What Capacity and Size of waste heat recovery boiler Do You Need?

Sizing a waste heat recovery boiler wrong is one of those mistakes that looks minor on a datasheet and catastrophic on the energy bill. Undersized units leave recoverable enthalpy venting straight to atmosphere — sometimes 2–5 MW of thermal energy just gone, depending on the process. Oversized units create partial-load condensation problems, accelerated corrosion in the economizer section, and capital tied up in steel that never earns its keep. Either way, the plant pays.

A waste heat recovery boiler should be sized to match the actual flue gas flow rate, inlet temperature, and required steam parameters of your specific process. Steam output typically ranges from 1 t/h to 130 t/h, with inlet flue gas temperatures anywhere from 300°C to 1,100°C depending on the heat source. Getting those three inputs right — flow, temperature, and end-use steam pressure — determines everything else about the unit’s physical size and configuration.

What makes this genuinely tricky is that the “right” capacity isn’t a fixed number you look up in a table. It shifts with your process load profile, whether you need saturated or superheated steam, how many tail-end surfaces you’re willing to pay for, and what your stack temperature limit is for corrosion control. The next sections work through each of those variables in the order that actually matters for selection.

Auditing Your Heat Source: Flue Gas Flow, Temperature, and Composition Before Anything Else

Before you specify a single tube diameter or drum pressure, you need three numbers from your process: how much gas is flowing, how hot it is, and what’s in it. Skip that audit — or rely on guesswork from a similar plant — and you’ll either build a boiler that starves for heat half the year or one that clogs, corrodes, and trips offline every few months. Both outcomes are expensive. The audit itself costs almost nothing compared to a redesign after fabrication.

Volumetric Flow Rate: Nm³/h vs. Actual m³/h

Always specify flow at normal conditions (0°C, 101.325 kPa) — that’s Nm³/h — not at the actual gas temperature. A cement kiln exit stream at 350°C that measures 80,000 actual m³/h is only around 47,000–52,000 Nm³/h once you correct for temperature, and that difference directly changes the heat exchanger surface area your supplier will calculate. Confusing the two is one of the more common errors I see in enquiry documents from first-time WHRB buyers.

If no flow meter exists on your exhaust duct — and on older plants, they often don’t — you can back-calculate from mass balance: fuel consumption rate × stoichiometric air factor × (1 + excess air coefficient α) gives you a reasonable working estimate of flue gas mass flow, which you then convert using the mixture’s molecular weight and density. It’s not as clean as a Pitot traverse measurement, but for initial sizing it’s usually within 8–12% depending on how well the fuel composition is known.

Inlet and Outlet Temperature Targets

Inlet temperature drives the duty calculation directly. The range is wide in practice — gas turbine exhaust typically runs 450–560°C, while a copper smelter reverberatory furnace can push 900–1,100°C, requiring a radiation chamber ahead of any convective surface just to protect the tubes. Your outlet target temperature (the flue gas temperature leaving the WHRB) is equally important and is often set not by thermodynamics but by corrosion constraints.

The Dew-Point Problem with Sulfurous Gases

For any source with SO₂ above roughly 500 mg/Nm³, set your minimum flue gas exit temperature at 160–180°C — the exact floor depends on SO₃ concentration and moisture content, so run the acid dew-point calculation before finalizing tube material selection. Low-alloy steels like 20G will fail within one or two seasons at sub-dew-point wall temperatures. Corten or 09CrCuSb (ND steel) adds maybe 15–25% to economizer material cost but buys years of service life. Worth it every time.

Dust Loading Dictates Layout, Not Just Cleaning Frequency

Dust content in g/Nm³ shapes the entire boiler configuration.

Get these measurements — or at minimum credible estimates with stated assumptions — documented before you send an enquiry to any boiler manufacturer. A reputable supplier will ask for all of them anyway. If they don’t, that itself tells you something.

Calculating Required Heat Absorption: The Core Sizing Equation Explained Step by Step

Before you can trust any vendor’s proposal, you need to run the heat balance yourself. It’s not complicated, but skipping it is how plants end up with a boiler that’s 30% undersized or one that can never reach design steam output because the economizer is steaming in summer.

The Fundamental Heat Balance

The starting equation is:

Q = ṁ_gas × Cp_gas × (T_in − T_out) × η_recovery

Where Q is recoverable heat (kW or kcal/h), ṁ_gas is the flue gas mass flow rate (kg/s or kg/h), Cp_gas is the specific heat of the flue gas mixture (roughly 1.05–1.15 kJ/kg·°C for most combustion exhaust, higher if CO₂-rich, lower if diluted with excess air), T_in and T_out are the inlet and target outlet flue gas temperatures, and η_recovery accounts for heat loss through casing radiation and incomplete surface utilization — typically 0.88–0.94 on a well-insulated unit.

