Experimental evaluation of air-side particulate fouling performance of heat exchangers.
In nearly all applications, heat exchangers are subject to fouling by undesirable formation of inorganic and/or organic deposits on surfaces over time. The deposits can reduce the heat transfer efficiency, increase the pressure drop across a heat exchanger, and correspondingly, increase energy consumption, resulting in extra maintenance and labor costs. Several primary modes of fouling have been observed in heat exchangers (Characklis et al., 1981), including: particulate fouling, crystalline or precipitation fouling, chemical reaction fouling, corrosion fouling, and biological fouling or biofouling. Air-side particulate fouling is a very common phenomenon occurring in applications where heat exchangers are installed in dusty environments such as HVAC systems in buildings exposed to natural dust storms and the heating and cooling systems of heavy-duty off-road vehicles and farming machinery.
The mechanisms that influence particle deposition on heat transfer surfaces include: impaction, gravitational settling, Brownian diffusion, electrostatic attraction, interception, diffusiophoresis, thermophoresis, and adhesion through van der Waals forces. Researchers have developed some theoretical models, based on one or more of the mechanisms, to predict the particulate fouling rates and performance impacts on heat exchangers (Bott, 1988, 1995; Marner, 1990; Epstein, 1999; Siegel and Nazaroff, 2002; Brahim et al., 2003). Siegel and Carey (2001) developed a particle deposition model and performed experimental tests using monodisperse particles. Their model suggests that the bulk of the HVAC coil fouling is due to particle sizes from 10 to 100 [mu]m. Another study by Siegel et al. (2002) concluded that the typical HVAC evaporator coils foul at a rate that can double the air-side pressure drop within 7.5 years. These studies show that developing an accurate predictive model for heat exchanger performance degradation due to fouling remains a challenge due to the complex nature of the fouling mechanisms and their direct relationship to the heat exchanger geometry. Therefore, empirical tests of heat exchangers in dusty environments are vital for understanding and evaluating their fouling performance in specific applications.
The purpose of this study was to develop a new protocol including reference test dusts and test procedures for experimental evaluation of heat exchanger performance in the presence of particulate fouling. This experimental protocol can serve as the basis for further investigations of particulate fouling mechanisms in typical fin configurations, which will facilitate the development of anti-fouling technologies and optimization of fin geometries for dusty environments. Preliminary evaluation results of four heat exchangers are presented based on the developed reference dusts and test procedures to demonstrate the experimental evaluation process for particulate fouling performance in both heating and cooling coils.
Actual Fouling Materials
Actual fouling materials encountered by heat exchangers vary with application and operating environment. Building HVAC heating and cooling coils mainly encounter indoor dust composed of small particulates and fibers ranging in size on the order of 0.01 to 100 [mu]m (Zhang, 2005), while radiators used in an agricultural combine mainly encounter organic debris and natural dust from loose soil ranging in size on the order of 0.01 [mu]m to several mm. For some heat exchangers, very different dusts may be encountered during their lives; e.g., a heat exchanger installed in a vehicle that may travel to many different geographical regions. Ahn and Lee (2005) collected some fouling materials from evaporators and condensers in the air-conditioning systems of an office space, a restaurant, and a seaside inn for a comparative compositional analysis. The fouling materials that they collected in the evaporator consisted of a combination of particulates and fibers, while the fouling materials in the condenser were primarily particulates. The average aerodynamic diameters of the fouling materials were 6.6 [mu]m for the office, 9.8 [mu]m for the restaurant, and 20.9 [mu]m for the seaside inn.
Fouling Test Dust Selection Criteria
Particle size, shape, density, and adhesion affect the deposition rates of dust on the surfaces of a heat exchanger. Therefore, it is very important to choose the right test dust to experimentally characterize the fouling performance of a heat exchanger with a high correlation to its actual fouling performance in situ. Ideally, the test dust should be the same as that a heat exchanger encounters in the actual application. However, this kind of dust is very difficult and expensive to re-create in bulk for repeated accelerated fouling tests. Instead, a more practical and feasible approach is to identify one or several reference test dusts that can be used in fouling tests of different heat exchangers with respect to the type of application.
