Analysis of airflow in a full-scale room with non-isothermal jet ventilation using PTV techniques.
Draft and non-uniform fresh air distribution are common problems in winter ventilation, especially for large animal buildings. Thermal-based anemometers have difficulties in accurately measuring non-isothermal and low-speed indoor airflow. A new technology, particle tracking velocimetry (PTV), which uses particles and their images to study indoor airflow, can overcome the traditional limitations in indoor airflow measurement. A PTV system was used to characterize indoor airflow in a full-scale ventilated room under nonisothermal ventilation conditions. Non-isothermal mild weather and winter ventilation conditions were simulated to analyze their effects on indoor airflow and air velocities in animal occupied zones and human breathing zones. It is found that winter ventilation created a totally reversed rotating airflow pattern. Air velocities in the animal occupied zone increased substantially compared with the corresponding isothermal ventilation conditions. Winter ventilation strategies for improvement of airflow distribution were studied: (1) increasing inlet air velocity, (2) increasing inlet air jet momentum by use of air recirculation devices, and (3) decreasing ventilation temperature difference.
Indoor air quality of large-scale animal production facilities is increasingly recognized as important to the health, well-being, and productivity of building occupants. Ventilation is one of the major means of controlling the indoor environment and indoor air quality. In summer, indoor air temperatures of animal buildings are generally within 3[degrees]C of inlet air temperatures. It is generally accepted that summer ventilation can be assumed "isothermal" for most practical engineering purposes. In mild and cold weather, room air temperatures tend to be much higher than the inlet air temperature due to heat production from animals and equipment. Supply air is not heated during winter for most production animal facilities and thus is much colder than the room air. Therefore, non-isothermal ventilation is typical in realistic buildings during heating seasons. Draft and non-uniform fresh air distribution are common problems with winter ventilation, especially for large-scale animal buildings. Understanding airflow patterns under non-isothermal ventilation conditions, especially those typical in winter, is important in evaluating ventilation system effectiveness and in developing environmental control strategies.
Ventilation research has been extensively studied over the past two decades (Nielsen et al. 1978; Timmons 1984; Sandberg 1987; Zhang et al. 1992; Jin and Ogilvie 1992; Riskowski et al. 1993; Heber and Boon 1993; Wang and Ogilvie 1996). It should be noted that most ventilation study cases simulated isothermal ventilation because it is simpler to measure, model, and analyze. Few studies were conducted on non-isothermal ventilation.
Under non-isothermal ventilation conditions, separation of the inlet air jet from the ceiling is mainly affected by the inertial force of the air jet, which is the result of jet momentum and buoyancy force caused by heating loads. The room inlet Archimedes number (Ar), defined in Equation 1, is the ratio of thermal buoyancy force to inertial force. Air jet separation behavior is determined by Archimedes number. Critical Archimedes number is a limit value at which the diffuser air jet drops immediately after entering the room. Zhang et al. (1992) indicated an equivalent critical Ar value of 0.005 for the test room.
[A.sub.r] = [g[beta][DELTA][T.sub.0][D.sub.0]]/[U.sub.0.sup.2] (1)
[beta] = thermal expansion coefficient of air, 1/K;
g = acceleration of gravity, 9.8 m/[s.sup.2];
[DELTA][T.sub.0] = air temperature difference of inlet air and room air, K;
[U.sub.0] = inlet air velocity, m/s; and
[D.sub.0] = inlet width, m.
Randal and Battams (1979) found that a corrected Archimedes number (Arc) of inlets can predict airflow patterns of buildings under non-isothermal ventilation conditions. Zhang et al. (1992) intensively studied non-isothermal ventilation with slot inlet and outlet and revealed that both the critical Archimedes number (Ar) and the critical Arc for full-scale building ventilation were affected by room height. Heber and Boon (1993) studied air velocity characteristics in a full-scale simulation livestock building with non-isothermal jet ventilation and a high ventilation rate by using three-dimensional ultrasonic anemometry. Wang and Ogilvie (1996) further studied non-isothermal ventilation and developed critical wall jet Archimedes numbers (Arm), which directly reflect the balance of initial force and thermal buoyancy force for full rotating airflow patterns. However, in previous research, non-isothermal ventilation was simulated merely by adding a heat load onto the floor while keeping a normal inlet temperature. The temperature differences were 7[degrees]C-15[degrees]C (45[degrees]F-59[degrees]F). Therefore, typical winter ventilation of animal facilities has not been fully studied. Zhang and Strom (1999) developed jet drop models for non-isothermal free jets without detailed airflow measurement.
