Fouling is a very undesirable phenomenon in the world of heat transfer and heat exchangers. In most heat exchangers, the fluid flowing is not completely free from dirt, oil, grease and chemical or organic deposits. In all cases, an unwanted coating can collect on the heat transfer surface, decreasing the heat transfer coefficient. The thermal efficiency of the heat exchanger will be reduced and the pressure drop characteristics may change.
This section discusses several types of fouling, the reasons for their occurrence and the preventive measures that can be taken avoid them. The different types of fouling discussed are:
- Scaling
- Particulate fouling
- Biological growths
- Corrosion
Scaling
Scaling is a type of fouling caused by inorganic salts in the water circuit of the heat exchanger. It increases the pressure drop and insulates the heat transfer surface, thus preventing efficient heat transfer. It occurs at high temperatures, or when there is low fluid velocity (laminar flow) and uneven distribution of the liquid along the passages and the heat transfer surface.
The likelihood of scaling increases with increased temperature, concentration and pH. Studies have shown that high turbulence and a small hydraulic diameter, such as with SWEP brazed plate heat exchangers, have beneficial effects on this type of fouling. Proper maintenance and treatment of the cooling water, e.g. pH treatment, greatly reduce the risk of scaling, especially in cooling towers.
Most scaling is due to either calcium carbonate (lime) or calcium sulfate (gypsum) precipitation. These salts have inverted solubility curves (see Figure 8.32), i.e. the solubility in water decreases with increasing temperature.
The salts are therefore deposited on the warm surface when the cold water makes contact with it. Pure calcium sulfate is very difficult to dissolve, which makes cleaning more complicated. In general, other types of scale are more easily removed.
The most important factors that influence scaling are:
- Temperature
- Turbulence
- Velocity
- Flow distribution
- Surface finish
- Composition and concentration of the salts in the water
- Water hardness
- pH
Scaling is more likely at a high pH, so a general approach to this problem is to keep the pH between 7 and 9. The risk of scaling generally increases with increasing water temperature. Experience shows that scale is seldom found where wall temperatures are below 65°C. This implies that the temperatures are usually not high enough to lead to scaling in refrigerant condensers. This is illustrated in the two examples below.
Example 1 – condenser
A typical R22 condenser works at Tin =85°C and the condensing temperature is Tcond=40°C on the refrigerant side. The water inlet temperature is 29°C and the leaving water temperature (LWT) is 36°C.
As shown in the temperature program in Figure 8.33, the maximum leaving water temperature (LWT) is only slightly above the condensing temperature (Tcond). This is due to the water temperature at the pinch point (circled in Figure 8.33) always being lower than the refrigerant temperature (Tcond) at the pinch point. No crossing of temperatures is possible, because the temperature difference is the driving force of the heat transfer (see chapter 1.1). Furthermore, the amount of heat transferred from the refrigerant gas to the water is relatively small. The discussion above shows that the temperatures in the bulk are not high enough to lead to scaling.
Figure 8.34 shows an example of the temperatures inside a condenser. Although the bulk gas temperature is as high as 85°C on the refrigerant side, the wall temperatures are determined by the bulk water temperature (36°C). This is because the film coefficient is much higher on the water side than on the refrigerant side. The maximum wall temperature is therefore 38°C on the water side and 38.6°C on the gas side, still below temperatures where scaling is a problem.
Example 2 – desuperheater/heat recovery
A typical R22 desuperheater works at Tin=85°C and Tout =45°C on the refrigerant side. No condensing occurs. The water inlet temperature is 10°C and the leaving water temperature is 50°C.
As can be seen in Figure 8.35, in this case there is no temperature pinch to prevent the leaving water temperature from rising. Nevertheless, there is no risk of scaling in this case because the LWT is designed to be only 50°C.
Figure 8.36 shows that although the gas temperature is high at the inlet (85°C), the maximum wall temperature will not be higher than 51°C, i.e. normally no risk of scaling. The reason for the relatively low wall temperature is the low heat transfer coefficient (αgas) on the gas side compared with the water side (αwater).