The reference test dust(s) should be easily made, cost-effective, and have the similar statistical characterization to the actual dusts in terms of particle density, size distribution, and adhesion factors. Because small variations in the particle size distribution may cause a significant effect on the fouling test results, the bulk test dusts must be consistent in terms of density, size distribution, texture, and quality in order to ensure repeatable results.
Reference Test Dust Selection
There are several types of commercially available reference test dusts, including: Arizona road dust (ARD), ASHRAE synthetic test dust, and MIL STD 810 test dust. Arizona road dusts are appropriate for fouling tests of heater cores installed in vehicles due to the particles' properties; ASHRAE synthetic test dust is appropriate for fouling tests of air coils used in indoor HVAC systems due to the higher content of fibers. MIL STD 810 test dust is used to help evaluate the ability of a material to resist the effects of dust that may obstruct openings and penetrate into cracks, crevices, bearings, and joints and to evaluate the effectiveness of filters. These standard test particles are too expensive to use for an array of coil/core fouling tests since a large of number of test particles is needed in each run of the tests.
Some other powder materials, including commercial cement and limestone powders, wheat flour and soy meal, sawdust, and BEE test dust (Zhang et al., 2005), have a much lower cost than the standardized test dusts mentioned above. These alternatives were considered potential candidates for the reference test dust(s) for heat exchanger fouling in this study. BEE test dust was previously developed and used in the Bioenvironmental Engineering (BEE) Lab at the University of Illinois at Urbana-Champaign. The properties of these materials were characterized in the BEE lab in terms of particle size distributions and particle density. The mass median diameter of the BEE test dust is 1067 [mu]m, and its bulk density is 400 kg/ [m.sup.3] (25.0 lb/f[t.sup.3]).
Among all the reference test dust candidates, Masons Hydrated limestone powder (type S) covers the particle size range of Arizona road dust and ASHRAE test dust, and it has a lower price of about $0.33/kg (~$0.15/lb) compared with Arizona road test dusts at $13, ~$88/kg ($6, ~$40/lb), ASHRAE synthetic test dust at about $20/kg (~ $9/lb), or MIL STD 810 test dust at about $15/kg (~$7/lb). The particle density of the limestone powder is 2344 kg/[m.sup.3] (146.3 lb/f[t.sup.3]), comparable to that of Arizona test dust, which is 2650 kg/[m.sup.3] (165.4 lb/f[t.sup.3]). Figure 1 shows the comparison of the particle size distributions between Masons Hydrated limestone powder (type S) and Arizona road test dusts. It can be seen that Masons Hydrated limestone powder is an appropriate substitute for Arizona road test dusts for heat exchanger fouling tests. If mixed with a specific kind and number of fibers like milled cotton linters, Masons Hydrated limestone powder (type S) can also serve as the reference test dust, like ASHRAE synthetic test dust, for fouling tests of coils working in the ambient air with significant number of fibers.
[FIGURE 1 OMITTED]
BEE test dust is appropriate for fouling tests of heat exchangers operating in the agricultural and off-road environments in terms of both particle properties and cost-effectiveness.
Our recommendation for the reference test dust selection is:
1. Masons Hydrated limestone powder (type S) for commonly used heat exchangers;
2. BEE test dust (ground oats) for heat exchangers installed in agricultural machines.
Figure 2 shows the schematic of the test facility. The airflow rate range of the chamber was from 0.14 to 1.89 [m.sup.3]/s (300-4000 cfm). The parameters sampled during tests included: upstream and downstream air temperatures, moisture content, liquid temperatures at the inlet and outlet, airflow rate, liquid flow rate, and air-side pressure drop across the tested heat exchanger. All the signals were logged automatically with a computer. The thermocouple array downstream of the heat exchanger included 25 type T thermocouples. The upstream air temperature and the liquid temperatures at the inlet and outlet were also measured with type T thermocouples. Two HVAC style low differential pressure transmitters were used to measure the air-side pressure drop across the tested heat exchangers and flow nozzles. The measurement range of the pressure transmitters was 0-5 in. [H.sub.2]O (0-1250 Pa) with an accuracy of [+ or -]1%. The temperature distributions on the front surface of a tested heat exchanger were also imaged with an infrared camera just before and after the dust-feeding process during the tests. Ambient conditions such as the barometric pressure, ambient air temperature and relative humidity, dust mass fed into the system, and mass deposited on the heat exchanger were also recorded.