Worked example: a cement kiln bypass exhaust at 55,000 Nm³/h, 520°C inlet, targeting 180°C outlet. Convert volumetric to mass flow using the actual flue gas density (around 0.38–0.42 kg/Nm³ at that temperature — check your gas composition). Call it 0.40 kg/Nm³, so ṁ_gas ≈ 22,000 kg/h. With Cp_gas = 1.08 kJ/kg·°C and η = 0.91:

Q = 22,000 × 1.08 × (520 − 180) × 0.91 ≈ 7,370 kW (roughly 6.34 Gcal/h)

That number is your design anchor. Everything else — tube surface area, drum sizing, circulation ratio — derives from it.

Converting Q to Steam Output

Once you have Q, steam output D (t/h) follows from:

Q = D × (h_steam − h_feedwater)

Enthalpy values come straight from steam tables. At 1.6 MPa saturated, h_steam ≈ 2,794 kJ/kg; at 3.82 MPa with 10°C superheat, h_steam ≈ 2,800 kJ/kg; at 9.8 MPa with 480°C superheat, h_steam ≈ 3,386 kJ/kg. Feedwater enthalpy depends on deaerator temperature — commonly 104°C, giving h_fw ≈ 436 kJ/kg.

Using the example above (Q = 7,370 kW = 26,532 MJ/h) at 3.82 MPa:

D = 26,532 / (2,800 − 436) ≈ 11.2 t/h

A vendor quoting 13 t/h from the same gas stream should be asked to show their enthalpy assumptions. Usually they’ve used optimistic outlet temperature or ignored duct heat loss.

Choosing Your Outlet Flue Gas Temperature

This is where engineers make or break the economics. Push T_out too low and you recover more heat — but risk sulfuric acid condensation on the cold-end economizer tubes, which destroys them within one to two heating seasons. Push it too high and you’re leaving recoverable energy in the stack.

Typical targets: 160–200°C for exhaust containing sulfur (diesel generators, sulfur-bearing coal, some waste gases); 120–140°C for clean natural gas exhaust. If your fuel or process gas varies in sulfur content seasonally — and in many plants it does — design to the worst-case dew point with a 15–20°C margin, not the average.

Pinch Point and Why It Controls Physical Size

The pinch point is the minimum temperature difference between the flue gas and the saturation temperature of steam at a given cross-section inside the boiler. Industry standard for WHRBs is 15–25°C. Go below 15°C and you need disproportionately more heat transfer surface — cost rises steeply, and there’s real risk of subcooled boiling in the economizer, which causes flow instability. Go above 25°C and you’re simply leaving recovery on the table.

Approach temperature (between economizer outlet water and saturation temperature) should stay at least 10–15°C to prevent steaming in the economizer at partial load — a condition that accelerates tube erosion and makes the drum level control behave erratically.

Superheater Section: When You Need It and What It Costs

If steam drives a turbine or a process that specifies minimum superheat, you need a dedicated superheater bank upstream of the evaporator section (higher gas temperature side). Tubes here see both high temperature and steam-side oxidation, so material typically steps up to 12Cr1MoV or TP304H, depending on steam temperature. That change alone can add 15–30% to tube bundle material cost and push lead time out by 6–10 weeks compared to a saturated steam unit.

Indicative Sizing Table

Ranges reflect variation in Cp_gas, feedwater temperature (95–130°C), and casing loss assumptions. A gas with high CO₂ or water vapor content will shift values upward; highly diluted exhaust (excess air > 200%) shifts them down noticeably.

Run these numbers before you open the bidding process. Vendors who can’t reconcile their proposal with your heat balance — with transparent assumptions — are a red flag.

Selecting Steam Pressure and Temperature to Match Your Process or Power Generation Requirement

Pressure and temperature aren’t just boiler specs — they define the entire mechanical design, the applicable pressure vessel code, your inspection and certification timeline, and ultimately what the steam can actually do once it leaves the drum. Getting this decision wrong late in a project is expensive. Getting it wrong early is just an engineering error you can fix on paper.

Low-Pressure Saturated Steam (0.4–1.6 MPa)

This is the workhorse range for most process industries: food and beverage sterilization, timber drying, rubber vulcanization, general plant heating. At these pressures, saturated steam is usually all you need. The boiler design stays relatively simple — natural circulation works reliably, drum wall thickness stays manageable, and you don’t need a superheater section at all. Fabrication and inspection costs are lower, lead times are shorter, and your operators have a less demanding certification requirement in most jurisdictions.

One practical point: if your waste heat source comes in below roughly 400°C, you may struggle to justify a superheater anyway, because the temperature differential across the superheating surface shrinks and the added surface area delivers diminishing returns. At that point, a natural-circulation saturated steam WHRB at 0.8–1.0 MPa is often the right call.

