KALTEC

BlogApplication notes

Why is my boiler getting scale and corrosion?

Type
Application notes
Published
20 November 2026
Prepared by
Technical team
Product group
Water & Waste Treatment

Acid-cleaning the tubes or adding more chemical is the usual response to boiler scale and corrosion, but if what comes in is left as it is, the same damage returns. What actually causes them can be hardness slipping past the softener, oxygen left in the feedwater, condensate turned acidic by carbonic acid, too little blowdown, or boiler water carried over with the steam, and one cause often brings on another.

Sometimes fuel use climbs from last month while steam demand stays the same, or an inspection finds a white crust inside the tubes and water seeping from joints in the condensate return. Because the boiler goes on making steam as usual, these changes tend to show up first on the fuel bill or as a leak repair request.

The loss starts with fuel. Scale conducts heat about an order of magnitude less well than bare steel, so even a thin layer acts as insulation, and the United States Department of Energy puts the extra fuel use from scale in fire-tube boilers, where hot gas runs through tubes surrounded by water, at up to 5%, with normal scale about 0.8 mm, or 1/32 inch, thick costing about 2% of fuel (DOE, 2012a). The larger issue is safety. When heat cannot pass into the water, tube metal runs hotter, loses strength and can in the end bulge and burst (DOE, 2012a; Kim et al., 2019). Corrosion ends the same way. Oxygen pits can perforate a tube wall even when overall metal loss is small (DoD, 2005; Qua et al., 2011), and acid in the condensate thins the return lines and drives up replacement costs (Department of the Army, 1986; DoD, 2005).

The principle is simple. A boiler sends out only steam, so most of what was dissolved in the feedwater stays behind and grows more concentrated (DOE, 2012b). Once a component passes the limit of what can stay dissolved, it hardens onto the heating surface as scale, while gases carried in with the water work towards dissolving the metal (DoD, 2005). So the aim of boiler water treatment is to remove harmful matter before it enters, and to hold what remains with chemicals and blowdown (ASME, n.d.; DoD, 2005).

In such cases the technical team sets up the five hypotheses below and rules them out one by one. The start is always the same: on the same day, measure softener outlet hardness, boiler water conductivity, which shows how much salt is dissolved, and condensate pH, and compare them with the operating log. That one comparison screens several hypotheses at once.

Hypothesis 1: hardness slip

If stack temperature climbs and fuel use rises at the same load, look at hardness first. Hardness means dissolved calcium and magnesium, which the softener normally takes out. This hypothesis shows up with softeners that regenerate on a timer rather than by water use, softeners whose salt depends on someone remembering to top it up, and plants whose raw water changes with the seasons. Think of a softener that ran for days with an empty brine tank. The boiler goes on making steam as if nothing were wrong, but weeks later the stack temperature at the same load is creeping up and the fuel log is rising.

Hardness salts behave the opposite way to most salts. Most salts dissolve better as water gets hotter, but the salts that make up hardness, such as calcium carbonate, calcium sulfate and magnesium hydroxide, dissolve less as temperature rises (DoD, 2005). So they come out first at the hottest spots, the tube surfaces facing the flame. Calcium bicarbonate in the feedwater breaks down with heat into calcium carbonate and sticks as white scale (DoD, 2005).

Hardness usually gets in through the softener. Its ion-exchange resin swaps calcium and magnesium for sodium, and once the resin is spent, hardness passes straight through (DoD, 2005). Public guidance names poor softener operation combined with too little sludge conditioner or dispersant as the typical scaling condition (DoD, 2005). Scale with silica and iron mixed in blocks more heat for the same thickness (DOE, 2012a).

Checking starts at the softener outlet. Outlet hardness should stay below 1 ppm, and the resin is regenerated when it goes over; test it daily or each shift, depending on softener capacity (DoD, 2005). Next comes the trend. A stack temperature that climbs over weeks at the same firing rate and flue gas oxygen level signals scale building up (DOE, 2012a). When the boiler is shut down, put a drop of dilute acid on any white or off-white deposit. If it fizzes, it is calcium carbonate; if not, think calcium sulfate, calcium phosphate or silica, and confirm the exact type by deposit analysis (DoD, 2005).

Closing the way in for hardness comes before stripping the scale (DOE, 2012a). Put softener outlet hardness on the shift log, and on days it exceeds 1 ppm, check the regeneration timing and the salt level together. In the boiler water, keep an internal treatment at its set residual so that leftover hardness turns into sludge that blowdown can remove (DoD, 2005). If stack temperature rises while the hardness log is clean, suspect silica or iron and send a deposit for analysis. Remove scale already formed mechanically or by acid cleaning, and reset the treatment programme afterwards (DOE, 2012a).

