- Email: [email protected]
Experimental investigation of spray dehumidification process in moist air
International Communications in Heat and Mass Transfer 97 (2018) 163–171
Contents lists available at ScienceDirect
International Communications in Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ichmt
Experimental investigation of spray dehumidification process in moist air a
a,⁎
T
b
S. Gumruk , M.K. Aktas , F. Kasap a b
TOBB University of Economics and Technology, Department of Mechanical Engineering, Ankara, Turkey ARCELIK, Cayırova Campus, Istanbul, Turkey
A R T I C LE I N FO
A B S T R A C T
Keywords: Direct contact condensation Spray dehumidification Spray cooling
This study represents an experimental investigation of spray dehumidification process of hot and moist air in turbulent flow. An experimental setup is constructed in order to produce hot and humid air flow in a circular tube. Sub-cooled water at a temperature below the dew point is injected into air stream. The effects of spray water flow rate, water temperature, spray injection direction (parallel/counter flow), nozzle type (hydraulic/air atomization) and air inlet conditions on dehumidification performance are investigated. The results of the present study show that improved dehumidification performance is obtained for highly humid air.
1. Introduction Air or water cooled condensers are frequently used in order to enhance cooling capacity and reduce moisture content of air in various industrial processes and household appliances. Well known disadvantages of moisture condensers are long process times, high energy and/or water consumption. Thus developing new drying techniques and systems with superior dehumidification performance has been a continuous effort. One proposed technique is the spray dehumidification. In the spray dehumidification process direct contact condensation phenomenon takes place between sprayed water and moist air. The direct contact condensation is utilized in various industrial applications such as emergency cooling systems of nuclear reactors, desalination and air-conditioning [1]. Since indirect contact processes lead to a resistance to heat transfer, in general direct contact condensation systems are known to be more effective in dehumidifying. In one application of spray dehumidification system, sub-cooled water is injected into hot and moist air [2]. As long as the water temperature stays below the dew point temperature of mixture, condensation occurs; otherwise water spray leads to increase of mixture humidity. When the water spray contact with the mixture, sensible heat of the mixture starts to decrease and condensation induces latent heat release. Brown [3] experimentally studied the vapor condensation on sub-cooled water droplets. The study revealed that decreasing droplet diameter results an enhancement in heat transfer. Ford and Lekic [4] obtained a correlation which shows the growth of the droplet diameter in the direct contact condensation of steam on water droplets. The correlation results were in good agreement with the experiments. Kulic and Rhodes [5] developed a model to determine the temperature
⁎
Corresponding author. E-mail address: [email protected] (M.K. Aktas).
https://doi.org/10.1016/j.icheatmasstransfer.2018.07.011
0735-1933/ © 2018 Elsevier Ltd. All rights reserved.
distribution during direct contact condensation of air-vapor mixture on a droplet. It is found that heat and mass transfer model estimations are in good agreement with experimental data. Crowe et al. [6] analyzed the velocity and temperature fields in a spray cooling application for a vertical channel flow. The inlet temperature of air and water was 20°C and 60°C, respectively. The relative humidity was 30% and the mass flow rate of water droplets was ten times that of air. Lee and Tankin [7] developed a model to investigate the behavior of water spray in a steam environment. They found that average droplet diameter is greater in vapor compared to air environment, due to the condensation of vapor on water droplets. Sundararajan and Ayyaswamy [8] investigated the heat and mass transfer mechanism of vapor condensation on droplets, numerically. It was concluded that droplet bulk temperature increases with time and results were in good agreement with experiments. Celata et al. [9, 10] investigated the direct contact condensation of saturated steam on subcooled water droplets, experimentally. The condensation efficiency and local heat transfer coefficients were reported as functions of droplet diameter and velocity. Mayinger and Chavez [11] experimentally investigated the growth of a sub-cooled spray droplet in a saturated vapor with pulsed laser holography technique. Increasing mass flow rate of water spray led to the decrease of droplet diameter. It is also found that as spray velocity increases, heat transfer enhances asymptotically and reaches a maximum value. Akira et al. [12] analyzed direct contact condensation phenomena in vapor-subcooled water interface using three different models. Reindl [13] investigated the heat and mass transfer to chilled water sprays in parallel, counter and cross flow configurations. The inlet temperature of air and water was 35°C and 0.6°C, respectively. The relative humidity was 40%. The ratio of mass flow rate of air to water varied from 0.2 to 0.95. Kachhwaha et al.
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
assembly can be run simultaneously or separately. Table 1 summarizes the measurement instrumentation of the experimental apparatus. The temperature and the relative humidity (RH) of air are measured by five probes (Rotronic Hygroflex4-HF4-Universal Transmitter) located at critical sections of the air circulation loop. The probes are inserted into the air channel in cross flow configuration. Due to relatively large diameter (15 mm) and length (20 mm) of the probes, the temperature and RH readings are assumed as crosssectional average measurements. The experimental setup is shown in Fig. 2. The air circulation loop is thermally insulated against heat losses. Five measurement probes are numbered as; Probe 1: spray inlet, Probe 2: spray outlet, Probe 3: condenser inlet, Probe 4: condenser outlet and Probe 5: cooling air inlet/outlet for condenser. The spray dehumidification module is attached by removable mechanism (see the red circle in Fig. 2). In order to avoid water interaction waterproof filters (Rotronic NSPPOM-FD2) are used in the probes. A dosing pump (Grundfos DDE 15–4) supplies filtered water to the spray nozzle at a constant flow rate. The water is provided from a container. Ice cubes and an immersion thermometer are used in order to adjust the container temperature within the range of ± 1 °C. A SCADA system (Fig. 3) was designed in order to control the system and collect data. The SCADA system measures the temperature and the relative humidity at spray inlet by one second intervals and brings the system to desired setting by adjusting the water heater power. The detachable spray dehumidification apparatus supplies water spray by a nozzle in to the air loop in parallel or counter flow configuration (Fig. 4). The nozzle exit is located on the pipe centerline.
[14] carried out experiments in order to investigate the heat and mass transfer in hollow cone water sprays in parallel and counter air flow configurations. Air and water temperatures were varied between 35°C – 42°C and 26°C–33°C, respectively. The maximum relative humidity of air was 60%. Takahashi and Nayak [15] analytically and experimentally studied the direct contact heat transfer between vapor and subcooled water spray. It is found that heat transfer in liquid sheet region is underestimated with pure conduction model. El-Morsi [16] investigated the optimum performance parameters of spray cooling and dehumidification process by employing experimental and analytical techniques. The inlet temperature of air and water was 30°C and 2°C, respectively. The relative humidity was 25%. The ratio of the mass flow rate of air to water was 0.85. The results showed that increasing mass flow rate of water spray results in enhancement of spray dehumidification performance. Malet and Lemaitre [17] analyzed water spray and air-vapor mixture interaction with a heat and mass transfer model. Experimental results showed that after water - spray interaction, humidity and temperature of air decreases. It is noted that experimental results are lower than the numerical data since the droplet evaporation in the numerical study was neglected. The aforementioned review reveals that there have been very few experimental studies on spray dehumidification of moist air. Mostly the air conditioning applications with moderate air temperature and relative humidity values were studied. The focus of the present work is a hot and moist air dehumidification system with an attached water spray injection nozzle. Relatively higher temperature (65°C – 75°C) and relative humidity (80%) values, compared to conventional air conditioning applications, were considered. To the authors' best knowledge, no experimental investigation was performed for the air temperature and relative humidity range considered in this study. In the experimental investigation the parameters affecting the performance of the spray dehumidification process are explored. The parameters under investigation are the water mass flow rate, the water temperature, the spray injection direction, the spray cone angle and the moist air inlet conditions (temperature and relative humidity). Air to water mass flow rate ratio values of the present investigation also significantly differ from the experiments reported in the literature. In the previous study [5] the air – water ratio varied from 0.1 to 1. In the present experiments relatively lower water flow rates were considered and the air – water mass flow rate ratio varied from 0.7 to 7.4. Preliminary results of the present research was presented by Gumruk and Aktas [18]. Unlike the present manuscript, the performance of the spray dehumidification system in parallel and counter flow configurations at different air temperature and humidity values were reported for limited number of cases. The present manuscript considers wider range of temperature and humidity values along with additional critical spray parameters (water temperature, water flow rate, nozzle type and spray cone angle) in order to fully characterize the system performance.
