1、4717 Effect of Ventilation System on Particle Spatial Distribution in Ventilated Rooms Xinlei Wang, Ph.D. Yuanhui Zhang, Ph.D., P.E. Ted L. Funk, Ph.D., P.E. Member ASHRAE Member ASHRAE- Lingying Zhao, Ph.D. Associate Member ASHRAE ABSTRACT Airborne particulate matter has been implicated as a major
2、contributor to the increased incidence of respiratory disorders amongpeople working in livestock buildings. A clear understanding of particle spatial distribution can provide important information for improvement of ventilation system design and control strategies. In this study, the dust mass spati
3、al distributions in three different ventilation systems were measured using a multi-point sampler in a full-scale mechan- ically ventilated laboratory room under controlled conditions. The experimental results showed that the particle mass spatial concentrations varied widely as a result of ventilat
4、ion. Increasing the ventilation rate within the same ventilation system reduced the overall mean particle concentration. At the same ventilation rate, the ventilation effectiveness varied widely with diferent ventilation systems. The experimental results also showed that the air outlet location had
5、a substan- tial efect on the dust spatial distribution and the overall dust mass concentration. Ventilation system design was therefore shown to be critical to dust control in a mechanically ventilated airspace. Positioning the air outlet at the dustiest location can substantially improve dust remov
6、al effectiveness. The ventila- tion efectiveness factor was used to quantitatively describe the effectiveness of a ventilation system for removal ofpollutants in a ventilated airspace. INTRODUCTION Dust has been shown to have direct negative effects on the health of operators working in livestock bu
7、ildings (DeBoer et al. 1991; Dosman et al. 1988; Donham et al. 1989; Zejda et al. 1994; Senthilselvan et al. 1997a). It has also been shown that reducing dust concentration within buildings resulted in Gerald L. Riskowski, Ph.D., P.E. Member ASHRAE improvement in human respiratory responses (Senthil
8、selvan et al. 1997b; Zhang et al. 1998). Ventilation is effective in the control and dilution of gaseous contaminants (Albright i 990). It is also generally believed that ventilation systems have a direct effect on dust concentrations (Breum et al. 1990; Carpenter 1986; Lavoie et al. 1997; Maghirang
9、 et ai, 1994; Qi et al. 1992; Vant Klooster et al. 1993). As dust is not uniformly distributed within a ventilated airspace, dust control by ventilation is more complex than gas control. Qi et al. (1 992) reported that using a higher ventilation rate increased both respirable and total particle gene
10、ration rates in a commer- cial poultry house. They observed that the total particle gener- ation rate and the respirable particle generation rate were 79.3% and 84.8% higher, respectively, for the high ventilation rate than for the low ventilation rate. Dawson (1 990) also indi- cated that increasin
11、g ventilation was not a practical method of restricting dust levels. Breum et al. (1990) evaluated the effectiveness of two different types of exhaust ventilation systems in a swine build- ing-upward and downward ventilation airflow. They found that the ventilation effectiveness of upward airflow wa
12、s supe- rior to downward airflow at maximum ventilation rate, but the airflow direction did not significantly affect the dust concen- tration at the minimum ventilation rate. Vant Klooster et al. (1 993) studied the effect of locations of the air inlet and outlet on the dust concentration in swine n
13、ursery rooms. They reported that installing an indirect air inlet near the human breathing level and the air outlet underneath the slats could reduce the dust exposure of stockmen by 40% compared to a reference room with a typical ventilation system. This indi- cates that improving ventilation effec
14、tiveness, as opposed to increasing ventilation, can be an important strategy to reduce Xinlei Wang is an assistant professor, Yuanhui Zhang is a professor, and Ted L. Funk is an assistant professor in the Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champai
15、gn, Urbana, III. Lingying Zhao is an assistant professor in the Department of Food, Agricultural m meters). I- 73. Figure2 Schematic side views of the three ventilation systems, all dimensions in meters: (a) System A, (b) System AB, and (c) System B. The ventilation system (B) was further modified f
16、rom venti- lation system AB. System B had a continuous ceiling slot inlet with a hinged baffle and a slot outlet on the same sidewall (Figure 2c). This air inlet opening could be adjusted based on required inlet velocity. The air outlet was kept constant at 200 mm wide-the same as ventilation system
17、 AB. Even with the minor change of air inlet from system A to system B, the inlet air jets and, thus, the air distribution were expected to be very different. Because the air jet in system A was a free jet, while the air jet in system B was a confined ceiling jet, the throw of ASHRAE Transactions: R