Hypothesis 2: residual oxygen

Oxygen becomes the suspect when holes and leaks matter more than scale. It turns up in boiler houses that do not log deaerator pressure and temperature, at sites that stop and restart boilers often, and in boilers left standing with the water open to air. Picture a boiler house where someone has nearly closed the deaerator vent valve to save steam. No plume has shown at the vent for a long time, and one year an inspection finds deep pits clustered inside the economiser tubes.

Dissolved oxygen does not wear carbon steel down evenly. It concentrates at small spots and digs deep; this is called pitting, and it can perforate a tube wall even when total metal loss is small (DoD, 2005). The corrosion rate of oxygenated water roughly doubles with every 10 °C rise (DoD, 2005), so the economiser, which preheats feedwater with flue gas heat, and the feedwater inlet are especially exposed (Qua et al., 2011). In a study of an actual economiser tube leak, dissolved oxygen in the feedwater was confirmed as the cause of the pitting inside the tubes (Qua et al., 2011).

Oxygen is removed in two steps. First, the deaerator mixes feedwater with steam and heats it to boiling, so oxygen and other gases come out and leave through the vent. A chemical oxygen scavenger such as sodium sulfite then ties up the trace that remains (DoD, 2005). If either step slips, oxygen reaches the boiler. Water exposed to air during a shutdown does the same, so how a boiler is stored while idle is also a cause (DoD, 2005).

Whether steam shows at the deaerator vent is the first check. A plume of steam should always be visible at the vent; if it is not, gases are not being driven off (DoD, 2005). Gauge pressure should be at least about 3 psig, or 0.2 bar, and outlet water within 1 °C of the saturation temperature at that pressure (DoD, 2005). A low temperature means oxygen has not been fully removed. Measure the scavenger residual in the boiler water daily, and on days it falls below target, look at feedwater flow and the dosing pump together. At inspection, deep pits clustered at the top or waterline of the steam drum, or inside economiser tubes, point first to oxygen corrosion (DoD, 2005; Qua et al., 2011).

Once the problem narrows to oxygen, work on the deaerator before the chemicals. Make vent, pressure and outlet temperature fixed items in the operating log, and if outlet temperature keeps running below saturation, fix the steam supply and the vent valve first. Feed the scavenger continuously, but when the residual drops, check the deaerator before raising the dose. A scavenger cannot make up for poor deaeration indefinitely. Hydrazine is classed as a potential occupational carcinogen (NIOSH, n.d.), so in low-pressure industrial boilers consider other scavengers first. For a stop of more than a few days, choose either wet lay-up, with the boiler full and sulfite and alkalinity held high, or complete drying (DoD, 2005).

Hypothesis 3: acidic condensate

This hypothesis comes to mind when the condensate return lines suffer before the boiler itself. It stands out where high-alkalinity raw water is used as make-up, in plants that return little condensate and so feed a lot of make-up, and on large sites with long return lines. Say the piping near the boiler house is sound, yet leaks keep being repaired only at threaded joints in the return line of the farthest building, and the condensate in the return tank sometimes looks red.

The starting point, perhaps surprisingly, is the feedwater. The main source of carbon dioxide in condensate is feedwater alkalinity, the acid-neutralising matter, which in boiler feedwater is mostly carbonate and bicarbonate ions (DoD, 2005). Heat and pressure in the boiler break these ions down into hydroxide and carbon dioxide gas. Carbon dioxide is volatile, so it leaves with the steam and dissolves again as carbonic acid wherever the steam condenses (DoD, 2005). Condensate is close to pure water with almost nothing dissolved, so there is almost nothing in it to absorb this acid.

This carbonic acid can corrode carbon steel faster than a strong acid at the same pH, because the dissolved acid keeps supplying fresh hydrogen ions to the metal surface. An iron carbonate film can form and protect the surface, but how well it does so depends on temperature and pH (Nešić, 2007). On site it shows up as even wall thinning rather than pits, worst along the bottom of the pipe where condensate runs and at threaded joints (DoD, 2005). The dissolved iron returns to the boiler and settles as iron oxide deposits, so condensate corrosion turns back into a scale problem (DoD, 2005).

