In the operation and maintenance of gas turbines, filter element replacement is one of the most frequent and costly operations. When the lifespan of a power plant’s filter elements suddenly drops, the problem is often no longer attributed to “poor air quality,” but rather points to a long-neglected physicochemical process— wet dust accumulation.
Stringent Air Quality Requirements of Gas Turbines
The intake air quality of a gas turbine directly determines the health of hot-channel components. Compressor blades rotate at linear velocities exceeding 300 m/s. Particles larger than 5 micrometers in diameter in the air are sufficient to cause abrasion at the leading edge of the blades; while fine particles smaller than 2 micrometers easily melt and deposit on the high-temperature blade surface, forming difficult-to-remove scale, leading to reduced flow area and decreased efficiency.
Therefore, modern gas turbine power plants generally employ a three-stage filtration system: coarse filtration (G4 standard) intercepts large particles and lint;medium-efficiency filtration (F7-F8 standard) captures medium-sized dust particles; and high-efficiency filtration (F9-H13 standard) acts as the final barrier, ensuring that the cleanliness of the air entering the compressor meets the permissible standards for industrial gas turbines. The initial design intention of this system is to allow each stage of the filter to work collaboratively, ensuring filtration efficiency while extending the lifespan of the final high-efficiency filter element through tiered interception.
Wet Dust CondensationAn Underestimated Physicochemical Process
Wet dust caking refers to the process where, when the relative humidity of the air exceeds a certain threshold (usually above 85%), dust particles intercepted by the filter element absorb moisture and undergo a physical state change. This process is not simply “dust getting wet,” but involves complex mechanisms such as capillary agglomeration, deliquescence, and interparticle bridging.
At the microscopic level, when water molecules adsorb onto the surface of dust particles to a critical value, a capillary effect forms between the particles, drawing moisture into the particle contact points. At this point, the previously loosely packed particles begin to form liquid bridges. If the dust contains soluble salts (such as sulfates and chlorides), these salts will dissolve to form an electrolyte solution, further reducing the surface tension of water and exacerbating moisture penetration. As the liquid bridges gradually solidify, the binding force between particles changes from weak van der Waals forces to stronger capillary forces, eventually forming an irreversible, hardened crust.
This process is fatal to the filtration system. The crust forms a dense “mud shell” on the filter element surface, with an air permeability resistance more than 10 times that of the clean state. Even more seriously, the pulse jet cleaning system commonly used in gas turbines—which uses compressed air to blow air from the inside of the filter element outwards—is effective for dry dust filter cakes, but for the sticky layer formed by wet dust crusts, the pulse energy often only creates a few holes on the surface, failing to achieve complete removal. The crust continues to accumulate until it completely clogs the filter element.
Typical Failure Modes: From Environmental Conditions to Filter Element Failure
A case study of a coastal power plant is typical. Located in a subtropical monsoon climate zone, the power station experiences consistently high relative humidity (above 90%) during spring and summer, and contains a certain concentration of sea salt aerosols. Initially, the power station was equipped with a standard three-stage filtration system: a G4 plate filter for coarse efficiency, an F7 bag filter for medium efficiency, and an F9 cartridge filter for high efficiency.
Initially, the differential pressure curve appeared normal. However, after the onset of the humid season, the differential pressure of the medium-efficiency filter rapidly increased from an initial 150 Pa to an alarm value of 800 Pa within two weeks. Shutdown inspection revealed that the F7 bag filter surface was covered with a layer of hard, grayish-brown crust, 5-8 mm thick, with a moisture content as high as 35%. Further analysis indicated that sea salt particles (sodium chloride) in the air deliquesced on the filter surface, absorbing moisture to form a brine solution. This solution, when mixed with industrial dust, formed a hydrated, solidified structure similar to cement.
This failure mode forces the power plant to replace a batch of medium-efficiency filter cartridges every three months, eight times the design frequency, increasing direct spare parts costs by over €400,000 per year. Furthermore, each replacement requires the gas turbine to operate at reduced load or undergo a short shutdown, resulting in power generation losses.
IV. Technical Response Path: From Material Innovation to System Restructuring
Simply increasing the filtration level or replacement frequency is not an economically effective solution to the problem of wet dust caking. TrennTech‘s filtration technology research indicates that solving this problem requires a rethinking of both material surface properties and system configuration.
At the material level, hydrophobic and oil-repellent treatment is key. By coating the surface of polyester (PET) or glass fiber filter media with a low surface energy fluorocarbon compound or polytetrafluoroethylene coating, the filter media’s ability to adsorb moisture and oil mist can be significantly reduced. The principle behind this treatment is to change the contact angle of the filter media fibers, causing water droplets to form spherical shapes on the fiber surface rather than spreading and wetting, thereby inhibiting capillary condensation. Filter cartridges treated with hydrophobic coatings can reduce the caking rate by more than 60% under the same high humidity conditions.
At the structural level, introducing composite gradient filtration is an effective strategy. For example, adding a pre-filter layer with high dirt-holding capacity at the front end of a medium-efficiency filter specifically intercepts highly hygroscopic fine particles, delaying the formation of a caking layer on the main filter element. Simultaneously, optimizing the filter element’s geometry, employing a design with shallower pleats and wider pleat spacing, reduces dust accumulation in the deeper layers of the filter element, allowing pulse backflushing energy to be more effectively transferred to the filter element surface.
An industrial filtration testing center in Ludwigshafen am Rhine, Germany, conducted a comparative experiment: under simulated high-humidity coastal conditions, an untreated F9 filter element showed significant caking after 300 hours, with the pressure differential rising to 1200 Pa; while a filter element with hydrophobic treatment and optimized pleat design, under the same conditions, maintained a stable pressure differential below 450 Pa after 1000 hours of operation, and no hard caking layer was observed.
V. Adjustment and Optimization of Operating Strategies
In addition to hardware improvements, adjusting operating strategies is equally important. During humid seasons, appropriately increasing the frequency and pressure of pulse backflushing can effectively suppress the formation of initial liquid bridges. Experimental data shows that when the pulse interval is shortened from 30 minutes to 10 minutes, the moisture content of the dust layer on the filter element surface can be reduced by 20%-30%, because the more frequent backflushing interrupts the migration and accumulation of moisture within the dust layer.
Furthermore, the application of intake air heating technology is also worth considering. Under extremely high humidity conditions, increasing the intake air temperature by 2-3°C through waste heat recovery or electric heating can reduce the relative humidity by 10-15 percentage points, reducing moisture entering the filtration system at the source. Although this consumes a small amount of energy, the overall economic benefits are often superior compared to the losses from frequent filter element replacements.
The case of frequent filter element replacements in power plants warns us that in high humidity environments, wet dust caking is a failure mechanism that cannot be ignored. It is not a simple physical clogging problem, but a complex process involving gas-solid-liquid multiphase interactions, material surface chemistry, and fluid dynamics coupling. Only by deeply understanding its microscopic mechanisms and conducting systematic reflection from the three levels of material innovation, structural design, and operation strategy can we truly solve this problem and provide reliable guarantees for the long-term stable operation of gas turbines.