18、esearch 259 these two jets could be quite different, and the systems A and B can therefore be considered to be two different air inlet systems and their air distribution patterns were expected to be very different (ASHRAE 1997). The discharge coefficient (or coefficient of discharge) is used to desc
19、ribe the combined effect of contraction and friction of flow from an air inlet. It is defined as the ratio of actual airflow to ideal airflow through an inlet. The values for discharge coefficients are usually determined empirically. Because System A and System B were two different air inlet systems
20、, their discharge coeffi- cients were different, with a higher discharge coefficient in System B. To account for this, the inlet opening width in System B was smaller than that in System A at the same air exchange rate and the same inlet velocity. To study the effect of the locations of the air outl
21、et and the inlet on the dust spatial distribution in a mechanically venti- lated airspace, three cases were tested (Table I). Airflow patterns were measured using particle image velocimetry (PIV) (Zhao et al. 1999). Inlet temperature and room temper- ature were measured using T-type thermocouples. F
22、or each case, three replications were performed for dust concentration measurement. i- O .48 Sampling i- Dust Generation and Concentration Measurement A dust generation and distribution system using Arizona dust was developed to generate dust uniformly along the floor, which is the primary source of
23、 dust in livestock buildings O .48 O .48 Side wall i- i. r + + i -+ Floor (Wang et al. 1999). A rotating-table dust generator was devel- oped to feed the dust to the dust emission system at a constant rate. Dust was uniformly emitted to the room airspace from 25 emitting ports evenly distributed ove
24、r the entire floor area. The outlets of these ports were downward-facing, and some dust was collected on the floor. The tubes distributing dust were 12 mm above the floor. Each emission port had the same diameter of 1.6 mm. In order to maintain the same air pressure at each port, the total opening a
25、rea of the five ports in each branch tube was one-fifth of the cross-sectional area of the tube. The air pressures were 7.1 kPa at both ends of the tube. In System A at a ventilation rate of 19.5 ACH, the airborne dust size distri- bution of fifteen dust samples was measured using an aerody- namic p
26、article sizer (APS) (Wang et al. 2000) with an inlet sampling velocity of 0.3 ms. The count median aerodynamic diameter of the airborne dust was 1.63 mm with geometric standard deviation (sg) of 1.59, and the estimatedmass median diameter was 3.1 mm. The density of the test dust was measured in the
27、factory as 2.65 g/cm3. Dust concentrations were measured at 25 points across the central cross section within the testing room. Since the room airflow field was considered to be approximately two-dimen- sional, measurement points were uniformly distributed in the central cross section (Figure 1) in
28、the test room, as shown in Figure 3. The dust collector located downstream of each criti- cal venturi orifice was a 37 mm diameter (0.8 mm porosity) Table 1. The Experimental Cases of Ventilation System Effect on the Dust Spatial Distribution ACH = air changes per hour Figure 3 Twenty-Jive sampling
29、points were evenly distributed across the cross section along the centerline of the room (all dimensions in meters) (side viao). 260 ASHRAE Transactions: Research filter housed in a holding cassette. As the surrounding air veloc- ity in most of the sampling points was less than 0.5 ds, the dust samp
30、ling inlet was placed perpendicular to the airflow to main- tain all sampling close to isokinetic conditions according to the criteria of still air sampling (Hinds 1999; Wang 2000). Filters were dried in a desiccant drier for 24 hours and weighed on a precision electronic balance before and after th
31、e dust collec- tion. RESULTS AND DISCUSSION Effect of Outlet Location on Dust Spatial Distribution Case 1 was conducted in ventilation system A and Case 2 in ventilation system AB. The ventilation rates in Case 1 and Case 2 were the same at 19.5 ACH (0.264 m3/s). All other conditions were also the s
32、ame, except the outlet location. The dust spatial distribution is shown in Figure 4. 18 o 17 o 16 O 15 o 14 O 13 O 12 o 11 o 10 o 90 70 60 50 40 30 20 10 O0 ea 176 168 160 152 144 136 i28 120 112 104 96 88 80 72 64 56 48 40 32 24 10 O8 O0 Figure 4 Efect ofthe outlet location on dust spatial distribu
33、tion (mg/m3); (a) Case I: the outlet is 1.3 m fromjoor on the right side wall; (6) Case 2: the outlet is next to thejoor on the left side wall; (c) dust spatial concentration change of Case 2 compared with Case 1 PA). ASH RAE Transactions: Research 261 Comparing the dust spatial distributions betwee