[FIGURE 2 OMITTED]
The operating procedure of the fouling tests was similar to that used by Yang et al. (2007) and Mason et al. (2006) and is listed as follows:
1. Weigh the clean heat exchanger sample unit. Clean and install the test equipment and the test heat exchanger to guarantee the test conditions are consistent for each test run;
2. Set the test condition parameters: air inlet temperature and relative humidity, airflow rate or face velocity, water/coolant inlet temperature, water/coolant flow rate, dust suspension density or dust-feeding rate;
3. Select the test mode: constant airflow rate or constant fan speed. In this study the constant airflow rate mode was used;
4. Start running the test system. After the stable required test conditions are achieved, start to feed the test dust to the airflow;
5. After the test period of 4 hours or the point at which the air-side pressure drop has doubled, stop feeding dust and shut down the test system;
6. Remove and weigh the heat exchanger to obtain the total mass of dust collected. Then clean the heat exchanger to return it to its original condition.
A series of fouling tests was conducted with the test system and the selected reference test dusts (Masons Hydrated limestone powder and BEE dust). The heat exchanger samples used in the preliminary tests are given in Table 1.
Table 1. Heat Exchanger Core Samples Tested Heat Exchanger Configuration Air-Side Heat Transfer ID Area HX1 (8F) 3 row round tubes and 8 2.340 [m.sup.2] (25.19 flat plate fins/in. f[t.sup.2]) HX2 (12F) 3 row round tubes and 12 3.510 [m.sup.2] (37.78 flat plate fins/in. f[t.sup.2]) HX3 (15W) 3 row round tubes and 15 4.387 [m.sup.2] (47.22 wave plate fins/in. f[t.sup.2]) HX4 (18W) 3 row round tubes and 18 5.265 [m.sup.2] (56.67 wave plate fins/in. f[t.sup.2])
The working faces of all the heat exchanger samples were 0.254 x 0.254 m (10.0 x 10.0 in.). Each heat exchanger was tested under two different applications (heating or cooling) with two different test dusts (Masons Hydrated limestone powder or BEE test dust), respectively. Thus, there were 16 experimental cases in total. The face velocity of air was 3.0 m/ s (590 fpm), which corresponded with an airflow rate of 0.194 [m.sup.3]/s (410 cfm). During the test, the airflow rate was kept constant and recorded. The dust feed rate was 11.6 g/min (179 grains/min) to keep the airborne dust concentration inside the test chamber upstream of the tested coil at 1.0 g/[m.sup.3] (0.001 oz/ f[t.sup.3]). The flow rate of the liquid (mixture of 50% water and 50% glycol) was kept at 5.7 L/min (1.5 G/min). The liquid temperature at the inlet of the tested heat exchangers was 75[degrees]C (167[degrees]F) when used for heating and 10[degrees]C (50[degrees]F) for cooling. The test period for each fouling test was 4 hours or to the point where the pressure drop across the tested heat exchanger increased to twice of the initial value (whichever occurred first).
PERFORMACE METRICS OF HEAT EXCHANGERS
In this study, the following metrics were used to examine the performance of the tested heat exchanger samples:
1. Air-side pressure drop with respect to time during dust exposure;
2. Test fouling time: the duration of the fouling test until the air-side pressure drop became twice as large as the initial pressure drop at the clean conditions;
3. Heat transfer effectiveness with respect to time during dust exposure;
4. Overall heat transfer coefficient with time during dust exposure.
The heat transfer effectiveness was calculated as follows:
[epsilon] = q/[q.sub.max] (1)
where q and [q.sub.max] are the actual and maximum possible heat transfer rates, respectively.
Given the test system settings in this study, air had the smaller heat capacity rate and was used for the calculation of the maximum possible heat transfer rate. When the heat exchangers were used in a cooling mode, condensation of moisture from the air was observed, and the moist air's heat capacity was still smaller than that of the coolant. To simplify the calculation of the heat transfer effectiveness for cooling cases, the actual and the maximum possible heat transfer rates in equation (1) were calculated with the following equations:
q = [m.sub.coolant][C.sub.pcoolant]([T.sub.coolant,out]-[T.sub.coolant,in]) (2)
[q.sub.max] = [m.sub.air][C.sub.pair]([T.sub.air,in]-[T.sub.coolant,in]) (3)
where [m.sub.coolant] and [C.sub.pcoolant] are the mass flow rate and specific heat capacity of the coolant (mixture of 50% water and 50% glycol), respectively; [m.sub.air] and [C.sub.pair] are the mass flow rate and specific heat capacity of the air (water vapor in the air was ignored), respectively.
