Knowledge / Ship systems

Exhaust economizers: heat recovery, fouling and soot fires

Read exhaust temperature, gas pressure loss, circulation and deposits together to distinguish performance loss from soot-fire mechanisms.

On this page

An exhaust economizer recovers useful heat from engine exhaust. Soot deposits can reduce heat transfer, increase gas-path resistance and create a combustible deposit. Assessment therefore extends beyond tonnes of steam produced: the exhaust side, the fluid receiving heat and the conditions under which deposits can burn must be understood together.

The name does not establish the circuit arrangement

Alfa Laval’s economizer overview describes transferring exhaust heat to water or thermal oil. Some marine arrangements include evaporating surfaces and a separate steam drum; others produce hot water or heat a thermal fluid. Similar names can therefore conceal different circulation arrangements and temperature limits. Begin with the actual flow diagram.

Heat passes from gas through the tube wall to water or another fluid. A clean gas side does not establish adequate circulation on the receiving side. Likewise, an apparently running circulation pump does not directly prove that every heated surface remains wetted and cooled as intended. Keep the physical boundary represented by each measurement explicit.

Why can fouling appear in two different measurements?

A deposit can add resistance to heat transfer. With similar gas and receiving-side conditions, less heat extraction may appear as a higher outlet gas temperature. If deposits also restrict flow area, pressure loss can rise. Temperature and pressure need not change for the same reason, however: engine load, gas flow or steam demand may have changed.

The Aalborg technical bulletin explains interactions between scrubber installation, exhaust conditions and soot accumulation, and the importance of a clean temperature/pressure baseline. The engineering implication is to compare before-and-after data at matching engine load and comparable operating conditions. Maintenance frequencies specific to the bulletin should not become universal vessel instructions.

A worked example of lost heat recovery

Assume steady exhaust flow of 8.0 kg/s, mean specific heat of 1.10 kJ/(kg·K), inlet temperature of 320 °C and outlet temperature of 210 °C. Neglecting losses for this first estimate, Q̇ = ṁ c_p ΔT = 8.0 × 1.10 × 110 = 968 kW. The temperature difference is 110 K; a Celsius temperature difference has the same numerical magnitude.

If flow, specific heat and inlet temperature remain unchanged but the outlet rises to 250 °C, calculated transfer falls to 616 kW. The difference is 352 kW, about 36.4% of the initial value. This can be consistent with fouling, but cannot be assigned uniquely to fouling without checking receiving-fluid temperature and control behaviour.

The steam side needs an enthalpy difference. For example, producing 0.40 kg/s of steam with an assumed feedwater-to-steam enthalpy rise of 2 400 kJ/kg requires 960 kW of useful transfer. Phase change cannot be represented by c_p ΔT alone. The agreement in scale between the two sides is illustrative consistency, not verification of actual instrumentation.

Pressure loss belongs to the whole exhaust route

MAN’s project-guide section for specified two-stroke engines treats economizer pressure loss together with other exhaust components. Its example limit cannot be transferred to every engine. A scrubber, silencer, spark arrester, transition pieces and duct route may all draw on the same back-pressure allowance.

If a clean differential pressure of 1.2 kPa later becomes 1.8 kPa at comparable gas flow and density, the increase is 50%. That ratio alone does not establish that an allowable limit has been exceeded. Different flow changes the comparison: where pressure loss varies approximately with velocity squared, increased engine load can raise the differential without fouling. Blocked pressure tappings can also create a misleading trend.

Loss of circulation and soot accumulation are different faults

Alfa Laval’s XWi description discusses water retention in one natural-circulation product and circulation difficulties in some forced-circulation arrangements. This manufacturer description does not make all natural-circulation equipment immune to fire. The physical question is whether enough heat can be removed from the heated surface during the relevant load change.

In a steam-generating circuit, circulation, drum level, pressure and evaporation interact. Pump current alone should be read alongside measurements that establish flow and the equipment’s specific limits. A normal value at one sensor cannot rule out local drying elsewhere without understanding the geometry and measurement positions.

A soot fire is not simply high exhaust temperature

Combustible deposits, an oxidising environment and sufficient local heating can initiate burning. Unburnt oil mixed with soot can change its ignition behaviour. Measured bulk gas temperature does not fully represent a fin tip’s metal temperature or conditions inside a thick deposit. A single universal ignition temperature cannot establish safe operation.

The Aalborg EX product description shows cleaning and firefighting arrangements as separate features of a particular thermal-fluid economizer. It does not establish a water-application or extinguishing sequence for other equipment. The actual response depends on equipment condition and the ship’s approved emergency arrangements; routine cleaning and response to an established fire are different operations.

Lower outlet temperature also has limits

The MAN V28/33D STC project guide links exhaust-path surfaces below the dew point with corrosion and deposit problems. Thus, the lowest possible outlet gas temperature is not always the best objective. A single acid-dew-point value cannot be adopted without the fuel and exhaust composition and the relevant surface temperatures.

The aim of recovering more heat must be considered with engine back pressure, surface temperature, condensation risk, steam demand and access for cleaning. A new scrubber or a different fuel does not establish that the former clean baseline remains applicable. The limits and comparison conditions deserve renewed assessment after a material change.

Records become useful when read together

Engine load and fuel condition, inlet/outlet gas temperatures, differential pressure, steam pressure and production, feedwater and circulation information form a stronger explanation when aligned in time. A return towards the expected curve after cleaning is useful evidence; it does not independently establish that every problem has been resolved.

A useful diagnosis identifies which measurement separates the candidate causes: increasing gas-side resistance, a change in receiving-fluid temperature, or a faulty measurement path. The energy balance, pressure trend and physical examination complement each other. Connecting them is more informative than asking one temperature alarm to explain every possible failure mechanism.

Sources