Bobbin might flag an indication; RPC tells you whether it’s a through-wall crack, a cluster of pits or a single deep pit.

The failure mode I’ve seen most often on new inspectors is defaulting to manufacturer-recommended frequencies without verifying the response on real calibration standards matching the tube material, wall thickness and support geometry.

Calibration Realities

ASME V mandates the use of an ASME reference standard with four through-wall holes at 100%, 60% and 20% through-wall, plus flat-bottom holes and a groove for circumferential flaw response. That’s fine for setting up the instrument, but it doesn’t tell you how your system will respond to a field of 40% deep pits distributed over a 25 mm axial zone.

For one 316L acid heat exchanger I worked on, we manufactured custom reference tubes with EDM notches at specific depths and known pit geometries simulating the expected damage mode. The bobbin amplitude response to a cluster of shallow pits was dramatically different to a single deeper pit of equivalent volumetric material loss — the former under-called, the latter over-called against a standard ASME reference.

The lesson: for high-consequence inspections, invest in custom reference standards that reflect the actual damage mechanism. Generic calibration standards are a floor, not a ceiling.

The Role of Videoscope and 3D Simulation

One of the things I pushed for in the South Australia paper was combining ET data with internal videoscope inspection of flagged tubes — and in high-consequence cases, importing the ET-derived defect geometry into a 3D simulation to visualise the discontinuity in its actual tube geometry.

Why does this matter? Because plant operators, engineering managers and metallurgists are not NDT specialists. Showing them a Lissajous figure and a Voltage Depth Curve doesn’t help them make maintenance decisions. Showing them a 3D model of the pitted tube wall with the defect clearly visible, next to a videoscope still of the same location, does.

This is particularly valuable for FFS (fitness-for-service) decisions. When the question is “can this tube survive another 18-month campaign?”, the engineering team needs confidence in the defect dimensions and morphology — not just a signal amplitude reading.

Brass Condensers — Different Beast

Brass is a good conductor, which compresses the phase angle range compared to stainless. The practical effect is that the phase shift between OD and ID defects is smaller, making defect position harder to determine from phase alone. Amplitude ratios and multi-frequency analysis become essential.

Dezincification is also a detection challenge. It’s a selective corrosion mechanism that removes zinc from the brass, leaving a porous copper matrix. The ET response is subtle — not a sharp signal like a pit, more of a gradual baseline shift over a long axial zone. Easy to miss if you’re only looking for classic defect signatures.

For dezincification-prone service, I now always include a comparison tube (one removed from the bundle and cross-sectioned metallographically) in the procedure qualification phase. This calibrates the inspector’s eye to what degraded brass actually looks like in the signal, not just what a reference standard says.

Reporting — Where Most Inspections Fail

The signal analysis is only half the job. The reporting is where inspections either add value to the asset owner or become filing cabinet fodder.

My standard deliverable for heat exchanger inspections includes:

Executive summary — pass/fail by tube, number of tubes plugged/monitored/replaced recommended, highest risk tubes called out by position
Tube sheet map — graphical representation showing the location of all flagged tubes, colour-coded by severity
Signal data — all flagged tubes with bobbin and RPC traces, amplitude, phase, axial position, through-wall percentage
Damage mechanism assessment — what the defect morphology suggests about the underlying failure mechanism
Recommendations — not just “tube X at location Y is at 60% wall loss” but “tubes in this zone show a pattern consistent with inlet erosion; consider sleeving or flow redistribution at next turnaround”

The inspection report is an engineering document, not just a data dump. The inspector who can translate signals into operational recommendations is worth ten who just log amplitudes.

Final Thought

Eddy current testing works — when the procedure, probes, calibration and interpretation all align with the specific asset and damage mechanism. Default settings and standard procedures are fine for commodity inspections. For high-consequence heat exchangers, invest in the engineering work upfront: custom references, multi-frequency analysis, 3D reporting, and most importantly, an inspector who understands both the signal and the metallurgy behind it.

Have a heat exchanger inspection programme that needs an independent review? Get in touch — I work with operators globally on ET procedure development and programme optimisation.