Medium-Pressure Steam (1.6–5.3 MPa)

Once you’re driving a backpressure turbine for CHP, feeding an absorption chiller, or supplying refinery process requirements, you’re in medium-pressure territory. Superheated steam becomes necessary here — not just for efficiency, but because saturated steam entering a turbine causes blade erosion within months. Expect to add a superheater section and, depending on flue gas temperature and mass flow, possibly a forced-circulation design review. Natural circulation is still often feasible at the lower end of this range, but your thermal hydraulics engineer needs to verify it, not assume it.

The absorption chiller case is worth flagging specifically: many plant engineers specify 0.8 MPa for an absorption chiller when the machine actually calls for 0.6–0.7 MPa. That gap seems trivial until you’re throttling steam constantly and wondering why your chiller COP is drifting.

High-Pressure Steam (5.3–9.8 MPa)

At these pressures — typically for condensing or extraction turbines exporting power to the grid — the boiler design changes substantially. Drum walls thicken, materials move into higher-grade alloy territory, and natural circulation becomes increasingly marginal above roughly 7 MPa, depending on boiler height and circulation ratio. Forced circulation or once-through designs are common. Critically, this pressure range triggers the most demanding regulatory tier in every major code: Class A boiler certification under China’s GB/T 16507 and TSG G0001, ASME Section I for North American and many Middle Eastern projects, or PED Category IV under EU Directive 2014/68/EU. Inspection hold points multiply, third-party witness requirements apply, and material traceability documentation becomes non-negotiable. Budget the extra time — in my experience, moving from a medium-pressure design to a high-pressure design can add 8–14 weeks to a project schedule, depending on the certifying body’s workload.

Hot Water WHRB: The Overlooked Alternative

When your end use is district heating, process water circuits, or building HVAC, you don’t need steam at all. A hot water waste heat boiler with an outlet temperature of 95–150°C eliminates the steam drum, the steam certification requirement, and the operational complexity of blowdown management and steam trapping. Regulatory burden drops noticeably — hot water boilers below certain pressure-temperature thresholds fall outside the heaviest inspection tiers in most codes. Operators with less steam experience also tend to run them more reliably.

The tradeoff is heat transfer distance. Hot water systems need adequate pump sizing and pipe insulation to maintain temperature over long runs. For a compact single-building plant, this is rarely a problem.

Thermal Oil WHRB: High Temperature Without High Pressure

Some processes genuinely need 250–320°C heat delivery but can’t tolerate the steam pressure that would correspond to those saturation temperatures — roughly 4–11 MPa. Asphalt plants, bitumen storage, certain chemical reactors, and some textile operations fall into this category. A thermal oil (heat transfer fluid) WHRB circulates a synthetic or mineral oil through the coil circuit instead of water, delivering high-temperature heat at near-atmospheric system pressure.

Construction standards, coil geometry, fluid velocity requirements, and fire safety classifications are all different from steam boiler work. Specify the two as separate packages. Don’t let a procurement team consolidate them under one steam boiler spec sheet.

Pressure Vessel Code Selection for Export Projects

The code you specify locks in your material procurement, your welding procedure qualifications, your NDE requirements, and your authorized inspection body. For projects destined for North America or many parts of the Middle East, ASME Section I is essentially the default, and it requires an ASME-authorized inspection agency present at the shop. European projects typically require EN 12952 (water-tube) with CE marking under PED. China-origin boilers built to GB/T 16507 can be adapted for export, but the adaptation process — including third-party review of design documents to ASME or EN standards — takes time and adds cost that needs to be in the budget from day one, not discovered during submittal review.

A compact comparison:

Confirm the applicable code before the thermal design is frozen. Changing codes mid-design isn’t impossible, but it usually means requalifying weld procedures, re-sourcing some pressure-part materials, and, occasionally, redesigning the drum entirely.

Physical Configuration and Layout: Fire-Tube vs. Water-Tube and How Dust Loading Decides

The structural choice between fire-tube and water-tube isn’t academic — get it wrong and you’re either over-spending on a water-tube unit for a clean, low-volume gas stream, or you’re watching a fire-tube unit choke on dust-laden kiln off-gas within the first operating season.

Fire-Tube (Shell-and-Tube) WHRB

In a fire-tube design, hot flue gas passes through tubes submerged in the water/steam-filled shell. Simple to fabricate, compact, and relatively inexpensive — a unit in the 2–8 t/h range can be shop-assembled and shipped as a single piece, which procurement teams tend to appreciate when site access is tight.