Study of non-isothermal ventilation was limited because of the complexity of the problem and the lack of proper measurement technologies. Most research used a hot-wire anemometer to measure airflow. Traditional airflow measurement technology, thermal based anemometers, had inherent difficulty in measuring low-speed and non-isothermal airflow. High temperature sensing heads of thermal anemometers cause significant amounts of free convection and need accurate temperature compensation. For the non-isothermal indoor airflow case studies, both low-speed airflow and temperature fluctuation in the flow field causes measurement difficulties. Zhang et al. (1992) concluded a 25% uncertainty for air velocity measurement using a hot-wire anemometer. Furthermore, thermal anemometers can only measure limited locations with no direct airflow direction measurement.
Particle tracking velocimetry (PTV) technologies use particle streak tracking to measure air velocities of an entire flow field simultaneously. This technology is not limited by low-speed airflow and is not significantly affected by the air temperature. Airflow characteristics of non-isothermal ventilation can be measured more accurately using PTV techniques.
Therefore, the objectives of this study were
* to quantify airflow patterns and air velocity distribution in animal buildings under non-isothermal ventilation conditions using PTV technologies for evaluation of numerical models,
* to characterize airflow of non-isothermal ventilation under typical mild weather and winter weather conditions, and
* to analyze strategies for improvement of airflow distribution under winter ventilation conditions.
EXPERIMENTAL DESIGN AND MEASUREMENT
A room ventilation simulator (RVS) (Wu et al. 1990) was used to simulate the ambient environmental conditions. The RVS consists of a 9.1 x 12.2 x 3.6 m (30 x 40 x 12 ft) outer room, which can simulate weather conditions from -25[degrees]C (-13[degrees]F) to 40[degrees]C (104[degrees]F) any time during the year. Humidity in the outer room can be controlled by a humidifier and dehumidifier within a range of 20% to 90%. A full-scale, adjustable testing room having dimensions of 5.5 x 3.7 x 2.4 m (18 x 12 x 8 ft) and equipped with two ventilation plenums with dimensions of 1.1 x 3.7 x 2.4 m (3.6 x 12 x 8 ft) and 0.7 x 3.7 x 2.4 m (2.3 x 12 x 8 ft), respectively, was constructed within the RVS. This served to simulate typical mild weather and winter non-isothermal ventilation under two typical ventilation schemes, crossflow and return flow ventilation (Figure 1). Air inlet and outlet were simulated on two sidewalls. In this study, air inlet refers to room air diffusers where fresh air enters the room and air outlet refers to room exhaust where mixed air was exhausted from the room. One long side wall of the test room is made of glass to permit convenient optical access. The other two side walls contain two glass slits to transmit light. The other walls, floor, and ceiling surfaces are painted with black and nonreflective paint to form a good optical background. The configuration and top view of the testing room was presented in Zhao et al. (1999).
[FIGURE 1 OMITTED]
Non-Isothermal Ventilation Simulation
Ventilation system A (Figure 1a) was constructed to simulate cross-flow ventilation of a typical swine building. The air inlet was 2 in. (50 mm) wide and the air outlet was 8 in. (200 mm) wide. Non-isothermal airflow in mild weather was simulated with System A. This simulation was identical to the non-isothermal ventilation simulation of Zhang et al. (1992) under which detailed airflow patterns, air velocity distribution, and turbulence intensity distribution were measured using a hot-wire anemometer. The test case was to verify the capability of the PTV measurement technology to study nonisothermal ventilation. The study results can be closely compared with the results of Zhang et al. (1992).
Typical swine grower-finisher buildings have air inlets on top of the two side walls and air outlets through the manure pits. Two air jets meet in the center of the room and turn down toward the floor. Based on the assumption that the typical room layouts are symmetric, System B (Figure 1b) simulated one-half of a typical ventilation setting for swine grower-finisher buildings. It represents return-flow ventilation. A 5 in. (130 mm) wide adjustable slot air inlet adjusted by 8 in. (200 mm) wide baffles was configured on the top of the wall. A linear air outlet with a width of 8 in. (200 mm) was configured at the bottom of the wall. Non-isothermal ventilation in winter weather was simulated with System B.
The test plan is summarized in Table 1. With System A, Test 1 and Test 2 simulated mild weather ventilation conditions with a small temperature difference of 8[degrees]C (46[degrees]F) between inlet air and indoor air and its relevant isothermal ventilation condition, respectively. For mild weather ventilation conditions, the inlet temperature was controlled at 24[degrees]C (75[degrees]F). The air within the room was heated by a floor panel heating system and was controlled at 32[degrees]C (90[degrees]F). Air exchange rates for both cases were 19.5 ach (air changes per hour).