3. Results Table 2 lists the operation parameters of the dehumidification system. The dry bulb air temperatures and relative humidity values were measured during the experiments. In a typical experiment, sufficient amount of time is given to overcome the initial transient of the setup. A sample temperature data at four different locations is shown in Fig. 5. During the first 40 min of the experiment rather strong temperature variations were observed. At this time the humidifier is turned on and experiment starts when the temperature values stabilize. Around 140 min from the start of the experiment nearly steady-state behavior is achieved. By this time the target air temperature and relative humidity value at the inlet (Point 1 in Fig. 2) is obtained. At this time the spray nozzle is opened and the effect of water droplets in air flow is studied. With the injection of the water droplets into the air stream, a sudden decrease in the air temperature is measured. This is due to the sensible heat transfer from air to water spray. The decrease is more pronounced right after the spray nozzle (Probe 2). Fig. 6 depicts the temporal variation of the relative humidity at Probe locations 1, 2, 3 and 4 in the air circulation pipe. The initialization of vapor release from the humidifier near 40 min is obvious. The relative humidity values reach steady-state behavior in nearly 140 min. With spray nozzle opening the relative humidity significantly increases due to increase of the moisture content and air temperature decrease. Table 3 presents the results of the experiments. The results also demonstrate the repeatability of the experiments. In experiments 1 and 2 the inlet conditions (Probe 1) of air are 65°C and % 80 relative humidity (RH). The volumetric flow rate of sprayed water is 10 l/h. The water spray is provided by a hydraulic nozzle which atomizes water by using water line pressure. The water spray injection is opposite to the air flow (counter flow configuration). The experiments were performed with water at 15°C. The air humidity ratio (ω) is calculated based on measured relative humidity and temperature values by using Eqs. (1)–(2).
2. Material and method An experimental setup is designed and constructed in order to generate hot and moist air-vapor mixture stream in a closed loop. A schematic drawing of the experimental setup is presented in Fig. 1. Air is circulated by using a fan in the loop at a constant flow rate of 20 l/s. Overall length and height of the setup are 1.5 m and 1.35 m, respectively. The air circulation loop is made of circular stainless steel pipe. The inner diameter of the pipe is 65 mm. The average air velocity in the pipe reaches 6 m/s and corresponding Reynolds number is nearly 20,000. The air temperature is controlled by resistance heaters. The setup consists a custom made water boiler that serves as humidifier. An air cooled condenser is utilized in order to evaluate the combined effect of the spray dehumidifier and conventional condenser on dehumidification process. The condenser and spray dehumidifier 164
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 1. Schematic drawing of the experimental setup.
here h is the heat transfer coefficient and A is the droplet surface area. The latent heat transfer as a function of latent heat of condensation (hfg) and mass transfer coefficient (hd) is given by;
Table 1 Measurements on the experimental apparatus. Measurement
Relative humidity and temperature
Water flow rate
ω = 0.622∙
Pw PB − Pw
Device
Rotronic Hygroflex4-HF4Universal Transmitter with Hygromer IN-1 probe Grundfos DDE 15–4
Measurement range
Accuracy
−50–100 °C 0…100% rh
± 0.2°C ± 1% rh
15 ml/h–15 l/h
± 5%
Ql̇ = hd Aρair hfg (ω − ωs )
where ρair is the dry air density and ωs is the humidity ratio at the water droplet surface. ωs is evaluated at 100% relative humidity and water temperature. For 15°C spray water temperature, approximately 5.4 g watervapor/kg dry-air humidity ratio decrease is calculated when spray dehumidification module is in function. Since vapor in moist air condenses on the water droplets, a reduction in the humidity ratio is obtained. The condenser unit of the setup provides additional condensation. The overall dehumidification significantly enhances when the spray module and air cooled condenser work together.
(1)
here PB is the local atmospheric pressure and Pw is the vapor pressure:
Pw =
Pws ∙RH 100
(2) 3.1. Uncertainty analysis
In Eq. (2), Pws indicates the saturated vapor pressure which is function of temperature. The sensible heat transfer from moist air to a water droplet can be expressed as;
Qṡ = hA (Tair − Twater )
(4)
The overall uncertainty of the experimental results were computed by considering the measurement errors. The moist air temperature and the relative humidity are the major source of the errors. Rotronic
(3)
Fig. 2. Experimental apparatus. 165
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 3. Interface of the SCADA system.
For a typical experiment with 65 °C and 80% RH inlet conditions, the overall uncertainty of the humidity ratio calculations was estimated to be 0.0034 kg/kg based on Eq. (5). Thus, a typical experimental error associated with the humidity ratio values is nearly 2.2%. Table 4 shows the standard deviation of repeatability tests. The results indicate that the experimental results are reproducible by % 0.6 maximum deviation. The parameters affecting the dehumidification performance of the spray module are investigated in the following sections. The performance of the spray dehumidification system at different air inlet conditions was studied first. Then the experiments were carried out in order to explore the influence of spray injection direction, spray droplet diameter, spray cone angle, water flow rate and temperature on the dehumidification process. Fig. 4. Spray apparatus.
3.2. Spray dehumidification at different air inlet conditions Hygroflex4 probes provide ± 1% relative humidity and ± 0.2°C temperature measurement accuracy. The experimental uncertainty [19] in the humidity ratio is given by Eq. (5).
wω =
2 2 ⎛ ∂w wRH ⎞ + ⎛ ∂w wT ⎞ RH T ∂ ∂ ⎝ ⎠ ⎝ ⎠
In order to evaluate the performance of the spray dehumidification process at different air inlet RH values, two cases (Exp. 3 & 4) with RH = % 80 and RH = % 50, both having 65 °C inlet temperature, were considered. The temperature of the water spray was 19 ± 1 °C for both cases. The inlet RH and temperature data were taken at nearly 30 cm above (Probe 1) the spray module. The computed humidity ratio values
(5)
Table 2 Experimentally studied parameters. Experiment
Temperature (°C)
Relative Humidity (%)
Flow Direction
Water Temperature (°C)
Water Mass Flow Rate (l/h)
Spray Nozzle Type
Spray Cone Angle (°)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
65 65 65 65 65 65 65 65 65 65 65 75 75 75 65 65 75 75
80 80 80 50 80 80 80 80 80 80 80 80 80 80 80 80 80 80
Counter Counter Counter Counter Counter Counter Parallel Parallel Counter Counter Parallel Counter Counter – Counter Counter Counter Counter
15 15 18 20 22 22 23 23 22 23 22 10 10 – 10 25 15 25
10 10 10.9 10.3 10 10 10 10 10 5 5 1 10 0 10 10 1 1
Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Air Atomizer Hydraulic – Hydraulic Hydraulic Air Atomizer Air Atomizer
65 65 65 65 65 65 65 65 105 65 65 30 65 – 65 65 30 30
166
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 5. Moist air temperature as a function of time.
were presented in Table 3. The spray dehumidifier performs well for RH = % 80 case. At high relative humidity, the vapor pressure is higher. Thus higher mass transfer rate is achieved between moist air and water droplets. Relatively reduced vapor pressure for RH = % 50 induces significantly low mass transfer. Here it is interesting to note that the performance of the condenser does not significantly differ for low and high RH cases. For both cases the amount of dehumidification at the condenser is approximately 10%. Table 5 shows the variation of the moist air temperature after the spray injection at different relative humidity values. For the same inlet temperature and nearly constant water temperature and flow rate in Exp. 3 & 4, the temperature drop at the spray module is noticeably higher for low RH (% 50) case. This is due to the fact that the evaporation rate of the droplets is greater at low RH. For both high and low RH cases, the temperature drop at the air cooled condenser rather small in comparison to the spray dehumidification module. In order to fully identify the effect of spray dehumidification process on the relative humidity, RH values were evaluated at the inlet temperature of the humid air (artificial reheat process to inlet temperature) by using the humidity ratio values presented in Table 3. The results are reported in Table 6. The relative humidity decrease at the spray module is 1.7 and 1.1 for RH = % 80 and RH = % 50 cases (Exp. 3 & 4), respectively. It is obvious that the spray dehumidifier performs better in terms of relative humidity reduction at high RH. Eqs. (6)–(7) are used to calculate the required thermal energy for reheating the dehumidified air to inlet temperature (65°C).
Table 3 Spray dehumidification performance at different conditions. Experiment
1 2 3 4 5 6 7 8 9 10 12 13 14
Average humidity ratio (g/kg) Air inlet
Spray outlet
Condenser outlet
Difference (spray)
Difference (spray + condenser)
152.7 153.2 153.2 87.4 152.9 152.9 152.7 153.2 152.9 152.9 271.9 271.5 271.5
146.9 148.2 149.2 85.6 145.1 146.7 146.2 149.0 145.1 146.7 265.3 249.4 –
134.5 135.2 136.0 77.7 132.9 133.2 133.5 135.5 132.9 133.2 245.9 233.3 254.1
5.8 5.0 4.0 1.8 7.7 6.2 6.6 4.2 7.7 6.2 6.6 22.1 –
18.2 18.0 17.2 9.7 20.0 19.7 19.3 17.7 20.0 19.7 26.0 38.2 17.4
Table 4 Standard deviation in the experiments. STANDARD DEVIATION (%) Exp. 1&2
Air inlet 0.2
Fig. 6. Relative humidity as a function of time. 167
Spray outlet 0.6
Condenser Outlet 0.4
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
state of 65°C in Exp. 3. When the spray module and the air cooled condenser run simultaneously, the required heat rate increases to 177 W.