The practical ceiling is roughly 600°C inlet gas temperature. Push beyond that and you’re stressing tube-to-tubesheet joints, and thermal fatigue becomes a maintenance headache inside two or three years. Capacity is realistically limited to about 10 t/h; anything larger gets physically unwieldy in shell form. Dust loading is the other hard constraint. Even moderate particulate concentrations — say, above 2–3 g/Nm³ — cause tube-side fouling that’s genuinely difficult to clean because you can’t easily access the inside of hundreds of small-bore tubes. Fire-tube WHRBs belong on clean gas sources: engine exhaust, natural-gas furnace flue gas, some chemical process off-gases where the stream has already been de-dusted upstream.

Water-Tube WHRB

Flip the arrangement — water and steam inside the tubes, hot gas flowing across the outside — and almost everything changes. You can handle inlet temperatures up to 1,100°C with proper front-end design, accommodate heavy dust loading because the gas-side surface is accessible for cleaning, and scale capacity up to 130 t/h or beyond. Water-tube units dominate cement kilns, electric arc furnaces, non-ferrous smelters, and glass tanks for exactly these reasons.

The trade-off is cost and complexity. A water-tube WHRB requires a proper steam drum, circulation system (natural or forced, depending on pressure and geometry), and more sophisticated engineering for tube bundle arrangement. But for anything above roughly 10 t/h or above 600°C inlet, it’s not really a choice — it’s the only sensible option.

Radiation Chamber: Non-Negotiable Above 700°C

When inlet gas comes in above roughly 700°C — typical for electric arc furnace off-gas, copper flash smelting, or certain glass furnace exhausts — a bare convection bank will overheat. The radiation chamber (sometimes called a radiant cooling section) is a membrane-wall or bare-tube enclosure that absorbs heat by radiation before the gas reaches the convection tube banks. It’s not optional at these temperatures; it’s a protection system. Skipping it to save capital cost is a short route to warped tube headers and premature failure. For arc furnace applications, the radiation section alone can account for 30–45% of total boiler length.

Soot Blowers and Cleaning Mechanism Selection

Dust loading dictates the cleaning system almost as directly as it dictates the structural type. For light, dry particulate — natural gas combustion residue, light process dust — rotary steam soot blowers spaced across the convection bank do the job adequately. For sticky or high-adhesion deposits, which you see in cement kiln WHRBs and glass tank units where alkali condensation occurs on tube surfaces, rapping mechanisms or acoustic (sonic) horns are worth specifying. In practice, some cement kiln operators run both: rapping for the front passes where adhesion is worst, steam blowers further back. It adds cost but avoids the alternative, which is manual cleaning outages every few weeks.

Horizontal vs. Vertical Arrangement

Large-capacity units — generally above 15–20 t/h — are almost universally horizontal, with multiple tube passes arranged in series. This gives better access for maintenance, more flexibility in placing soot blowers, and simpler steam drum placement. Vertical natural-circulation designs are common for cement kiln WHRBs in the 5–20 t/h range and wherever plot footprint is the binding constraint. Vertical layout keeps the drum at the top and the gas inlet at the bottom or side, which suits preheater-bypass configurations on dry-process kilns.

Modular Shop-Fabricated vs. Field-Erected

For units up to roughly 25 t/h, shop fabrication is almost always preferable. Assembly under controlled conditions, with qualified welders working on accessible joints, produces better weld quality and tighter dimensional tolerances than site work. Transport limits — typically 4.2 m width and around 30–35 t per section for road shipping, varying by country and route — are the practical ceiling. Modular units at this scale typically reduce site civil and mechanical construction time by 40–60% compared to field-erected equivalents. Above 25 t/h, or wherever transport restrictions force you to, field erection is the only path. Plan for it properly: site supervision by the manufacturer’s engineers, weld procedure qualification at site, and hydrostatic testing before commissioning. Trying to cut corners on field erection supervision is where projects accumulate expensive rework.

Auxiliary Systems That Determine Real-World Boiler Performance: Feedwater, Blowdown, and Controls

A waste heat recovery boiler doesn’t operate in isolation. The main pressure vessel gets all the attention during procurement, but in practice, the auxiliaries determine whether that boiler runs reliably for 15 years or starts failing tubes in year three. Feedwater treatment, blowdown management, bypass routing, and control architecture are where most real-world performance gaps show up — and where scope creep quietly inflates project costs if you don’t nail them down early.

Feedwater Quality: The Constraint That Breaks Tubes

Once you’re operating above roughly 2.5 MPa, dissolved oxygen and hardness become genuinely serious constraints, not just checklist items. Dissolved oxygen must stay below 7 ppb; total hardness below 0.03 mg/L. Exceed those thresholds consistently and you’re looking at pitting corrosion on the waterwall tubes, magnetite layer breakdown, and eventually pinhole leaks — usually in the worst possible location, like a bent section or a weld seam near the drum.