The overall heat transfer coefficient, U, was calculated as follows:
U = q/A[[increment of T].sub.m] (4)
where A is the heat transfer surface area and [[increment of T].sub.m] is the logarithmic mean temperature difference.
RESULTS AND DISCUSSION
Heating While Exposed to Aerosolized Limestone Powder
Figure 3 shows the air-side pressure drop across the four heat exchangers in a heating mode with limestone powder as the reference test dust.
[FIGURE 3 OMITTED]
The air-side pressure drop across all the four heating coils increased slowly over time due to dust exposure and particle deposition. Table 2 shows the average initial pressure drop, final pressure drop, and the dust mass retained in the heat exchangers after four-hour tests. It can be seen that the initial air-side pressure drop increased significantly with respect to increasing fin density. This effect is especially pronounced for heat exchangers with wave plate fin shapes, where pressure drop increased at a larger rate with respect to dust exposure due to their higher fin densities. Overall, the increases observed in the air-side flow resistance were small and may not significantly affect the fan power required to operate these heat exchangers in a heating mode while exposed to dust similar to limestone in particle size, distribution, density, and shape.
Table 2. Results from Four-Hour Fouling Tests of Heat Exchangers Used in Heating and Exposed to Limestone Powder Heat Initial Pressure Final Relative Dust Exchanger Drop, Pa Pressure Pressure Deposited ID (in.[H.sub.2]O) Drop, Pa (in. Drop in HX, g [H.sub.2]O) Increase (oz) HX1 (8F) 42.8 (0.171) 43.7 (0.175) 2.1% 90.7 (3.19) HX2 (12F) 51.7 (0.207) 52.8 (0.211) 2.1% 75.9 (2.67) HX3 (15W) 105.7 (0.423) 110.4 4.4% 74.9 (0.442) (2.64) HX4 (18W) 128.3 (0.513) 136.9 6.7% 84.8 (0.548) (2.99)
Figure 4 shows the heat transfer effectiveness and the overall heat transfer coefficients of the four heat exchangers when used for heating and exposed to aerosolized limestone powder. The heat transfer effectiveness of all the coils decreased slowly with respect to the dust exposure time. The heat transfer effectiveness of a heat exchanger with higher fin density decreased at a slower rate than those with a lower fin density. In contradiction with the pressure drop results above, this indicates that heat exchangers with higher fin densities perform better with respect to heat transfer in the presence of dust deposition. Therefore, higher fin densities are more resistant to fouling from a heat transfer standpoint but lead to a trade-off in terms of clogging and increased rate of added air-side pressure drop. Insight into this result can be gained by comparing the ratio of heat transfer surface area to dust mass collected for the fixed exposure time. For heat exchangers HX1 and HX2 (8 and 12 fins per inch, respectively) the ratio of dust mass to fin area was 38.76 g/[m.sup.2] (0.1270 oz/f[t.sup.2]) and 21.62 g/[m.sup.2] (0.0708 oz/f[t.sup.2]), while for HX3 and HX4 (15 and 18 fins per inch, respectively) the ratio was 17.06 g/[m.sup.2] (0.0559 oz/f[t.sup.2]) and 16.12 g/[m.sup.2] (0.0528 oz/f[t.sup.2]). Thus the heat exchangers with a higher fin density fouled at a slower rate because each unit of surface area contained less deposited dust than those with a lower fin density. This is supported by the relative performance of the four heat exchangers shown in Figure 4, where HX1 had the greatest reduction in heat transfer effectiveness, followed by HX2, HX3, and HX4, in order of their respective ratios of dust collected to fin area.
The overall heat transfer coefficients have similar trends with the testing time to the heat transfer effectiveness. The overall heat transfer coefficient of HX4 (18W) had no obvious decrease with time. With the same airflow rate, the air velocity through HX4 (18W) was the highest and so was the turbulent intensity, which might be the reason to keep the overall heat transfer coefficient of HX4 (18W) consistent. For the other three heat exchangers, their overall heat transfer coefficients decrease with dust feeding time; i.e., their fouling resistance increases with time.