With System B, Test 3 and Test 4 simulated cold winter ventilation conditions with a temperature difference of 26[degrees]C (47[degrees]F) and its relevant isothermal ventilation condition, respectively. For winter ventilation conditions, the inlet temperature and indoor air temperature were controlled to remain at -2[degrees]C (28[degrees]F) and 24[degrees]C (75[degrees]F). Air exchange rates simulate typical low winter ventilation rates (5-10 ach) and were controlled at 8.6 ach. Test 5 is a purely experimental case to study ventilation strategies that may help to achieve better mixing airflow patterns for the improvement of winter ventilation.
Indoor Air Measurement System
A two-dimensional PTV system was used to measure airflow patterns and air velocity distribution. The PTV measurement system consists of a test room, an illumination system, a particle seeding system, an image acquisition system, an image processing and interpolating system, and a data analysis system. Helium-filled soap bubbles with neutral buoyancy in the air were seeded into the ventilation plenum of the test room and then allowed to enter the test room with the inlet airflow. As inlet air jets traveled through the test room, bubbles filled the airspace of the test room. One center section of the room was illuminated with a light sheet. After 5 to 10 minutes of flow seeding, the bubble distribution reached its steady state. Tracks of traveling bubbles were then recorded by the image acquisition system. Proper exposure time was used to capture bubble pictures as streaks. An image shifting technology was used to resolve any ambiguity in the direction of the bubbles' travel. Streak images of bubbles were then processed by the image processing system to extract coordination and magnitude of the streaks. Air velocity vector maps and velocity distribution contour maps were developed to describe airflow patterns and air velocity distribution in the test room. Regional air velocity can be extracted to examine airflow characteristics of certain zones, such as human and animal occupied zones. The PTV methodology and detailed measurement procedures were discussed in Zhao et al. (1999, 2001).
Temperature Distribution Measurement System
Twenty-nine T-type thermocouples and a data acquisition system were used for temperature measurement. The data acquisition has 30 analog inputs, and the input voltage ranged from 0 to 10 V. The measurement channels can be programmed to be scanned in any order. Acquisition parameters include sampling rate, acquisition period, trigger time, and data destination.
Uniform measurement grids were used to measure temperature distribution in the airflow field of the entire room. Since the temperature gradients are not as large as the velocity gradient, air temperatures at 29 points of the flow field (Figure 2), including inlet, outlet, and two floor surface points, were measured.
Airflow Rate Measurement and Control System
The air delivery and measurement system consists of a measurement chamber and centrifugal exhaust fan. The air delivery capability ranges from 0.05 to 1.4 [m.sup.3]/s (106-2966 cfm), which can meet the designed ventilation requirement of 0.118 to 0.897 [m.sup.3]/s (250 to 1901 cfm), corresponding to 8.6 and 66 ach. The centrifugal fan was controlled by a variable-frequency controller. The air exchange rates were determined by measuring the drop in static pressure across sharp orifices on a perforated plate. The perforated plates were calibrated with a fan test chamber that is designed according to ASHRAE Standard 51-1985 (ASHRAE 1985) and maintained to the current ASHRAE Standard 51-1999 (ASHRAE 1999). The entire air delivery system was also calibrated with the fan test chamber. The test chamber was described by Hughes et al. (1988).
Cooling System and Floor Heating
In non-isothermal ventilation conditions, the internal heat was provided with 24 0.6 m x 1.2 m (2 ft x 4 ft) heating panels uniformly arranged on the floor of the test room. Each heating panel consisted of an electrical heating element sandwiched between two pieces of 0.25 in. (6 mm) thick plywood and then covered with 24-gauge galvanized steel to improve the temperature uniformity on the surface of the heating panel. Each heating panel can provide 150 W with 240 V AC power supply. The temperature on the surface of the heating panel is 106.4 [+ or -] 1.5[degrees]F (41.3 [+ or -] 1[degrees]C). The 24 panels were switched on and off using a thermostat set to maintain essentially constant room air temperature. The heating panels have the capacity to raise the room air temperature to 32[degrees]C (90[degrees]F) when the air exchange rate is 19.5 ach and the inlet air temperature is 24[degrees]C (75[degrees]F).