The fix is a deaerator combined with continuous chemical dosing: oxygen scavenger (typically sodium sulfite or hydrazine alternatives for higher-pressure units) plus an anti-scalant/phosphate program. What I’ve seen cause more WHRB tube failures than anything else is an undersized or poorly commissioned water treatment skid — the boiler itself was correctly specified, but whoever scoped the project treated water treatment as a commodity line item and cut corners. Don’t do that.

Deaerator Sizing: Margin Matters More Than You Think

Size the thermal deaerator to at least 120% of maximum continuous evaporation rate. That 20% margin sounds conservative until you have a sudden load swing and the deaerator can’t keep up — then you’re feeding oxygen-rich water directly into a hot boiler, which accelerates corrosion faster than most operators realize.

For boilers below about 1.6 MPa, an atmospheric deaerator usually suffices and costs meaningfully less. Above that threshold, a pressurized deaerator (typically operating at 0.12–0.20 MPa, around 104–120°C) does a better job of scrubbing oxygen down to acceptable levels, and the elevated feedwater temperature also reduces thermal shock at the economizer inlet. That last point matters more in waste heat applications because flue gas temperatures can swing significantly with upstream process load changes.

Blowdown: Don’t Leave Heat on the Floor

Continuous blowdown rates for a well-operated WHRB typically run 1–3% of evaporation, though units on harder makeup water or with marginal treatment systems can drift toward 5%. Every kilogram of blowdown you discharge carries enthalpy with it — which is recoverable. A blowdown flash vessel paired with a small heat exchanger can push 75–85% of that thermal energy back into feedwater preheat or the deaerator, depending on system layout. Payback on this equipment is often under 18 months at current energy prices, though the exact figure depends on blowdown rate, local fuel cost, and operating hours per year.

Bypass Damper and Diverter Valve: Process Continuity Is Non-Negotiable

If the boiler goes offline for inspection or an emergency, the upstream process — a cement kiln, a glass furnace, a gas turbine exhaust — cannot simply stop. The bypass damper and diverter valve assembly is what keeps that process running. Pneumatic actuators respond faster and are generally preferred where fail-safe positioning is critical; the standard fail-safe for most applications is “divert to bypass stack” on loss of instrument air or power, keeping the process clear of back-pressure. Electric actuators are simpler to integrate into a DCS but need a reliable UPS if fail-safe behavior is required. Specify the fail-safe direction explicitly in your datasheet — leaving it to the vendor’s default is a mistake.

Instrumentation, Control, and Safety Valve Certification

Drum level control on a WHRB should be three-element: measuring steam flow, drum level, and feedwater flow simultaneously rather than relying on level alone. Single-element control is technically adequate at very low load but causes hunting and poor transient response when flue gas temperature fluctuates — common in intermittent industrial processes. Steam pressure and temperature PID loops need tuning specifically for your heat source’s characteristic response time, which varies considerably between, say, a steady gas turbine exhaust and the cyclic output of an electric arc furnace.

For plant-wide integration, confirm DCS protocol compatibility early: Modbus RTU/TCP is ubiquitous and low-risk; Profibus DP is common in older European plant environments; OPC-UA is increasingly specified for new greenfield projects with modern SCADA platforms. Sorting this out after the panel is built costs real money.

Safety valve sizing must comply with the applicable pressure vessel code for your jurisdiction — GB 150 for China-registered projects, ASME Section VIII Div. 1 (with UV stamp) for North American and many export projects, EN 13445 for European installations. Set pressure, discharge capacity, and minimum lift must all be calculated and documented, not just copied from a catalog. Inspectors check this, and getting it wrong on a high-pressure unit is not a paperwork problem — it is a safety problem.

Industry-Specific Sizing Benchmarks: Cement, Steel, Glass, Gas Turbine, and Chemical Plant Applications

Before any detailed thermal calculation, it’s worth checking whether your project sits in a recognizable band for your industry. These benchmarks won’t replace engineering, but they will tell you immediately if a vendor proposal is in the right ballpark — or if something is badly off.

Cement Kiln Preheater Exit Gas

Clinker kilns are among the most productive waste heat sources in heavy industry, and also among the dirtiest. Preheater exit gas typically runs 150,000–400,000 Nm³/h at 300–380°C — temperatures that sound modest until you account for the dust loading, which usually falls between 80 and 120 g/Nm³ depending on meal feed rate and cyclone efficiency. That dust is the real design driver. A WHRB on a cement kiln almost always needs a gravity settling chamber or coarse cyclone upstream, followed by wide-pitch bare tubes in the boiler proper; trying to run finned tubes here accelerates fouling to the point where you’re cleaning every few days instead of every few weeks.