Table 5 Variation of air temperature after spray injection at different relative humidity values, water flow rates and water temperatures. Average temperature (°C) Experiment
Spray outlet
Condenser outlet
3 4 6 8 10 11 15 16
58.9 50.6 59.3 59.6 59.8 60.0 59.1 59.3
57.9 50.2 58.3 58.8 59.5 59.6 58.1 58.6
3.3. Effect of injection direction on the spray dehumidification process The injection configuration of water spray into moist air affects the dehumidification process only to a limited extent. Table 3 also reports the humidity ratio values at two different flow configurations. In parallel setting (Exp. 7 & 8) the water is injected at the same direction to the air flow. For counter flow configuration (Exp. 5 & 6) the injection is opposite to air flow direction. The volumetric flow rate of water is kept constant at 10 l/h during the experiments. The humidity ratio reduction at the spray module is slightly higher for the counter flow configurations. Between the inlet and spray outlet, on the average 5.4 g/kg and 7 g/kg reduction in the humidity ratio was measured for parallel and counter flow configurations, respectively. Thus the spray module at counter flow configuration performs slightly better due to greater relative velocity of spray droplets in air. El-Morsi [16] reported an increase of the mass transfer coefficient by increasing droplet relative velocity. The results of the present experiments are consistent with theoretical findings. The behavior is more obvious at a longer distance. Between the inlet and condenser outlet, the reduction in the humidity ratio was increased from 18.5 g/kg to 19.8 g/kg when counter flow configuration was utilized instead of the parallel one.
Table 6 Spray dehumidification performance in terms of relative humidity. Relative humidity at the Inlet Temperature (%) Experiment
Spray outlet
Condenser outlet
Difference (spray)
Difference (spray + condenser)
3 4 12 13 14 15 16
78.3 48.9 78.4 75.0 – 74.8 78.7
72.7 44.9 74.3 71.5 76.2 70.3 73.1
1.7 1.1 1.6 5.0 – 5.2 1.3
7.3 5.1 5.7 8.5 3.8 9.7 6.9
3.4. Effect of water spray cone angle on the spray dehumidification process
Q̇ = ṁ m cp, m ∆T
(6) The effect of water spray cone angle on the spray dehumidification performance is studied by utilizing two nozzles having different cone angles, 105° (Exp. 9) and 65° (Exp. 6). The volumetric flow rate of water is kept constant at 10 l/h during the experiments. The results presented in Table 3 reveal that increasing water spray cone angle slightly augments the dehumidification at the spray module. This can be attributed to the larger interfacial area between moist air and water droplets. Larger total interfacial area enhances both heat and mass transfer. However the dehumidification enhancement can be considered to be marginal.
here, ṁ is the mass flow rate of the humid air and ΔT is the temperature increase. The temperature increase was calculated for both spray outlet and the condenser outlet. The specific heat of the air-vapor mixture at constant pressure calculated by;
cp, m = cp, a + ω∙cp, w
(7)
The subscripts m, a and w denote the air-vapor mixture, the dry air and the water vapor, respectively. When the spray module operates alone, 152 W heat transfer rate is required in order to bring the humid air at the spray outlet to initial
Fig. 7. Effect of water flow rate on dehumidification process in terms of average humidity ratio. 168
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
3.5. Effect of water flow rate on the spray dehumidification process
Spray surface area ∝ ṁ water /d
In order to investigate the effect of water mass flow rate on the spray dehumidification process two different mass flow rate are considered, 10 l/h and 5 l/h. Both parallel and counter flow conditions were studied for 65 °C temperature and 80% relative humidity air inlet conditions. The spray water temperature was 22.5 ± 0.5 °C in all experiments. Fig. 7 depicts the humidity ratio values achieved by 5 l/h and 10 l/h flow rates. As the water spray mass flow rate increases, the humidity ratio of air significantly decreases due to larger total droplet surface area. An increase of water flow rate from 5 l/h to 10 l/h increases the amount of dehumidification approximately 4 g/kg at the spray module for both parallel and counter flow configurations. The increase of the humidity ratio reduction for the counter flow configuration (reported in Table 3) between the inlet and condenser outlet sections was not measured for 5 l/h spray water flow rate. A temperature decrease of air is also measured by increasing water flow rate due to sensible heat transfer for both counter (Exp. 10 & 6) and parallel flow (Exp. 11 & 8) cases (Table 5). An increase of spray water flow rate from 5 l/h to 10 l/h reduces the temperature by 0.5 °C and 0.8 °C at spray and condenser outlets, respectively for both parallel and counter flows. The total temperature decrease of humid air between the inlet and condenser outlet sections is approximately 5.5 °C for 5 l/h flow rate. When the flow rate was increased to 10 l/h, the temperature decrease increased to 6.2 °C for parallel flow configuration. For counter flow setting the temperature reduction is even higher (6.7 °C).
Since smaller droplets form greater spray surface area, rate of vapor condensation on droplet surfaces enhances when the air atomization nozzle is used. El-Morsi (2002) reported a decreasing spray chamber effectiveness by increasing droplet diameter. Table 6 shows the relative humidity values (for the moisture content given in Table 3) calculated at 75°C inlet temperature. Without the spray module (Exp. 14) the relative humidity reduces from 80% to 76.2%. When the spray module is utilized (Exp. 13) 71.5% relative humidity is achieved. If the total (spray module plus air cooled condenser) humidity reduction values are compared, it is obvious that the air atomizer nozzle is as effective as hydraulic nozzle with only 1/10th of water consumption. 3.7. Effect of water spray temperature on the spray dehumidification process 10 °C (Exp. 15), 15 °C (Exp. 1) and 25 °C (Exp. 16) sprayed (by hydraulic nozzle) water temperatures were considered in order to explore the effect of water temperature on dehumidification. The results are presented in Fig. 8 and Table 5. When the spray module injects water at 25 °C, nearly 6 g-water vapor/kg-dry air condensation is obtained while 12 g-water vapor/kg-dry air condensation is achieved by water at 10 °C. As the water spray temperature decreases, the difference between the dew-point temperature and droplet temperature increases. The difference between the moist air humidity ratio and humidity ratio at the droplet surface also increases. Thus the mass transfer (condensation) rate enhances. As Table 5 demonstrates the variation of the temperature drop with water temperature at the spray module is insignificant. This indicates the stronger influence of latent heat transfer (Eq. 4) compared to sensible heat exchange (Eq. 3). For 10°C water temperature cases, humidity ratio at the droplet surface is lower. Thus the decrease in the humidity ratio is more pronounced (nearly 12 g water-vapor/kg dry-air) due to higher mass transfer rate. If the relative humidity is evaluated at 65°C based on the humidity ratio values presented in Fig. 8, it is found that remarkably high condensation at the spray module is obtained with 10°C water (Table 6). The effect of water temperature was examined for the dehumidification performance of spray module utilizing air atomization nozzle as well (Fig. 9). For these group of experiments the spray water flow rate was 1 l/h. 10 °C (Exp. 12), 15 °C (Exp. 17) and 25 °C (Exp. 18) sprayed water temperatures were considered. For the spray module injecting water at 25 °C, nearly 3 g-water vapor/kg-dry air condensation is obtained. The condensation increases to approximately 6 g-water vapor/ kg-dry air by water at 10 °C. The water temperature has similar influence on condensation for air atomization nozzle and hydraulic nozzle. Figs. 10–13 depict the effect of various parameters on the variation of humidity ratio along the process line where the condensation occurs.
3.6. Effect of spray nozzle type on the spray dehumidification process In order to investigate the effect of spray nozzle type on dehumidification process, in addition to hydraulic nozzle, air atomizer nozzle was utilized at the spray module in Exp. 12. The hydraulic nozzle atomizes liquid water by using water line pressure while the air atomizer nozzle uses pressurized air for droplet formation. The spray droplet diameter varies with water mass flow rate. In this study, the average droplet diameter is nearly 100 μm for hydraulic nozzle at 10 l/ h. The average droplet diameter reduces to nearly 15 μm for air atomizer nozzle at 1 l/h [20]. Because of the fact that decreasing mass flow rate of water spray leads to reduction in droplet size, the air atomizer nozzle was used at the minimum flow rate. The experiments are performed at identical water spray temperatures for the hydraulic (Exp. 13) and the air atomizer (Exp. 12) nozzles in counter flow direction and the results are tabulated (Table 3). Table 3 also includes the result of an experiment where the spray module was not utilized (Exp. 14). This case corresponds a traditional dehumidification process provided by an air cooled condenser only. The dehumidification at the spray module is 22.2 g-water vapor/kgdry air with the hydraulic nozzle (Exp. 13). In Exp. 14, the dehumidification at the condenser is 17.4 g-water vapor/kg-dry air without the spray module. This value is 6.6 g/kg for the air atomizer nozzle. Here it is important to note that air atomizer nozzle consumes one tenth of the water the hydraulic nozzle injects. Thus it can be concluded that the humidity reduction per unit water flow rate is much higher for the air atomizer nozzle. The surface area of the spray is proportional to the square of the droplet diameter and the total number of droplets.