The outer room heating, ventilating, and air-conditioning system, a climate simulator, provided cold air. The climate simulator consists of an air-cooled condenser, compressors, evaporator, electric heaters, a supply fan, and a control system. The capacity of the climate simulator control system is:
* Temperature control from -25[degrees]C (-13[degrees]F) to 38[degrees]C (100[degrees]F) when ambient temperature is 35[degrees]C (95[degrees]F) and -19.5[degrees]C (-3[degrees]F), respectively.
* Humidity control from 20% to 90%.
[FIGURE 2 OMITTED]
The cold air was uniformly distributed in the inlet plenum of the test room by a 10 in. perforated polyethylene tube. Cold air was led down to the floor from air discharge holes of the distribution tube, which was hung near the ceiling of the inlet plenum. Thus, the cold air was well mixed in the plenum before being introduced into the test room.
RESULTS AND DISCUSSIONS
Mild Weather Non-Isothermal Ventilation
Temperature Distribution. Figure 3 shows spatial temperature distribution of Test 1: non-isothermal ventilation conditions with a small temperature difference. Temperature distribution is determined by air jets and room air distribution. The air entered the room at a temperature of 24[degrees]C (75[degrees]F). As the air jet traveled, adjacent warmer room air heated the cold air jet by mixing and thermal diffusion. Air near the floor was heated by the floor heating system. The reverse flow pulled air near the floor from the outlet side to the inlet side. So warmer air accumulated near the inlet wall and formed a high-temperature region. The temperature distribution also showed that the air jet separates at the middle of the room.
Airflow Patterns. Figures 4 and 5 show the airflow patterns of Test 1 and Test 2, respectively. A flow pattern common to both was that the air jet attached to the ceiling after entering the room, due to Coanda effect, which is the phenomenon that when an air jet enters a room through inlets close to the ceiling, then the air jet will bend toward the ceiling and finally attach to the ceiling due to the pressure difference between the two sides of the air jet caused by air entrainment and limited space between the air jet and the ceiling. The air jet traveled along the ceiling and then either reached the opposite wall (Test 2, isothermal ventilation) or separated from the ceiling and dropped toward the floor (Test 1, non-isothermal ventilation). Because of air jet separation in Test 1, a clockwise rotating airflow pattern formed near the inlet side of the room. Another weaker and smaller eddy formed near the outlet side of the room. The airflow on the right top corner of the room was much more turbulent and not stable. Sometimes airflow moved toward the outlet, and sometimes it moved toward the inlet. This may be because of variation in jet momentum and internal heating. A primary recirculation zone occupied approximately two-thirds of the room area and the temperature distribution in that zone was relatively constant and similar to the room temperature, as desired. Comparing these results with the flow patterns of Test 2, the correlated isothermal case, it is clear that even this relatively small temperature difference caused the air jet to fall sooner (approximately at three-fifths width of the test room). Under non-isothermal ventilation conditions, the major forces exerted on the air jet are inertial forces, thermal buoyancy forces, and friction force generated by surfaces. Thermal buoyancy force caused the air jet to fall down early (Zhang et al. 1992).
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
The general structure of these airflow patterns agrees with those of Zhang et al. (1992), which had exactly the same experimental settings but used a hot-wire anemometer for air velocity and a smoke gun for airflow pattern. One exception is that the PTV airflow patterns have more detailed structure and definitive airflow direction data. Another exception is that Zhang et al. (1992) showed that a reverse airflow pattern formed below the air jet under the isothermal ventilation, whereas the PTV-measured airflow pattern (Figure 6) showed not reverse airflow but a stagnant zone. This discrepancy may be caused by limitations of the smoke gun flow observation in low airspeed zones and the high turbulence level of the low airspeed zone. In summary, these results confirmed that the room airflow PTV system overcomes difficulties of traditional airflow measurement and visualization, yielding clearer and more quantitative airflow pattern results. It can effectively be used to study airflow of non-isothermal ventilation.
[FIGURE 6 OMITTED]
Air Velocity Distribution. Figure 6 shows the air velocity distribution of Tests 1 and 2 over the center section of the test room. The non-isothermal ventilation case with a small temperature difference had the same range of air velocities in the air jet and in the animal occupied zone as its isothermal counterpart. However, because of the temperature difference, the air jet was heavier than the room air. The air jet did not travel the whole room width and dropped early. According to ASHRAE (2005), air velocity greater than 25-30 fpm (0.1-0.15 m/s) with 60% turbulence intensity will cause cold draft conditions for 15% of the population. Indoor airflow of large animal buildings is fully turbulent, with turbulence intensity of up to 97% (Zhang et al. 1992). Therefore, when air velocity in the human breathing zone increases to 0.3 m/s (59 fpm) due to the air jet drop, it is likely that a cold draft will form at the human breathing zone in the center of the room. This also can be verified from temperature distribution (Figure 3). However, the jet drop did not significantly affect air velocities near the floor, which is normally recognized as the animal occupied zone in animal buildings.