Typical electric equivalent output via ORC or back-pressure steam turbine runs roughly 5–25 MW, with the spread depending heavily on kiln throughput and whether you’re recovering from the cooler vent as well. Plants running 3,000 t/d clinker are usually at the lower end; 6,000–10,000 t/d lines can push toward the top.

Electric Arc Furnace Off-Gas

EAF waste heat is genuinely difficult. The off-gas volume — typically 60,000–150,000 Nm³/h at 900–1,100°C — is impressive on paper, but the heat release is violently intermittent: high during the melt-down phase, near zero during charging and tapping. A WHRB sized for peak gas conditions will spend a lot of time doing nothing useful.

The standard engineering response is a radiation cooling chamber (sometimes called a drop-out box) followed by an elevated-pressure steam drum large enough to act as a thermal accumulator. Capacity figures of 15–40 t/h steam are common on medium to large EAF shops, but the drum sizing conversation is usually more important than the evaporator surface area calculation. Getting that wrong gives you pressure swings and safety valve chatter on every heat cycle. A bypass damper to the quench tower is not optional — it’s your insurance when the EAF goes into an abnormal sequence.

Gas Turbine Exhaust (HRSG)

This is the cleanest, most predictable waste heat stream most engineers will encounter. Gas turbine exhaust runs 100,000–600,000 Nm³/h at 450–580°C with negligible particulate, which means you can justify finned-tube economizers, multi-pressure drum arrangements, and superheating sections without worrying about fouling every other month. Steam output for combined-cycle applications typically spans 10–80 t/h, with larger aeroderivative or heavy-frame machines at the top of that range.

Glass Furnace Regenerator Exhaust

Glass melting exhaust is moderate in volume — 20,000–60,000 Nm³/h — but the chemical environment is hostile. SO₂ and NOₓ concentrations are elevated enough that your economizer tube material matters enormously. ND steel (a low-alloy acid-resistant grade) is the minimum; 316L stainless is worth considering on the coldest sections where you’re closest to the acid dew point. Exit gas temperature from the WHRB must stay above roughly 160–170°C to avoid condensate corrosion, which limits how aggressively you can extract heat. SCR or SNCR for NOₓ control often gets integrated into the same ductwork, so the layout discussion with your engineering team needs to happen before the boiler is sized, not after.

Sulfuric Acid Plant Converter and Tail Gas

In a contact-process sulfuric acid plant, the converter outlet gas — typically 400–550°C — feeds a WHRB that must maintain exit temperature above the sulfuric acid dew point, which is usually in the 155–165°C range depending on SO₃ content and moisture. Practical capacity runs 2–10 t/h steam, which sounds small but represents significant process heat integration. The exit temperature constraint is non-negotiable; operators who push it too low in search of extra steam tonnage end up with severe cold-end corrosion and tube failures within a single campaign.

Diesel Generator Exhaust and Jacket Water — Combined Recovery

Remote sites — mines, island facilities, off-grid industrial camps — often run large diesel generator sets and discard two separate heat streams: exhaust gas at 350–450°C and jacket cooling water at 80–95°C. A dual-medium WHRB that handles both streams in one unit is a real product, not an exotic option, and it’s the right choice when you need to minimize equipment footprint and connection points. Typical output is modest: 0.5–3 t/h steam or the equivalent in hot water, depending on generator size. For a 500–1,000 kW genset serving a remote camp, that heat can cover domestic hot water, space heating, and light process loads entirely — which changes the economics of the whole site utility system.

These benchmarks are starting points. Any real project will have enough site-specific variables — altitude, feedwater quality, downstream steam header pressure, available maintenance capability — that the final sizing will drift from these norms. But if a vendor’s proposal lands well outside the range for your industry without a clear technical explanation, that’s worth a direct question before you accept it.

Procurement and EPC Checklist: What to Include in Your Technical Inquiry to Get Comparable Quotations

Getting three quotations back that you can actually compare against each other is harder than it sounds. Without a complete technical inquiry package, one vendor quotes a bare-shell unit with no economizer, another includes a full tail surface package with soot blowers, and a third prices in local third-party inspection while the others exclude it entirely. The resulting spread in numbers tells you almost nothing useful. The checklist below is what experienced EPC project managers send out — not a wish list, but the minimum data set.

Mandatory Process Data Sheet Fields

Start with flue gas flow in Nm³/h. State both the normal operating rate and the design maximum — the boiler has to handle peak load without going into thermal stress, and vendors who only see the average figure will undersize the pressure parts. Inlet temperature needs to be the worst-case sustained value, not a momentary spike; if your process swings 80–100°C depending on production rate, say so explicitly.