However, it is important in desuperheating/heat recovery applications to have a high constant flow on the water side. If the water flow is reduced or turned off, the temperature will rise and there will be a risk of scaling.
Types of scale
Calcium carbonate (CaCO3) can be formed when calcium or bicarbonate alkalinity (HCO3 -, CO3 2- and OH- ions in the water) is present. An increase in heat and/or an increase in pH will cause precipitation of calcium carbonate according to the equation:
Calcium sulfate (CaSO4) is 50 times more soluble than calcium carbonate and will therefore precipitate only after calcium carbonate has been formed. This type of scale can exist in various forms, and its formation depends strongly on the temperature. An increase in temperature decreases the solubility of this salt and increases the risk of scaling.
Water scaling tendency
In order to estimate the scaling tendency of natural water, several parameters must be analyzed and determined:
- pH
- Calcium content
- Alkalinity
- Ionic strength of the water
The first three parameters are relatively straightforward to determine. However, the ionic strength depends on the total amount of dissolved, dissociated compounds, i.e. salts and acids, as well as the relative concentrations of the various salts and acids.
The Langlier saturation index, Is, is calculated from the amount of total dissolved solids (TDS), calcium concentration, total alkalinity, pH and solution temperature. It shows the tendency of a water solution to precipitate or dissolve calcium carbonate. In this method, the pHs (the pH at the equilibrium state) is calculated from the total salt content (pS), the alkalinity (pAlk) and the calcium content (pCa). The pHs is then compared with the actual pH for the water, giving the Langlier Index, Is:
where
The pH measurement is straightforward and is performed routinely. Because the pH may vary with the season and the climatic conditions, it should be measured on several different occasions. The calcium content, pCa, is normally expressed as the concentration of calcium either as calcium carbonate (CaCO3) or as calcium ion (Ca2+). The bicarbonate alkalinity, pAlk, can be determined by titrating the water with an acid and a suitable indicator (e.g. methyl orange). The result is expressed in various ways, e.g. as the equivalent CaCO3. The corresponding pAlk is obtained from the Langlier diagram (see Figure 8.37). The relative proportions of the various salts are fairly constant in naturally occurring water. Langlier uses the total salt content (mg/l), i.e. TDS (Total Dissolved Solids), as a measurement of the ionic strength. The corresponding amount of total solids, pS is obtained from the Langlier diagram (see Figure 8.37). All these measurements can be obtained from a general water analysis.
Please note:
- For water analysis, mg/l is equal to ppm
- The relationship between calcium and calcium carbonate is:
- TDS = Salt content (mg/l), or possibly the conductivity x 0.63 (μS/cm).
If Is is negative, the water has a tendency to be corrosive. This corrosivity is valid for carbon steel and, to a lesser extent, copper, but not for 316 stainless steel. If Is is positive, the water has a tendency to cause scaling.
Figure 8.37
Example of the use of the Langlier diagram
When analyzing the water sample, the following values were obtained:
Using the results from the water analysis above, the pCa, pAlk and pS can be found in the Langlier diagram as follows:
pCa
On the Langlier diagram (Figures 8.37 and 8.38), note the measured Ca concentration of 120 mg/I CaCO3 (or 120 ppm). Read off the pCa value at the point where the calcium concentration meets the diagonal line for Ca/pCa. This gives pCa=2.92.
pAlk
On the Langlier diagram (Figure 8.37 and Figure 8.38), note the measured alkalinity value of 100 mg/l CaCO3 (or 100 ppm). Read off the pAlk value at the point where the alkalinity value meets the diagonal line for CaCO3/pAlk. This gives pAlk=2.70.
pS
On the Langlier diagram (Figure 8.37 and Figure 8.39), note the measured TDS concentration of 210 mg/I. Read off the pS value at the point where the TDS concentration meets the temperature line (in this case 49°C). This gives pS=1.70.