Typical Winter Ventilation Conditions
Figure 7 shows the temperature spatial distribution of Test 3, which represents typical winter ventilation conditions. The inlet air temperature was only -2[degrees]C (28[degrees]F) when it entered the room. The temperature distribution pattern with low air temperature near the inlet wall and floor showed that the cold air jet dropped and traveled in a counterclockwise direction. The air temperature at the center space of the room rose quickly due to turbulent mixing and thermal diffusion between cold jet air, the heated floor, and warm indoor air. There is a high-temperature zone in the room near the air inlet.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Airflow Patterns. Figure 8 shows the airflow patterns of Test 3, a typical winter non-isothermal ventilation test case. Figure 9 shows the airflow pattern of Test 4, which is the correlated isothermal ventilation case of Test 3. When a heated hallway is used in a swine building to preheat the inlet air, the winter ventilation case will likely be as in Test 4, an isothermal ventilation case with a low air ventilation rate.
Under typical non-isothermal ventilation conditions, the air jet dropped immediately after it entered the test room, and this resulted in a strong counterclockwise rotating airflow pattern. This phenomenon was detected by several previous researchers using flag airflow indictor (Wang and Ogilvie 1996), smoke gun (Zhang et al. 1992), and temperature sensors (Zhang and Strom 1999). The PTV-measured airflow pattern not only clearly showed the dropping jet but also revealed the resulting airflow patterns in the animal occupied zone and human breathing zone and the overall airflow pattern of the whole room airspace. The cold air jet dropped immediately after it entered the room because a cold air jet is heavier than room air, and then negative buoyancy forces caused the air jet to fall. The Archimedes number was 0.292, which is much larger than the Critical Archimedes number of 0.023 for the full-scale test room by Zhang et al. (1992).
[FIGURE 9 OMITTED]
The air jet directly traveled down to floor level, with little entrainment or mixing with warm indoor air. A big vortex was formed at the other side of the test room. This location is the center of typical rooms. Likely in this ventilation mode, the contaminants will not be effectively removed. The floor-level air velocity was much higher in the non-isothermal ventilation conditions than in the isothermal ventilation conditions and will likely cause cold drafts. The overall airflow was not stable and fluctuated over a larger range. This is likely due to the temperature difference between the floor surface and room air. Thermal buoyancy due to the heat production from the floor contributes to the increase in room air turbulence (Zhang et al. 1992).
Under the same ventilation settings but with isothermal conditions, the airflow formed a desired clockwise rotating airflow pattern (Figure 9). The air jet travels along the ceiling and drops after reaching the opposite wall. The air jet entrains the room air and achieves good mixing before dropping down to the animal zone. The overall airflow pattern is much smoother than that of the non-isothermal ventilation case.
Air Velocity Distribution. Figure 10 indicates the air velocity distribution of Tests 3 and 4 over the center section of the test room. In comparison with air velocity distributions under isothermal ventilation, the floor-level air velocities increased from 0.1 to 0.2 m/s (20 to 39 fpm) due to counter rotating airflow patterns and buoyancy under this typical nonisothermal winter ventilation. Air velocities at the human breathing level also increased from 0.05 to 0.15 m/s (10 to 30 fpm). These changes from isothermal ventilation caused cold drafts in the animal occupied zones. Depending on where the human walkway is located, it will likely cause cold drafts in the human zone as well, especially when the walkway is at the inlet side. In addition, with this ventilation system, counter rotary airflow patterns will likely cause short-circuiting of air because the outlet is right underneath the air inlet on the same side wall.
Air Velocities at the Human Breathing Zone and Animal Occupied Zone. Air velocity directly affects temperature distribution, thermal comfort, and air quality. Ultimately, the goal in studying airflow patterns of typical animal buildings is to optimize the thermal and air quality environments around animals and human workers. Regional characteristics of airflow, especially in the animal occupied zone and at the human breathing level, are most critical.
The ASAE (2001) Standard D321.2DEC94 provides typical dimensions for various animal species. For pigs, 0.7 m is defined as the highest point of a standing pig. Many researchers used 0.0 to 0.6 m (0 to 2 ft) as the animal occupied zone in swine buildings. Based on the ASAE standard and other research literature, the animal occupied zone was defined as 0.0 to 0.6 m (0 to 2 ft) from the floor and the human breathing level 1.4 to 1.7 m (4.6 to 5.6 ft) above the floor in this study.