Spray surface area ∝ d2N
(10)
(8)
The number of droplets depends on the mass flow rate of water and the volume of a droplet;
N∝
ṁ water d3
(9) Fig. 8. Variation of the humidity ratio with sprayed water temperature for hydraulic nozzle.
Thus the total surface area of the spray increases by the water mass flow rate and decreases by the droplet diameter; 169
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 9. Variation of the humidity ratio with sprayed water temperature for air atomization nozzle.
Fig. 12. Effect of water spray mass flow rate and temperature on relative humidity.
Fig. 10. Effect of air inlet conditions on spray dehumidifier performance.
In Fig. 10, the humidity reduction at 50% RH is much slower along the process line. For 80% RH case, the spray module and the air cooled condenser have nearly same effect on dehumidification. The humidity ratio values along the process line are negligibly affected by the flow configuration (counter/parallel) as shown in Fig. 11. For the nearly same moist air inlet conditions, the calculated humidity ratio values after the spray module for parallel and counter flow settings differs by maximum 1.5% deviation at low and high water flow rate cases. Similar behavior was observed at the condenser outlet section. On the other hand, the dehumidification process is strongly affected by the spray water flow rate and the water temperature as shown in Fig. 12. In order to clearly demonstrate the effect, the results are presented in terms of relative humidity evaluated at 65 °C inlet temperature. Thus
Fig. 13. Effect of nozzle type on humidity ratio.
the values correspond to dehumidified and reheated moist air conditions. The moist air enters the inlet section at 65°C and 80% relative humidity in all cases. The spray (with hydraulic nozzle) setting is counter flow configuration. For the spray flow rate experiments the water temperature was 22.5 ± 0.5°C. For the spray temperature experiments the flow rate was 10 l/h. As the figure indicates the water temperature is the most influential parameter on the dehumidification. The relative humidity reduces to nearly 70.3% when water at 10°C was injected to the air stream with 10 l/h flow rate. The dehumidification process is also affected by the nozzle type as
Fig. 11. Effect of water spray injection direction on humidity ratio. 170
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Ql Qs Ra RH ΔT T T· b V ω ρm ρda
reported in Fig. 13. The moist air enters the inlet section at 75°C and 80% RH for both hydraulic and air atomization utilized experiments. The spray water temperature was 10°C. The decrease of humidity ratio at the condenser outlet is approximately 38 g/kg and 26 g/kg for hydraulic and air atomization nozzles, respectively. Thus, when the water consumption was considered the spray module employing air atomization nozzle has superior dehumidification performance. 4. Conclusions The spray dehumidification process in a turbulent channel flow experimentally investigated by a parametric study. The experiments results revealed that the spray dehumidification module can be as effective as a conventional air-cooled condenser in dehumidifying of moist air. Based on the measurements of the present research one may conclude:
Acknowledgements The authors gratefully acknowledge the financial support for this work by Ministry of Science, Industry and Technology of Turkey under Grant# 01488.STZ.2012-2.
- Water flow rate and the temperature are the most critical parameters on the dehumidification performance of the spray module. - An increase of water flow rate from 5 l/h to 10 l/h increases the total amount of dehumidification between the inlet and condenser outlet sections by 24% in counter flow configuration. For parallel flow configuration, the increase is 11%. - Spray water temperatures significantly affect the dehumidification performance. A decrease of sprayed water temperature from 25 °C to 10 °C nearly doubles the amount of dehumidification at the spray module for both hydraulic and air atomization nozzles. - The injection direction of the spray have relatively small effect on dehumidification performance. For the nearly same moist air inlet conditions, approximately 7% higher dehumidification was achieved between the inlet and condenser outlet with 10 l/h spray flow rate. For 5 l/h spray flow rate, flow configuration did not alter the amount of dehumidification. - The spray cone angle has relatively small effect on dehumidification performance. - The spray module utilizing air atomizer nozzle has superior dehumidification performance per unit water flow rate.
References [1] Y. Kim, J. Park, C. Song, Investigation of the stem-water direct contact condensation heat transfer coefficients using interfacial transport models, Int. Commun. Heat and Mass Transf. 31 (2004) 397–408. [2] M. El-Morsi, S.A. Klein, D.T. Reindl, Air washers: a new look at a vintage technology, ASHRAE J. 45 (10) (2003) 32–36. [3] G. Brown, Heat Transmission by Condensation of Steam on a Spray of Water Drops, Proc. of General Discussion on Heat Transfer, Institution of Mechanical Engineers, 1951, pp. 49–52. [4] J.D. Ford, A. Lekic, Rate of growth of drops during condensation, Int. J. Heat Mass Transf. 16 (1972) 61–64. [5] E. Kulic, E. Rhodes, Direct contact condensation from air-steam mixtures of a single droplet, Can. J. Chem. Eng. 55 (1977) 131–137. [6] C.T. Crowe, M. Sharma, D.E. Stock, The particle source-in-cell (psi-cell) model for gas droplet flows, J. Fluids Eng. 99 (1977) 325–352. [7] S.Y. Lee, R.S. Tankin, Study of liquid spray (water) in a non-condensable environment (air), Int. J. Heat Mass Transf. 27 (1983) 331–363. [8] T. Sundararajan, P.S. Ayyaswamy, Hydrodynamics and heat transfer associated with condensation on a moving drop: solutions for intermediate Reynolds numbers, J. Fluid Mech. 149 (1984) 33–58. [9] G.P. Celata, M. Cumo, G.E. Farello, G. Focardi, A comprehensive analysis of direct contact condensation of saturated steam on subcooled liquid jets, Int. J. Heat Mass Transf. 32 (1989) 639–654. [10] G.P. Celata, M. Cumo, E. D'Annibale, G.E. Farello, G. Focardi, A theoretical and experimental study of direct-contact condensation on water in turbulent flow, Experim. Heat Transf. 2 (1989) 129–148. [11] F. Mayinger, A. Chávez, Measurement of direct-contact condensation of pure saturated vapour on an injection spray by applying pulsed laser holography, Int. J. Heat Mass Transf. 35 (1992) 691–702. [12] M. Akira, H. Fiji, S. Takamoto, Prediction of heat transfer by direct contact condensation at a steam-subcooled water Interface, Int. J. Heat Mass Transf. 35 (1) (1992) 101–109. [13] D.T. Reindl, Combustion Turbine Inlet Air Cooling Using Thermal Storage & DirectContact Sprays, Proceedings of the EPRI International Sustainable Thermal Energy Storage Conference, 147–150, Minneapolis, MN, August, (1996). [14] S.S. Kachhwaha, P.L. Dhar, S.R. Kale, Experimental studies and numerical simulation of evaporative cooling of air with a water spray-I. Horizontal parallel flow, Int. J. Heat Mass Transf. 41 (2) (1998) 447–464. [15] M. Takahashi, K.A. Nayak, Heat transfer direct contact condensation of steam to subcooled water spray, J. Heat Transf. 123 (2001) 703–710. [16] M.S. El-Morsi, Optimization of Direct Contact Spray Coolers, Ph.D. thesis University of Wisconsin- Madison, 2002. [17] J. Malet, P. Lemaitre, Water Spray Interaction with Air-Steam Mixtures Under Containment Spray Conditions: Comparison of Heat and Mass Transfer Model With the TOSQAN Spray Tests, the 11th International Topical Meeting on Nuclear Reactor Thermal-Hydraulics, Avignon, France, (2005). [18] S. Gumruk, M.K. Aktas, Experimental Study of Direct Contact Condensation of Steam on Water Droplets, Proceedings of the World Congress on Engineering 2015 Vol II, July 1–3, 2015, London, UK, (2015). [19] J. Holman, Experimental Methods for Engineers, 8th Edition, McGraw-Hill, 2016. [20] Spraying Systems, Spray Technology Reference Guide Understanding Drop Size, (1996).
The results of the present investigation may be useful in the design of energy efficient drying or dehumidifying processes for industrial applications and household appliances such as tumble dryers. A future investigation will focus on the spray dehumidification process utilizing air atomization nozzles that can be used only at special applications since they require pressurized air. Nomenclature A cpm cpa cpw d h hd hfg ṁ m N Pws Pw PB
Latent heat transfer (W) Sensible heat transfer (W) Gas constant (J/kgK) Relative humidity Temperature difference (°C) Temperature (°C) Bulk temperature (°C) Volumetric flow rate of air-vapor mixture (l/s) Humidity ratio (kg/kg or g/kg) Density of air-vapor mixture (kg/m3) Density of dry air (kg/m3)
Surface area (m2) Specific heat of air-vapor mixture at constant pressure (J/ kgK) Specific heat of dry air at constant pressure (J/kgK) Specific heat of water vapor at constant pressure (J/kgK) Droplet diameter (m) Heat transfer coefficient (W/m2K) Mass transfer coefficient (m/s) Latent heat of condensation (J/kg) Mass flow of air-vapor mixture (kg/s) Number of droplets Saturated vapor pressure (Pa) Vapor pressure (Pa) Atmospheric pressure (Pa)
171
Contents lists available at ScienceDirect
International Communications in Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ichmt
Experimental investigation of spray dehumidification process in moist air a
a,⁎
T
b
S. Gumruk , M.K. Aktas , F. Kasap a b
TOBB University of Economics and Technology, Department of Mechanical Engineering, Ankara, Turkey ARCELIK, Cayırova Campus, Istanbul, Turkey
A R T I C LE I N FO
A B S T R A C T
Keywords: Direct contact condensation Spray dehumidification Spray cooling
This study represents an experimental investigation of spray dehumidification process of hot and moist air in turbulent flow. An experimental setup is constructed in order to produce hot and humid air flow in a circular tube. Sub-cooled water at a temperature below the dew point is injected into air stream. The effects of spray water flow rate, water temperature, spray injection direction (parallel/counter flow), nozzle type (hydraulic/air atomization) and air inlet conditions on dehumidification performance are investigated. The results of the present study show that improved dehumidification performance is obtained for highly humid air.