Using the results extracted from the Langlier diagram, pHs can be calculated, and then the saturation index Is:
Because Is>0, the water in this example has a tendency to cause scaling.
Determine whether scaling has occurred
To be able to clean the heat exchanger unit easily, it is important to note the signs of scaling before the unit is completely clogged. This can be done by measuring the inlet and outlet temperatures of the heat exchanger, which indicate whether fouling has occurred. Fouling of the heat transfer surface decreases the heat transfer, resulting in a temperature difference smaller than specified. Another way to detect fouling is by measuring the pressure drop over the heat exchanger. Because fouling restricts the passages, and thus increases the velocity, the pressure drop will increase. When using this method, make sure that the water flow rate is as specified, because changes in the flow rate will of course also affect the temperature change and the pressure drop.
Prevention of Scaling
The formation of calcium carbonate can be controlled by adding acids or specific chemicals (phosphate compounds, e.g. AMP, or organic polymers, e.g. polyacrylates) tailored to inhibit the precipitation of the compound.
However, water treatment is not an easy task, and a water specialist should be consulted in order to determine the correct treatment. Improper use of acids can cause severe corrosion of the brazed plate heat exchanger in a very short time. Calcium sulfate scaling can be controlled most effectively with chemicals such as polyacrylates or AMP.
Removing the scale that has been formed restores the operating efficiencies of the equipment and the heat transfer surfaces. Other benefits from removing the scale are that it lowers the pressure drops, reduces the power consumption and extends the lifetime of the equipment.
Particulate fouling
Particulate fouling is caused by suspended solids (foulants) such as mud, silt, sand or other particles in the heat transfer medium. Important factors that affect particulate fouling are:
- velocity
- distribution of the flow
- roughness of the heat transfer surface
- size of the particles
Velocity
The velocity is an important factor in the sense that it controls whether the flow is turbulent or laminar. Turbulent flow is desirable for several reasons. Turbulent flow will keep particles in the fluid in suspension, i.e. no particles are allowed to collect on the surface, which will avoid surface fouling. Another very important reason, of course, is that turbulent flow improves the heat transfer.
Brazed plate heat exchangers have a high degree of turbulence, and the fluid has a scouring action that keeps the heat transfer surface clean. This is due to the unique design of brazed plate heat exchangers. As the fluid passes through the channels, it constantly changes its direction and velocity. This ensures turbulent flow even at very low flow rates and pressure drops.
For shell and tube (S&T) heat exchangers, a much higher velocity is required to reach turbulent flow.
In an S&T, the water can flow either inside the tubes or outside the tubes. When the water passes through a tube, the maximum velocity is at the center of the tube. The turbulence at the walls is too low to keep particles in the fluid in suspension. These particles are allowed to precipitate and collect on the tube wall, which causes fouling of the heat transfer surface. When the water flows outside the tubes, the flow rate is lower and low-flow areas are created, which increases the risk of fouling. This means that S&T heat exchangers are much more sensitive to fouling than plate heat exchangers. When designing S&T heat exchangers, the use of so-called fouling factors is recommended to account for the risk of fouling and the consequent decrease in performance.
Distribution of the flow
It is of great importance for the flow over the heat transfer surface to be well distributed to maintain uniform velocity. The flow distribution depends very much on the plate pattern. A special distribution pattern in the port areas of SWEP brazed plate heat exchangers ensures a well-distributed flow. In other heat exchangers (S&T, coaxial and other brazed heat exchanger brands), there can be areas of low velocity (resulting in laminar flow) due to uneven distribution of the fluid through the exchanger. These sections are sensitive to fouling. The fouling starts at the low velocity areas and spreads over the heat transfer surface.
Roughness of the heat transfer surface
Rough surfaces are known to encourage fouling by collecting particulate matter. The material used in every SWEP brazed plate heat exchanger is AISI-316 stainless steel, and the smooth surface of this material minimizes fouling. The round shape of the brazing points ensures that no pockets of stagnant water can be formed.