Measuring pH at one nearby point is not enough. Test the condensate not only at the return tank but also at the use point farthest from the boiler, and expect corrosion wherever it falls below 7.5 in the return system (DoD, 2005). Log make-up alkalinity and the condensate return rate as well. The higher the make-up alkalinity and the lower the return rate, the more raw material for carbonic acid enters the boiler (DoD, 2005; EPA, 2012). Red or cloudy condensate, or deposits that are mostly magnetic iron oxide, mean the return lines are corroding (DoD, 2005).

With carbonic acid as the cause, the remedy runs two ways. One is to feed a volatile neutralising amine, which travels with the steam, continuously to hold condensate pH at 7.5 to 9.0, ideally 7.5 to 8.5 (DoD, 2005). Each amine splits differently between steam and water, so amines with different properties are sometimes combined to protect both near and far piping (Department of the Army, 1986; DoD, 2005). The other is to cut the raw material. If make-up alkalinity is so high that neutralising takes a large dose, consider dealkalisation (DoD, 2005). Where air leaks in and brings oxygen too, neutralising alone falls short, and a film-forming treatment that coats the metal surface can be considered alongside it (IAPWS, 2019).

Hypothesis 4: low blowdown

Too little blowdown shows two signs together: boiler water conductivity stays above target, and sludge is found on the bottom at shutdown. It tends to arise in boilers where bottom blowdown is done by hand, in boilers whose steam use has grown while the blowdown setting stayed the same, and where blowdown was cut to save fuel and water. One example is a boiler where, with fewer people on shift, bottom blowdown starts to be skipped on some days. Nothing happens at first; then the gauge glass begins to swing more often, and at the next overhaul a thick layer of sludge lies on the drum bottom.

Water leaves as steam, but the solids stay. Suspended solids settle as sludge that cuts heat transfer, and dissolved solids encourage foaming and carryover (DOE, 2012b). Blowdown is the practice of draining part of the boiler water to keep these below set levels. Too little leads to deposits and carryover; too much wastes energy, water and chemicals (DOE, 2012b; EPA, 2012).

A spot with deposits becomes more dangerous. The deposit blocks heat, the metal beneath it overheats, the water that seeps into the deposit boils off, and sodium hydroxide from the boiler water is left behind at very high strength. A failure analysis of a power plant boiler tube found this sequence: overheating from internal deposits, evaporation of the water, then caustic corrosion by sodium hydroxide, ending in a bulging rupture (Kim et al., 2019). This kind of alkaline attack is chiefly a concern in high-pressure boilers (DoD, 2005).

To put a number on it, work out the cycles of concentration. Cycles means how many times stronger the boiler water is than the feedwater; blowdown conductivity divided by feedwater conductivity gives the cycles, and its inverse is the share of feedwater drained as blowdown (EPA, 2012). Work it out daily and check it against the target range. The blowdown rate is typically 4 to 8% of feedwater flow, rising to 10% when make-up solids are high (DOE, 2012b). Conductivity that runs high while the rate is well below this means blowdown is too low. Bottom blowdown is a manual task of a few seconds every few hours (DOE, 2012b), so check the records too, to see whether each shift actually did it.

Correcting low blowdown starts with setting limits. The upper limits for boiler water conductivity, alkalinity and silica depend on pressure and duty, so set industry consensus limits that match the operating pressure (ASME, n.d.; DoD, 2005). Controlling continuous surface blowdown automatically from conductivity cuts both the variation between operators and the waste (DOE, 2012b). If conductivity often runs over the limit, look at returning more condensate to cut make-up before adding blowdown. Condensate carries almost no solids, so it lowers the blowdown needed in the first place (EPA, 2012). If tubes show bulging or local wall loss, get a deposit analysis and an expert inspection.

Hypothesis 5: carryover

Carryover is boiler water leaving with the steam, and it is the suspect when white deposits build up in the equipment that uses steam, or the steam comes out wet. It becomes a problem in processes with sudden load changes, in boilers with finely tuned feedwater control, in boilers whose water is heavily concentrated, and wherever steam goes to a turbine. Suppose several processes open their steam valves at once at morning start-up. The drum gauge glass surges, and weeks later white deposits begin to collect in traps and valves at the far end of the steam users.

Boiler water reaches the steam by four routes. First is mist, fine droplets thrown up as bubbles burst at the surface, that get past the steam separator. Second is foaming, where high alkalinity, dissolved or suspended solids, or treatment chemicals acting together make foam that passes into the steam. Third is priming, where a sudden jump in steam demand or oversensitive feedwater control makes the drum level surge and water spill over, a mechanical cause (DoD, 2005). Fourth is silica that volatilises straight into the steam without any droplets, which grows with pressure (DoD, 2005; IAPWS, 2013).

