Undertube Deposits in a Cooling-Water Cooler: Under-Deposit Corrosion or MIC?
A carbon steel shell-and-tube lube oil cooler on an open cooling tower loop was opened after tube failures. Tubes are 3/4 in OD, 12 BWG (0.109 in wall), water on the tube side at 3 ft/s bulk velocity, outlet 112°F. About 60% of the tube-side cross-section is covered with a black, foul-smelling deposit containing iron oxide and biological slime; the inlet strainer is fouled and 18 of 220 tubes are already plugged. Under the deposits, pits reach 0.055 in deep and are hemispherical with some tunneling; clean tube surfaces show essentially no general metal loss. Nozzle-to-channel pitting is minimal. Cooling water: 320 ppm chloride, 2.1 ppm total iron, 0.3 ppm sulfide, and ORP logs that swing between 150 mV and 480 mV because chlorination is intermittent and manually controlled. An adjacent 304 stainless bundle in the same loop shows pitting under similar deposits.
1. Evaluate the competing mechanisms. Why is the very low general corrosion rate on clean surfaces misleading, and what does the 304 stainless evidence add?
Read the worked answer
Both under-deposit corrosion and microbiologically influenced corrosion (MIC) are operating, and they are not mutually exclusive - MIC is a common driver of localized under-deposit attack.
The low rate on clean surfaces is misleading because corrosion in this system is not area-averaged. Deposits create a differential aeration/ion-concentration cell: the shielded metal becomes anodic, the bulk water cathodic, and attack concentrates in a small fraction of the surface. A bundle that looks 'fine' by general-thickness measurements can still perforate in months. Key indicators of MIC specifically:
- Foul-smelling, organic-rich, iron-sulfide-bearing deposit with slime.
- Localized hemispherical pits and tunneling under deposits, not uniform thinning.
- ORP swinging 150–480 mV - intermittent, manually dosed chlorination means long periods with no effective biocide residual, which is exactly the window in which sessile biofilms establish.
- The 18 plugged tubes and fouled strainer raise local velocity in remaining tubes while also allowing deposit accumulation in low-flow regions.
The 304 bundle pitting under the same deposits supports a shared root cause (deposit + biological activity) rather than a carbon-steel-specific corrosion issue. At 112°F and 320 ppm chloride, chloride SCC of the 304 is unlikely (that generally needs sustained temperatures well above this), so pitting under deposits is the more defensible interpretation for the stainless as well.
2. Specify the field and laboratory work needed to confirm MIC and to separate it from a purely abiotic under-deposit mechanism, and state what each result would change in your mitigation plan.
Read the worked answer
Request, on a still-wet deposit sample:
- Sessile/planktonic bacteria counts and ATP (general aerobic/anaerobic bacteria, acid-producing and sulfate-reducing groups). Elevated sessile counts or high ATP relative to bulk water confirms an active biofilm and justifies a biocide program change; sterile findings point toward a deposits/fouling control problem instead.
- Deposit chemistry - organic carbon, sulfur species, iron sulfide, and nutrient content; plus bulk water analysis (iron, sulfide, chloride, suspended solids, LSI/saturation index). Corrosive or scaling water chemistry changes the corrosion-control program, not just the biocide.
- Metallographic/scanning examination of the pit morphology - tunneling, subsurface cavities, and a 'comet-tail' penetration pattern with distinct colony-like features support MIC; smooth hemispherical pits with clean rims under a simple scale layer support abiotic under-deposit attack.
- Corrosion-product analysis - iron sulfide and biomass distribution.
How results change the plan: confirmed MIC → continuous, automated, alternating oxidizing/non-oxidizing biocide dosing with residual and ORP control, deposit dispersant, and a formal biofilm-monitoring program. Abiotic under-deposit → focus on filtration, strainer reliability, and side-stream filtration. In both cases the fouled strainer and low-velocity/plugged-tube hydraulics must be corrected, since deposits are the enabling condition.
3. The cooler must run another two years before replacement. What inspection and mitigation program would you specify, and what thickness measurement approach is appropriate?
Read the worked answer
- Do not rely on spot UT thickness on cleaned tube IDs. The damage is localized and under deposits; a general-thickness survey will average it away. Use full-length internal rotary/IRIS or eddy current on the carbon steel tubes with a defined reporting threshold, plus visual borescope of every opened tube for deposit coverage.
- Sample and map the worst tubes - pull a tube, section it at the deepest pit and measure remaining wall against the calculated minimum required thickness, then use it as a calibration point for the NDE.
- Fix the feed condition first - repair/replace the inlet strainer, restore design tube velocity, adjust plugging strategy so remaining tubes are not over- or under-velocity, and institute continuous biocide control with ORP/residual records rather than manual chlorination.
- Increase cleaning frequency and verify effectiveness by deposit weight or visual coverage, not by 'it looks cleaner'.
- Upgrade for the residual life - consider an upgraded tube alloy or a coatings/internal lining only if the deposit environment can be controlled simultaneously; coating over an active biofilm environment often fails.
- Set a re-inspection interval short enough that the observed pit growth (0.055 in in this service interval) cannot perforate before the next inspection, i.e., roughly half the time it took to reach that depth, and re-evaluate on the second inspection rather than assuming linear growth.