Pre-cooling water applied to porous plates of evaporative cooling systems. Pre-cooling water applied to porous plates of evaporative cooling systems

: This study assessed the ways the thermal environment is influenced by pre-cooling the water employed in wetting the porous plates present in the evaporative cooling systems (ECS). The experiment was performed using two physical models constructed on a distorted and reduced scale, which simulated closed agricultural facilities equipped with ECS made up of porous cellulose panels. In one physical model, the plates were made wet using chilled water (ECS cw ), while in the other they were moistened using natural water, at environment temperature (ECS nw ). Both inside the physical models and outside, in the external environment, the air dry-bulb temperature (t db ), black globe temperature (t bg ) and air relative humidity (RH) were measured, at 10 sec intervals. Later, the environmental indices, ECS nw cooling effectiveness and ECS cw performance factor were assessed. When the porous ECS plates were thoroughly wet using pre-cooled water, lowering of the thermal variables and comfort indices was seen to be greater than when wetting the plates was done using water at room temperature. An empirical equation was proposed to determine the cooling performance factor applied to the ECS cw related to water temperature, t db and air wet-bulb temperature.


INTRODUCTION
The problems involved in controlling the thermal environment inside agricultural facilities has directed research towards the state of ambience (CURI et al., 2014;WATANABE et al., 2018). This control; although, crucial in regions experiencing hot climates, continues to be stressful for breeders.
However, for the achievement of positive outcomes in the agricultural sector, such control becomes imperative, in order to ensure that the animals throughout their production cycle, can be reared in their thermal comfort zone, thus ensuring higher rates of production (SCHIASSI et al., 2015;LOPES et al., 2020).
This sector has witnessed a gradual transition from the popularly termed conventional Fernandes et al. open breeding systems and the employment of artificial temperature control only when required, to systems enclosed by curtains or masonry that, besides the climate control processes, provide the advantages of lowering the internal temperature of the animal growth environments.
Evaporative techniques, using nebulization or moistened porous plates have been proven as the most effective and economically feasible options in closed aviaries to decrease the internal temperatures (DAMASCENO et al., 2010). The wet porous plates, more than the nebulization method, have higher aircooling capacity, which may differ, based on the various materials used in their construction (TINOCO et al., 2002;PANAGAKIS and AXAOPOULOS, 2006).
From the commercial perspective, cellulose is the most prominent of the different materials used to manufacture the evaporative cooling plates. However, good alternative substitutes for these pads, are expanded clay and vegetable fibers, to ensure higher performance or lower expenditure (TINÔCO et al., 2002;ROSA et al., 2011).
As the air temperature increases and the air humidity outside the premises decreases, the cooling effectiveness also rises. However, works that assessed the temperature of the water used to wet the porous plates employed in animal husbandry to facilitate better thermal indices are very few in the literature.
Therefore, the goal of this research was to determine the effect on the thermal environment, which pre-cooling the water utilized to wet the porous plates of the evaporative cooling systems could exert (ECS).

MATERIALS AND METHODS
The experiment was carried out using two identical physical models constructed on a site, at the geographic coordinates of 21º13' 45.2'S latitude; 44º58'32.85''W of longitude and 918 m of altitude. Experiencing the Cwa type of climate (based on the Köppen classification), the place revealed rainy and temperate weather conditions (mesothermal), with dry winters and rainy summers, subtropical, with higher than 22 °C temperature during the warmest month (DANTAS et al., 2007).

Physical models and experimental treatments
The two physical models, identical in their construction properties, are typical of the enclosed agricultural facilities in animal husbandry ( Figure 1). The physical models were constructed on the scales mentioned: width (W), length (L) and height (H) of 1:10, 1:25 and 1:2, respectively, thus ensuring the internal dimensions of 1.5 m (W) x 6 m (L) x 1.5 m (H). The physical models had roof ridges oriented in the true East-West direction. The main aim of the physical models, constructed in a suitable site, was to facilitate the placement of evaporative porous plates and exhaust fans, ensuring free airflow through them. The first physical model had a cellulose pad cooling system using natural water (ECS nw ), while the second had the cellulose pad cooling system but with chilled water (ECS cw ) being used instead.