Figure 11a shows a comparison of air velocities at the human breathing level under winter non-isothermal and isothermal conditions. It is clear that under non-isothermal ventilation conditions, due to cold drafts and floor heat, the air velocities at the human breathing level were much higher than for the isothermal ventilation cases. In winter, this velocity increase is not desirable. Because of counter primary rotating airflow patterns in the non-isothermal ventilation cases, the human breathing level has a reverse airflow. The reverse airflow resulted in air entrainment and higher air velocities at the human breathing level as it traveled toward the inlet wall. The air velocity fluctuated due to turbulence activities. Air velocities dropped dramatically at one place near the opposite wall because of the large vortex. Under the correlated isothermal ventilation case, because of primary rotating airflow patterns, air jets attached to the ceiling and gradually entrained room air. Therefore, air velocities at the human breathing level increased as the air jet traveled from the air inlet along the width of the room.
[FIGURE 10 OMITTED]
Figure 11b shows a comparison of air velocities in the animal occupied zone under winter isothermal ventilation and the correlated isothermal ventilation. Large differences in air temperature resulted in higher air velocities and large velocity fluctuations in the animal occupied zone. Because of a counterclockwise rotary airflow pattern, air velocities at the animal occupied zone decreased as the airstream traveled from the inlet wall to the opposite wall. In the isothermal ventilation case, airflow in the animal occupied zone was dominated by reverse airflow of the primary rotating airflow patterns. Because of friction of the floor and viscosity of airflow, the air velocity decreased as the reverse airflow progressed.
Evaluation of Strategies to Improve Winter Ventilation
Room airflow patterns are mainly affected by the air jets present in the mixing ventilation. Under non-isothermal ventilation conditions, air jet separation behavior is determined by Archimedes number. Critical Archimedes number (Ar) is a limit Ar value at which the diffuser air jet drops immediately after entering the room. An Ar less than the critical Ar is needed to create full rotary airflow in a ventilation air space. By analysis of the Archimedes number definition (Equation 1), we can learn that there are three ways to decrease Archimedes number: (1) decreasing temperature difference, (2) decreasing inlet opening to increase air velocity, and (3) increasing inlet air velocity by increasing airflow rate. In winter, the minimum ventilation rate is determined by maximum conservation of heat energy and optimum control of moisture and air quality.
The temperature difference is caused by ambient weather conditions and indoor thermal comfort requirements. Decreasing temperature difference is an effective strategy to solve winter ventilation difficulties. Warming up the inlet air by means of a heated hallway and passing the ambient air through an attic are practical ways to increase inlet air temperature, thus decreasing the difference in air temperature. Air-to-air heat exchangers or geothermal heat pumps may also be feasible options with today's high energy cost environment.
To evaluate the strategy for improving airflow patterns under winter ventilation conditions by reducing the inlet opening and simultaneously increasing the inlet air velocity, Test 5 was designed. Test 5 ventilation settings are similar to those of Test 3, except that the inlet opening decreased to a minimum limit of 6 mm (0.24 in.) so that the inlet air velocity was increased to 3.3 m/s (650 fpm), similar to that of a typical winter inlet opening of a swine building.
Figure 12 shows the airflow patterns of Test 5. Contrasting with the airflow patterns of Test 3 (Figure 8), it was found that the air jet was lifted up to the ceiling and traveled about 1.5 m (5 ft) from the inlet wall before it began to drop. Due to insufficient jet momentum and temperature difference with the resulting buoyancy effects, however, the inlet air jet still separated shortly after it was discharged into the test room and the air jet fell into the animal occupied zone. The draft was strong in the room. Two major rotating zones formed: one a clockwise rotating flow and the other a counterclockwise rotating flow. In the clockwise rotating zone, there were a few weak vortexes showing that the airflow was more turbulent. The overall floor-level velocity was also larger than that which occurred under isothermal ventilation conditions. A clear vortex formed on the other side of the test room.
[FIGURE 11 OMITTED]
By analyzing the winter non-isothermal ventilation cases, one can appreciate that it is not easy to design and manage winter ventilation systems for animal facilities. From Test 5, which was designed to improve airflow patterns in winter ventilation conditions, it was found that there is a limit to increasing inlet air velocity by decreasing the inlet openings. Inlet openings of 6 mm (0.24 in.), the minimum in realistic buildings, resulted in a maximum inlet air velocity of 3.3 m/s (650 fpm) with a typical 8.6 ach air exchange rate. The 3.3 m/s (650 fpm) inlet air velocity is not large enough to form a stable airflow pattern. An inlet air velocity of 1.6 m/s (315 fpm) was reported in Zhang et al. (1992) as the limit in forming stable airflow patterns under non-isothermal ventilation conditions with an 8[degrees]C (15[degrees]F) temperature difference. Therefore, inlet air velocities forming stable airflow patterns depend on the temperature difference. Because of the minimum inlet opening limit, this strategy did not resolve the winter ventilation challenge.