In applications where a cooling tower or other open system is used, the cooling water will be rich in oxygen. This can cause the corrosion of materials such as the carbon steel used in conventional heat exchangers. This corrosion is usually in the form of iron oxide scale on the carbon steel surface, but loose iron oxide can be deposited elsewhere as well. The stainless steel used in the SWEP brazed plate heat exchanger is not subject to the uniform corrosion that causes fouling problems. However, SWEP brazed plate heat exchangers are not completely immune to corrosion under certain conditions.
The size of the particles
Particulate fouling can influence the performance of the heat exchanger in two ways, depending on the particle size. First, if the particles are large (>1 mm) or have a fibrous structure, they may lodge in the inlet of the heat exchanger or clog the channels. The result is an increased pressure drop in the water circuit of the heat exchanger. Clogged channels also mean low water velocities, which can result in freezing when using the brazed plate heat exchanger as an evaporator. Second, particles may adhere to the heat transfer surface and build up a layer of low thermal conductivity material. Initially, this leads to reduced heat transfer, and a higher pressure drop in severe cases of fouling.
Prevention of Particulate Fouling
Clean cooling water
The best way to avoid particulate fouling is to keep the cooling water clean and thereby prevent particles from entering the heat exchanger. However, in all cooling systems, and especially when using open cooling systems (with cooling towers), there will always be particles present in the cooling water. The correct maintenance of cooling towers will dramatically reduce the risk of fouling, including particulate fouling, scaling and corrosion. The evaporation of water in cooling towers is unavoidable, and they must be re-filled with make-up water. However, it is very important to bleed (discharge) water from the tower, otherwise impurities will accumulate and soon reach dangerous concentrations. This bleed water is called blowdown.
Strainer
A strainer is recommended before the inlet of the heat exchanger. A strainer will prevent large particles (>1 mm) from entering the heat exchanger.
The recommended size of strainer for this purpose is 16-20 mesh or a mesh size of 0.5 to 1 mm. If a smaller mesh size is used, this will of course result in better filtration of the water, but the system will also need more frequent cleaning. It also creates an unwanted pressure drop.
Side-stream filtration
When the make-up water for the cooling tower contains significant suspended matter, it is advantageous to use side-stream filtration. A filtration unit (several types are available) is connected to the cooling tower basin. Water from the basin is then pumped through the filtration unit and back again. In general, passing a few percent of the re-circulating water through the side-stream filter will reduce the suspended solids by 80-90%.
Adequate flow rates
High flow rates will keep particles in suspension and prevent them from depositing on the heat transfer surface.
Chemical water treatment
Chemical treatment of water can also be an effective method of controlling suspended particulates. As in the prevention of scaling, polyacrylates disperse foulants (suspended solids) very efficiently. Concentrations of a few milligrams per liter are required in open re-circulating systems.
Biological Growths
Fouling through biological growths (also called biofouling) occurs when living matter grows on the heat transfer surface. In many cases, re-circulating cooling systems are ideal for promoting the life of microorganisms.
Three types of living organisms are considered here: algae, fungi and bacteria.
- Algae are easily detected by their green color. They need oxygen and sunlight to grow, and they can therefore exist in cooling towers. In addition to reducing the thermal performance, algae can also have a severe impact on metal corrosion by providing conditions that increase the risk of corrosion.
- Fungi are similar to algae but do not require sunlight. They require moisture and air and exist on nutrients found in water or on the material they are attached to, for example bacteria, algae or wood.
- Bacteria can live with or without oxygen. Water and other wet environments with organic content are suitable for the growth of bacteria. Heat exchangers can therefore provide an excellent environment for this type of micro-organism, which will reduce the heat transfer. Bacteria can also initiate pitting corrosion.
Prevention of Biological Growths
Biocides are the most practical and efficient method of controlling the growth of micro-organisms in cooling water systems. Biocides kill or inhibit the growth of the organisms. Although these chemicals inhibit biofouling, they will not remove material already adhering to surfaces. This emphasizes the importance of a clean system from the start. A number of methods and chemicals are available.