Acclimatization systems installed on physical models
Each physical model had, at one end, a porous commercial cellulose pad (1 m wide x 1 m high x 0.15 m thick) installed, while at the other end three exhausts (flow rate of 4200 m³ h -1 each) were provided. This acclimatization system is normally employed in the agricultural sector, as in the wind tunnel type commercial aviaries. The evaporative cooling system was designed based on VANTRESS (2018).
In the ECS cw , the water-cooling system installed included a hermetic compressor, air condenser (0.56 W), thermal box (77 L capacity) and a pump (0.37 kW and 2400 L h -1 ). Water flow rates of 0.17±0.004 L s -1 and 0.17±0.003 L s -1 , respectively for the ECS cw and ECS nw , were pumped to the upper part of the cellulose pad. Water volume in excess was returned to the reservoir to be reused. In the ECS cw physical model, water cooling was controlled using the digital thermostat model TIC-17RGTi), manufactured by Full Gauge Controls (±1 °C accuracy).

Measurements and instrumentation
To analyze the cooling capacity of the air flow through the moistened porous plates, the internal and external thermal environments were evaluated, for each physical model.
Inside the physical models, constructed on a reduced scale, five recording thermo-hygrometers were positioned, model U12-001, manufactured by Onset, to ensure the measurement of the air dry-bulb temperature (t db , °C) and air relative humidity (RH, %), with ±0.35 °C and ±2.5% accuracy for the RH values in the 10 to 90 % range, respectively. The black globe temperature (t bg , °C) was measured at the center of the porous plate alone (Figure 1), while an external temperature sensor, model TMC6-HD, manufactured by Onset, was positioned in the center of a black plastic globe, 3.6 cm in diameter (SANTOS et al., 2005), and painted with matte black paint on the outside. This external temperature sensor was hooked up to the external channel of the recording thermohygrometer. The responses from the thermometers installed in the plastic black globe were confirmed against a standard black globe thermometer, built using a copper plate 15 cm in diameter and 0.5 mm thick, painted in matte black color.
Five thermo-hygrometers were fitted 30 cm from the porous plate, one each at 20 cm from both ends and one in the center. The thermo-hygrometers were placed 30 cm from the porous plates to ensure constructed with a cellulose board cooling system using natural water (ECS nw ) and chilled water (ECS cw ).
The dots in Section AB indicate the measurement positions of the temperatures on the air dry-bulb and black globe (t db and t bg , respectively) and air relative humidity (RH). Detailing without scale.
that the condition of the air immediately post cooling after passing through the wet porous plates, is not wet with the water droplets from the plates. Apart from the external channels of these logging instruments, the wetting water temperatures (t water ) of the cellulose plates inside both the ECS cw and ECS nw reservoirs were also recorded. Using the recording thermohygrometer in the external environment, the t db , t bg and RH were determined. All the specified measurements cited earlier were taken at 10 second intervals.
Using a Highmed manufactured model HM-385, hot-wire anemometer (accuracy ±5% +0.1 m s -1 ) the values of V air were noted in nine distinct positions, within each physical model ( Figure 1) at 75 cm height, always at the commencement of each experiment day.
Measurements of the surface temperature of the external surface of the moistened cellulose panels (t ps ) were taken at 1-minute intervals, with an infrared thermographic camera, model Ti 55 manufactured by Fluke (accuracy of 0.05 ºC).
The THI is a thermal comfort evaluation index dependent upon the t db and air dew-point temperature (t dp ) (THOM, 1958). In a single value, the BGHI surveys the effects of t db , radiation and air flow via the t bg and air humidity through the t dp Table 1 -Equations for calculation of the thermal environment evaluation indices, thermodynamic properties of the air and effectiveness of the evaporative cooling of the air.

Nº
Equation Legend 1 THI: Temperature humidity index (dimensionless); tdb: air dry-bulb temperature (º C); tdp: air dew-point temperature (° C) 2 BGHI: black globe temperature index and humidity (dimensionless); tbg: black globe temperature (° C); tdp: air dew-point temperature (° C) 3 h: specific enthalpy of air (kJ kgdry air -1 ); tdb: air dry-bulb temperature (ºC); RH: air relative humidity (%); pB: local barometric pressure (mmHg) 4 ε = ε: Evaporative cooling effectiveness; tdbi: air dry-bulb temperature at the evaporative plates' inlet (°C); tdbo: air dry-bulb temperature at the evaporative plates'outlet (°C); twbi: air wet-bulb temperature at the evaporative plates' inlet (°C) 5 F: Evaporative cooling performance factor (decimal); hi: enthalpy of the air at the evaporative plates' inlet (kJ kgair dry -1 ); ho: enthalpy of the air at the evaporative plates' outlet (kJ kgair dry -1); ha: saturated air enthalpy calculated considering tdb equal to the water temperature at the evaporative plates'inlet (kJ kgair dry -1 ) 6 VPD: vapor pressure deficit (kPa); RH: Relative air humidity (%); Pws: air saturation pressure proposed by Tetens for air dry-bulb temperatures above 0º C (equation 6);  (BUFFINGTON et al., 1981). Enthalpy is a physical quantity, expressed in the realm of thermodynamics, which quantifies the energy of a system (ALBRIGHT, 1990). The cooling effectiveness or saturation efficiency indicates the ratio between the lowered t db recorded and the maximum that can be theoretically achieved (CAMARGO, 2008;RIANGVILAIKUL & KUMAR, 2010). The performance factor F, pertinent to non-adiabatic cooling processes is the ratio between the enthalpy reduction observed and the maximum theoretically achievable (ASHRAE, 1997). The VPD refers to the difference between the water vapor pressure during saturation conditions and the current water vapor pressure at a specific temperature (YUAN et al., 2019). The air wet-bulb temperature (t wb ) was calculated analytically by the equations that define the psychrometric properties of the air (WILHELM, 1976).