Photographic and thermal images were taken of the heat exchanger's upstream and downstream faces before and after dust exposure tests for visual examination of the fouling performance (Zhang et al., 2010). Only two thermal images (Figure 5) are shown in this paper for brevity. The aerosolized limestone powder did not significantly influence the face temperature distribution for all the four heating coils tested, an example of which is shown in Figure 6 for HX1. After each four-hour test, more deposited dust was observed on the down-stream face of each heat exchanger than on the upstream face. Thus, the thickness of the dust layer increased through the depth of the heat exchanger.
[FIGURE 5 OMITTED]
Overall, fouling is not a very serious problem for these heat exchangers when they are used for heating in environments containing dust similar to limestone powder. The impact from fouling in this scenario is only seen as slight increases in pressure drop on the order of about 5% from the initial value and minor reduction in overall heat transfer coefficients on the order of about 2%.
Heating While Exposed to BEE Test Dust
Figure 6 shows the air-side pressure drop across the four heat exchangers operating as heating coils with the BEE dust as the reference challenge dust. It can be seen that the pressure drop across each of the four heating coils increased rapidly, and again the higher the fin densities of HX3 and HX4 caused a steep increase in pressure drop with respect to duration of dust exposure. Table 3 lists the time it took to clog each heat exchanger to the point of doubling its initial pressure drop along with the respective mass of dust retained in each core. This indicates that a heat exchanger with a higher fin density will have a higher initial air pressure drop and will clog sooner with the BEE test dust, even though it will retain less dust.
Table 3. Time to Double the Initial Air-side Flow Resistance with Bee Dust and Mass of Dust Retained in the Heat Exchangers in Heating Mode Heat Exchanger Time to Double Pressure Dust Retained, g ID Drop, min (oz) HX1 (8F) 76.8 154.6 (5.44) HX2 (12F) 31.5 135.4 (4.77) HX3 (15W) 16.3 93.3 (3.28) HX4 (18W) 11.4 63.8 (2.25)
[FIGURE 6 OMITTED]
Figure 7 shows the heat transfer effectiveness and the overall heat transfer coefficients of the four heating coils that were exposed to the BEE test dust. The heat transfer effectiveness of HX1 (8F) and HX2 (12F) coils decreased slowly at first, and then became stable with continued dust feeding. In the other two heating coils with higher density wave plate fins, the heat transfer effectiveness appeared to increase steadily with the dust feeding time before clogging (clogging was defined as the point where the air pressure drop became twice the initial pressure drop). The phenomenon that the heat transfer effectiveness did not decrease much or even increased slightly with fouling might be explained in the following way: because a variable frequency controller was used to maintain a set airflow rate through the heat exchanger, the fouling only increased the air-side pressure drop but did not decrease the airflow rate. Thus, the air velocity increased through the channels of the HX core as the face area was reduced by clogging. This made the air flowing through the clogged channels more turbulent, which potentially increased the heat transfer, effectively offsetting some of the adverse effects caused by fouling. In most applications where a constant-speed fan is used, the airflow rate will decrease significantly as the pressure drop increases, which will then decrease the heat transfer performance of the heat exchanger.
As shown in Figure 7b, the overall heat transfer coefficients of HX1 (8F) and HX2 (12F) decreased at first, reaching their lowest at about 37% of their respective clogging times. The overall heat transfer coefficients increased slightly with the dust exposure time. For HX3 (15W) and HX4 (18W), both of the overall heat transfer coefficients increased with the dust exposure time.
The BEE dust made the face temperature distributions of all four heating coils significantly uneven. On the upstream faces, HX2 (12F) held more BEE dust and had a larger area of relatively cold spots than HX1 (8F) as shown in Figure 5b. More dust can be observed on the downstream side of HX1 (8F), and HX1 (8F) trapped more dust internally than HX2 (12F). The thermal images of HX3 (15W) and HX4 (18W) look similar, with relative hot spots located in the middle of the face area and offset toward the hot water inlet. HX3 (15W) and HX4 (18W) held less dust on their upstream faces than did HX1 (8F) and HX2 (12F). This was due to the larger BEE dust particles that could easily get lodged in the wider fin gaps of the first two heat exchangers, while the higher density wave fins tended to block more of the particles from depositing. The amount of dust deposited on the upstream faces of HX1 and HX2 was much greater than that seen from their downstream sides, while the observed differences of the deposited dust mass between the upstream and downstream surfaces for HX3 and HX4 were fairly small.