34、n Case 2 and Case 1, as shown in Figure 4, there is a significant differ- ence. The overall dust concentration in Case 2 is substantially lower than in Case 1. The dust concentration in most of the room is below 0.8 mg/m3 in Case 2, while the dust concentra- tion in most ofthe room in Case 1 is abov
35、e 2.0 mg/m3. Further- more, the overall average dust concentration in Case 2 is only 0.90 mg/m3, which is much lower than 2.30 mg/m3 in Case 1. It appears that this result is due to the outlet of Case 2 being located next to the high dust concentration area, which enables the exhaust air to remove m
36、ore dust from the room than in Case 1. This implies that outlet location has a significant effect on the dust spatial distribution and overall dust levels in a venti- lated airspace. This information is very important in improv- ing the ventilation effectiveness. A quantitative comparison of the dus
37、t spatial concentrations between Case 1 and Case 2 is shown in Figure 4c. This map shows the dust reduction percentage from Case 1 to Case 2. The dust concentration was reduced by 45% to 75% in most of the area within the test room. Effect of Inlet Location on Dust Spatial Distribution The effect of
38、 inlet location on the dust spatial distribution was studied in Case 3 by keeping the same outlet as in Case 2 but moving the inlet up to the ceiling. The ventilation rate was maintained at 19.5 ACH (0.264 m3/s). All other experimental conditions were also the same as Case 2, with the only differ- e
39、nce being the inlet location. A comparison of dust spatial distributions between Case 3 and Case 2 is shown in Figure 5. Comparing the dust spatial distributions between Case 3 and Case 2, it was found that there was little difference. The overall average dust concentration in Case 3 was 0.83 mg/m3,
40、 which is slightly lower than the 0.90 mg/m3 in Case 2. In these two case studies, the ventilation rate, the inlet velocity, and the calculated inlet jet momentum were the same. One possible reason for the differences in dust concentration is the different jet momentum loss after the free inlet jet
41、enters the airspace. The inlet air jet of Case 3 was directed along the ceiling imme- diately after it entered the airspace, which caused minimum jet momentum loss. But in Case 2, the air jet entered the airspace not directly attached to the ceiling and traveled a short distance before it was direct
42、ed to the ceiling because of the Coanda effect (Albright 1990). This effect could cause jet momentum loss in the inlet area and, consequently, a difference in the flow pattern and the dust spatial distribution. The quantitative comparison of dust spatial concentration change from Case 2 to Case 3 is
43、 shown in Figure 5c. The dust concentrations in most of the area in Case 3 are lower than those in Case 2, but in one area in the left side of the room, the dust concentration in Case 3 is higher than that in Case 2. Comparing with Case 2, the overall average dust concentration was 10% lower in Case
44、 3. These data indicate that the inlet location had a minor effect on the dust spatial distribution and overall dust level of a ventilated airspace in these two cases. Further study of other ventilation layouts is needed because there was not a substantial difference in the air inlet location in Cas
45、es 2 and 3. For example, if the air inlet is placed in the center of the room, it is likely that the dust spatial distri- bution could be very sensitive to inlet location. Mean Dust Concentration and Ventilation Effectiveness To study the overall difference between System A and System B, twelve case
46、s were tested in the test room-six in System A and six in System B-at ventilation rates from 8.6 ACH to 66 ACH. For each case, three replication measure- ments were taken. All experimental parameters are summa- rized in Table 2. During these tests, it was difficult to control the inlet air temperatu
47、re and room air temperature at 24C as in the experimental design because of limitations in the heating and cooling capacities of the room ventilation simulator. The temperature was within the range of 26*6“C. It was also difi- cult to control the relative humidity of the room air in the full- scale
48、test room. The relative humidity was within the range of 50*20%. Therefore, the measured dust spatial distribution also included the effects of room air temperature and room air rela- tive humidity. Despite this, analysis shows that the effect of room air temperature and the room air relative humidi
49、ty within the test range (temperature: 26* 6 OC, RH: 50* 20%) was not significant. For example, of the three tests in case C, relative humidity in test 2 and test 3 was 48% and 33%, respectively, but the measured overall mean dust concentration were almost the same (1.53 mg/m3 vs. 1.52 mg/m3). Another example showing that the ventilation system, rather than relative humidity, plays the major role in the mean dust concentration reduction is the comparison between the results of Case E and Case K, which both had ventilation rates of 56 ACH. One test in each case was performed at similar re