Experimental phases, experimental design and statistical analysis
To confirm the presence of equivalence between the two ECS fitted in the physical models, experiments were conducted prior, for three nonconsecutive days, where similar conditions required for the system to function were maintained. This step was termed the 'pre-test phase'.
Subsequently, nine non-consecutive days, the systems (ECS nw and ECS cw ) were evaluated, during the hottest part of the day (from 12:00 to 15:45), termed the 'test phase'. The t water values during the course of the tests were in the 18.3 to 21.5 °C range for the ECS nw and from 8.8 to 16.1 °C for the ECS cw . Analysis was done each day, for three operating cycles of the ECS, with and without cooling the water used to wet the plates. Each cycle took 1 hour to complete, which included 15 min of ECS operation and 45 min of waiting, to cool the water down to as close to the desired temperature as possible.
From among all the 15 min of data collected, only the data drawn during the intermediate 5 min were statistically analyzed. This procedure enabled higher stabilization of the ECS because in the initial stage the evaporative cooling plate was kept wet and, during the final 5 minutes, there was a trend of the surface temperature of the plate to rise due to the inability of the water cooling system to maintain the water at constant. All the data recording was done in 10 second intervals.
The completely randomized design was adopted (CRD) with the water temperature represented as a single factor, subdivided into two levels (ECS nw and ECS cw ), used in the physical models. The response variables, pertaining to the thermal comfort in these two situations, were evaluated, giving 27 repetitions in all, for each one.
Using the Student's t test for the independent data with 5% significance level these variables were analyzed between the two levels of the factor. The fit of the models was assessed based on the regression analysis of variance and the coefficient of determination (R²). Adjustment was done for the multiple linear regression models to evaluate the cooling performance factor for the ECS cw as a function of the t water , thermal environment variables, thermal environment evaluation indices and enthalpy, and the correlation coefficients were determined.
Employing the R statistical computational system (R DEVELOPMENT CORE TEAM, 2019), all the statistical analyses were done.