Chlorination
Chlorine (Cl2) is an excellent bactericide and algaecide. Chlorination can be either continuous or of the shock type. When adding chlorine continuously, levels of 0.1 to 0.2 mg/l (ppm) are recommended. For the best biocide effect, the pH should be between 6.5 and 7.5. A lower pH will accelerate corrosion, while a higher pH will have less impact on bioorganisms.
The shock chlorination method uses a chlorine concentration approximately 10 times higher, but only for a few brief periods every day. The advantage of this method is lower chlorine consumption.
Because chlorine decomposes into chloride (Cl-) ions, there is a risk of pitting corrosion of the 316 stainless steel used in brazed plate heat exchangers. For this reason, it is very important to bleed off cooling water from the cooling tower to avoid the accumulation of chloride ions and dangerous chloride concentrations.
Due to the risk of corrosion, chlorine should always be added as far from the heat exchanger as possible.
Fouling due to corrosion
In some cases, fouling can be due to corrosion. The added layer of corrosion products on the heat transfer surface will reduce the heat transfer efficiency. The degree of corrosion depends very much on the water quality.
Prevention of corrosion
The main fouling risk is due to corrosion products from other parts of the system. These particles will be carried by the water and may adhere to the heat transfer surface or lodge inside the heat exchanger. This type of fouling should be considered as particulate fouling, and prevented as for particulate fouling.
Fouling Resistance due to SWEP's brazed plate heat exchanger design
SWEP's brazed plate heat exchangers provide good resistance against surface fouling for several reasons. The unique design of a SWEP brazed plate heat exchanger allows the heat exchanger to operate at extremely low velocities while maintaining a turbulent flow. As the fluid passes through the channel, its direction constantly changes, which disturbs the boundary layer and ensures turbulent flow even at extremely low velocities. A SWEP brazed plate heat exchanger is actually much less prone to fouling than other heat exchangers. This is because of its internal geometry (which ensures evenly distributed fluid), the higher turbulence and the hardness and smoothness of the stainless steel in the channel plates. With laminar flow, the velocity of the fluid close to the plate surface is very low, which means that the suspended particles are allowed to settle (cf. Figure 8.40).
- The smooth surface of the channel plate material has a positive effect on minimizing fouling. Rough surfaces are known to encourage fouling, because they collect particulate matter by giving it a chance to adhere to the surface.
- The design of SWEP brazed plate heat exchangers ensures that no dead zones (where fouling compounds can settle) are created. In a dead zone, the liquid is stagnant and the suspended material has the chance to settle and accumulate.
- The particles are kept in suspension by the very high turbulence, even at low flow rates, caused by the corrugations in the plates. Turbulent flow and a small hydraulic diameter, such as with SWEP brazed plate heat exchangers, are important to prevent the suspended particles from settling. With laminar flow, the particles have a much higher tendency to settle.
Optimization of factors that affect surfaces under various conditions
Some factors that affect the surfaces of a heat exchanger are discussed below:
- Use the highest possible water pressure drop. A high pressure-drop implies higher shear stresses, and large shear stresses are always beneficial if there is any scale. The shear stresses work as descalers by constantly imposing forces on the adhered material that pull the particulate material away from the surface. The shear stresses also help keep the particles in suspension (see Figure 8.41).
- For a heat exchanger with a temperature above 70°C on the hot side and/or very hard water (and hence a danger of scaling), the pressure drop should be increased as much as possible on the cold water side and reduced on the hot side. This reduces the wall temperature on the cooling water side and increases the shear stresses, thus making it more difficult for the scaling compounds to adhere.
- Consider the use of co-current instead of counter-current flow. The warmest part of the hot side, the inlet, will then face the coldest part of the cold side. This normally decreases the maximum wall temperature on the cooling water side, which automatically limits the outlet water temperature.
- The normal practice is to let the cold water enter the lower port. This arrangement should be used whenever possible, because if the cold water enters through the upper port, it could encourage debris to enter the channels.