Overall, fouling is a serious problem for these heat exchangers when they are used as heating coils operating in the environments with large particular matter like the BEE test dust. This kind of dust can increase the air-side pressure drop rapidly and clog the heat exchangers to the point of inoperability.
Cooling While Exposed to Aerosolized Limestone Powder
Figure 8 shows the air-side pressure drop across the four heat exchangers as cooling coils with limestone powder as the challenge test dust. As in the case of the heat exchangers operating as heating coils with the BEE dust but in a faster way, the air pressure drops across all four heating coils increased rapidly. Table 4 lists the dust exposure time required for the pressure drop of each of the four heat exchangers to reach twice their initial values along with the mass of dust and condensate retained in each coil. All four heat exchangers clogged more rapidly when operating as cooling coils exposed to limestone powder dust than they did when operating as heating coils and exposed to the BEE test dust. A limestone slurry was formed when the powder impinged on the wetted fin surfaces and mixed with condensed moisture. The slurry deposited inside the coils and cemented in place. As a result, much more limestone powder and water slurry were trapped in the cooling coils than BEE dust was in the heating coils.
Table 4. Time to Double the Initial Air-Side Resistance with Limestone Powder and Mass of Condensate and Dust Retained in the Heat Exchangers in Cooling Mode Heat Exchanger Time to Double Pressure Condensate and Dust ID Drop, min Retained, g (oz) HX1 (8F) 21.1 297.8 (10.48) HX2 (12F) 8.4 513.1 (18.07) HX3 (15W) 9.2 427.8 (15.06) HX4 (18W) 6.1 268.2 (9.44)
[FIGURE 8 OMITTED]
Figure 9 shows the heat transfer effectiveness and the overall heat transfer coefficients of the four cooling coils with limestone powder as the challenge test dust, respectively. The fluctuation in the cooling coil heat transfer effectiveness and overall heat transfer coefficients are much larger than those in the heating coil performance measurements. This was caused by the larger fluctuation of the liquid temperature at the inlet of the tested heat exchangers. The chiller controller somehow did not generate a steady temperature of the coolant and made the temperature fluctuate much more wildly in the cooling cases than in the heating cases.
Both the heat transfer effectiveness and the overall heat transfer coefficient of the four cooling coils kept nearly constant or slightly increased with the duration of dust exposure. In the same way as in the previous case, the increased air pressure drop made the airflow more turbulent, which would offset the adverse effect on thermal conduction caused by the fouling layers on heat transfer surfaces, and the overall effect of fouling on the heat transfer performance was ignorable when the airflow rate was kept constant.
The temperature distribution of the upstream face area on all four cooling coils was more uniform after the tests than their initial distributions due to an evenly distributed layer of limestone powder on fin surfaces. More deposited dust was seen on the upstream faces of all the cooling coils than on the downstream faces.
Overall, fouling needs to be seriously considered in the applications of these heat exchangers when they are used as cooling coils operating in environments containing smaller particles in the range of 0.5 to 100 [mu]m in moist air, as seen with the limestone powder experiments. A lot of dust and condensed moisture will be trapped internally and quickly clog the heat exchangers due to the increased adhesion of particles to the wetted surfaces. In the case of water soluble particles such as calcium carbonate, the reaction between the solid and liquid phases can create a cement-like slurry that when dried can be very difficult to remove.
Cooling While Exposed to BEE Test Dust
Figure 10 shows the air-side pressure drop across the four heat exchangers operating as cooling coils with the BEE test dust. Table 5 lists the time required for pressure drop of each of the four heat exchangers to reach twice the initial values, along with the mass of condensate and dust retained in each. The times required for the cooling coils to double in pressure drop when exposed to BEE dust were comparable to those with the limestone powder.