The droplets carry solids at boiler water strength, so they leave sodium salt deposits and contamination in steam lines, condensate and any process that uses the steam (DoD, 2005). Where there is a steam turbine, silica deposits can become serious enough to unbalance the blades (DoD, 2005; IAPWS, 2013).

The clue comes first from condensate conductivity. Below 25 µS/cm, carryover is usually insignificant; above it, trace the cause (DoD, 2005). The procedure for measuring how much boiler water the steam carries can follow international guidance (IAPWS, 2008). If hardness shows up along with high conductivity, raw water or a heat exchanger leak into the condensate is more likely than carryover (DoD, 2005). Steam-side deposits that are mostly sodium salts point to carryover, silica to silica volatilisation, and magnetic iron oxide to condensate corrosion (DoD, 2005). Watch, too, whether the gauge glass swings and the steam turns wet at sudden load changes or right after manual blowdown (DoD, 2005).

The response depends on the route. Foaming is usually held by keeping total alkalinity below 20% of dissolved solids and suspended solids below 8%, with an antifoam if needed (DoD, 2005). Priming cannot be fixed with chemicals, so correct feedwater control, load swings and blowdown procedure. Keep boiler water silica below the limit for its operating pressure, and if it goes over, raise blowdown or remove silica from the make-up (DoD, 2005). If the steam drives a turbine, set and track separate steam purity limits (IAPWS, 2013).

Testing the hypotheses

Five hypotheses do not call for five separate investigations. Split the problem first, then rule them out one at a time, starting with what is easy to measure.

The first step is to sort the symptom. Rising fuel use and stack temperature point towards scale, leaks and holes towards corrosion, and contaminated steam users towards carryover (DOE, 2012a; DoD, 2005). Then take the three readings from the start, softener outlet hardness, boiler water conductivity and condensate pH, on the same day. Outlet hardness above 1 ppm makes hypothesis 1 the strong one, and condensate pH below 7.5 at any point, near or far, sends you to hypothesis 3 (DoD, 2005). Cycles worked out from boiler water conductivity that run above target point to hypothesis 4 (DOE, 2012b; EPA, 2012).

If holes are the problem and those three readings are clean, go to the deaerator. If any of the vent plume, pressure, outlet temperature or scavenger residual is off, treat oxygen, hypothesis 2, as the main cause (DoD, 2005). If the steam side is suspect, look at condensate conductivity and hardness together. High conductivity alone points to hypothesis 5; hardness as well means look first for a heat exchanger or raw water leak rather than carryover (DoD, 2005).

If that still does not settle it, take deposits at shutdown and send them for analysis. Calcium carbonate that fizzes in dilute acid points to hypothesis 1, magnetic iron oxide to condensate corrosion, hypothesis 3, and sodium salts on the steam side to hypothesis 5 (DoD, 2005). Caustic corrosion under the deposit in a bulged tube means hypothesis 4 (Kim et al., 2019). After the fix, it is done only when outlet hardness, deaerator conditions, condensate pH and boiler water conductivity stay in range over several checks, and stack temperature at the same load does not climb again.

One more thing to remember: the hypotheses often overlap. Iron from condensate corrosion settles in the boiler, alkali concentrates under that deposit, and too little blowdown adds foam on top, so hypotheses 3, 4 and 5 run into one another (DoD, 2005; Kim et al., 2019). In such cases, fix what comes in first: hardness, oxygen and the iron that returns with the condensate. Unless the inflow is stopped, blowdown and chemicals cannot catch up downstream.

Limits and cautions

The control values are general figures from public guidance for low- and medium-pressure industrial steam boilers; actual limits are set from the boiler maker's recommended range and industry consensus limits for the operating pressure (ASME, n.d.). High-pressure power boilers and heat recovery boilers have their own treatment schemes and limits, so follow the guidance for those (IAPWS, 2013, 2015). When a tube has bulged or burst, or leaks appear in several places within a short time, do not decide on floor judgement alone; get an expert inspection and a failure analysis.

Safety matters too. Boiler chemicals include strong alkalis, acids and toxic substances, so check the product's safety data sheet, the SDS, and wear eye protection and chemical-resistant gloves. Hydrazine is a potential occupational carcinogen (NIOSH, n.d.). Where steam touches food or people directly, the use of neutralising amines is restricted by regulation, so check those rules first (DoD, 2005). Leave acid cleaning, tube failure investigation and repairs to pressure parts to qualified specialists.

