Carbonate Scale Cleaning on the Circulating-Water Side of Shell-and-Tube Heat Exchangers

Abstract: When carbonate scale builds up on the circulating-water side of a shell-and-tube exchanger, the first symptom is a widening terminal difference and falling circulation rate. This article follows the field sequence: sample the deposit from the header, tube sheet and behind the baffles after shutdown and depressurisation, type it by laboratory analysis, confirm a tube-side loop with the shell side isolated, run closed-loop circulation monitored by acidity, iron and turbidity, neutralise, rinse and passivate without a gap, and accept the job on tube-sheet inspection, before-and-after terminal difference and a complete record set. Parameters specific to this unit are marked as pending confirmation; no efficiency gain is assumed.

1. Sample the deposit first: carbonate scale cannot be typed by colour

The deposit on a cooling-water side is rarely a single compound. After a long run the carbonate layer usually covers rust, mud and a little biological slime. Dosing for carbonate alone wastes the acid on the surface layer, the rust and slime underneath do not loosen, and the result is the familiar picture of clean tube mouths with a blocked mid-section.

Sample after shutdown, depressurisation and header removal, covering the header wall, the tube-sheet face, the tube mouths and behind the baffles. Record the location, appearance and thickness at each point (measured thickness at each point: [TO BE CONFIRMED: deposit thickness at each sampling point]) and photograph the coupons. Location matters more than quantity: behind the baffles the velocity is lowest, the layer is usually thickest, and that is exactly where acceptance problems appear.

The laboratory should at least split the sample into carbonate (as CaCO3), iron oxide, and mud or biological slime, and where possible add acid-insoluble matter and oil content ([TO BE CONFIRMED: laboratory composition of the deposit]). Under GB/T 25146-2010, carbonate at 60% or more as calcium carbonate may be judged as carbonate scale, and iron oxide at 80% or more as rust scale; the ratio changes the chemistry completely.

Carbonate-dominant deposits are treated mainly by dissolution. Where rust dominates, complexing and dispersion come in, sometimes after a loosening stage. Where slime is heavy, a biocide and dispersant stage comes first, otherwise the acid is wrapped in slime and consumed at the surface. No firm chemical quantity or cleaning duration is quoted before the analysis is in hand.

2. Confirm the cleaning side and isolate the rest

Walk the process drawing against the actual pipework before touching anything: which side carries circulating water and which carries process medium, confirmed from the nameplate and line markings rather than experience. Copy the nameplate data into the file ([TO BE CONFIRMED: nameplate material, tube count and design conditions]), because it fixes the tube material, tube count and the compatibility limits of the cleaning solution.

Only the circulating-water side is cleaned. Where the tube side carries the water, the cleaning loop connects to the tube side; where the shell side carries process medium such as oil, steam or product, its valves are closed, tagged and locked, with blinds where needed. Cross-contamination does not merely spoil a cleaning job, it contaminates the whole process system.

The loop is assembled from temporary pipework: mixing tank, circulation pump, temporary strainer, drain and neutralisation point. The pump must hold a steady flow through the bundle, with head and flow calculated from bundle resistance and loop length ([TO BE CONFIRMED: pump flow and head]); a small pump pushed against a full bundle leaves dead zones behind the baffles.

A clean-water trial circulation follows: check flanges, hose joints, valves and the pump seal for leaks, confirm a steady level and no weeping, and record the pressure test ([TO BE CONFIRMED: test pressure and hold result]) before any chemical is mixed. Skipping this step means stopping the wash and rebuilding the loop if a joint fails.

3. Mixing and circulation: concentration and temperature follow the monitoring

Mix the bath to the analysis once the trial circulation passes. Dosing order and concentration follow the supplier instructions and the laboratory beaker test, and the beaker-test record is filed before the bath is scaled up. No dosing ratio is given here: change the deposit type on the same unit and the ratio has to be recalculated, and copying another plant's recipe is the quickest way to damage a bundle.

Main cleaning runs as a closed loop, fed low and returned high with venting at the high point so that gas pockets do not create dead zones. The flow direction is reversed once according to the plan so that both tube ends and the area behind the baffles see fresh solution; stop the pump, release pressure, then reconnect hoses, never breaking a joint under pressure.

Sample the loop at fixed intervals for three things: free acid concentration, total and ferric iron, and turbidity. Falling acidity means the chemical is still being consumed; when iron stops rising and acidity is steady, the soluble deposit has largely reacted. The interval follows deposit thickness and unit size ([TO BE CONFIRMED: sampling interval]) and every reading goes into the monitoring sheet, because a log that says normal proves nothing at acceptance.

Temperature follows the material and the chemical: carbon steel tolerates a higher bath to speed the reaction, while stainless steel and copper alloys need a firm ceiling ([TO BE CONFIRMED: cleaning temperature and tolerance]). Overheating costs the inhibitor first and raises the corrosion rate, which cannot be felt by hand afterwards, only weighed on coupons.