Another way would be to increase inlet air velocity by introducing recirculation air into the inlet airflow. This would increase airflow rate through the inlet and still keep the fresh airflow rate low, conserving energy.
The distance from jet separation point to the air inlet is considered the jet separation distance. The jet separation is defined as where the jet centerline starts to curve downward. Li et al. (1995) developed the following equation to predict the air jet separation distance:
[X.sub.s] = C[square root of ([A.sub.0]/[A.sub.r])] (2)
[FIGURE 12 OMITTED]
[X.sub.S] = jet separation distance,
C = a constant,
[A.sub.r] = Archimedes number, and
[A.sub.0] = effective area of inlet, [m.sup.2].
In Test 5, [X.sub.s] is approximately 1.5 m (5 ft). If we consider adding a recirculation air volume of Q = 8.6 ach, then [X.sub.snew] = 2.82 [X.sub.s], which is approximately 4.24 m (14 ft). If the recirculation air volume is increased to one and a half times the winter ventilation rate, which is about 12.9 ach, then [X.sub.snew] = 3.86 [X.sub.s], which is approximately 5.78 m (19 ft) and larger than the room length. This means that by adding 1.5 times the recirculation airflow to the winter inlet airflow, the jet momentum will be enlarged and a stable and fully rotary airflow pattern will form. Winter ventilation problems such as cold drafts at the occupied zones will likely be resolved.
Particle tracking velocimetry could be an effective tool to measure non-isothermal indoor airflow. It can clearly visualize airflow patterns and quantitatively determine air velocity distribution in a full-scale room and low-velocity regions, which are needed for evaluation or development of numerical simulation models.
Non-isothermal ventilation conditions, especially with large temperature differences between the supply air and room air, present difficulty in maintaining a primary rotary (thus mixing) airflow pattern. It was found that even with a relatively small temperature difference (8[degrees]C [15[degrees]F]), the air jet will separate from the ceiling earlier than an isothermal jet will. Early dropping of inlet jet resulted in reversed rotary airflow pattern or several subrotating zones. Airflow patterns with subrotating zones are not effective for contaminant removal. Early jet separation also could cause short-circulation of the fresh air, which violates the ventilation purposes. It also causes cold drafts that result in animal discomfort and may hurt their performance.
Non-isothermal ventilation conditions generally result in higher air velocities with larger fluctuations in the animal occupied zone than those of isothermal ventilation at the same rate. The air velocity increment is well correlated to the temperature difference between inlet air and room air.
There are ways to mitigate the above problems of the nonisothermal ventilation conditions: (1) decreasing the temperature difference, (2) increasing inlet air velocity by maintaining proper inlet opening and pressure, and (3) adding an air recirculation device to increase inlet air jet momentum. If the temperature difference is small, increasing inlet air velocity is an effective and simple way to achieve better mixing ventilation. If the temperature difference is large in winter, increasing inlet air velocity only may not be sufficient and an additional air recirculation device may be needed to increase inlet air jet momentum and maintain the primary rotating airflow patterns. Preheating of the inlet airflow is also an effective strategy to achieve a better mixing ventilation mode.
This work is supported by the Illinois Council for Food and Agricultural Research via an Internal Competitive Grant IDA CF 98I-40-5.
ASAE. 2001. ASAE Standards on Engineering Practices Data. St. Joseph, MI: American Society of Agricultural Engineering.
ASHRAE. 1985. ASHRAE Standard 51-1985, Laboratory Methods of Testing Fans for Rating. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
ASHRAE. 1999. ANSI/ASHRAE Standard 51-1999, Laboratory Methods of Testing Fans for Aerodynamic Performance Rating (AMCA Standard 210-99). Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
ASHRAE. 2005. 2005 ASHRAE Handbook--Fundamentals. Atlanta: American Society of Heating, Refrigerting and Air-Conditioning Engineers, Inc.
Heber, A.J., and C.R. Boon. 1993. Air velocity characteristics in an experimental livestock building with non-isothermal jet ventilation. ASHRAE Transactions 99 (1):1139-1151.