Table 5. Time to Double the Initial Air-Side Resistance with Bee Test Dust and Mass of Condensate and Dust Retained in the Heat Exchangers in Cooling Mode Heat Exchanger Time to Double Pressure Condensate and Dust ID Drop, min Retained, g (oz) HX1 (8F) 21.1 243.2 (8.56) HX2 (12F) 13.9 331.8 (11.68) HX3 (15W) 7.9 114.0 (4.01) HX4 (18W) 7.6 160.6 (5.65)
[FIGURE 10 OMITTED]
Figure 11 shows the heat transfer effectiveness and the overall heat transfer coefficient of the four cooling coils during the tests with BEE dust. These values remained nearly constant with respect to the duration of dust exposure, except the heat transfer effectiveness and the overall heat transfer coefficient of HX2 (12F) decreased gradually in the last two minutes of the test. There was a high similarity for the four cooling coils with the two kinds of test dusts.
The BEE dust made the face temperature distribution of all the four cooling coils look more uneven after the tests but not as much as in the heating coil cases with the BEE dust. More deposited dust was seen on the upstream faces of all the cooling coils than on the downstream faces. Additionally, the relative order of dust mass observed on the downstream faces of the heat exchangers from large to small was: 1) HX4 (18W), 2) HX3 (15W), 3) HX2 (12F), and 4) HX1 (8F). Note that the pressure drop is not the same as the dust mass retaining order, indicating the effect of the fin types.
Overall, fouling will significantly degrade the performance of these heat exchangers when they are used as cooling coils operating in the environments with the BEE test dust, like particles in moist air. Dust and condensed water will clog the heat exchangers after a short period of time.
A fouling test system and an operating procedure have been developed based on the literature review, our experience, and existing test operating procedures. Several candidates of reference dust have been characterized. The reference dusts selected for fouling tests are: (1) Masons Hydrated Limestone powder (type S) for commonly used heat exchangers and (2) BEE test dust (ground oats) for heat exchangers installed in agricultural machinery.
Four heat exchangers with different fin densities and shapes were experimentally tested in both heating and cooling modes with aerosolized limestone powder and BEE test dust. The experimental results showed that the impact of air-side particulate fouling on the performance of a heat exchanger is mainly attributed to increased airflow resistance, while the resulting increase in thermal resistance played a less significant role. When a heat exchanger is used as a heating coil operating in an environment with small and not-sticky particles suspended in the air, fouling is usually not a serious problem even when the duration of dust exposure is long (on the order of months). The airflow resistance across the heat exchanger will increase very slowly, on the order of 5 to 10 percent, and the thermal conductivity of the heat exchanger surfaces will degrade only slightly, by a few percent. However, when particles suspended in the air are large relative to the fin spacing, then a heat exchanger in a heating application can readily clog. The increasing airflow resistance will degrade the heat transfer performance significantly in actual applications. When a heat exchanger is used as a cooling coil in a moist and dusty environment, particles and condensed moisture will deposit inside the cooling coil, and the airflow resistance will increase quickly. The higher the fin density, the faster the airflow resistance increases, and the more rapidly the heat transfer performance degrades. If the airflow rate is kept constant, the increased turbulence of the airflow will offset the adverse effect on thermal conduction caused by the fouling layers on heat transfer surfaces. The overall effect of fouling on the heat transfer performance is negligible, while the pressure drop across the heat exchanger will increase significantly. In our experiments, the reduction of heat exchanger performance due to particulate fouling was primarily observed as an increased air-side flow resistance, and increased resistance to heat transfer by an insulating particulate deposition layer on fin surfaces played lesser role in performance degradation.
All of the tests presented in this paper were conducted at a constant airflow rate through the use of a variable-frequency fan controller that automatically compensated for increased pressure loads on the system. As indicated in the data analysis, reduced face area from clogged internals of the heat exchangers led to increased air velocities, which helped overall heat transfer and compensated for some of the heat transfer resistance caused by the fouling. In an application where a variable frequency drive is not in use, the airflow rate will decrease as dust builds up in the heat exchanger. The decrease of the airflow rate will mainly depend on the performance curve of the fan used in the system. When the pressure drop increases, the airflow rate will decrease. Therefore, the turbulence effect on the overall heat transfer performance can be negligible in this type of application, and the degradation of the performance of the heat exchangers will be more serious than observed in a flow-rate controlled system.
[FIGURE 10 OMITTED]
More fouling tests for more types of heat exchangers with different fin densities and shapes in different applications are needed for developing fouling performance criteria to serve as an optimal design tool for heat exchangers. The fouling and clogging mechanism of different heat exchangers with different particles in different environmental conditions should be investigated comprehensively.