Flue gas composition matters more than most buyers realize. Specify each major component by volume percent: CO₂, O₂, N₂, H₂O, SO₂, HCl if relevant. For cement or steel applications, dust loading in g/Nm³ (dry basis, at inlet conditions) must be stated — this single figure drives the decision between water-tube bare-tube passes and convection banks with wide pitch, and it affects soot blower specification and cleaning cycle frequency. A number above roughly 30–50 g/Nm³ almost always changes the mechanical layout, and vendors cannot price this correctly if you omit it.

Required steam output in t/h, or total heat duty in kW if you are generating hot oil or hot water instead — do not leave this ambiguous. Add steam pressure (MPa, gauge) and whether you need saturated or superheated steam; if superheated, state the target temperature in °C. Feedwater inlet temperature is often forgotten and it directly affects economizer sizing and, by extension, total boiler length. Site altitude is needed for fan and stack draught calculations; a plant at 1,800 m above sea level needs materially different fan sizing than a coastal site.

Scope of Supply Battery Limits

Define where the vendor’s scope starts and stops. The flue gas inlet battery limit should state whether the diverter damper and bypass stack are included or supplied by others. On the steam side, specify whether the outlet stops at the boiler steam nozzle or extends to the main steam header, including isolation valves and instrumentation. Feedwater supply point — pressure and temperature at the boiler feed pump outlet — and the electrical supply point (voltage, frequency, whether soft-starters or VFDs are in or out of scope) should both be written down. Scope gaps here are where contract disputes start.

Certification and Inspection Requirements

State the applicable pressure vessel code upfront: ASME Section I, GB/T standards, EU PED, or whichever your end-user or insurer requires. Name the acceptable third-party inspection agencies — TÜV, Bureau Veritas, Lloyd’s, SGS are the common ones for export projects. Specify whether you require a witnessed hydrostatic test and, if so, whether your own inspector travels to the shop or the agency acts as proxy. Leaving this open leads to cost surprises late in the project.

Utility Data and Site Logistics

Cooling water supply temperature and pressure affect condenser or blowdown cooler sizing. Instrument air pressure matters for pneumatic control valve selection. These are small line items that accumulate into scope gaps if unspecified.

For shipping, state the maximum piece dimensions and weight per lift based on your port and inland transport constraints. Site cranage capacity sets the maximum single-piece weight. If your erection window is narrow — say, a four-week planned shutdown — put that in writing; it affects whether the vendor can offer a pre-assembled module or must design for field assembly.

Performance Guarantee Terms

Require vendors to state guaranteed heat recovery efficiency and the exact test conditions under which the guarantee applies. Reference the flow measurement standard (ISO 10780 pitot traverse is widely accepted; some projects use ASME PTC 4 adapted to WHRB geometry). Steam metering calibration standard should be stated. A guarantee that reads “85% thermal efficiency” without specifying the reference inlet temperature, flue gas composition, and feedwater conditions at test is essentially unenforceable.

A complete inquiry package takes an extra half-day to prepare. In practice, it cuts the clarification round-trip by two to three weeks and produces quotations you can actually score against each other — which is the whole point.

Frequently Asked Questions About Waste Heat Recovery Boiler Sizing and Selection

What is the minimum flue gas temperature at which a WHRB becomes economically viable?

Generally, you need inlet flue gas above 250°C to justify steam generation from a WHRB. Below that threshold, the heat exchanger surface area required grows disproportionately, capital cost climbs, and payback periods stretch past the point where most plant owners will approve the budget. A low-temperature hot water circuit — a plate or shell-and-tube exchanger feeding process washing water or space heating — is almost always the smarter call under 250°C.

The exception worth knowing: if your site is evaluating an Organic Rankine Cycle for power generation, inlet temperatures as low as 200°C can be viable because ORC working fluids have much lower boiling points than water. This is common in geothermal-adjacent applications and some light industrial exhaust streams, though ORC unit costs are substantially higher per installed kWe than conventional steam paths.

Can one WHRB serve multiple heat sources simultaneously?

Yes, and it is done fairly often in facilities with several furnaces or kilns operating on staggered production schedules. The arrangement uses a common flue gas header with individual isolation dampers on each source branch. The boiler must be sized for the maximum combined gas flow — not the average — because you have to handle the case where every source runs simultaneously.

The controls side is where this gets genuinely complicated. Each source may deliver flue gas at a different temperature and with different composition, so the boiler inlet condition can swing significantly as dampers open and close. A single-loop temperature controller is not adequate here; you need feedforward logic that anticipates the inlet shift before drum pressure reacts. Skimping on controls for a multi-source configuration is a reliable path to steam quality problems and accelerated thermal cycling on tubes.

How do I decide between natural circulation and forced circulation?