Hughes, W.C., L.L. Christianson, K. Helmink, R. Korthals, and C. Gates. 1988. Fan performance as affected by partial obstructions, ASAE Paper No. 88-4546. St. Joseph, Mich.: American Society of Agricultural Engineers.
Jin, Y., and J.R. Ogilvie. 1992. Isothermal airflow characteristics in a ventilated room with a slot inlet opening. ASHRAE Transactions 98 (1):296-306.
Li, Z., L.L. Christianson, and J.S. Zhang. 1995. Separation Distance of Non-Isothermal Air Jets. Research Triangle Park, NC.
Nielsen, P.V., A. Restive, and J.H. Whitelaw. 1978. The velocity characteristics of ventilated rooms. J. of Fluids Eng. 100 (9):291-98.
Randall, J.M., and V.A. Battams. 1979. Stability criteria for airflow patterns in livestock buildings. J. Agr. Eng. Res. 24:36l-74.
Riskowski, G.L., J.B. Priest, S.E. Ford, and L.L. Christianson. 1993. Environment in animal regions. Livestock Environment IV, Fourth International Symposium, University of Warwick, Coventry, England, pp. 411-18.
Sandberg, M. 1987. Velocity characteristics in mechanically ventilated offices rooms. In Room Vent '87, Proc. of 1st International Conference of Air Distribution in Ventilated Spaces, Stockholm, Sweden, 10-12 June, Session 2A.
Timmons, M.B. 1984. Internal air velocities as affected by the size and location of continuous diffuser slots. Trans. ASAE 27 (5):1514-17. American Society of Agricultural Engineering.
Wang, J., and J.R. Ogilvie, 1996. Airflow distributions at floor level in a slot-outlet and slot-inlet ventilated room. ASHRAE Transactions 102(2):347-58.
Wu, G.J., L.L. Christianson, J.S. Zhang, and G.L. Riskowski. 1990. Adjustable room ventilation simulator for room sir and air distribution modeling. In Indoor Air '90, Proc. 5th International Conference on Indoor Air Quality and Climate, Toronto, Canada, July 29-August 3 4:237-42.
Zhang, J.S., G.J. Wu, L.L. Christianson, and G.L. Riskowski. 1992. Detailed measurements of room air distribution for evaluating numerical simulation models. ASHRAE Transactions 98(2):58-65.
Zhang, G., and J.S. Strom. 1999. Jet drop models for control of non-isothermal free jets in a side-wall multi-inlet ventilation system. Trans. ASAE 42(4):1121-26. American Society of Agricultural Engineers.
Zhao, L.Y., Y. Zhang, X. Wang, G.L. Riskowski, and L.L. Christianson. 2001. Measurement of two-dimensional air velocities in a full-scale room using particle image velocimetry. ASHRAE Transactions 107(2):434-44.
Zhao, L.Y., Y. Zhang, X. Wang, G. L. Riskowski, and L. L. Christianson. 1999. Development of PIV techniques to measure airflow patterns in ventilated airspaces. ASHRAE Transactions 105(2):1098-107.
Lingying Zhao, PhD
Yuanhui Zhang, PhD, PE
Xinlei Wang, PhD
Gerald L. Riskowski, PhD, PE
Lingying Zhao is an assistant professor in the Department of Food, Agricultural and Biological Engineering, Ohio State University, Columbus. Yuanhui Zhang is a professor and Xinlei Wang is an assistant professor in the Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champaign. Gerald L. Riskowski is a professor in the Department of Biological and Agricultural Engineering, Texas A & M University, College Station, TX.
Table 1. Summary of the Test Cases Test Ventilation Inlet Width Inlet Temperature Case System mm in. [degrees]C [degrees]F Test 1 A 50 2 24 75 Test 2 A 50 2 24 75 Test 3 B 41 1.6 -2 28 Test 4 B 41 1.6 24 75 Test 5 B 6 0.24 -2 28 Test Ventilation Outlet Temperature [DELTA]T Case System [degrees]C [degrees]F [degrees]C [degrees]F Test 1 A 32 90 8 15 Test 2 A 24 75 0 0 Test 3 B 24 75 26 47 Test 4 B 24 75 0 0 Test 5 B 24 75 26 47 Test Ventilation Air Exchange Rate Inlet Velocity Case System ach* m/s fpm Test 1 A 19.5 1.78 350 Test 2 A 19.5 1.78 350 Test 3 B 8.6 0.74 146 Test 4 B 8.6 0.74 146 Test 5 B 8.6 3.30 650 * ach = air changes per hour