1. Introduction Air or water cooled condensers are frequently used in order to enhance cooling capacity and reduce moisture content of air in various industrial processes and household appliances. Well known disadvantages of moisture condensers are long process times, high energy and/or water consumption. Thus developing new drying techniques and systems with superior dehumidification performance has been a continuous effort. One proposed technique is the spray dehumidification. In the spray dehumidification process direct contact condensation phenomenon takes place between sprayed water and moist air. The direct contact condensation is utilized in various industrial applications such as emergency cooling systems of nuclear reactors, desalination and air-conditioning [1]. Since indirect contact processes lead to a resistance to heat transfer, in general direct contact condensation systems are known to be more effective in dehumidifying. In one application of spray dehumidification system, sub-cooled water is injected into hot and moist air [2]. As long as the water temperature stays below the dew point temperature of mixture, condensation occurs; otherwise water spray leads to increase of mixture humidity. When the water spray contact with the mixture, sensible heat of the mixture starts to decrease and condensation induces latent heat release. Brown [3] experimentally studied the vapor condensation on sub-cooled water droplets. The study revealed that decreasing droplet diameter results an enhancement in heat transfer. Ford and Lekic [4] obtained a correlation which shows the growth of the droplet diameter in the direct contact condensation of steam on water droplets. The correlation results were in good agreement with the experiments. Kulic and Rhodes [5] developed a model to determine the temperature
⁎
Corresponding author. E-mail address: [email protected] (M.K. Aktas).
https://doi.org/10.1016/j.icheatmasstransfer.2018.07.011
0735-1933/ © 2018 Elsevier Ltd. All rights reserved.
distribution during direct contact condensation of air-vapor mixture on a droplet. It is found that heat and mass transfer model estimations are in good agreement with experimental data. Crowe et al. [6] analyzed the velocity and temperature fields in a spray cooling application for a vertical channel flow. The inlet temperature of air and water was 20°C and 60°C, respectively. The relative humidity was 30% and the mass flow rate of water droplets was ten times that of air. Lee and Tankin [7] developed a model to investigate the behavior of water spray in a steam environment. They found that average droplet diameter is greater in vapor compared to air environment, due to the condensation of vapor on water droplets. Sundararajan and Ayyaswamy [8] investigated the heat and mass transfer mechanism of vapor condensation on droplets, numerically. It was concluded that droplet bulk temperature increases with time and results were in good agreement with experiments. Celata et al. [9, 10] investigated the direct contact condensation of saturated steam on subcooled water droplets, experimentally. The condensation efficiency and local heat transfer coefficients were reported as functions of droplet diameter and velocity. Mayinger and Chavez [11] experimentally investigated the growth of a sub-cooled spray droplet in a saturated vapor with pulsed laser holography technique. Increasing mass flow rate of water spray led to the decrease of droplet diameter. It is also found that as spray velocity increases, heat transfer enhances asymptotically and reaches a maximum value. Akira et al. [12] analyzed direct contact condensation phenomena in vapor-subcooled water interface using three different models. Reindl [13] investigated the heat and mass transfer to chilled water sprays in parallel, counter and cross flow configurations. The inlet temperature of air and water was 35°C and 0.6°C, respectively. The relative humidity was 40%. The ratio of mass flow rate of air to water varied from 0.2 to 0.95. Kachhwaha et al.
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
assembly can be run simultaneously or separately. Table 1 summarizes the measurement instrumentation of the experimental apparatus. The temperature and the relative humidity (RH) of air are measured by five probes (Rotronic Hygroflex4-HF4-Universal Transmitter) located at critical sections of the air circulation loop. The probes are inserted into the air channel in cross flow configuration. Due to relatively large diameter (15 mm) and length (20 mm) of the probes, the temperature and RH readings are assumed as crosssectional average measurements. The experimental setup is shown in Fig. 2. The air circulation loop is thermally insulated against heat losses. Five measurement probes are numbered as; Probe 1: spray inlet, Probe 2: spray outlet, Probe 3: condenser inlet, Probe 4: condenser outlet and Probe 5: cooling air inlet/outlet for condenser. The spray dehumidification module is attached by removable mechanism (see the red circle in Fig. 2). In order to avoid water interaction waterproof filters (Rotronic NSPPOM-FD2) are used in the probes. A dosing pump (Grundfos DDE 15–4) supplies filtered water to the spray nozzle at a constant flow rate. The water is provided from a container. Ice cubes and an immersion thermometer are used in order to adjust the container temperature within the range of ± 1 °C. A SCADA system (Fig. 3) was designed in order to control the system and collect data. The SCADA system measures the temperature and the relative humidity at spray inlet by one second intervals and brings the system to desired setting by adjusting the water heater power. The detachable spray dehumidification apparatus supplies water spray by a nozzle in to the air loop in parallel or counter flow configuration (Fig. 4). The nozzle exit is located on the pipe centerline.
[14] carried out experiments in order to investigate the heat and mass transfer in hollow cone water sprays in parallel and counter air flow configurations. Air and water temperatures were varied between 35°C – 42°C and 26°C–33°C, respectively. The maximum relative humidity of air was 60%. Takahashi and Nayak [15] analytically and experimentally studied the direct contact heat transfer between vapor and subcooled water spray. It is found that heat transfer in liquid sheet region is underestimated with pure conduction model. El-Morsi [16] investigated the optimum performance parameters of spray cooling and dehumidification process by employing experimental and analytical techniques. The inlet temperature of air and water was 30°C and 2°C, respectively. The relative humidity was 25%. The ratio of the mass flow rate of air to water was 0.85. The results showed that increasing mass flow rate of water spray results in enhancement of spray dehumidification performance. Malet and Lemaitre [17] analyzed water spray and air-vapor mixture interaction with a heat and mass transfer model. Experimental results showed that after water - spray interaction, humidity and temperature of air decreases. It is noted that experimental results are lower than the numerical data since the droplet evaporation in the numerical study was neglected. The aforementioned review reveals that there have been very few experimental studies on spray dehumidification of moist air. Mostly the air conditioning applications with moderate air temperature and relative humidity values were studied. The focus of the present work is a hot and moist air dehumidification system with an attached water spray injection nozzle. Relatively higher temperature (65°C – 75°C) and relative humidity (80%) values, compared to conventional air conditioning applications, were considered. To the authors' best knowledge, no experimental investigation was performed for the air temperature and relative humidity range considered in this study. In the experimental investigation the parameters affecting the performance of the spray dehumidification process are explored. The parameters under investigation are the water mass flow rate, the water temperature, the spray injection direction, the spray cone angle and the moist air inlet conditions (temperature and relative humidity). Air to water mass flow rate ratio values of the present investigation also significantly differ from the experiments reported in the literature. In the previous study [5] the air – water ratio varied from 0.1 to 1. In the present experiments relatively lower water flow rates were considered and the air – water mass flow rate ratio varied from 0.7 to 7.4. Preliminary results of the present research was presented by Gumruk and Aktas [18]. Unlike the present manuscript, the performance of the spray dehumidification system in parallel and counter flow configurations at different air temperature and humidity values were reported for limited number of cases. The present manuscript considers wider range of temperature and humidity values along with additional critical spray parameters (water temperature, water flow rate, nozzle type and spray cone angle) in order to fully characterize the system performance.