Natural circulation — where density difference between steam-water mixture and downcomer water drives flow — works reliably up to roughly 7 MPa and suits most industrial WHRB applications where heat flux across the tube banks is reasonably uniform. It is simpler, has no circulation pump to maintain, and tolerates brief feedwater upsets better than people expect.

Forced circulation becomes necessary above 7 MPa, where the density differential shrinks to the point that natural driving head is insufficient. You also want pump-assisted circulation any time the flue gas flow is horizontal and heat distribution across tube rows is uneven — a layout common in retrofitted systems squeezed into existing plant structures. Once-through supercritical designs are a separate category entirely and rarely appear in general industrial WHRB specifications; they belong in large combined-cycle power blocks.

What causes WHRB tube failures most frequently in practice?

Three failure modes dominate field experience. Acid dew-point corrosion on economizer tubes is the most common — it happens when exit gas temperature drops below the sulfuric or hydrochloric acid dew point, typically 130–160°C depending on fuel sulfur content and chlorine load. Keeping exit gas 15–25°C above the calculated dew point is standard practice, but it is frequently ignored during commissioning when the team is focused on maximizing heat recovery.

Oxygen pitting from inadequate feedwater deaeration is the second. A poorly maintained deaerator or a chemical dosing system that runs dry overnight will show up as pitting on economizer and evaporator tube bores within months. Third is thermal fatigue in intermittent-source applications — electric arc furnace exhaust is the classic case, where the boiler cycles from cold to full load and back every 40–60 minutes. Tube-to-header welds accumulate fatigue damage faster than in continuous-source designs, and inspection intervals should reflect that reality.

How long does engineering, manufacturing, and commissioning take for a mid-size unit?

For a shop-fabricated WHRB in the 10–30 t/h range, expect 14–22 weeks from contract signing to factory acceptance test. The range depends mainly on pressure class — a 1.6 MPa saturated unit moves faster through the shop than a 5.0 MPa superheated unit that requires more weld inspection stages — and on how quickly the buyer approves drawings.

Add 4–8 weeks for sea freight and site erection, depending on destination port and whether civil foundation work is already complete when the unit arrives. Total project schedule lands in the 18–30 week window in most cases. Projects that compress this timeline usually do so by overlapping civil work with manufacturing, which requires issuing certified foundation loads from the vendor early in the process — worth requesting explicitly at inquiry stage.

Is it possible to retrofit a superheater into an existing saturated-steam WHRB?

Technically possible, but rarely worth pursuing. A superheater needs to sit in the highest-temperature gas zone, upstream of the evaporator bank, and it requires inlet gas above roughly 450°C to generate any meaningful superheat. If the structural space exists and the inlet temperature qualifies, you still need a full thermal re-analysis of the existing evaporator and a mechanical review of the drum and header connections to handle the changed heat distribution. That engineering cost, plus the fabrication and downtime, usually makes replacement of the entire unit the more economical path. The one scenario where retrofit makes sense is when the shell and pressure parts are relatively new and the original specification simply undershooted the superheat requirement by a small margin.

What efficiency gain can I realistically expect from adding an economizer?

A well-designed economizer that pulls exit gas temperature down by 100°C — say from 300°C to 200°C — will typically increase overall heat recovery by 8–12 percentage points. The actual gain depends on gas flow rate, specific heat capacity, and the moisture content of the flue gas. That improvement translates directly into reduced steam cost per tonne, and on a 20 t/h unit running 7,000 hours per year, the fuel or recovered-heat value of that incremental efficiency is meaningful enough that economizer payback periods of 18–36 months are common even accounting for the added tube surface maintenance.

One practical caveat: if your process flue gas carries significant particulate loading, the economizer fin geometry matters. Bare tubes foul more slowly than finned tubes in high-dust streams, and the cleaning access between rows needs to be specified explicitly — not left to the vendor’s standard design.

References

1. Industrial Waste Heat Recovery Basics and Project Considerations — U.S. Department of Energy

2. How to Specify Steam Generator Capacity and Operating Conditions — Babcock & Wilcox

3. Methods for Estimating Industrial Steam Consumption — Spirax Sarco

4. Understanding Boiler Capacity and Steam Ratings — Spirax Sarco

5. Using Steam Accumulators to Handle Peak Demand — Spirax Sarco

6. Calculating Steam Consumption for Industrial Heat Exchangers — Spirax Sarco

7. Gas Turbine Exhaust and Heat Recovery Steam Generation — U.S. Department of Energy

8. Combined Heat and Power Technology Selection Resources — U.S. Environmental Protection Agency

9. HRSG Components, Modules and Engineering Considerations — Babcock & Wilcox

10. Boiler Heat Transfer, Water Quality and Operating Efficiency — Babcock & Wilcox