RESULTS AND DISCUSSION
The equivalence present between the cellulose board cooling system using natural water (ECS nw ) and the cellulose board cooling system utilizing chilled water (ECS cw ) was obvious during the pre-test phase (Table 2) because, on analysis, all the variables were statistically equal (P > 0.05, t test). The reduction of t db , THI, BGHI and h showed values of -7.1 and -6.7 ºC, -6.2 and -6.0, -10.0 and -10.4, -3.5 and -3.1 for the ECS nw and ECS cw , respectively. The ε values for the ECS nw and ECS cw were 79.7 and 75.5, respectively, clearly indicating that the responses of the cooling systems were similar and that it is the exclusively the water temperature variations that induce the changes.
During the test phase, statistical differences were confirmed, revealing improved ECS cw performance when the means of the t db , t dp and t water variables, and the THI and BGHI indexes and the thermodynamic property h (P < 0.01, test t) were analyzed. The t bg also displayed identical profile (P < 0.05, t-test). In turn, the variables UR and t ps were statistically equal (P > 0.05, t test).
During the days that the experiments were performed, when the t db and RH mean values of the external environment were 31.2 ºC and 40.9%, respectively, the ECS nw induced a 6.8 ºC decrease in the t db , with a corresponding 29.1 % increase in the RH, while the ECS cw caused the t db to decline at 8.1ºC and the RH to show a rise of 28.3 % in (Table 2). For a closely similar thermal condition (32 ºC and 40 %), a system made up of a cellulose evaporative plate can lower by 7.7 ºC the internal temperature of aviaries, while causing a 34.6% increase in the RH (VANTRESS, 2018). Fernandes et al. In their investigations of a variety of configurations for the evaporative cooling system using computer simulations, CARVALHO et al. (2009) concluded that in September, for the study duration, a tunnel mode ventilation system related to a porous material type evaporative cooling system, moistened at 70% efficiency, lowered the THI by 1.8±1.4 when compared to a negative pressure wind tunnel ventilation system lacking the evaporative cooling. In the current study, the ECS cw induced a 2.0±0.8 reduction when compared to the ECS nw , for the same index being investigated.
In their research, in Brazil, TINÔCO et al. (2002) compared a few of the available alternative porous materials, and reported BGHI values roughly 2.5 lower in the expanded clay-filled slabs than for those filled with sawdust, vegetable fiber and charcoal. A difference of 2.2, on average, was achieved favoring the ECS cw , suggesting this as another possibility to boost the cooling performance.
In their endeavor to determine the distribution of illuminance and enthalpy in two aviaries used for rearing broilers, FAUSTINO et al.
(2021) recorded a 0.5 kJ kg -1 reduction, induced by a clay tile-covered brick shed when compared to a straw-covered wooden shed. In this study, the ECS cw registered a 3.9±0.8 kJ kg -1 reduction compared to the ECS nw for the same index investigated.
For ECS nw , the mean value of ε was found to be 72.5 % (Table 2), the range being 52.4 to 83%. ROSA et al. (2011), recorded 78.1 % as the average value of ε for cellulose. Of note, that the value of ε  Δtdb: Difference between indoor and outdoor air dry-bulb temperature (°C). ΔRH: Difference between indoor and outdoor air relative humidity (%). ΔTHI: Difference between indoor and outdoor temperature-humidity index. ΔBGHI: Difference between internal and external black globe-humidity index. Δh: Difference between internal and external enthalpies (kJ kgdry air -1 ). Δtbg: Difference between internal and external black globe temperatures (°C). Δtdp: Difference between internal and external air dew-point temperatures (°C). ε: Evaporative cooling effectiveness (%). F: Evaporative cooling performance factor (%). tps: Surface temperature on the external surface of the wet porous plate (°C). twater: Water temperature (ºC). Significance levels by Student's t test: * (P < 0.05) ** (P < 0.01).
shows variations as a function of the air that flows through the porous plates, the water mass that flows over the porous material surface and its uniformity, besides other variables. For the ECS cw , the mean performance factor (F) was found to be 0.237±0.054. During the tests, the outdoor and indoor air dry-bulb temperatures were in the 28.2 to 35.6 °C range (average of 31.2° C) and from 21.5 to 27.1° C (average of 23.1° C), respectively. The relative humidity of the outdoor air was in the range of 25.8 to 51.4 % (average 40. %) while, for the indoor air, the variation hovered between 53.2 and 76.8%, with 69.1% as the average value. The water temperature was in the 8.8 to 16.1 °C range, with the average at 12.9 °C. In fact, AL-BADRI & AL-WAALY (2017), in their assessment of a direct evaporative cooler for inlet air temperatures from 24.5 to 33.6 °C (average 31.0° C) and water temperatures from 15.8 to 21.0 °C (average 18.7° C), observed that F values occurred between 0.10 and 0.41, showing a mean of 0.214. Thus, among other factors, the greatest value achieved by the average performance factor acquired in this research can be credited to the lowest values attained by the wetting water temperature on the porous plates. Further, these results imply an escalation in the performance factor with lowering of the water temperature, findings which concur with those reported by these authors.
On investigating the multiple regression models among the variables, the best fit model identified to determine the F values was seen as a function of the t water , t db and t wb (equation 8), achieving R² of 0.9753 and with all the variables showing a minimum of 97.52% significance. The standard errors corresponding to the coefficients of the equation for t water , t db and t wb were 0.004267, 0.004019 and 0.005922, respectively. F = -0.010218 • t water -0.017841 • t db + 0.042595 • t wb

CONCLUSION
In the present study, under the specific conditions, pre-cooling the water used to wet the porous plates present in the evaporative cooling systems induced an additional drop in the thermal variables (of 1.4 °C and 1.4 °C for t db and t bg , respectively). With respect to the evaluation of the indices of the thermal environment and the thermodynamic property of the air values of 2.1, 2.2 and 4.0 kJ kg dry air -1 (for THI, BGHI and h, respectively) were observed, related to the wetting of the porous plates, using water at room temperature.
In order to estimate the evaporative cooling performance factor as a function of the temperature of wetting water of the porous plates, as well as the air dry-and wet-bulb temperatures, an empirical equation was fitted.
The electrical energy consumed for cooling the water in the chilled water cellulose board cooling system (ECS cw ) was not determined in the current study because the investigations were performed only on a smallscale model; however, in future it has been suggested to conduct the research in commercial aviaries.

DECLARATION OF CONFLICT OF INTEREST
The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.