3. Results Table 2 lists the operation parameters of the dehumidification system. The dry bulb air temperatures and relative humidity values were measured during the experiments. In a typical experiment, sufficient amount of time is given to overcome the initial transient of the setup. A sample temperature data at four different locations is shown in Fig. 5. During the first 40 min of the experiment rather strong temperature variations were observed. At this time the humidifier is turned on and experiment starts when the temperature values stabilize. Around 140 min from the start of the experiment nearly steady-state behavior is achieved. By this time the target air temperature and relative humidity value at the inlet (Point 1 in Fig. 2) is obtained. At this time the spray nozzle is opened and the effect of water droplets in air flow is studied. With the injection of the water droplets into the air stream, a sudden decrease in the air temperature is measured. This is due to the sensible heat transfer from air to water spray. The decrease is more pronounced right after the spray nozzle (Probe 2). Fig. 6 depicts the temporal variation of the relative humidity at Probe locations 1, 2, 3 and 4 in the air circulation pipe. The initialization of vapor release from the humidifier near 40 min is obvious. The relative humidity values reach steady-state behavior in nearly 140 min. With spray nozzle opening the relative humidity significantly increases due to increase of the moisture content and air temperature decrease. Table 3 presents the results of the experiments. The results also demonstrate the repeatability of the experiments. In experiments 1 and 2 the inlet conditions (Probe 1) of air are 65°C and % 80 relative humidity (RH). The volumetric flow rate of sprayed water is 10 l/h. The water spray is provided by a hydraulic nozzle which atomizes water by using water line pressure. The water spray injection is opposite to the air flow (counter flow configuration). The experiments were performed with water at 15°C. The air humidity ratio (ω) is calculated based on measured relative humidity and temperature values by using Eqs. (1)–(2).
2. Material and method An experimental setup is designed and constructed in order to generate hot and moist air-vapor mixture stream in a closed loop. A schematic drawing of the experimental setup is presented in Fig. 1. Air is circulated by using a fan in the loop at a constant flow rate of 20 l/s. Overall length and height of the setup are 1.5 m and 1.35 m, respectively. The air circulation loop is made of circular stainless steel pipe. The inner diameter of the pipe is 65 mm. The average air velocity in the pipe reaches 6 m/s and corresponding Reynolds number is nearly 20,000. The air temperature is controlled by resistance heaters. The setup consists a custom made water boiler that serves as humidifier. An air cooled condenser is utilized in order to evaluate the combined effect of the spray dehumidifier and conventional condenser on dehumidification process. The condenser and spray dehumidifier 164
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 1. Schematic drawing of the experimental setup.
here h is the heat transfer coefficient and A is the droplet surface area. The latent heat transfer as a function of latent heat of condensation (hfg) and mass transfer coefficient (hd) is given by;
Table 1 Measurements on the experimental apparatus. Measurement
Relative humidity and temperature
Water flow rate
ω = 0.622∙
Pw PB − Pw
Device
Rotronic Hygroflex4-HF4Universal Transmitter with Hygromer IN-1 probe Grundfos DDE 15–4
Measurement range
Accuracy
−50–100 °C 0…100% rh
± 0.2°C ± 1% rh
15 ml/h–15 l/h
± 5%
Ql̇ = hd Aρair hfg (ω − ωs )
where ρair is the dry air density and ωs is the humidity ratio at the water droplet surface. ωs is evaluated at 100% relative humidity and water temperature. For 15°C spray water temperature, approximately 5.4 g watervapor/kg dry-air humidity ratio decrease is calculated when spray dehumidification module is in function. Since vapor in moist air condenses on the water droplets, a reduction in the humidity ratio is obtained. The condenser unit of the setup provides additional condensation. The overall dehumidification significantly enhances when the spray module and air cooled condenser work together.
(1)
here PB is the local atmospheric pressure and Pw is the vapor pressure:
Pw =
Pws ∙RH 100
(2) 3.1. Uncertainty analysis
In Eq. (2), Pws indicates the saturated vapor pressure which is function of temperature. The sensible heat transfer from moist air to a water droplet can be expressed as;
Qṡ = hA (Tair − Twater )
(4)
The overall uncertainty of the experimental results were computed by considering the measurement errors. The moist air temperature and the relative humidity are the major source of the errors. Rotronic
(3)
Fig. 2. Experimental apparatus. 165
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 3. Interface of the SCADA system.
For a typical experiment with 65 °C and 80% RH inlet conditions, the overall uncertainty of the humidity ratio calculations was estimated to be 0.0034 kg/kg based on Eq. (5). Thus, a typical experimental error associated with the humidity ratio values is nearly 2.2%. Table 4 shows the standard deviation of repeatability tests. The results indicate that the experimental results are reproducible by % 0.6 maximum deviation. The parameters affecting the dehumidification performance of the spray module are investigated in the following sections. The performance of the spray dehumidification system at different air inlet conditions was studied first. Then the experiments were carried out in order to explore the influence of spray injection direction, spray droplet diameter, spray cone angle, water flow rate and temperature on the dehumidification process. Fig. 4. Spray apparatus.
3.2. Spray dehumidification at different air inlet conditions Hygroflex4 probes provide ± 1% relative humidity and ± 0.2°C temperature measurement accuracy. The experimental uncertainty [19] in the humidity ratio is given by Eq. (5).
wω =
2 2 ⎛ ∂w wRH ⎞ + ⎛ ∂w wT ⎞ RH T ∂ ∂ ⎝ ⎠ ⎝ ⎠
In order to evaluate the performance of the spray dehumidification process at different air inlet RH values, two cases (Exp. 3 & 4) with RH = % 80 and RH = % 50, both having 65 °C inlet temperature, were considered. The temperature of the water spray was 19 ± 1 °C for both cases. The inlet RH and temperature data were taken at nearly 30 cm above (Probe 1) the spray module. The computed humidity ratio values
(5)
Table 2 Experimentally studied parameters. Experiment
Temperature (°C)
Relative Humidity (%)
Flow Direction
Water Temperature (°C)
Water Mass Flow Rate (l/h)
Spray Nozzle Type
Spray Cone Angle (°)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
65 65 65 65 65 65 65 65 65 65 65 75 75 75 65 65 75 75
80 80 80 50 80 80 80 80 80 80 80 80 80 80 80 80 80 80
Counter Counter Counter Counter Counter Counter Parallel Parallel Counter Counter Parallel Counter Counter – Counter Counter Counter Counter
15 15 18 20 22 22 23 23 22 23 22 10 10 – 10 25 15 25
10 10 10.9 10.3 10 10 10 10 10 5 5 1 10 0 10 10 1 1
Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Hydraulic Air Atomizer Hydraulic – Hydraulic Hydraulic Air Atomizer Air Atomizer
65 65 65 65 65 65 65 65 105 65 65 30 65 – 65 65 30 30
166
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 5. Moist air temperature as a function of time.
were presented in Table 3. The spray dehumidifier performs well for RH = % 80 case. At high relative humidity, the vapor pressure is higher. Thus higher mass transfer rate is achieved between moist air and water droplets. Relatively reduced vapor pressure for RH = % 50 induces significantly low mass transfer. Here it is interesting to note that the performance of the condenser does not significantly differ for low and high RH cases. For both cases the amount of dehumidification at the condenser is approximately 10%. Table 5 shows the variation of the moist air temperature after the spray injection at different relative humidity values. For the same inlet temperature and nearly constant water temperature and flow rate in Exp. 3 & 4, the temperature drop at the spray module is noticeably higher for low RH (% 50) case. This is due to the fact that the evaporation rate of the droplets is greater at low RH. For both high and low RH cases, the temperature drop at the air cooled condenser rather small in comparison to the spray dehumidification module. In order to fully identify the effect of spray dehumidification process on the relative humidity, RH values were evaluated at the inlet temperature of the humid air (artificial reheat process to inlet temperature) by using the humidity ratio values presented in Table 3. The results are reported in Table 6. The relative humidity decrease at the spray module is 1.7 and 1.1 for RH = % 80 and RH = % 50 cases (Exp. 3 & 4), respectively. It is obvious that the spray dehumidifier performs better in terms of relative humidity reduction at high RH. Eqs. (6)–(7) are used to calculate the required thermal energy for reheating the dehumidified air to inlet temperature (65°C).
Table 3 Spray dehumidification performance at different conditions. Experiment
1 2 3 4 5 6 7 8 9 10 12 13 14
Average humidity ratio (g/kg) Air inlet
Spray outlet
Condenser outlet
Difference (spray)
Difference (spray + condenser)
152.7 153.2 153.2 87.4 152.9 152.9 152.7 153.2 152.9 152.9 271.9 271.5 271.5
146.9 148.2 149.2 85.6 145.1 146.7 146.2 149.0 145.1 146.7 265.3 249.4 –
134.5 135.2 136.0 77.7 132.9 133.2 133.5 135.5 132.9 133.2 245.9 233.3 254.1
5.8 5.0 4.0 1.8 7.7 6.2 6.6 4.2 7.7 6.2 6.6 22.1 –
18.2 18.0 17.2 9.7 20.0 19.7 19.3 17.7 20.0 19.7 26.0 38.2 17.4
Table 4 Standard deviation in the experiments. STANDARD DEVIATION (%) Exp. 1&2
Air inlet 0.2
Fig. 6. Relative humidity as a function of time. 167
Spray outlet 0.6
Condenser Outlet 0.4
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
state of 65°C in Exp. 3. When the spray module and the air cooled condenser run simultaneously, the required heat rate increases to 177 W.