Sources

  1. American Society of Mechanical Engineers [ASME]. (n.d.). Consensus on operating practices for the control of feedwater and boiler water chemistry in industrial and institutional boilers. Research and Technology Committee on Water and Steam in Thermal Systems, Water Technology Subcommittee. First published 1979; edition revised after 1994 and 1998. https://asme.org/publications-submissions/books/find-book/consensus-operating-practives-control-feedwater-boiler-water-chemistry-industrial-institutional-boilers
  2. Department of the Army, Office of the Chief of Engineers. (1986). Use of diethylaminoethanol, morpholine, and cyclohexylamine for condensate return line corrosion prevention (Technical Note 86-3). https://apps.dtic.mil/sti/tr/pdf/ADA168297.pdf
  3. International Association for the Properties of Water and Steam [IAPWS]. (2008). TGD1-08, Procedures for the measurement of carryover of boiler water into steam. https://iapws.org/techguide.html
  4. International Association for the Properties of Water and Steam [IAPWS]. (2013). TGD5-13, Steam purity for turbine operation. https://iapws.org/techguide/Purity.html
  5. International Association for the Properties of Water and Steam [IAPWS]. (2015). TGD3-10(2015), Volatile treatments for the steam-water circuits of fossil and combined cycle/HRSG power plants. https://iapws.org/techguide.html
  6. International Association for the Properties of Water and Steam [IAPWS]. (2019). TGD11-19, Application of film forming substances in industrial steam generators. https://iapws.org/techguide.html
  7. Kim, Y. S., Kim, W. C., & Kim, J. G. (2019). Bulging rupture and caustic corrosion of a boiler tube in a thermal power plant. Engineering Failure Analysis, 104, 560–567. https://doi.org/10.1016/j.engfailanal.2019.06.022
  8. National Institute for Occupational Safety and Health [NIOSH]. (n.d.). NIOSH pocket guide to chemical hazards: Hydrazine. https://www.cdc.gov/niosh/npg/npgd0329.html
  9. Nešić, S. (2007). Key issues related to modelling of internal corrosion of oil and gas pipelines – A review. Corrosion Science, 49(12), 4308–4338. https://doi.org/10.1016/j.corsci.2007.06.006
  10. Qua, H. C., Khaw, C. C., Tan, C. S., Wang, X., & Ooi, J. B. (2011). Fast identification of O2 corrosion in economiser tubes. Engineering Failure Analysis, 18(8), 2201–2210. https://doi.org/10.1016/j.engfailanal.2011.07.014
  11. U.S. Department of Defense [DoD]. (2005). UFC 3-240-13FN, Industrial water treatment operation and maintenance (archived). https://www.wbdg.org/FFC/DOD/UFC/ARCHIVES/ufc_3_240_13fn_2005.pdf
  12. U.S. Department of Energy [DOE]. (2012a). Clean firetube boiler waterside heat transfer surfaces (Energy Tips: Steam, Steam Tip Sheet #7, DOE/GO-102012-3597). Advanced Manufacturing Office. https://www.energy.gov/sites/prod/files/2014/05/f16/steam7_surfaces.pdf
  13. U.S. Department of Energy [DOE]. (2012b). Minimize boiler blowdown (Energy Tips: Steam, Steam Tip Sheet #9, DOE/GO-102012-3406). Advanced Manufacturing Office. https://www1.eere.energy.gov/manufacturing/tech_assistance/pdfs/steam9_blowdown.pdf
  14. U.S. Environmental Protection Agency [EPA]. (2012). WaterSense at work: Best management practices for commercial and institutional facilities, Section 6.5, Boiler and steam systems. https://19january2025snapshot.epa.gov/system/files/documents/2023-05/ws-commercial-watersense-at-work_Section_6.5_Boilers.pdf

Product and site consultation

Leave your company, site and the product groups in use. Our engineer will get in touch to go through your site conditions and concerns. Materials, equipment, operating conditions and downstream steps show what should change and what can stay. If nothing needs to change, we will say so.

Contact us
After you contact us
  1. 01
    Site setup

    We confirm process conditions and needs first.

  2. 02
    Spec design

    We set the product spec and selection criteria.

  3. 03
    Comparison

    We compare matching candidates and propose.

  4. 04
    Tailored supply

    We supply on time and manage the stock.

  5. 05
    Site check

    We visit, check conditions of use and record.

  6. 06
    Use management

    We track use and change cycles to catch issues.

  7. 07
    Regular analysis

    We analyze the records and send a yearly report.