4. Neutralisation, rinsing and passivation: clean is not the same as ready to run

Drain the spent bath to the designated collection tank, neutralise it to the environmental requirement before discharge, and keep the volume and disposal record. Rinse with clean water in repeated displacements until the outlet is near neutral with no foam or suspended solids, then open the headers and inspect the tube sheet and tube mouths, marking residue positions on the point map.

Carbon steel is left active after pickling and rusts quickly in humid air. Passivation or pre-filming follows as soon as rinsing passes, with the sequence written into the cleaning plan in advance, so there is no waiting period after the wash. The film on carbon steel is judged by the red-dot test, with no dot earlier than 5 s for a continuous film.

Stainless steel is judged differently: the blue-dot test allows no more than 8 dots in 10 min, and chloride in the cleaning solution is held at 25 mg/L from the mixing stage rather than corrected halfway through. Copper alloy bundles depend on chemical compatibility, and any formulation risk to copper is excluded during the beaker test.

Renew the gaskets before refitting the headers, tighten the bolts in diagonal passes to an even load, and repeat a hydrostatic or leak test afterwards, recording the result on the acceptance sheet. Refitting quality that is not recorded leaves a later leak impossible to attribute between cleaning and assembly.

5. Acceptance on three things: tube mouths, terminal difference, records

First, visual inspection: after the headers come off, check whether the deposit on the tube-sheet face and inside the tube mouths has gone, record the position and extent of minor dead-zone residue, and agree on the spot whether a local re-clean is needed. File the sampling point map and the before-and-after photographs as field evidence of removal.

Second, operating data: at the same load after start-up, compare inlet and outlet temperatures, flow and terminal difference with the pre-cleaning figures. Keep both sets; without a before-and-after comparison there is no basis for a performance statement, and this article assumes no efficiency gain in advance - a quotation that fixes a percentage is difficult to substantiate at acceptance.

Third, the document set: deposit analysis report, bath mixing and dosing records, monitoring sheet, waste disposal record, leak and pressure test records and the acceptance sheet, filed as one package. Missing documents reduce acceptance to visual inspection, and any later dispute about heat-transfer performance has nothing behind it.

Where plugged tubes are found during acceptance, count them and record their positions separately ([TO BE CONFIRMED: number and position of plugged tubes]). If plugging is outside the scope of the chemical cleaning, state the scope and responsibility on the acceptance sheet so that a mechanical blockage is not confused with cleaning performance.

FAQ

The deposit analysis says carbonate, but the rust layer underneath is significant. Why is one acid wash not enough?

Once the carbonate layer dissolves, the rust underneath reacts differently to the same chemical and needs complexing and dispersion, often after a separate loosening step. A single pass typically leaves the tube mouths clean and the mid-section still coated, so the bath is designed in stages according to the layer-by-layer analysis rather than one chemistry run to the end.

The tube side carries circulating water and the shell side carries process oil. Why must the cleaning loop be connected to the tube side only?

The carbonate scale sits on the tube-side wall that the circulating water touches, so the cleaning solution has to enter the tubes to reach it. The shell side carries process oil, where a connected loop would risk cross-contamination of the process system, so its valves are closed, tagged and, when needed, blinded.

Iron readings have levelled off during circulation, but hard scale remains behind the baffles. What is done?

That is usually dead-zone residue where the flow is low and the solution exchanges poorly. The location and extent are recorded, the area is given extended circulation or the flow direction is switched and re-cleaned, and the inspection sheet names the re-cleaned area after the headers come off again.

No pre-cleaning terminal difference data was kept. How is the acceptance comparison handled?

Without same-load data from before the wash, no figure for the improvement can be stated. The post-cleaning terminal difference and flow are logged in full as the baseline for the next cleaning, and this acceptance covers only the open-header inspection and the process records.

Why must a carbon steel exchanger be passivated or pre-filmed after acid cleaning?

Pickling leaves the steel surface active, so flash rust appears quickly in humid air and corrosion products later travel with the water. Passivation follows as soon as rinsing passes; the film on carbon steel is judged by the red-dot test, with no dot earlier than 5 s, and the sequence is written into the plan in advance.

Related reading

• Cleaning process selection for heat exchangers
• How many condenser cleaning methods are there
• All cleaning articles

Deposit typing, removal rate and corrosion limits follow GB/T 25146-2010, Acceptance specification for industrial equipment chemical cleaning quality.

Service coverage: Jiangsu (Nanjing, Suzhou, Wuxi, Changzhou, Zhenjiang, Yangzhou) · Zhejiang (Hangzhou, Ningbo) · Shanghai · Anhui · Shandong · Henan

© 2026 Danyang Blue Star Anti-corrosion Cleaning Co., Ltd. | Source: www.lanxingqingxi.com