Table 5 Variation of air temperature after spray injection at different relative humidity values, water flow rates and water temperatures. Average temperature (°C) Experiment
Spray outlet
Condenser outlet
3 4 6 8 10 11 15 16
58.9 50.6 59.3 59.6 59.8 60.0 59.1 59.3
57.9 50.2 58.3 58.8 59.5 59.6 58.1 58.6
3.3. Effect of injection direction on the spray dehumidification process The injection configuration of water spray into moist air affects the dehumidification process only to a limited extent. Table 3 also reports the humidity ratio values at two different flow configurations. In parallel setting (Exp. 7 & 8) the water is injected at the same direction to the air flow. For counter flow configuration (Exp. 5 & 6) the injection is opposite to air flow direction. The volumetric flow rate of water is kept constant at 10 l/h during the experiments. The humidity ratio reduction at the spray module is slightly higher for the counter flow configurations. Between the inlet and spray outlet, on the average 5.4 g/kg and 7 g/kg reduction in the humidity ratio was measured for parallel and counter flow configurations, respectively. Thus the spray module at counter flow configuration performs slightly better due to greater relative velocity of spray droplets in air. El-Morsi [16] reported an increase of the mass transfer coefficient by increasing droplet relative velocity. The results of the present experiments are consistent with theoretical findings. The behavior is more obvious at a longer distance. Between the inlet and condenser outlet, the reduction in the humidity ratio was increased from 18.5 g/kg to 19.8 g/kg when counter flow configuration was utilized instead of the parallel one.
Table 6 Spray dehumidification performance in terms of relative humidity. Relative humidity at the Inlet Temperature (%) Experiment
Spray outlet
Condenser outlet
Difference (spray)
Difference (spray + condenser)
3 4 12 13 14 15 16
78.3 48.9 78.4 75.0 – 74.8 78.7
72.7 44.9 74.3 71.5 76.2 70.3 73.1
1.7 1.1 1.6 5.0 – 5.2 1.3
7.3 5.1 5.7 8.5 3.8 9.7 6.9
3.4. Effect of water spray cone angle on the spray dehumidification process
Q̇ = ṁ m cp, m ∆T
(6) The effect of water spray cone angle on the spray dehumidification performance is studied by utilizing two nozzles having different cone angles, 105° (Exp. 9) and 65° (Exp. 6). The volumetric flow rate of water is kept constant at 10 l/h during the experiments. The results presented in Table 3 reveal that increasing water spray cone angle slightly augments the dehumidification at the spray module. This can be attributed to the larger interfacial area between moist air and water droplets. Larger total interfacial area enhances both heat and mass transfer. However the dehumidification enhancement can be considered to be marginal.
here, ṁ is the mass flow rate of the humid air and ΔT is the temperature increase. The temperature increase was calculated for both spray outlet and the condenser outlet. The specific heat of the air-vapor mixture at constant pressure calculated by;
cp, m = cp, a + ω∙cp, w
(7)
The subscripts m, a and w denote the air-vapor mixture, the dry air and the water vapor, respectively. When the spray module operates alone, 152 W heat transfer rate is required in order to bring the humid air at the spray outlet to initial
Fig. 7. Effect of water flow rate on dehumidification process in terms of average humidity ratio. 168
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
3.5. Effect of water flow rate on the spray dehumidification process
Spray surface area ∝ ṁ water /d
In order to investigate the effect of water mass flow rate on the spray dehumidification process two different mass flow rate are considered, 10 l/h and 5 l/h. Both parallel and counter flow conditions were studied for 65 °C temperature and 80% relative humidity air inlet conditions. The spray water temperature was 22.5 ± 0.5 °C in all experiments. Fig. 7 depicts the humidity ratio values achieved by 5 l/h and 10 l/h flow rates. As the water spray mass flow rate increases, the humidity ratio of air significantly decreases due to larger total droplet surface area. An increase of water flow rate from 5 l/h to 10 l/h increases the amount of dehumidification approximately 4 g/kg at the spray module for both parallel and counter flow configurations. The increase of the humidity ratio reduction for the counter flow configuration (reported in Table 3) between the inlet and condenser outlet sections was not measured for 5 l/h spray water flow rate. A temperature decrease of air is also measured by increasing water flow rate due to sensible heat transfer for both counter (Exp. 10 & 6) and parallel flow (Exp. 11 & 8) cases (Table 5). An increase of spray water flow rate from 5 l/h to 10 l/h reduces the temperature by 0.5 °C and 0.8 °C at spray and condenser outlets, respectively for both parallel and counter flows. The total temperature decrease of humid air between the inlet and condenser outlet sections is approximately 5.5 °C for 5 l/h flow rate. When the flow rate was increased to 10 l/h, the temperature decrease increased to 6.2 °C for parallel flow configuration. For counter flow setting the temperature reduction is even higher (6.7 °C).
Since smaller droplets form greater spray surface area, rate of vapor condensation on droplet surfaces enhances when the air atomization nozzle is used. El-Morsi (2002) reported a decreasing spray chamber effectiveness by increasing droplet diameter. Table 6 shows the relative humidity values (for the moisture content given in Table 3) calculated at 75°C inlet temperature. Without the spray module (Exp. 14) the relative humidity reduces from 80% to 76.2%. When the spray module is utilized (Exp. 13) 71.5% relative humidity is achieved. If the total (spray module plus air cooled condenser) humidity reduction values are compared, it is obvious that the air atomizer nozzle is as effective as hydraulic nozzle with only 1/10th of water consumption. 3.7. Effect of water spray temperature on the spray dehumidification process 10 °C (Exp. 15), 15 °C (Exp. 1) and 25 °C (Exp. 16) sprayed (by hydraulic nozzle) water temperatures were considered in order to explore the effect of water temperature on dehumidification. The results are presented in Fig. 8 and Table 5. When the spray module injects water at 25 °C, nearly 6 g-water vapor/kg-dry air condensation is obtained while 12 g-water vapor/kg-dry air condensation is achieved by water at 10 °C. As the water spray temperature decreases, the difference between the dew-point temperature and droplet temperature increases. The difference between the moist air humidity ratio and humidity ratio at the droplet surface also increases. Thus the mass transfer (condensation) rate enhances. As Table 5 demonstrates the variation of the temperature drop with water temperature at the spray module is insignificant. This indicates the stronger influence of latent heat transfer (Eq. 4) compared to sensible heat exchange (Eq. 3). For 10°C water temperature cases, humidity ratio at the droplet surface is lower. Thus the decrease in the humidity ratio is more pronounced (nearly 12 g water-vapor/kg dry-air) due to higher mass transfer rate. If the relative humidity is evaluated at 65°C based on the humidity ratio values presented in Fig. 8, it is found that remarkably high condensation at the spray module is obtained with 10°C water (Table 6). The effect of water temperature was examined for the dehumidification performance of spray module utilizing air atomization nozzle as well (Fig. 9). For these group of experiments the spray water flow rate was 1 l/h. 10 °C (Exp. 12), 15 °C (Exp. 17) and 25 °C (Exp. 18) sprayed water temperatures were considered. For the spray module injecting water at 25 °C, nearly 3 g-water vapor/kg-dry air condensation is obtained. The condensation increases to approximately 6 g-water vapor/ kg-dry air by water at 10 °C. The water temperature has similar influence on condensation for air atomization nozzle and hydraulic nozzle. Figs. 10–13 depict the effect of various parameters on the variation of humidity ratio along the process line where the condensation occurs.
3.6. Effect of spray nozzle type on the spray dehumidification process In order to investigate the effect of spray nozzle type on dehumidification process, in addition to hydraulic nozzle, air atomizer nozzle was utilized at the spray module in Exp. 12. The hydraulic nozzle atomizes liquid water by using water line pressure while the air atomizer nozzle uses pressurized air for droplet formation. The spray droplet diameter varies with water mass flow rate. In this study, the average droplet diameter is nearly 100 μm for hydraulic nozzle at 10 l/ h. The average droplet diameter reduces to nearly 15 μm for air atomizer nozzle at 1 l/h [20]. Because of the fact that decreasing mass flow rate of water spray leads to reduction in droplet size, the air atomizer nozzle was used at the minimum flow rate. The experiments are performed at identical water spray temperatures for the hydraulic (Exp. 13) and the air atomizer (Exp. 12) nozzles in counter flow direction and the results are tabulated (Table 3). Table 3 also includes the result of an experiment where the spray module was not utilized (Exp. 14). This case corresponds a traditional dehumidification process provided by an air cooled condenser only. The dehumidification at the spray module is 22.2 g-water vapor/kgdry air with the hydraulic nozzle (Exp. 13). In Exp. 14, the dehumidification at the condenser is 17.4 g-water vapor/kg-dry air without the spray module. This value is 6.6 g/kg for the air atomizer nozzle. Here it is important to note that air atomizer nozzle consumes one tenth of the water the hydraulic nozzle injects. Thus it can be concluded that the humidity reduction per unit water flow rate is much higher for the air atomizer nozzle. The surface area of the spray is proportional to the square of the droplet diameter and the total number of droplets.
Spray surface area ∝ d2N
(10)
(8)
The number of droplets depends on the mass flow rate of water and the volume of a droplet;
N∝
ṁ water d3
(9) Fig. 8. Variation of the humidity ratio with sprayed water temperature for hydraulic nozzle.