This project was funded by the Air Conditioning and Refrigeration Center at the University of Illinois at Urbana-Champaign.
Ahn, Y.-C., and J.-K. Lee. 2005. Characteristics of air-side particulate fouling materials in finned-tube heat exchangers of air conditioners. Particulate Science and Technology 23: 297-307.
Bott, T.R. 1988. Gas side fouling, in Fouling science and technology, eds. L.F. Melo, T.R. Bott, and C.A. Bernardo, pp. 191-206. Dordrecht: Kluwer Academic Publishers.
Bott, T.R. 1995. Fouling of heat exchangers. Amsterdam, The Netherlands: Elsevier.
Brahim, F., W. Augustin, and M. Bohnet. 2003. Numerical simulation of the fouling process. International Journal of Thermal Sciences 42: 323-334.
Characklis, W.G., N. Zelver, M. Turakhia, and M.J. Nimmons. 1981. Fouling and heat transfer, In HTD-Vol. 17 Fouling in Heat Exchange Equipment, eds. J.M. Che-noweth and M. Impagliazzo, pp. 1-15. The 20th ASME/AIChE Heat Transfer Conference, Milwaukee, Wisconsin, USA, 2-5 August.
Epstein, N. 1999. Particle deposition and its mitigation. In Proceedings of Engineering Foundation Conference on Understanding Heat Exchanger Fouling and its Mitigation. ed. T.R. Bott, pp. 3-21. Begell House, New York.
Marner, W.J. 1990. Progress in gas-side fouling of heat-transfer surfaces. Applied Mechanics Review 43(3): 35 - 66.
Mason, D.J., M.R. Heikal, and N. Douch. 2006. Air side fouling of compact heat exchangers. International Journal of Heat Exchangers 7: 1-14.
Siegel, J., I. Walker, and M. Sherman. 2002. Dirty air conditioners: Energy implications of coil fouling. In Proceedings of the 2002 ACEEE Summer Study on Energy Efficiency in Buildings 1: 287-300.
Siegel, J., and V.P. Carey. 2001. Fouling of HVAC fin and tube heat exchangers. In Heat Exchanger Fouling: Fundamental Approaches & Technical Solutions, pp. 425 - 433. The 4th International Conference of Fouling.
Siegel, J.A., and W.W. Nazaroff. 2002. Modeling particle deposition on HVAC heat exchangers. In Proceedings of the 9th International Conference on Indoor Air and Climate I: 521-526.
Yang, L., J.E. Braun, and E.A. Groll. 2007. The impact of fouling on the performance of filter-evaporator combinations. International Journal of Refrigeration 30: 489 - 498.
Zhang, Y. 2005. Indoor Air Quality Engineering. CRC Press. Boca Raton, Florida 33431.
Zhang, Y., X. Wang, S. Ford, and J. Su. 2005. Air cleaning technologies for off-road machinery cooling systems phase I: Characterization of physical properties and fouling mechanisms of different types of debris, unpublished report to Deere & Company. University of Illinois at Urbana-Champaign.
Zhang, Y., Y. Sun, S. Ford, D. Barker, and W. Yan. 2010. Characterization of heat exchanger fouling: Develop reference dust and investigate fouling physics, final report to The Air Conditioning and Refrigeration Center (ACRC). University of Illinois at Urbana-Champaign.
Yigang Sun, PhD
Yuanhui Zhang, PhD, PE
Associate Member ASHRAE
Yuanhui Zhang is a professor, Yigang Sun is a senior research engineer, Douglas Barker is a graduate research assistant, and Steve Ford is a research engineer in the Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champaign, Urbana, IL. Mark Johnson is a product development supervisor at the Modine Manufacturing Company, Racine, WI.
|Printer friendly Cite/link Email Feedback|
|Author:||Sun, Yigang; Zhang, Yuanhui; Barker, Douglas; Ford, Steve; Johnson, Mark|
|Date:||Jan 1, 2012|
|Previous Article:||Measurements of frost growth on louvered folded fins of microchannel heat exchangers, part 1: experimental methodology.|
|Next Article:||Laboratory testing of saddle-tap tees to determine loss coefficients.|