Thus the total surface area of the spray increases by the water mass flow rate and decreases by the droplet diameter; 169
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Fig. 9. Variation of the humidity ratio with sprayed water temperature for air atomization nozzle.
Fig. 12. Effect of water spray mass flow rate and temperature on relative humidity.
Fig. 10. Effect of air inlet conditions on spray dehumidifier performance.
In Fig. 10, the humidity reduction at 50% RH is much slower along the process line. For 80% RH case, the spray module and the air cooled condenser have nearly same effect on dehumidification. The humidity ratio values along the process line are negligibly affected by the flow configuration (counter/parallel) as shown in Fig. 11. For the nearly same moist air inlet conditions, the calculated humidity ratio values after the spray module for parallel and counter flow settings differs by maximum 1.5% deviation at low and high water flow rate cases. Similar behavior was observed at the condenser outlet section. On the other hand, the dehumidification process is strongly affected by the spray water flow rate and the water temperature as shown in Fig. 12. In order to clearly demonstrate the effect, the results are presented in terms of relative humidity evaluated at 65 °C inlet temperature. Thus
Fig. 13. Effect of nozzle type on humidity ratio.
the values correspond to dehumidified and reheated moist air conditions. The moist air enters the inlet section at 65°C and 80% relative humidity in all cases. The spray (with hydraulic nozzle) setting is counter flow configuration. For the spray flow rate experiments the water temperature was 22.5 ± 0.5°C. For the spray temperature experiments the flow rate was 10 l/h. As the figure indicates the water temperature is the most influential parameter on the dehumidification. The relative humidity reduces to nearly 70.3% when water at 10°C was injected to the air stream with 10 l/h flow rate. The dehumidification process is also affected by the nozzle type as
Fig. 11. Effect of water spray injection direction on humidity ratio. 170
International Communications in Heat and Mass Transfer 97 (2018) 163–171
S. Gumruk et al.
Ql Qs Ra RH ΔT T T· b V ω ρm ρda
reported in Fig. 13. The moist air enters the inlet section at 75°C and 80% RH for both hydraulic and air atomization utilized experiments. The spray water temperature was 10°C. The decrease of humidity ratio at the condenser outlet is approximately 38 g/kg and 26 g/kg for hydraulic and air atomization nozzles, respectively. Thus, when the water consumption was considered the spray module employing air atomization nozzle has superior dehumidification performance. 4. Conclusions The spray dehumidification process in a turbulent channel flow experimentally investigated by a parametric study. The experiments results revealed that the spray dehumidification module can be as effective as a conventional air-cooled condenser in dehumidifying of moist air. Based on the measurements of the present research one may conclude:
Acknowledgements The authors gratefully acknowledge the financial support for this work by Ministry of Science, Industry and Technology of Turkey under Grant# 01488.STZ.2012-2.
- Water flow rate and the temperature are the most critical parameters on the dehumidification performance of the spray module. - An increase of water flow rate from 5 l/h to 10 l/h increases the total amount of dehumidification between the inlet and condenser outlet sections by 24% in counter flow configuration. For parallel flow configuration, the increase is 11%. - Spray water temperatures significantly affect the dehumidification performance. A decrease of sprayed water temperature from 25 °C to 10 °C nearly doubles the amount of dehumidification at the spray module for both hydraulic and air atomization nozzles. - The injection direction of the spray have relatively small effect on dehumidification performance. For the nearly same moist air inlet conditions, approximately 7% higher dehumidification was achieved between the inlet and condenser outlet with 10 l/h spray flow rate. For 5 l/h spray flow rate, flow configuration did not alter the amount of dehumidification. - The spray cone angle has relatively small effect on dehumidification performance. - The spray module utilizing air atomizer nozzle has superior dehumidification performance per unit water flow rate.
References [1] Y. Kim, J. Park, C. Song, Investigation of the stem-water direct contact condensation heat transfer coefficients using interfacial transport models, Int. Commun. Heat and Mass Transf. 31 (2004) 397–408. [2] M. El-Morsi, S.A. Klein, D.T. Reindl, Air washers: a new look at a vintage technology, ASHRAE J. 45 (10) (2003) 32–36. [3] G. Brown, Heat Transmission by Condensation of Steam on a Spray of Water Drops, Proc. of General Discussion on Heat Transfer, Institution of Mechanical Engineers, 1951, pp. 49–52. [4] J.D. Ford, A. Lekic, Rate of growth of drops during condensation, Int. J. Heat Mass Transf. 16 (1972) 61–64. [5] E. Kulic, E. Rhodes, Direct contact condensation from air-steam mixtures of a single droplet, Can. J. Chem. Eng. 55 (1977) 131–137. [6] C.T. Crowe, M. Sharma, D.E. Stock, The particle source-in-cell (psi-cell) model for gas droplet flows, J. Fluids Eng. 99 (1977) 325–352. [7] S.Y. Lee, R.S. Tankin, Study of liquid spray (water) in a non-condensable environment (air), Int. J. Heat Mass Transf. 27 (1983) 331–363. [8] T. Sundararajan, P.S. Ayyaswamy, Hydrodynamics and heat transfer associated with condensation on a moving drop: solutions for intermediate Reynolds numbers, J. Fluid Mech. 149 (1984) 33–58. [9] G.P. Celata, M. Cumo, G.E. Farello, G. Focardi, A comprehensive analysis of direct contact condensation of saturated steam on subcooled liquid jets, Int. J. Heat Mass Transf. 32 (1989) 639–654. [10] G.P. Celata, M. Cumo, E. D'Annibale, G.E. Farello, G. Focardi, A theoretical and experimental study of direct-contact condensation on water in turbulent flow, Experim. Heat Transf. 2 (1989) 129–148. [11] F. Mayinger, A. Chávez, Measurement of direct-contact condensation of pure saturated vapour on an injection spray by applying pulsed laser holography, Int. J. Heat Mass Transf. 35 (1992) 691–702. [12] M. Akira, H. Fiji, S. Takamoto, Prediction of heat transfer by direct contact condensation at a steam-subcooled water Interface, Int. J. Heat Mass Transf. 35 (1) (1992) 101–109. [13] D.T. Reindl, Combustion Turbine Inlet Air Cooling Using Thermal Storage & DirectContact Sprays, Proceedings of the EPRI International Sustainable Thermal Energy Storage Conference, 147–150, Minneapolis, MN, August, (1996). [14] S.S. Kachhwaha, P.L. Dhar, S.R. Kale, Experimental studies and numerical simulation of evaporative cooling of air with a water spray-I. Horizontal parallel flow, Int. J. Heat Mass Transf. 41 (2) (1998) 447–464. [15] M. Takahashi, K.A. Nayak, Heat transfer direct contact condensation of steam to subcooled water spray, J. Heat Transf. 123 (2001) 703–710. [16] M.S. El-Morsi, Optimization of Direct Contact Spray Coolers, Ph.D. thesis University of Wisconsin- Madison, 2002. [17] J. Malet, P. Lemaitre, Water Spray Interaction with Air-Steam Mixtures Under Containment Spray Conditions: Comparison of Heat and Mass Transfer Model With the TOSQAN Spray Tests, the 11th International Topical Meeting on Nuclear Reactor Thermal-Hydraulics, Avignon, France, (2005). [18] S. Gumruk, M.K. Aktas, Experimental Study of Direct Contact Condensation of Steam on Water Droplets, Proceedings of the World Congress on Engineering 2015 Vol II, July 1–3, 2015, London, UK, (2015). [19] J. Holman, Experimental Methods for Engineers, 8th Edition, McGraw-Hill, 2016. [20] Spraying Systems, Spray Technology Reference Guide Understanding Drop Size, (1996).
The results of the present investigation may be useful in the design of energy efficient drying or dehumidifying processes for industrial applications and household appliances such as tumble dryers. A future investigation will focus on the spray dehumidification process utilizing air atomization nozzles that can be used only at special applications since they require pressurized air. Nomenclature A cpm cpa cpw d h hd hfg ṁ m N Pws Pw PB
Latent heat transfer (W) Sensible heat transfer (W) Gas constant (J/kgK) Relative humidity Temperature difference (°C) Temperature (°C) Bulk temperature (°C) Volumetric flow rate of air-vapor mixture (l/s) Humidity ratio (kg/kg or g/kg) Density of air-vapor mixture (kg/m3) Density of dry air (kg/m3)
Surface area (m2) Specific heat of air-vapor mixture at constant pressure (J/ kgK) Specific heat of dry air at constant pressure (J/kgK) Specific heat of water vapor at constant pressure (J/kgK) Droplet diameter (m) Heat transfer coefficient (W/m2K) Mass transfer coefficient (m/s) Latent heat of condensation (J/kg) Mass flow of air-vapor mixture (kg/s) Number of droplets Saturated vapor pressure (Pa) Vapor pressure (Pa) Atmospheric pressure (Pa)
171